Document PDF

Injury Prevention and Control
At least five million people die each year from injuries, and about half the deaths in the
10–24 years age group are accountable to them. This is a major health problem, for which
a number of strategies for prevention and control can be developed.
This book presents a series of the plenary and state-of-the-art presentations from the
5th World Conference on Injury Prevention and Control. There is a focus on
transportation, the workplace, sport and leisure, domestic sectors and violence. There is
also an exploration of the legal, medical, environmental, safety and governmental issues
which play a part in the subject.
The contributors have examined their themes from an international perspective and
also suggested guidelines for the future. Practitioners and researchers in a variety of
activities (including epidemiology and public health, occupational safety and health,
ergonomics and product design, medicine, criminology, engineering and physical
sciences, and the behavioural sciences) should find this a useful and challenging work.
Edited by Dinesh Mohan and Geetam Tiwari, Indian Institute of Technology, Delhi.
Injury Prevention and Control
Edited by
Dinesh Mohan
and
Geetam Tiwari
London and New York

First published 2000 by Taylor & Francis
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada
by Taylor & Francis
29 West 35th Street, New York, NY 10001
Tailor & Francis is an imprint of the Taylor & Francis Group
This edition published in the Taylor & Francis e-Library, 2005.

"To purchase your own copy of this or any of Taylor & Francis or
Routledge's collection of thousands of eBooks please go to
www.eBookstore.tandf.co.uk."
© 2000 Taylor & Francis
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and recording,
or in any information storage or retrieval system, without permission in
writing from the publishers.
Every effort has been made to ensure that the advice and information in
this book is true and accurate at the time of going to press. However,
neither the publisher nor the authors can accept any legal responsibility
or liability for any errors or omissions that may be made. In the case of drug
administration, any medical procedure or the use of technical equipment
mentioned within this book, you are strongly advised to consult the
manufacturer’s guidelines.
Publisher’s Note
This book has been produced from camera-ready copy supplied by the authors.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
A catalog record for this book has been requested
ISBN 0-203-30533-7 Master e-book ISBN
ISBN 0-203-34380-8 (Adobe e-Reader Format)

ISBN 0-748-40959-9 (Print Edition)
Contents
Contributors vii
Preface x
Acknowledgements xi
1. Injury Control and Safety Promotion: Ethics, Science and Practice 1

DINESH MOHAN
2. On the Escape of Tigers: An Ecological Note 14
WILLIAM HADDON, JR.
3. Where Have We Been and Where are We Going With Injury Control? 20
SUSAN P.BAKER
4. Reducing Injury Losses: What Private Insurers Can and Cannot Do 28
BRIAN O’NEILL
5. Methodological Consideration in the Analysis of injury Data: A Challenge for 37
the Injury Research Community
SHRIKANT I.BANGDIWALA
6. Assessing the Burden of Injury: Progress and Pitfalls 52
TED R.MILLER
7. Traffic Flow and Safety: Need for New Models for Heterogeneous Traffic 73
GEETAM TIWARI
8. Folkore and Science in Traffic Safety: Some New Directions 92
MURRAY MACKAY
9. Demands to Vehicle Design and Test Procedures for Injury Control of 104
Vulnerable Road Users in Traffic Accidents
DIETMAR OTTE
10. Risky Business: Safety Regulations, Risk Compensation and Individual 123
Behaviour
JAMES HEDLUND
11. Aging and Transportation: Mobility or Safety? 148
LIISA HAKAMIES-BLOMQVIST
12. Adolescents’ Risk-taking Behaviour, Myth or Reality: Evidence from 158
International Data
ANNE TURSZ
13. Injury Prevention and Children’s Rights 177
JAMES NIXON
14. Evidence-based Injury Prevention and Safety Promotion: State-of-the-art 192
LEIF SVANSTRÖM
15. Community Psychology and Safety: A Psychospiritual Perspective 210
MOHAMED SEEDAT
16. Labour and Safety: Science, Technology and Society 223
A.K. (DUNU) ROY
17. Achievements in Consumer Product Safety and the Challenges of Globalization 235
WIMROGMANS
18. Technologies, Therapies, Emotions and Empiricism in Pre-hospital Care 262
MATHEW VARGHESE
19. The Burden of Violence: An International Public Health Perspective 278
ETIENNE KRUG
20. A New Vision for Suicide Prevention: The Public Health Approach 287
MARK L.ROSENBERG AND JAMES A.MERCY
21. Barbarism and Solitude: Reflections on Globalization and Violence 300
EMMANUEL ROZENTAL-KLINGER
Index 311
Contributors
Susan P.Baker, M.P.H., is Professor in the Department of Health Policy and Management
and was the founder and first Director of the Johns Hopkins Injury Prevention Center,
Baltimore, USA. Her research interests include motor vehicle injuries, carbon
monoxide poisoning, drowning, asphyxiation, house fires and clothing-ignition burns,
falls in the elderly, homicide, suicide, fatal occupational injuries, medical care for the
injured, and injury severity scoring. She holds an honorary doctorate from the
University of North Carolina.
Shrikant I.Bangdiwala, Ph.D., is Research Associate Professor in the Department of
Biostatistics, School of Public Health and Injury Prevention Research Center,
University of North Carolina at Chapel Hill, USA. He is also Director of the Biometry
Core of The Injury Prevention Center. His research interests include biostatistical
methodological research in design, conduct and analysis of clinical trials, especially in
cardiovascular disease and in gas tro-intestinal biology and disease.
Liisa Hakamies-Blomqvist is Senior Research Scientist with the Swedish National Road
and Transport Institute. She is also Professor in the Swedish School of Social Science,
University of Helsinki, Finland. Her research interests include different aspects of
traffic and aging, and life-span psychology. She has also been involved in several
national and international projects on epidemiological and behavioural aspects of aging
and transportation. She is currently involved in research dealing with developmental
psychology in early infancy and health psychology in adults.
William Haddon, Jr., M.D. (1927–1985), was the first Administrator of the National
Highway Traffic Safety Administration, USA and then the first President of the
Insurance Institute for Highway Safety, Washington, DC. A pioneer in injury control
research, he developed the first systematic methods of identifying a complete range of
options for reducing crash losses.
James Hedlund, Ph.D., is principal of Highway Safety North—traffic safety policy and
research consultants. He has directed and participated in studies on impaired driving,
speed, motorcycle helmets, and policy and management studies. As Associate
Administrator of Traffic Safety Programs, National Highway Traffic Safety
Administration, he has directed a wide variety of traffic research, development, and
implementation and other evaluation activities.
Emmanuel Rozental-Klinger is a general and colo-rectal surgeon. He is at present
practising at the North York General Hospital in Toronto, Canada. He is founder and
co-ordinator of Universitu Coalitions for Partnerships with Colombia, which is aimed
at the protection of human rights, development institutions and their workers. He is
also Professor at the Universidad del Valle in Cali, where he was Director of
CISALVA, a WHO collaborating centre in violence prevention.
Etienne Krug, MD, MPH, is Medical Officer Violence and Injury Prevention with the
World Health Organization. He co-ordinates WHO’s activities in injury epidemiology.
He is involved in a number of international projects in this area including the World
Report on Violence and Health. He has been Country Director for Medecines sans
Frontiers and a researcher at Centers for Disease Control (CDC) where he has
conducted research on firearm injuries, youth violence and self-inflicted violence.
Murray Mackay, OBE, FREng., is Professor Emeritus of Transport Safety at the
University of Birmingham, UK. He established the Birmingham Accident Research
Centre in 1964 and ran it for 33 years. He has specialized in multi-disciplinary, in-
depth crash investigation research with special interests in vehicle design and the
biomechanics of impact injuries. He is also the Founding Director of the Parliamentary
Advisory Council for Transport Safety in UK and the European Transport Safety
Council in Brussels. His research interests include rational science-based counter-
measures for traffic injury reductions in developing countries.
James A.Mercy, Ph.D., is Associate Director for Science of the Division of Violence
Prevention at the National Center for Injury Prevention and Control of the Centers for
Disease Control and Prevention (CDC). He has conducted studies on the epidemiology
of youth suicide, family violence, homicide and firearm injuries.
Ted R.Miller, Ph.D., is Principal Research Scientist at the Pacific Institute for Research
and Evaluation. A safety economist, he brought modern costing approaches to injury
control in the United States, New Zealand, Canada and Australia. He has pioneered
methods for valuing quality of life lost in non-fatal injury. His research interests
include assessing unintentional injury, substance abuse and violence costs or the
savings from prevention.
Dinesh Mohan, Ph.D., is Professor at the Indian Institute of Technology, Delhi. He is also
the Co-ordinator of Transportation Research and Injury Prevention Programme
(TRIPP) and Head of the WHO collaborating Centre on Research and Training in
Safety Technology. He has been involved in research on the human body’s tolerance to
injury, biomechanics of injury, safety of vulnerable road users, agricultural injuries and
international issues concerning safety.
James Nixon, M.SocWk., Ph.D., is Associate Professor in the Department of Paediatrics
and child Health at the University of Queensland and Statewide Paediatric
Rehabilitation Service, Royal Children’s Hospital. His research interests include
intentional and unintentional injury to children, impact of injury on the families of
injured children and the rehabilitation of children following injury.
Brian O’Neill is president of the Insurance Institute for Highway Safety, and Highway
Loss Data Institute, both insurer supported research and communications organizations
dedicated to reducing deaths, injuries and property damage from motor vehicle crashes.
His research interests include all aspects of highway safety particularly human, vehicle
and environmental factors. He is co-author of The Injury Fact Book.
Dietmar Otte, Dipl.-Ing (FH), Dipl.-Ing.(TU) is the technical manager of the Accident
Research Unit at the Medical University of Hanover, Germany. He is an expert in bio-
mechanics and accident reconstruction and investigation. In 1996 he received the
Award for Engineering Excellence in Traffic Safety in USA.
Wim Rogmans is the Director of the Consumer safety Institute in the Netherlands. He is
the General Secretary of the European Consumer Safety Association (ECOSA). He is
involved in international collaboration through the World Health Organization
(representing the Consumer Safety Institute as WHO Collaborating institute for Injury
Prevention) and the European Community Commission. He is also Editor-in-Chief of
the International Journal for Consumer and Product Safety.
Mark L.Rosenberg, M.D., M.P.P., is the Executive Director of the Task force for Child
Survival and Development. He is also Program Director for the Collaborative Center
for Child Well-being, Atlanta. He has served in the US Public Health Service as
Assistant Surgeon General and is the Founding Director of the National Center for
Injury Prevention and Control at CDC Atlanta, USA.
A.K.(Dunu) Roy, M.Tech, is Dean Research at the People’s Science Institute, Dehradun,
India. A chemical engineer with 30 years experience in rural development,
environmental planning and safety training. He is involved in running the Hazard
Centre which provides technical consultancy to community organizations on hazard
and pollution control.
Mohamed Seedat is Director of the Institute for Social and Health Sciences at the
University of South Africa. His special interests are best practice approaches to
violence prevention and safety promotion. His research interests include community
resilience, youth development and health.
Leif Svanström, M.D., Ph.D., Professor and Chair of the Research Group on Safety
Promotion and Injury Prevention at the Division of Social Medicine, Department
Public Health Sciences at the Karolinska Institutet, Sweden. His research has
concentrated in developing evaluation models on process and outcome of Safe
Community programmes.
Geetam Tiwari, Ph.D., is Assistant Professor in the Applied Systems Research
Programme of the Department of Electrical Engineering at The Indian Institute of
Technology, Delhi. Her area of specialization includes urban transportation planning
and travel demand modelling including analytical and simulation models of
environment, energy and traffic safety.
Anne Tursz, M.D., is Research Director, National Institute for Health and Medical
Research. She directs epidemiological research projects,with a multi-disciplinary
focus, on child and adolescent health, with particular interest in the use of health care
services, injury and violent deaths. Her research interests include methodological
aspects of reliability and quality of data. She is also a WHO expert on Injury
Prevention and Control and a member of the French National Interministerial
Committee for Road Safety.
Mathew Varghese, MBBS, MS, is the Head of Department of Orthopaedics and Director ,
St. Stephen’s Hospital, Delhi. He has specialized in trauma care with particular
emphasis on reconstructive surgery for complex trauma to the musculo-skeletal
system. He is associated with the Transportation Research and Injury Prevention
Programme at the Indian Institute of Technology, Delhi as an expert on injury control
and epidemiology of injuries.
Preface
This book is being published on the occasion of the Fifth World Conference on Injury
Prevention and Control to be held in the first year of the twenty-first century.
According to all estimates, non-communicable diseases will be a major cause of
morbidity and mortality in all parts of the world in the next century. Intentional and
unintentional injuries will be associated with a significant proportion of this burden of
health. The control of this problem would depend on how much we can learn from each
other, from different disciplines, from different income groups and different parts of the
world.
Papers included in the volume reflect a wide variety of experiences, philosophies and
ways of doing things. All except one will be presented at the Fifth World Conference as
plenary or state-of-the-art lectures. We have included William Haddon’s paper because of
his lasting and important contribution to the field of injury control.
Haddon insisted that injury is a public health problem and must be dealt with as such.
Many societies, but not enough, have already recognized this and included injury control
in their health policies. Such policies are already showing positive results. Other
researchers have also stressed the role of societal and organizational structures in
determining the level of safety in our world. It is clear that just blaming the individual or
trying to reform him cannot solve the problem. To ensure a safer environment around us
we need to pay much more attention to changing designs, structures and the relative
power of different stakeholders in society.
Our objective in putting these papers together is to bring a broad understanding of the
state-of-the-art knowledge in the field and to explore future directions. The chapters fall
into three broad groupings: injury control and societal imperatives; survey of current
knowledge and possibilities for future action; use of different theoretical techniques in
injury control and the need for innovation in these techniques.
It is clear to us that injury control research needs to move beyond routine collection of
data and analysis. All safety counter-measures have to be put in place by law makers and
accepted by the people. This is why the interface between science, society and
technology is critical in this endeavour. Obviously, this increases the complexity of our
task. We hope that this volume helps professionals in meeting the challenges of injury
control in the new millennium.
Dinesh Mohan
Geetam Tiwari
Acknowledgements
The editors gratefully acknowledge the support of:

The International Organising Committee and all other committee members and
advisors of the Fifth World Conference on Injury Prevention and Control.
Insurance Institute for Highway Safety, Arlington, Virginia.
Mahesh K.Gaur for data processing.
1
Injury Control and Safety Promotion: Ethics,
Science and Practice
Dinesh Mohan
INJURY CONTROL, SOCIETY AND ETHICS
Injury control, safety of individuals and societal arrangements are all interlinked. William
Haddon wrote seminal pieces on the folly of focusing on ‘human error’ as the main cause
in the occurrence of accidents (Haddon, 1968; 1970; 1972; 1973; 1974; 1980). He did not
like the use of the word ‘accident’ as he thought that this leads to a feeling of inevitability
in the occurrence of these incidents. Further, he was convinced that the term ‘accident
prevention’ was too limiting and prevented the evolution of other safety counter-
measures useful in limiting the severity of injury and in injury management after the
event. Instead, he promoted the use of the phrase ‘injury control’ as being more neutral
and scientific. But, he did not address the issues of ideology and the power of elites that
societies are influenced by.
Perrow (1994, 1999) on the other hand, agrees with Haddon that individuals cannot
always be held responsible for ‘human error’ under the system they operate in but
provides a more sophisticated model of systemic imperatives: ‘I wish to point away from
the basic and pervasive sin identified by those who casually examine organizational
failures, that of operator error; this is given as the cause of about 80 per cent of the
accidents in risky systems. I would put it at under 40 per cent. I will suggest that what is
attributed to operator error stems primarily from the structure they operate in, and thus,
stems from the actions of elites. Elite errors and elite interests stem from their class and
historical power positions in society, and changes in these positions are glacial’ (Perrow,
1994).
Obviously, societal responsibility in the control of injuries becomes paramount when
the problem is stated in these terms. Morbidity and mortality due to injuries have always
existed in the past but their recognition as a public health problem is a phenomenon of the
mid-twentieth century. Policy makers and safety professionals in every country find it
very difficult to institute changes which actually result in a dramatic decrease in fatalities
due to injuries. This is mainly because experience shows that individuals do not follow
all the instructions given to them to promote safety. Attempts to educate people regarding
safety are also not very effective and wide variations are found between people’s
knowledge and their actual behaviour (Robertson, 1983). This is particularly true for
those situations where we cannot select the people who will be involved in a particular
activity. For example, almost everyone is involved in domestic chores, in road use and
working in offices, factories or on farms. It is not possible to select people who will
Injury prevention and control 2
always be careful in performing these activities. While some control can be exercised in
licensing drivers of motor vehicles, almost no control is possible in selection of
pedestrians and bicyclists. At the work place, only some very specialized jobs allow
careful selection and monitoring. This makes it very difficult to promote safety by relying
on improvements in individual behaviour and makes injury control a very complex
process. This is illustrated by using road traffic as an example.
Almost all the persons in the school-going and working age groups have to be on the
road at least twice a day in every country. This forces many individuals to use the road
even when they are not adequately equipped to do so. These situations would include
individuals with any of the following problems:
• Those who unable to concentrate on the road because they have suffered a
personal tragedy recently, such as death of a loved one, loss of a job, failure in
an important examination, monetary loss.
• Those who are disturbed because of problems in personal relationships with
a spouse, parent, sibling or close friend.
• Persons taking medication or drugs which alter behaviour and perceptual
abilities, or those who are under the influence of alcohol.
• Children whose cognitive and locomotor abilities make it difficult for them
to understand or follow instructions given to them.
• Elderly people whose motor and cognitive functions are impaired.
• Disabled persons who have to be a part of regular traffic if they have to earn
a living.
• All psychologically disturbed persons who may not be able to function as
desired on the road but who cannot be singled out from participation in traffic.
therefore, we have a societal and moral responsibility to design our products,

environment and laws so that people find it easy and convenient to behave in a
safe manner without sacrificing their needs to earn a living and fulfill their other
societal obligations
If we add up the total number of individuals who could be included in these categories on
any given day it would amount to a significant proportion of people on the road (say, 20–
30 per cent). These individuals cannot always be identified or prevented from using the
road space. At the same time it is also a fact that their presence on the road is not out of
choice, but a compulsion. In our modern ways of living we have to use products and do
things at places and at times which are determined by someone else or by the society at
large. The same holds true for activity at the work place or even at home. A large number
of us have little choice in the design of the home we live in, the design of the tools we
use, or the work place where we spend a major part of the day.
Therefore, we have a societal and moral responsibility to design our products,
environment and laws so that people find it easy and convenient to behave in a safe
manner without sacrificing their needs to earn a living and fulfill their other societal
obligations. The systems must be such that they are safe not only for ‘normal’ people but
also for those individuals who might belong to any of the groups of people having
Injury control and safety promotion 3
problems. These kinds of designs, rules and regulations would reduce the probability of
people hurting each other or themselves even when they make mistakes. Such systems
are very often referred to as ‘forgiving’ systems.
Such systems cannot be put in place unless there is a societal and political
understanding about the ethical and moral responsibility of the state and civil society to
ensure the right to life of all its citizens. This right to life includes living in good health
according to currently available knowledge and technology. A document prepared
recently by safety professionals summarizes some of these issues as follows (Maurice et
al., 1998):
• Safety is a fundamental human right.

• Safety is a state in which hazards and conditions leading to physical,
psychological or material harm are controlled in order to preserve the health and
well-being of individuals and the community. It is an essential resource for
everyday life that an individual and a community need in order to realize their
aspirations.
• There are two dimensions to safety: one is objective and assessed by
behavioural and environmental objective parameters and the other is subjective
and appreciated according to the feeling of safety of the population. Both
dimensions can influence each other either positively or negatively. It is
therefore necessary to consider these two dimensions to improve the safety of
the population
• Safety is a pre-requisite to the maintenance and improvement of the well-
being and health of the population. It is the result of a dynamic balance that is
established between the different components within a specific setting.
INJURY CONTROL, PUBLIC HEALTH AND TECHNOLOGY
Once we accept that injury control is a public health problem and that we have an ethical
responsibility to arrange for the safety of individuals, then it follows that we also
incorporate the lessons learned from our experience of the past few decades. We know
that drinking water should be purified at its source; it is unreasonable to expect everyone
to boil water before drinking it. Those societies which depend upon individuals to purify
their own drinking water suffer from much higher rates of communicable diseases than
those which purify water at source. Ironically, it is quite common to create a product or
environment which is likely to cause injury, warn the user to be careful, and then blame
the user if a mishap occurs. We would never tolerate a person who introduced cholera
germs in the city water supply and then ‘educated’ every citizen to boil water before
drinking it with the argument that those who knowingly do not do so would then be
responsible for getting sick. This is the argument we all to often use when dealing with
matters concerning safety. We put in place hazardous roads, vehicles and driving rules,
and then expect road users to be safe by behaving in some ideal manner.
Once we are clear that injury control activities involve the same principles as any other
public health problem, then we can institute policies and programmes for
institutionalizing safety promotion. However, most models of safety promotion and
community action have their origins in the high income countries (HIC) and it is assumed
that similar measures would work in the low income countries (LIC) also. Many of these
policies are heavily dependent on introduction of expensive technologies and difficult
regulation and legislation enforcement systems. Therefore, the transfer of ‘knowledge’
from HICs to LICs is sometimes almost impossible. However, we forget that many
advances in public health in the control of communicable diseases took place before the
invention of the modern definitive disease control drugs and vaccines.
Figure 1 Decrease in disease specific mortality rates in USA 1900–1980 (Data

from Sathyamala et al., 1986)
Figure 1 shows the decrease in disease specific mortality rates in USA, 1900–1980. This
data show that medical interventions for measles, scarlet fever, tuberculosis and
poliomyelitis were introduced when the incidence rates had already declined considerably
(Sathyamala et al., 1986). None of these death rates, except polio, show as marked a
decline after the medical intervention as before. There is enough other evidence available
to show that improvements in public health involve much more than just the introduction
of new technologies and treatment methods. For example, in India the crude death rate
declined from 47.2 per thousand per year in the decade of 1911–1921 to 27.4 for the
decade 1941–1951 (Government of India, 1983). This 42 per cent reduction in death rates
over three decades in India took place when most medical technologies were not
available to a vast majority of the country’s population. These reductions took place
because of improvements in environmental and other social conditions.
safety is a pre-requisite to the maintenance and improvement of the wellbeing and

health of the population
Disease is the product of social, economic and technological environments that people
live in. Within this environment the power to take decisions regarding choices available
for one’s own well being and the power to influence other people’s lives plays an
important role in what health benefits are available to the society as a whole. The same
holds true for injury control and safety promotion.
Injury control activities will not be successful around the world unless we address
these issues of social, economic and technological environments and the power available
to people to influence decision making regarding their own well being. The lower the
income of a society, the more important it becomes for injury control counter-measures to
be instituted at the societal and system level. Many of the current approaches in injury
control do not give enough importance to these aspects of the problem. They focus on
control at the product or environmental design levels or in the establishment of safety
standards without looking at the ability of a society to influence or sustain these
activities. Success in promoting safety in LICs is further limited because LIC systems
happen to be more complex than those obtaining in HICs. We cannot wait for all
communities to become rich before they can avail of the same systems. The history of
Public health shows that it should be possible for our lives to be safer in all communities
before the introduction of new expensive technologies and expensive societal
arrangements.
Societal complexity and injury control

The following factors played an important role in instituting safety programmes and
policies in HICs:
• Decline of mortality due to infections and contagious diseases. This made the
community more aware of injuries as a health problem and therefore gave
support to injury control initiatives as a priority.
• Development of a middle class society. By the mid-eighties a significant
majority of Europeans had incomes which would define them as ‘middle class’.
At the same time an equalization process took place which made most
professionals ‘equals’. This meant that policemen, school teachers, doctors,
nurses, lawyers, university professors could sit around a table and actually
communicate and respect each other as equals. Cooperation between various
interest groups, law enforcers, policy makers and policy implementers then
becomes more possible. These processes resulted in conglomerations of people
which could be called ‘communities’ in a real sense. Most counter-measures for
injury control benefit large proportions of the community. It can also be
assumed that particular counter-measures would not harm some sections of the
population since there are less conflicts of interest by different class categories.
• Acquisition of decision making powers by local self governments. Over time
local communities have been able to acquire decision making powers over most
aspects of community life, owing to national governments’ inclination to
decentralize policies that relate closely to the citizens’ well being. This gives
them the confidence to attempt changes.
• Establishment of institutions and organizations with high degree of
expertise. This makes it possible for reasonably accurate and reliable data to be
collected. These data can then be used for policy making purposes with support
from most sections of society.
• Laws can be enforced. Because of the relative egalitarian structure of society
it is assumed that most laws would affect most people in a similar manner. Since
the law enforcers belong to the same social stratum as the general public it
becomes possible to enforce laws more efficiently and more uniformly.
• Availability of safer technologies. Most technologies are developed and their
designs controlled by the wider society where they are needed. Such
technologies are more in tune with the needs of the community and can be
changed if necessary.
• Safety standards can be enforced. Since most production is centralized, it is
possible to make standards and enforce them.
In most LICs at present many of these conditions are not ment. Some
characterisitics of LICs are as follows:
• Heterogeneity. The post-war period has witnessed the emergence of a very
large number of independent nation-states in Africa and Asia. Most of these
nation-states had never existed in the present form ever before in history. Many
of these countries came into existence, most of them have very mixed
population. These population may differ in religions, languages, common law,
social customs and may not have shared values. The urban areas in these
countries house people with very diverse backgrounds and so there may be very
little homogeneity. In many cities in LICs people live in developments
characterized by ethnic and religions bonds.
infections, contagious diseases, and other health problems due to

malnutrition, air and water contamination, parasites, mosquitoes and
unsafe work conditions, dominate the attention of the public and policy
makers. Under such circumstances it becomes very difficult to arrive at a
consensus to consider injuries as an important public health problem
• Inadequate public health facilities. Most LICs have not been able to
institutionalize twentieth-century levels of hygiene and public health. Infant
mortality and maternal mortality indices remain much higher than those in HICs.
In addition, infections, contagious diseases, and other health problems due to
malnutrition, air and water contamination, parasites, mosquitoes and unsafe
work conditions, dominate the attention of the public and policy makers. Under
such circumstances it becomes very difficult to arrive at a consensus to consider
injuries as an important public health problem.
• Hierarchical societies. Most LICs have not been able to achieve high
enough levels of economic growth over the past four decades. Low economic
growth combined with nonegalitarian ideologies result in very low levels of
upward mobility. The poorer sections of society remain dominant in terms of
proportions of the population, but they have little influence on setting the policy
agenda. Within institutions the hierarchy also gets in the way of the dialogue
needed for smooth functioning. Teachers, nurses, policemen occupy low social
status as far as decision making is concerned. They hardly ever get to sit at the
same table where bureaucrats and experts discuss policies and take decisions.
• Inadequate control over technology. Most LICs import almost all
technological products and processes from HICs. Even aid projects ensure
movement of technology from the donor to the receiver. Very often this
technology is old or less expensive, and therefore, more hazardous. Local
communities have almost no control over the choice of these technologies. For
example, when a highway project is executed, the design and construction are
done by people who belong to the metropolis of that country aided by experts
from multilateral or bilateral international agencies and multinational
corporations. The local community can hardly influence the execution of these
projects except in the form of protests to halt the construction or change the
location of the highway. Most of the time they do not have the expertise or the
power to influence design. In addition, the local community may not possess the
expertise to evaluate the hazards implicit in the designs of products or
technologies being put in place.
• Increase of complexity in social and technological systems. Over the past
few decades standardization of and homogenization of technologies has resulted
in the reduction of complexity in many sectors in HICs. The roads have become
identical in layout and design, vehicles have become similar, variety of vehicles
has been reduced, school designs for most sections of the population are
becoming similar, technologies used in houses are similar and the labour
component in industry and farming has reduced. This reduction in complexity
has made it somewhat easier to institute safety counter-measures.
On the other hand, in most LICs, both social structures and technologies include a great
deal of variety which leads to more complex systems. The most modern vehicles share the
same road space with non-motorized transport, modern gadgets are used in a traditional
kitchen, inadequately trained labour is forced to handle high energy chemicals and
equipment, and mechanized systems co-exist with labour intensive ways of living. These
issues concerning increasing complexity in LICs is discussed in the following section.
Consequences of increasingly complex systems in LICs

Systems which have unfamiliar feedback loops, many potential interactions, indirect or
inferential information sources, and limited understanding of some processes are
considered to be more complex than those with the opposite characteristics (Perrow,
1999). These characteristics of LICs show that these countries tend to have more complex
social and technological environments than those present in HICs. The most important
issue to be understood regarding increasingly complex systems in LICs is that these
societies face new problems which are different from those currently prevalent in HICs.
They also have little precedence in the past of the HICs. It is not usually possible to find
solutions from the past of HICs and transfer these old solutions to LICs today.
The complexity in the socio-political domain is a result of centralized decision making
systems of nation-states and local government bodies not being able to accommodate the
interests of the poorer sections of society. This happens because the individuals who take
decisions are getting increasingly insulated from the daily lives and concerns of
disadvantaged communities. Ease in international travel and instantaneous
communication links between elite groups around the world tends unifies their interests
and concerns. In earlier centuries and the first few decades of the present century there
was greater conflict between the elite groups across nation-states than there exists today.
This interaction and solidarity between richer sections of society in different communities
and the conflicts between the poor sections within and across communities is at the
expense of the interests of the poor.
The interests of poor communities can also be in direct conflict with the interests of the
richer ones. Providing safety at the work place to prevent a small number of injuries and
deaths may reduce the profits of the owner and the shareholders of the company. Slowing
down traffic and providing a larger number of safer pedestrian crossings annoy car
owners. Providing low-cost housing for low income groups takes away expensive land
for making larger houses for the richer ones.
These problems are further compounded by the fact that the global information
exchange makes poor people more aware of the latest happenings all over the world and
raises their expectations for fair play. This results in more conflict in society making
governance more difficult. Known counter-measures for safety demand the use of latest
technologies which may or may not be suitable for the problem at hand. Innovative
technologies need to be developed, but most LICs do not at present have the scientific
pool of manpower or the institutional structure to develop these new technologies. A
result of injury control work still not being perceived as an activity requiring
sophisticated scientific and academic input. So communities end up trying out one
unsuccessful ‘solution’ after another. This promotes a feeling of helplessness,
powerlessness and lack of trust in policy makers. Not an ideal situation for ‘community’
action.
What needs to be understood is that the theoretical base of injury control counter-
measures may have international applicability but the actual physical solutions may not.
There is clearly a poverty of theory for work around the globe. For example, most road
safety measures instituted in HICs have centred around the automobile and the
automobile occupant. Road and intersection designs are based only on car, bus, and truck
movement. Roads in LICs are dominated by motorcycles, human-powered vehicles,
pedestrians carrying loads and locally-designed vehicles. No traffic flow models and
computer programs are able to account for this mix. Even if all the solutions developed in
HICs were put in place on the roads of LICs, the decrease in fatality rates would not be of
the same magnitude as experienced in the HICs.
what needs to be understood is that the theoretical base of injury control counter-
measures may have international applicability but the actual physical solutions
may not. There is clearly a poverty of theory for work around the globe
A good example of this is the role of expressways in inter-city travel. When an

expressway is built through the countryside it divides the landscape into separate zones.
People from one side of the expressway cannot go to the other side of the expressway
easily on foot or on a bicycle. In HICs this does not pose a serious problem as most
people possess motorized transport. However, in LICs the countryside may be heavily
popukted on both sides of the expressway by people of low income who need to interact
with each other. They need to cross the expressway carrying or pulling heavy loads. In
such a situation they do not like to go long distances to cross the expressway at
designated overbridges or underpasses. They end up breaking the fences and crossing the
expressway at locations convenient for them. This makes the expressway much more
hazardous for everyone concerned. The decision makers come from a different strata of
society who are only concerned with increasing the flow of inter-city motor traffic and
see the villagers as impediments to ‘progress’. This is just one example showing the
inherent conflict among different ‘communities’ and the difficulties inherent in
promoting a safer environment for everyone.
SCIENTISTS, SOCIETY AND INJURY CONTROL
The discussion in the previous section highlights the complex issues involved in dealing
with public health problems at different income and organizational levels. A further
complication is the role of scientists in dealing with issues concerning changes in
technology and policy at the societal level. Accompanied by the ‘globalization’ of the
economy is the globalization of the measure of scientific competence irrespective of the
location where the professionals work. Pressures on scientists force them to try and
maintain ‘high standards’ judged by the concerns of HICs. At times this works against
the interests and the needs of the many in order to concentrate on the conditions of the
few holding economic power. It will not be easy to change this state of affairs without
understanding the changing role of scientists and their interaction with society.
Safety promotion requires a great deal of understanding between scientists, the public
and policy makers they seek to address. A very large number of findings in the injury
control area are counter-intuitive: limitations of education, lack of effectiveness of severe
punishment for violators of safety norms, low correlation between attitude and behaviour
of individuals, some safer technologies promoting riskier behaviour, etc. In this situation
large doses of ‘science’, especially through the mass media, does not enable people to
digest information adequately as experts disagree on many issues. Within this confusion,
it is very convenient for vested interests to introduce political and business agenda in the
guise of technical and cost-benefit analyses. The citizen is left with little except cynicism
as she has no competence to examine competing claims. Discussing the problems
inherent in the public understanding of science, Raina (1999) claims that, ‘The scientists’
understanding of science is a very limited aspect of a more complex web of
understanding that weaves together science and society… We all grew up with the image
of science as disinterested investigation. But this image of science has been bruised and
science is now projected as serving vested interests… Studies on scientific controversies
reveal that every time a controversy erupts, the reserves of trust available within the
community of science are exposed, and there is a corresponding change in the social
authority of science… Thus both trust and judgement, central and enduring
characteristics of academic science, are rendered suspect in practical circumstances.’
These concerns become particularly relevant for those involved in injury control and
safety promotion because very few safety policies can be put in place without political
support and public agreement. This necessitates the engagement of at least some
scientists in the public domain. Traditional research activity, especially in universities,
was confined to the academic world without frequent interaction with users of the
knowledge. However, effective injury prevention and control work cannot be done in the
in the confines of academia and needs the practitioner to engage professionals in other
disciplines along with civil society. This mode of operation is relatively new and has
been characterized by Gibbons et al. (1994) as ‘socially distributed knowledge’. The
differences between ‘traditional knowledge generation’ and ‘socially distributed
knowledge’ is shown in Table 1.
Table 1
Features of traditional and socially distributed knowledge generation (Adapted from

Gibbons et al., 1994).
TRADITIONAL KNOWLEDGE SOCIALLY DISTRIBUTED

GENERATION KNOWLEDGE
Production Problems set and solved by academic Context of application
sites interests of a specific community
Nature Disciplinary Transdisciplinary
Characteristic Homogeneity Heterogeneity
Organization Hierarchical Hetrarchical and transient
Social Individualistic, emphasizes autonomy More socially accountable and
feedback reflexive
Injury control scientists work on generation of socially distributed knowledge. Such

knowledge is difficult to produce and even more difficult to disseminate in an effective
manner. It is difficult to produce because of its transdisciplinary, heterogeneous,
hetrarchical and transient characteristics. The work requires not only people in different
disciplines working together but also the integration of knowledge generated by
professionals who have no personal contact with each other. The actual operation of
research projects may involve individuals who are not controlled by a single authority
figure. In addition these scientists could also be working under very different paradigms
and who may also have very heterogenous working methods. Scientific work under such
conditions does not have a long history and tradition and so the methods and approaches
used are in a nascent stage of evolution.
Injury control research fits this description of socially distributed knowledge rather
adequately. Such work needs very innovative working techniques and the same are being
evolved slowly. The present institutional structures, collaboration mechanisms for
intradisciplinary research, peer recognition methods, information dissemination and
sharing techniques, and structures for interaction between scientists and the public are
still somewhat weak. The better structures and methodologies will become apparent only
if we consciously evaluate experiences, successes and failures in widely different
societies and settings and spread our net much wider than at present.
effective injury prevention and control work cannot be done in the in the confines
of academia and needs the practitioner to engage professionals in other disciplines
along with civil society
THE WAY FORWARD
This re-focusing of our efforts will not be easy. The international scientific community
still does not view much of the work being done for injury control as ‘sophisticated’
enough. However,this view is likely to change as we become more adept at generating
‘socially distributed knowledge’ and our work leads to benefits for a larger proportion of
the population around the world. Though most of the principles we discover will have
universal applicability, many of the technologies and specific methods may not. Some
critics may still not regard research work on many of these non-global technologies to be
‘modern’ or ‘scientific’ enough. However, the contrary is true. The issues surrounding
these products are actually very modern. They are the products of the late twentieth
century—combination of new socio-economic living patterns, instantaneous global
communication, availability of sophisticated scientific knowledge and low per capita
incomes. Work on these technologies will require very innovative thinking, familiarity
with the latest scientific information, and packaging of products in ways which may
require combination of technologies already available with those developed by locally in
new settings. Unless we change our research and development activities in this direction
we are likely to end up with very inefficient technological systems in our society. Our
success in future will depend on how much we are willing to learn from each other and
blending perspectives rather than working in our narrow confines of ‘science’.
Such changes will require much greater efforts to integrate theory with practice.
Current trends in injury control are giving prime importance to data generation,
surveillance and epidemiology. More and more sophisticated statistical methods are
being used to analyse these data resulting in debates on micro-issues of methodology and
definitions. Such work is also rewarded by journal editors and the scientific peer group in
the form of published articles and recognition. However it need not result in deeper
insight needed for socially generated knowledge. It is difficult to write a prescription for
the future course of action, but the following guidelines may help us stumble onto the
right path.
• A conscious effort to examine the theoretical directions in disciplines which

do not have much involvement in injury control at present. Theories on
complexity in physics, mathematics and computer science may help us
understand the behaviour of the system that we want to make safer. Work in
evolution, biology and genetics may give us clues on how natural systems
behave. Linguists and philosophers of science could lead us to a better
understanding of the role of metaphors and symbols in the building of
knowledge. This transdisciplinary work would need the development of
arrangements for information exchange even when partners are not involved in
work on the same problem.
• Development of a better understanding of the needs, aspirations,
compulsions and operational mechanisms of people belonging to low income
communities. Only then would it be possible to promote safety policies which
are sustainable in such settings. Such communities form the majority of the
world population and also exist in a minority in HICs. They are the norm rather
than the exception.
• Institution of imaginative collaborative projects between researchers
working in low and high income communities. Since safety counter-measures in
LICs have to be low cost, such projects could provide to useful insight on
reduction of safety expenditures in HICs with evolution of innovative ideas and
products. Joint working in LICs and HICs would give a better understanding of
the characteristics of ‘natural’ systems.
• Exploration of societal arrangements and technologies which ensure safer
lives with minimum policing.
• Promotion of safety as a fundamental human right. Inclusion of the costs of
safety in the basic design of products, infrastructure, roads, buildings and work
places. These considerations would decide whether any of these items should be
put in place or not.
REFERENCES
Gibbons, M., Limoges, C., Noworthy, H., Schwartzman, S., Trow, M. and Scott, P., 1994,
The New Production of Knowledge: The Dynamics of Science and Research in
Contemporary Societies. (London: Sage).
Government of India, 1983, Health Statistics of India. Ministry of Health and Family
Welfare (New Delhi: Government of India).
Haddon W., Jr., 1970, On the Escape of Tigers: An Ecologic Note. American Journal of
Public Health, 60(12), pp. 2229–2234.
Haddon W., Jr., 1972, A logical framework for categorizing highway safety phenomena
and activity. Journal of Trauma, 12(3), pp. 193–207.
Haddon W., Jr., 1973, Energy damage and the ten countermeasure strategies. Human
Factors, 15(4), pp. 355–366.
Haddon W., Jr., 1980, Advances in the epidemiology of injuries as a basis for public
policy. Public Health Report, 95(5), pp. 411–421.
Haddon W., Jr., 1974, Editorial: Strategy in preventive medicine: passive vs. active
approaches to reducing human wastage. Journal of Trauma, 14(4), pp. 353–354.
Haddon, W., Jr., 1968, The changing approach to the epidemiology, prevention, and
amelioration of trauma: the transition to approaches etiologically rather than
descriptively based. American Journal of Public Health, 58(8), pp. 1431–1438.
Maurice, P., Lavoie, M., Charron, R.L., Chapdelaine, A., Bonneau, H.B., Svanstrom, L.,
Laflamme, L., Andersso, R. and Romer, C., 1998, Safety and Safety Promotion:
Conceptual and Operational Aspects. (Quebec, Cananda: Quebec WHO Collaborating
Centre for Safety Promotion and Injury Prevention).
Perrow, C., 1994, Accidents in high-risk systems. Technology Studies, 1(1), pp. 1–38.
Perrow, C., 1999, Normal Accidents: Living with High-Risk Technologies. (Princeton,
NJ: Princeton University Press).
Raina, D., 1999, Science and its publics. IIC Quarterly, 26(2), pp. 42–53.
Robertson, L.S., 1983, Injuries: Causes, Control Strategies and Public policy.
(Lexington, MA: Lexington Books).
Sathyamala, C., Sundharam, N. and Bhanot, N., 1986. Taking Sides: The Choices Before
the Health Worker. (Chennai: Asian Network for Innovative Training Trust).
2
On the Escape of Tigers: An Ecological Note
William Haddon, Jr.
Republished with permission of MIT Technology Review, from ‘On the Escape of Tigers:
An Ecological Note’, Volume 72 Number 7. Copyright 1970; permission conveyed
through Copyright Clearance Center, Inc.
A major class of ecologic phenomena involves the transfer of energy in such ways and
amounts, and at such rapid rates, that inanimate or animate structures are damaged. The
harmful interactions with people and property of hurricanes, earthquakes, projectiles,
moving vehicles, ionizing radiation, lightning, conflagrations, and the cuts and bruises of
daily life illustrate this class.
TEN STRATEGIES FOR REDUCING THESE LOSSES
Several strategies, in one mix or another, are available for reducing the human and
economic losses that make this class of phenomena of social concern. In their logical
sequence, they are as follows:
The first strategy is to prevent the marshalling of the form of energy in the first place:
preventing the generation of thermal, kinetic, or electrical energy, or ionizing radiation;
the manufacture of gunpowder; the concentration of U-235; the build-up of hurricanes,
tornadoes, or tectonic stresses; the accumulation of snow where avalanches are possible;
the elevating of skiers; the raising of babies above the floor, as to cribs and chairs from
which they may fall; the starting and movement of vehicles; and so on, in the richness
and variety of ecologic circumstances.
The second strategy is to reduce the amount of energy marshalled: reducing the
amounts and concentrations of high school chemistry reagents, the size of bombs or
firecrackers, the height of divers above swimming pools, or the speed of vehicles.
The third strategy is to prevent the release of the energy: preventing the discharge of
nuclear devices, armed crossbows, gunpowder, or electricity; the descent of skiers; the
fall of elevators; the jumping of would-be suicides; the undermining of cliffs; or the
escape of tigers. An Old Testament writer illustrated this strategy in the context both of
the architecture of his area and of the moral imperatives of this entire field: ‘When you
build a new house, you shall make a parapet for your roof, that you may not bring the
guilt of blood upon your house, if any one fall from it’ (Deuteronomy 22:8). This biblical
position, incidentally, is fundamentally at variance with that of those who, by conditioned
reflex, regard harmful interactions between man and his environment as problems
requiring reforming imperfect man rather than suitably modifying his environment.
On the escape of tigers 15
The fourth strategy is to modify the rate or spatial distribution of release of the energy
from its source: slowing the burning rate of explosives, reducing the slope of ski trails for
beginners, and choosing the re-entry speed and trajectory of space capsules. The third
strategy is the limiting case of such release reduction, but is identified separately because
in the real world it commonly involves substantially different circumstances and tactics.
The fifth strategy is to separate, in space or time, the energy being released from the
susceptible structure, whether living or inanimate: the evacuation of the Bikini islanders
and test personnel, the use of sidewalks and the phasing of pedestrian and vehicular
traffic, the elimination of vehicles and their pathways from community areas commonly
used by children and adults, the use of lightning rods, and the placing of electric power
lines out of reach. This strategy, in a sense also concerned with rate-of-release
modification, has as its hallmark the elimination of intersections of energy and
susceptible structure—a common and important approach.
there is no logical reason why the rank order…of loss-reduction counter-measures

generally considered must parallel the sequence, or rank order, of causes
contributing to the result of damaged people or property
The very important sixth strategy uses not separation in time and space but separation by
interposition of a material ‘barrier’: the use of electrical and thermal insulation, shoes,
safety glasses, shin guards, helmets, shields, armour plate, torpedo nets, antiballistic
missiles, lead aprons, buzz-saw guards, and boxing gloves. Note that some ‘barriers’,
such as fire nets and other ‘impact barriers’ and ionizing radiation shields, attenuate or
lessen but do not totally block the energy from reaching the structure to be protected.
This strategy, although also a variety of rate-of-release modification, is separately
identified because the tactics involved comprise a large, and usually clearly discrete,
category.
The seventh strategy, into which the sixth blends, is also very important—to modify
appropriately the contact surface, subsurface, or basic structure, as in eliminating,
rounding, and softening corners, edges, and points with which people can, and therefore
sooner or later do, come in contact. This strategy is widely overlooked in architecture
with many minor and serious injuries the result. It is, however, increasingly reflected in
automobile design and in such everyday measures as making lollipop sticks of cardboard
and making some toys less harmful for children in impact. Despite the still only spotty
application of such principles, the two basic requisites, large radius of curvature and
softness, have been known since at least about 400 BC, when the author of the treatise on
head injury attributed to Hippocrates wrote: ‘Of those who are wounded in the parts
about the bone, or in the bone itself, by a fall, he who fails from a very high place upon a
very hard and blunt object is in most danger of sustaining a fracture and contusion of the
bone, and of having it depressed from its natural position; whereas he that fails upon
more level ground, and upon a softer object, is likely to suffer less injury in the bone, or it
may not be injured at all…’ (On Injuries of the Head, The Genuine Works of
Hippocrates, trans. F. Adams [The Williams and Wilkins Co., Baltimore, 1939]).
The eighth strategy in reducing losses in people and property is to strengthen the
structure, living or non-living, that might otherwise be damaged by the energy transfer.
Common tactics, often expensively under-applied, include tougher codes for earthquake,
fire, and hurricane resistance, and for ship and motor vehicle impact resistance. The
training of athletes and soldiers has a similar purpose, among others, as does the
treatment of haemophiliacs to reduce the results of subsequent mechanical insults. A
successful therapeutic approach to reduce the osteoporosis of many post-menopausal
women would also illustrate this strategy, as would a drug to increase resistance to
ionizing radiation in civilian or military experience. (Vaccines, such as those for polio,
yellow fever, and smallpox, are analogous strategies in the closely parallel set to reduce
losses from infectious agents.)
The ninth strategy in loss reduction applies to the damage not prevented by measures
under the eight preceding-to move rapidly in detection and evaluation of damage that has
occurred or is occurring, and to counter its continuation and extension. The generation of
a signal that response is required; the signal’s transfer, receipt, and evaluation; the
decision and follow-through, are all elements herewhether the issue be an urban fire or
wounds on the battlefield or highway. Sprinkler and other suppressor responses, fire
doors, MAYDAY and SOS calls, fire alarms, emergency medical care, emergency
transport, and related tactics all illustrate this counter-measure strategy. (Such tactics
have close parallels in many earlier stages of the sequence discussed here, as, for
example, storm and tsunami warnings.)
The tenth strategy encompasses all the measures between the emergency period
following the damaging energy exchange and the final stabilization of the process after
appropriate intermediate and long-term reparative and rehabilitative measures. These may
involve return to the pre-event status or stabilization in structurally or functionally altered
states.
SEPARATION OF LOSS REDUCTION AND CAUSATION
There are, of course, many real-world variations on the main theme. These include those
unique to each particular form of energy and those determined by the geometry and other
characteristics of the energy’s path and the point or area and characteristics of the
structure on which it impinges—whether a BB hits the forehead or the centre of the
cornea.
One point, however, is of overriding importance: subject to qualifications as noted
subsequently, there is no logical reason why the rank order (or priority) of loss-reduction
counter-measures generally considered must parallel the sequence, or rank order, of
causes contributing to the result of damaged people or property. One can eliminate losses
in broken teacups by packaging them properly (the sixth strategy), even though they be
placed in motion in the hands of the postal service, vibrated, dropped, piled on, or
otherwise abused. Similarly, a vehicle crash, per se, need necessitate no injury, nor a
hurricane housing damage.
Failure to understand this point in the context of measures to reduce highway losses
underlies the common statement: ‘if it’s the driver, why talk about the vehicle.’ This
confuses the rank or sequence of causes, on the one hand, with that of loss-reduction
counter-measures—in this case ‘crash packaging’—on the other.
There are, nonetheless, practical limits in physics, biology, and strategy potentials. One
final limit is operative at the boundary between the objectives of the eighth and ninth
strategies. Once appreciable injury to man or to other living structure occurs, complete
elimination of undesirable end results is often impossible, though appreciable reduction is
commonly achievable. (This is often also true for inanimate structures, for example,
teacups.) When lethal damage has occurred, the subsequent strategies, except as far as the
strictly secondary salvage of parts is concerned, have no application.
There is another fundamental constraint. Generally speaking, the larger the amounts of
energy involved in relation to the resistance to damage of the structures at risk, the earlier
in the counter-measure sequence must the strategy lie. In the ultimate case, that of a
potential energy release of proportions that could not be countered to any satisfactory
extent by any known means, the prevention of marshalling or of release, or both,
becomes the only approach available. Furthermore, in such an ultimate case, if there is a
finite probability of release, prevention of marshalling (and dismantling of stockpiles of
energy already marshalled) becomes the only, and essential, strategy to assure that the
undesirable end result cannot occur.
the larger the amounts of energy involved in relation to the resistance to damage
of the structures at risk, the earlier in the counter-measure sequence must the
strategy lie
FOR EACH STRATEGY AN ANALOGOUS OPPOSITE
Although the concern here is the reduction of damage produced by energy transfer, it is
noteworthy that to each strategy there is an opposite focused on increasing damage. The
latter are most commonly seen in collective and individual violence-as in war, homicide,
and arson. Various of them are also seen in manufacturing, mining, machining, hunting,
and some medical and other activities in which structural damage often of a very specific
nature is sought. (A medical illustration would be the destruction of the anterior pituitary
with a beam of ionizing radiation as a measure to eliminate pathological hyper activity.)
For example, a maker of motor vehicles or of aircraft landing-gear struts—a product
predictably subject to energy insults—could make his product more delicate, both to
increase labour and sales of parts and materials, and to shorten its average useful life by
decreasing the age at which commonplace amounts of damage increasingly exceed in
cost the depreciating value of the product in use. The manufacturer might also design for
difficulty of repair by using complex exterior sheet metal surfaces, making components
difficult to get at, and other means.
The type of categorization outlined here is similar to those useful for dealing
systematically with other environmental problems and their ecology. In brief illustration,
various species of toxic and environment-damaging atoms (such as lead), molecules (e.g.
DDT), and mixtures (garbage and some air pollutants, among others) are marshalled, go
through series of physical states and situations, interact with structures and systems of
various characteristics, and produce damage in sequences leading to the final, stable
results.
Similar comments can be made concerning the ecology of some of the viral,
unicellular, and metazoan organisms that attack animate and inanimate structures; their
hosts; and the types and stages of damage they produce.*
Sufficient differences among systems often exist, however-for example, the ecology of
the agents of many arthropod-borne diseases is quite complex, and the life cycles of
organisms such as schistosomes require two or more different host species in sequence-to
preclude at this time many generalizations useful across the breadth of all environmental
hazards and their damaging interactions with other organisms and structures.
A SYSTEMATIC ANALYSIS OF OPTIONS
It has not generally been customary for individuals and organizations that influence, or
are influenced by, damage due to harmful transfers of energy to analyse systematically
their options for loss reduction, the mix of strategies and tactics they might employ, and
their cost. Yet, it is entirely feasible and not especially difficult to do so, although specific
supporting data are still often lacking. In fact, unless such systematic analysis is done
routinely and well, it is generally impossible to maximize the pay-offs both of loss-
reduction planning and of resource allocations.
Such analysis is also needed to consider properly the problems inherent in the use of
given strategies in specific situations. Different strategies to accomplish the same end
commonly have different requirements; in kinds and numbers of people, in material
resources, in capital investments, and in public and professional education, among others.
In the case of some damage-reduction problems, particular strategies may require
political and legislative action more than others. And, where the potential or actual
hazard exists across national boundaries, correspondingly international action is
commonly essential.
The types of concepts outlined in this note are basic to dealing with important aspects
of the quality of life, and all of the professions concerned with the environment and with
the public health need to understand and apply the principles
*Actual and potential birth control and related strategies and tactics can be somewhat similarly
categorized. Thus, in brief, beginning on the male line: preventing the marshalling of viable sperm
(by castration or certain pharmacological agents); reducing the amount of sperm produced;
preventing the release of semen (or of one of its necessary components, e.g. by vasectomy);
modifying the rate or spatial distribution of release of semen (as in hypospadias, a usually
developmental or traumatic condition in which the urethra opens on the underside of the penis,
sometimes near its base); separating semen release in space or time from the susceptible ovum
(e.g., continence, limiting intercourse to presumably non-fertile periods, coitus interruptus, and
preventing a fertile ovum from being present when sperm arrive); separation by interposition of a
material barrier (e.g. condoms, spermicidal creams, foams, jellies); increasing resistance of the
ovum to penetration; making the ovum infertile, even if penetrated; prevention of implantation of
the fertilized egg; abortion; and infanticide
involved and not in the haphazard, spotty, and poorly conceptualized fashion now
virtually universal. It is the purpose of this brief note to introduce the pathway along
which this can be achieved
SUGGESTED READINGS
W.Haddon, Jr., 1968, The changing approach to the epidemiology, prevention, and
amelioration of trauma: The transition to approaches etiologically rather than
descriptively based. American Journal of Public Health. 58, pp. 1431–1438.
W.Haddon, Jr., 1970, Why the issue is loss reduction rather than only crash prevention,
presented at the Automotive Engineering Congress, S.A.E., Detroit, Michigan, January
12, S.A.E. Preprint 700196.
W.Haddon, Jr., The prevention of accidents. In Textbook of Preventive Medicine, edited
by D.W.Clark and B.MaeMahon (Boston: Little, Brown and Company), pp. 591–621.
W.Haddon, Jr., E.A.Suchman, and D.Klein, Accident Research, Methods and
Approaches, Harper and Row, 1964. (See especially Chapters 9 and 10).
3
Where Have We Been and Where are We
Going With Injury Control?
Susan P.Baker
INTRODUCTION
I would like to begin this paper by discussing where we have been as we have travelled
the road to injury prevention. And although that is a fascinating story, I am even more
interested in discussing where we are now: what are we doing right, what are we doing
wrong? And, the most important part, where are we going? How can we best achieve
freedom from preventable injury, disability, and death for people throughout the world?
WHERE HAVE WE BEEN?
The history of injury control goes back many thousands of years. Prehistoric man may
not have understood the concepts of injury control, but he had an instinct for self-
preservation that led him to design tools with which he could not only slaughter wild
animals, but protect himself from being slaughtered in the process.
Somewhat later, people used amulets for self-protection. They may not have been
effective, but they reflected the need for protection against hazards that people
recognized yet could not prevent—lightning-induced forest fires, hurricanes, drought,
famine.
When man moved out of caves and into buildings he created new hazards. Dr. William
Haddon often pointed out that the Old Testament warns that there should be a parapet or
railing around the edge of a roof so that people could not fall off. (In Nepal and many
other places where animals are housed on the ground floor, children still play on roofs
with no walls, and injuries from falls are not uncommon.) Later, Leonardo da Vinci was
to advise that the floors of buildings should be strong enough to withstand the vibrations
when a building was shaken by people dancing.
Stephen Bradwell, in his 1633 book Helps for Suddain Accidents, provided an early
text of first-aid advice, somewhat limited by the knowledge of his day. He recommends,
for example, that if a viper creeps into your stomach, the remedy is ‘the smoake of burnt
old shooes received in at the mouth through a funnell’ (Bradwell, 1633).
Perhaps that sets the stage for looking at recent decades, during which we have begun
to take a far more scientific approach to injury control, following Dr. William Haddon’s
emphasis on not only the pre-event phase, that we used to call ‘accident prevention’, but
also the ability to prevent injury during the event phase while crashes, falls and shootings
Where are we going with injury control? 21
are actually occurring, and in the post-event phase when emergency services, acute care,
and rehabilitation can save lives and minimize disability (Haddon, 1972). Table 1
provides examples of strategies related to each of the phases.
Table 1
Examples of Factors Related to Injury Prevention on the Highway, in the Three Phases
PHASES FACTORS
Human Vehicle Physical and social environment
Pre-crash Alcohol intoxication Speed capability Signs, lighting
Fatigue Brakes Road design
Training, experience Centre of gravity Speed limits
Knowledge Alcohol laws
Crash Use of seat belts Airbags Median barriers
Helmet wearing Crashworthydesign Guardrails
Bone fragility Occupant containment Recovery areas
Post-crash Age Fuel system design EMS system
Physical condition Telephone Medical and rehabilita tiorcervices
First-aid knowledge
Haddon’s conceptual advances were important, but the greatest achievement of recent
decades, in my opinion, is the widespread recognition of injuries as a health problem and
the increasing numbers of health professionals who are dedicated to reducing this
problem.
epidemiologists too often overemphasize the personal risk factors that differentiate
between injured and uninjured people
WHERE ARE WE NOW?
A recent report of the Institute of Medicine, Reducing the Burden of injury (1999)
documents some of the current inadequacies of American efforts to control injuries.
These include the need for:
• Developing surveillance systems with uniform coding of the external cause

and nature of the injury
• Creating opportunities for training injury prevention researchers and
practitioners
• Planning, implementing, and evaluating trauma systems.
Most of these inadequacies are prevalent in other countries, both developed and
underdeveloped. Countries may differ markedly in their available resources; they
certainly differ in the specific circumstances of injury (for example, do young children
drown in swimming pools or irrigation ditches?). Countries also differ in the degree to
which people are protected by regulations and their enforcement.
A major similarity among countries, however, is in the importance of the injury
problem. Throughout the world, after early childhood the mortality from injuries dwarfs
other major causes of death for children and young adults. In the United States, China
and many other countries, injuries are the leading cause of death from about age 1 to after
age 40, yet in many countries injuries cause a far greater proportion of all deaths among
people age 15 or older than in the United States (Table 2).
Table 2
Number of injury deaths, and injuries as a percentage of deaths from all causes* in the
population 1 5 years in various Asian and Latin American countries with other
countries for comparison
ALL INJURIES YEAR

Number of deaths Percentage of all deaths
SOUTHEAST ASIA
China, rural 32,973 11 1994
China, urban 22,173 6 1994
Sri Lanka 11,421 14 1986
AMERICAS
Guatemala 3,548 12 1984
El Salvador 5,428 27 1984
Colombia 39,612 28 1991
Mexico 50, 107 15 1993
Honduras 1,823 16 1981
Ecuador 5,309 14 1988
Chile 7,571 11 1992
Cuba 7,979 12 1990
Dominican Republic 1,705 10 1982
Uruguay 1,676 6 1990
Argentina 14,391 6 1991
Venezuela 9,816 15 1989
COMPARISON
RussianFederation 347,600 16 1994

United States 136,990 6 1992
Sweden 4,420 5 1993
Australia 6,502 5 1993
Data from: World Health Statistics Annual 1985–1995 (WHO, 1985–1996)
* Includes all deaths from unintentional injury, homicide, suicide, and other violence (ICD-9
codes E800–999)
WHAT ARE WE DOING RIGHT?
One important improvement in recent years is the growing recognition that intentional
injuries, both assaultive and self-inflicted, are components of the injury problem and in
many respects can be prevented through the same means as unintentional injuries. We
know this in theory because the same gun that can be turned upon oneself or another
person by a teenager, or that can fire inadvertently when children are playing with it, can
be kept from harming anyone by a device that allows only the owner to fire it. There is
also evidence of the fact that a single preventive measure can prevent both intentional and
unintentional injuries: for example, there was a huge drop in both unintentional and
suicidal poisoning when poisonous coal gas was replaced by natural gas in the United
States (Baker et al., 1992). In England, it was shown that this advance almost eliminated
suicidal poisoning by gas, without a corresponding increase in suicide by other means
(Hassall and Trethowan, 1972). I emphasize the fact that a single approach can reduce
both intentional and unintentional injury because in countries such as Sri Lanka and
China, agricultural poisons are used as a means of suicide by many despondent woman,
at the same time killing curious children. I am optimistic that a single solution will be
found to address these huge problems. Larry Berger and Dinesh Mohan (1996) have
pointed to the potential value of national and multinational regulations that control the
manufacture and sale of pesticides to prevent both intentional and unintentional
poisonings.
the search for personal risk factors that would predict which members of this
homogenous group would be injured was doomed to failure, and distracted the
investigators from performing a descriptive study of the circumstances of injury
that would have had far greater potential for injury reduction
What else are we doing that is right? Certainly we have seen advances in many countries
in:
• The designs of vehicles to make them more crashworthy with seat belts, for
example,
• Safer toys that are less likely be choking hazards for children,
• Child proof fencing around swimming pools, which caused a nine-fold
difference in the child drowning rates between Canberra, which required the
fencing, and Brisbane, which did not.
We have made great advances in reducing alcohol-involved crashes with legal approaches
such as increasing the legal drinking age and lowering the legal alcohol level for drivers,
laws that in the United States have helped to substantially reduce teenage driver mortality
rates.
WHAT ARE WE DOING WRONG?
First, epidemiologists too often overemphasize the personal risk factors that differentiate
between injured and uninjured people. In a case-control study of back injuries in
municipal workers, my colleagues and I matched injured cases and their controls on the
basis of their departments and jobs, and then studied the differences between them in
hundreds of variables (Myers et al., 1999). Because we matched on the job, which is the
biggest determinant of back injury, we found that the greatest differences were in
personal factors such as the ratio of weight to height, and the workers’ attitude towards
their jobs. We purposely had not made it possible to analyse the relative importance of
their department and job as risk factors, because we knew that the huge difference among
departments in rates of back injury would overwhelm other risk factors. Unfortunately,
because of a design that obscured the effect of exposure to jobs that placed great strain on
the back, the results of our study placed great emphasis on personal risk factors rather
than aspects of the jobs that could be changed—inordinately heavy tree trunks and
moving equipment that had to be lifted on to trucks, uncomfortable stooping positions
from which they had to move or raise a load, etc. For purposes of prevention, it would
have been ideal to have placed greater emphasis on reducing the hazards.
Recently I talked with epidemiologists who were perplexed by their inability to find
differences between injured and uninjured military trainees. It turned out that the young
men in question had all been selected for an elite group of uniformly young, fit, intelligent
males to be trained for especially hazardous and demanding tasks. The training
programme was incredibly vigorous and had a high injury rate. But the search for
personal risk factors that would predict which members of this homogenous group would
be injured was doomed to failure, and distracted the investigators from performing a
descriptive study of the circumstances of injury that would have had far greater potential
for injury reduction. (A hypothetical example illustrating the same phenomenon would
look at 100 identical siblings in a building that caught fire and killed half of them and
then focus on personal risk factors rather than comparing injured vs. uninjured in terms of
where they were when the fire broke out in relation to the point of origin of the fire and
likely escape routes.)
Thus, we need to avoid the temptation to overemphasize personal risk factors when the
environment and circumstances of injury dictate not only who is hurt but also the best
means to prevent the injury.
WHAT ELSE ARE WE DOING THAT IS WRONG?
Too often, we design our physical environment for smart people who are highly
motivated to prevent injury. Virtually everything—from roads to hand tools to stoves to
poison containers—is designed with the assumption that people can and will read, follow
directions, proceed with caution, take no risks. Yet who are the people at highest risks?
Young children, teenage males, the elderly, the intoxicated—in other words, the people
who are hardest to influence with information and educational approaches.
Tragically, we have ignored the pressures that cause people to take risks:
• Peer pressure, that leads teenage drivers to take more risks when they have
passengers to distract or challenge them (Chen, 1999)
• The pressures of competition—you have all seen the risks taken by athletes
in the Olympics
• Productivity pressures—farmers rushing to harvest the hay before it rains
• Economic pressures that cause a poor man to accept the risks that go with a
hazardous job so he can feed his family
• The most important pressures are time pressures: no one likes to waste time.
That is why you see pedestrians darting across traffic lanes and drivers running
red lights; most people take the quickest way, even when it is not the safest way.
FUTURE DIRECTIONS
This leads me to my final question: where are we going with injury control?
First of all, I hope we are going to begin to design our environment and our tasks in a
manner that recognizes that most people have some kind of limitation—none of us is
perfect all of the time—and that most of us at one time or are under some form of
pressure that can lead us to take risks. This means, for example, that we should redesign
many of our jobs, so that workers have fewer incentives to work unsafely. When Swedish
lumbermen were paid by the hour, rather than by how many trees they cut or how much
lumber they sawed, their injury rate dropped dramatically (Sundstrom-Frisk, 1984). It
means designing roads that can be crossed safely by pedestrians because the safest way
has been made the easiest way, for example by raising the road slightly so that
pedestrians can pass beneath it rather that requiring them to climb many stairs with their
packages, bicycles, and children.
Second, I hope we are going to stop blaming the victim, finding fault with the injured
person so that we fail to reproach other people who might have prevented the injury-
designers, manufacturers, administrators, and other decision makers whose decisions can
determine the likelihood of injury for thousands of people.
too often, we design our physical environment for smart people who are highly
motivated to prevent injury. Tragically, we have ignored the pressures that cause
people to take risks
Asking the injured person about injury prevention can provide useful information, if the
question is asked the right way. If you ask an injured farmer how he could have kept from
losing an arm in a corn picker, he is likely to say ‘I shouldn’t have been so stupid.’ But if
you ask him how he might keep his best friend or his son from being hurt the same way,
he is more likely to say ‘Find a way to get the corn stalks unstuck without his even
having to get off the tractor!’
We took this approach with the back-injured municipal workers I mentioned earlier
and found they had sound suggestions, often the same ideas as the ones provided by a
trained ergonomist who evaluated the circumstances of injury. Some of the suggestions
offered by the workers on the sanitation crews when they were asked how to prevent
injuries to their friends were: ramps to help them load mowing machines onto trucks,
waist-high shelves for storing heavy boxes, and rollers to help them slide heavy food
trays.
Third, and finally, I hope we will continue to learn from one another, through meeting
and conferences as well as through collegial communication year-in and year out. We
each have much to share, to teach, and to learn. It is essential that we not only profit from
one another’s achievements, but also that we learn from one another’s mistakes. An
example of how we have failed to do that in the past is the tremendous toll taken by
motorization, in one country after another, because we have not heeded the fact that wide,
straight, undivided highways without reasonable and enforced speed limits invite the
mayhem of high-speed crashes.
An example of a way in which we can profit from other countries’ achievements is
through product safety regulation. Until each country adopts regulations that set standards
for safe products and prohibit the sale of hazardous products, whether domestic or
imported within its borders, its citizens will not be safe from many preventable injuries.
In the United States, we experienced one example of this when we failed to set standards
for infant pacifiers. In Maryland, our medical examiner determined that babies were
choking to death on undersized pacifiers. Through his efforts, Maryland set a standard
that prohibited the sale of pacifiers smaller than a certain size. Elsewhere in the United
States, however, small pacifiers could legally be sold, and these cheap imports were
popular. Tragically, despite petitions to the Consumer Product Safety Commission to set
a national standard, we did not get a national standard until more babies had died in other
states.
A type of product safety standard that might well be popular in any country is one that
keeps young children from being burned. No one who has ever seen a burned child wants
to see another. Most of you have seen cooking, heating, and lighting devices that are
extremely hazardous to little children: stoves that can be tipped over or that are so low
that children can easily reach them, for example, and unstable oil lamps that spill burning
oil when a child pulls on a table cloth. The sale of these and other products can be
regulated by the country in which they are sold, and is the kind of regulation most likely
to be supported by many people.
I hope that all countries will protect children of all ages and adults from hazardous
products by adopting regulations that ensure the safety of products sold within their
borders. If this gathering of injury prevention leaders were to focus on achieving a single
needed preventive measure, this one would have my vote.
REFERENCES
Baker, S.P., O’Neill, B., Ginsburg, M.J., and Li, G., 1992, The Injury Fact Book, Second
Edition. (New York: Oxford University Press).
Berger, LA., and Mohan, D., 1996, Injury Control: A Global View (New Delhi: Oxford
University Press).
Bradwell, S., 1633, Helps for Suddain Accidents. (London: Thomas Purfott).
Chen, L.H., 1999, Teenage Driver Crash Risk—The Effect of Passengers. (Baltimore,
Maryland: The Johns Hopkins University, School of Public Health).
Haddon, W., Jr., 1972, A Logical Framework for Categorizing Highway Safety
Phenomena and Activity. Journal of Trauma, 12, p. 193.
Hassall, C., and Trethowan, W.H., 1972, Suicide in Birmingham. British Medical
Journal, 1, pp. 717–718.
Institute of Medicine, 1999, Reducing the Burden of injury: Advancing Prevention and
Treatment, edited by Bonnie, R.J., Fulco, C.E., and Liverman, C.T. (Washington, DC:
National Academy Press).
Myers, A.H., Baker, S.P., Li, G., Smith, G., Wiker, S., Liang, K., and Johnson, J.V.,
1999, Back injury in municipal workers: A case-control study. American Journal of
Public Health, 89, pp. 1036–1041.
Sundstrom-Frisk, C., 1984, Behavioral control through piece-rate wages. Journal of
Occupational Accidents, 6, pp. 49–59.
4
Reducing Injury Losses: What Private
Insurers Can and Cannot Do
Brian O’Neill
INTRODUCTION
Injuries are a major public health problem worldwide and, increasingly, the focus of
prevention efforts based on science. Resources available for injury prevention are
minuscule. In USA about 150,000 injury deaths occur each year, but government
activities and funding of injury prevention programmes/research, including motor vehicle
crash injuries, are relatively insignificant when compared to those aimed at other major
health problems (Institute of Medicine, 1999).
The chronic problem of inadequate government commitment to injury prevention and
control has led professionals to seek additional involvement and support from the private
sector. One industry routinely identified as an obvious candidate is insurance, it is
assumed that insurers pay for injury losses, therefore, benefit financially from loss
reductions. This paper examines the validity of these assumptions and reviews some
successful injury prevention activities by insurers. The focus is on private insurers, not
insurance systems organized as government monopolies.
INSURANCE HISTORY
The basic concept of insurance, spreading risk of loss faced by an individual among a
larger group exposed to similar risks, is very old. Earliest forms of insurance provided
loans to fund risky businesses such as sending goods by caravan and ship, and loans were
repaid with interest only if the goods arrived safely.
By the mid-fourteenth century, marine insurance was widely used in Europe to provide
a mechanism for sharing risks associated with loss of ships and cargos. Lloyds of London
started as a marine insurer in the seventeenth century, becoming one of the first modern
insurers in the eighteenth century. To assess the risks it was considering, Lloyds collected
as much information as possible on ship owners, captains, ships, routes, and cargos.
Insurance premiums were adjusted to reflect anticipated risks.
In some cases, insurance was impossible to obtain because the risks were considered
too high. If risks could be substantially lowered, more ships could be insured and the
business could grow; so Lloyds began finding ways to prevent ships from colliding or
running aground. The intent was to reduce property losses, but a clear additional benefit
was to reduce the loss of life. Though a property insurer, in the early 1800s Lloyds
Reducing injury losses 29
funded the development of shore-based lifeboats designed to rescue crew and others from
shipwrecks. Lloyds also funded the deployment of these boats around the British Isles
and some other countries (Grey, 1922).
Fire insurance began to develop in England after the London fire of 1666. The first US
insurance company, formed in 1735, was a short-lived fire insurer. In 1752, Benjamin
Franklin founded the first successful fire insurance company in the US, the Philadelphia
Contributorship for the Insurance of Houses from Loss by Fire. Franklin campaigned to
remove fire hazards from houses and grounds, developed the lightning rod to prevent
fires. He designed the Franklin stove to provide more heat than an open fire at the same
time, reducing the risk of home fires (O’Neill, 1987).
Modern life insurance dates to the eighteenth century, but there is no history of loss
reduction activities by life insurers. Insurance against accidental injuries, which in some
sense was an extension of life insurance, began with coverage for injuries in railway
accidents but quickly extended to include all accidents. As with life insurance, these
policies were issued to individuals; insurers did not work to make railroad travel safer.
government activities and funding of injury prevention programmes/ research,

including motor vehicle crash injuries, are relatively insignificant when compared
toactivities and funds aimed at other major health problems
As this brief history indicates, insurance has taken many forms, and it still does. While
the general public may tend to think of this industry as a single entity, in reality it is not;
the many forms of insurance differ widely. Today the two principal categories of
insurance are ‘life’ and ‘property and casualty’. In some countries such as USA, there
also is general health insurance; this differs in many respects from traditional insurance in
that, in addition to spreading the risks of significant losses, a component also serves, in
effect, as prepayment for routine medical care.
The category of insurance that historically, and still today, has the most involvement in
loss control efforts affecting the risk of death and injury is property and casualty
insurance. Although this branch of insurance is primarily concerned with property losses,
as we shall see reducing such losses also can reduce deaths and injuries.
INSURANCE RISK AND LOSS REDUCTION
Insurance allows organizations and/or individuals to assume otherwise unacceptable

financial risks, which also may be related to injury risks. For example, the availability of
marine insurance in the fourteenth century resulted in a dramatic growth of commercial
shipping which, in turn, led to more shipwrecks because of greater exposure. This is one
of the anomalies associated with insurance; it makes risk-taking more affordable, so the
overall exposure to risk can be increased.
Recognition of this fact led insurers to focus beyond simply underwriting risks (i.e.,
charging an appropriate premium) to also finding ways to reduce risks. For example, it
was not in the best business interest of Lloyds to reduce its risks simply by charging so
much that it restricted the amount of shipping it insured. Instead its business could grow
if it identified ways to make marine insurance more affordable by making it ‘safer’ so
losses would be reduced.
Among the various types of property and casualty insurance, commercial property
insurance provides many examples of loss reduction activities. Insurers of commercial
buildings inspect boilers and other potential causes of loss and demand improvements
when deficiencies are identified. They also require sprinkler systems for fire suppression.
Although the focus of these kinds of activities is to reduce the risk of property damage to
buildings, such programmes also reduce risks to occupants. In California, insurers of
larger commercial properties work with building owners to ensure that buildings sustain
less damage in earthquakes; this can involve structural remodelling to increase
earthquake safety (Opferkuch and Frazier, 1987).
The fact that owners of larger commercial buildings can be pressured to make
structures more earthquake resistant surely has contributed to the relatively small loss of
life in recent US earthquakes, compared with those elsewhere involving the collapse of
large buildings and loss of life. I do not wish to claim that without the loss reduction
activities of US commercial insurers there would have been many more earthquake
deaths. Government regulations, building inspections, etc., play obvious roles, but the
contribution of insurers is not insignificant.
It is interesting to note that, in the case of individual homes and earthquake risks, the
potential role of insurers in California is much more limited because state law requires
insurers to offer earthquake coverage to all homeowners. Insurers cannot turn down poor
risks such as homes built on hillsides, poor soil areas, etc. This legislated availability of
earthquake insurance restricts the ability of home insurers to directly affect loss potential
by requiring improvements as a condition of insurance.
Commercial earthquake insurance is an example of how insurers can help to reduce
future losses by using the insurance pricing mechanism to directly influence risk. In
addition to demanding safer buildings as a condition of insurance, insurers also can offer
premium discounts for equipment that makes an insured home or product less risky, for
example, discounts on homeowner’s insurance if smoke detectors are installed or reduced
injury insurance premiums for cars equipped with airbags. Thus insurers can and do
directly influence policyholders with financial incentives to choose features or devices
that reduce their risks.
Much more limited is insurers’ ability to directly influence the behaviour of insured
individuals through pricing. For example, although auto insurance discounts for injuries
to motorists who promise to use safety belts all the time might seem attractive, this
approach is limited by problems verifying that the commitment is being kept. Motorists
in many European countries can earn sizable discounts on auto insurance if they file no
claims for several years; the thinking might be that this promotes ‘safer’ driving, but it is
likely that it encourages motorists not to file claims for crashes with relatively small
losses. It has been suggested that insurers could make premiums prohibitively expensive
for motorists with bad driving records, and although premium prices typically do reflect
driver history, insurers often are legally prevented from charging appropriately high
premiums for the poorest risks. In much of the US, special insurance pools with
subsidized premiums are mandated by, or in some cases run by, state authorities to ensure
that all licensed drivers have access to mandatory insurance coverage. The philosophy is
that if a motorist has a valid licence, affordable insurance should be available, regardless
of risk; otherwise many high-risk motorists would be likely to drive uninsured.
Direct loss reduction using insurance pricing and/or discounts can be limited by
government imposed regulations and complicated by the legal reparation system. The
system for homeowners’ fire losses is relatively straightforward—if there is an
unintentional loss the insured homeowner will be reimbursed—so insurers can use
pricing discounts to encourage homeowners to reduce fire risks by installing smoke
detectors without much complication. In contrast, automobile insurance typically
involves a more complicated reparation system. In most developed countries motorists
are required to purchase some form of liability insurance to provide reimbursement for
injuries and property damage they cause to others (third parties). Insurance to reimburse a
motorist for his or her own losses (first party) typically is not mandatory, although there
are a few exceptions. Some government insurance systems, Quebec, Canada’s, reimburse
insured motorists regardless of fault, and these are referred to as no-fault systems. To
further complicate matters, the automobile insurance systems in some US states are
called no-fault but, in reality, are only partial no-fault systems. In these jurisdictions, the
intent is for many less serious injuries sustained in two vehicle crashes (soft tissue
injuries) to be reimbursed by each motorist’s insurer without regard to fault. Liability
insurance, the most commonly mandated coverage, requires determination of who is ‘at
fault’ in two vehicle collisions. This, in turn, determines which of the two insurers
(presuming both vehicles are insured) will reimburse the ‘innocent’ victim. The ‘at fault’
motorist may or not be reimbursed for his or her losses, depending on what kind of
insurance was purchased.
there are both tangible and intangible benefits to insurers who embark on indirect
loss reduction activities
These potentially different reimbursement sources for motor vehicle crash injuries
influence insurers’ direct loss reduction possibilities. In USA, mandatory automobile
liability insurance for injuries typically is much more expensive than optional first party
injury insurance. This is partly because under the liability system, injured third parties
can recover not only their economic losses (medical bills, etc.) but also additional costs
for ‘pain and suffering’ associated with their injuries. In contrast, first party injury
insurance has maximum reimbursement limits that typically are only a few thousand US
dollars; there are no payments for ‘pain and suffering’. Thus, although US insurers offer
discounts for cars equipped with airbags, they only can do so on first party insurance
premiums, which are small in comparison with liability premiums. Another example is
that in most front-to-rear collisions the driver of the striking car is considered ‘at fault’
this means his or her insurer will pay for injuries such as whiplash that might occur in the
struck car (this would not be so under a few no-fault systems). If good head restraints in
the struck car prevent an injury, the insurer who benefits directly is the insurer of the
striking car, not the insurer of the car with the good safety equipment, so offering
discounts for good head restraints is not economically justified.
INDIRECT LOSS REDUCTION BY INSURERS
Property and casualty insurers have long recognized the benefits of using the pricing
mechanism to directly influence risk, but they also have a long tradition of indirect loss
reduction that is, influencing risk by methods other than pricing. Ben Franklin worked to
prevent fire hazards in eighteenth-century homes, and Lloyds funded lifeboats to rescue
people from shipwrecks.
These two examples provide an interesting contrast: to the extent that Franklin
succeeded in reducing home fire risks, his insurance company accrued direct benefits in
reduced claims for fires. In the case of the lifeboats, however, Lloyds insured property, so
it realized no direct benefits from the rescue of people from shipwrecks. It turns out that
the Lloyds’ investment in lifeboats occurred about the time the British public had been
shocked by a number of bad shipwrecks, including one involving an onshore crowd that
watched a crew drown without being able to offer aid. It appears that reason why Lloyds
funded lifeboats was a recognition that this would be good ‘public relations’. These
examples suggest there are both tangible and intangible benefits to insurers who embark
on indirect loss reduction activities.
Despite US fire insurance efforts, fire-related losses increased much more rapidly than
the population through the latter part of the nineteenth century and the beginning of the
twentieth. Major conflagrations were relatively rare, but fires, often related to newer
products and processes, occurred frequently. At the end of the nineteenth century, fire
insurers were increasingly dealing with electrical installations and products. This led to
the establishment by Chicago insurers of electrical product testing by an organization
called Underwriters Electrical Bureau, which became Underwriters Laboratories (UL) in
1894. Since then, UL has been involved in numerous evaluations of the safety of products
including gasoline pumps, oil burners, auto safety glass, lightning rods, life vests, hair
dryers (Bezane, 1994). Now UL laboratories worldwide examine how products are
constructed, test and evaluate electrical and other products, and develop product safety
standards. The most visible results are the ubiquitous UL certifications applied to more
than 14 billion products each year.
Another example of US insurer commitment to loss control is the work by the
Insurance Institute for Highway Safety (IIHS) and its affiliated organization, Highway
Loss Data Institute (HLDI). IIHS dates to the sixties, and its research founder was Dr.
William Haddon, Jr., a pioneer in scientific injury control. Their mission is to find out
and communicate what works and just as important, what does not work to reduce motor
vehicle crash losses. IIHS research focuses on potential counter-measures aimed at each
major factor involved in motor vehicle crashes—human, vehicle, and physical/legal
environment. Human factors research areas include graduated licensing systems for
young beginner drivers, alcohol-impaired driving, truck driver fatigue and safety belt use.
Vehicle factors research focuses on crash avoidance and crashworthiness. IIHS began
crash testing vehicles in 1969, evaluating the performance of bumpers in low-speed
impacts. In the seventies, IIHS conducted high-speed tests to illustrate the vulnerability of
fuel systems in rear impacts, the problem of cars under-riding trailers in rear-end crashes,
the effect of vehicle size in collisions between large and small cars and the importance of
airbags and safety belts in occupant protection.
This work expanded in 1992 with the Vehicle Research Center in central Virginia.
Frontal offset crash testing has become the focus of research at this world-class facility.
Initiated in 1995, offset tests evaluate the performance of popular new passenger vehicles
in 64.4 km/h frontal impacts into a deformable barrier to determine how well the vehicle
structure performs to protect its occupants in crashes. These tests are shown to large
audiences on national network television; the results are reported by print, broadcast, and
electronic media outlets. Such wide coverage is prompting automakers to improve the
crashworthiness of their passenger vehicles.
A brief list of other IIHS accomplishments:
• Research that provided the impetus to raise the legal minimum age for
purchasing alcohol to 21 in all states (DuMouchel et al., 1987; Williams, 1986;
Williams et al., 1983). This counter-measure alone has prevented thousands of
teenage deaths related to alcohol.
• Research that documented the effectiveness of laws providing for quick
administrative revocation of the licences of drivers who fail or refuse to take a
breath test for alcohol (Zador et al., 1989). The findings of this study were
released at a joint press conference with Mothers Against Drunk Drivers
(MADD), which adopted administrative licence suspension laws as its top
legislative priority. At the time of the press conference, only 23 US jurisdictions
had such laws; now 43 do.
• Research on the teenage driving problem has culminated in recent years
with the proliferation of graduated licensing laws, which are saving lives (see
Insurance Institute for Highway Safety and Traffic Injury Research Foundation,
1999; Mayhew et al., 1998; Ulmeret al., 1999; Williams, 1997, 1999).
• Research on the benefits of motorcycle helmet use laws isrepeatedly cited in
states that are considering either adoption or repeal of such laws (see Kraus et
al., 1976; Kraus et al., 1994; Kraus, Franti et al., 1975; Kraus, Riggins et al.,
1975; Lund, 1990; Lund et al., 1991; Muelleman et al., 1992; Sakar et al., 1995;
Watson et al., 1980, 1981).
insurance companies are financial institutions subject to government

regulations, and they are part of the legal reparation system. These and
other factors sometimes can limit loss control efforts
• A five-year programme to promote safety belt use in North Carolina

through well-publicized enforcement of the state’s belt law now is cited by the
US government as a model for increasing belt use (Foss et al., 1997). It is
because of these efforts that the US belt use rate now is about 70 per cent, up
from about 20 per cent in the mid-eighties.
• Research on urban crash problems, particularly at signalized intersections.
Researchers have found ways to time light sequences better to reduce crashes
(see Persaud et al., 1997; Retting, 1996; Retting and Greene, 1997; Retting et
al., 1995). The use of red light cameras in USA has been increased to
apprehend—better yet de ter-flagrant red light runners. IIHS research reveals 40
per cent reductions in red light running in two communities using such cameras.
Reductions occurred throughout the communities, not just at intersections where
cameras were located (see Retting et al., 1999a,b).
Besides IIHS, auto insurers in many countries have established research and testing
centres to reduce motor vehicle crash losses. Many of these centres belong to an
international coalition, Research Council for Automobile Repairs (RCAR), which exists
to reduce ‘insurance costs by improving motor vehicle damageability, repairability,
safety, and security’. Although the focus is on reducing property damage losses, most also
are involved in issues related to crash deaths and injuries.
A recent non-vehicle loss reduction initiative supported by US insurers is the Institute
for Business and Home Safety (IBHS), its mission to ‘reduce deaths, injuries, property
damage, economic losses, and human suffering caused by natural disasters’ Among its
early initiatives, IBHS has developed a programme to retrofit community child care
centres to protect against natural disasters. This organization also has designated
showcase communities and states to demonstrate what can and should be done to reduce
losses including deaths and injuries from hurricanes, tornadoes, floods, wildfires,
earthquakes, etc.
As these examples indicate, insurers take many opportunities to expand their role
beyond spreading risks and compensating people for losses. Insurers also work to reduce
injury losses both directly and indirectly. This is not only good business for insurers but
also a benefit to society; insurers’ financial interest in loss reduction directly parallels the
public’s interest in fewer deaths and injuries. Insurers’ efforts worldwide have resulted in
thousands of lives saved and injuries prevented. The world is undoubtedly safer and freer
from risks because of programmes begun and financed by private insurers. But at the
same time, insurance companies are financial institutions subject to government
regulations, and they are part of the legal reparation system. These and other factors
sometimes can limit loss control efforts.
REFERENCES
Bezane. N., 1994, The Inventive Century: The Incredible Journey of Underwriters
Laboratories, 1894–1994. (Northbrook, IL).
DuMouchel, W., Williams, A.F. and Zador, P.L., 1987, Raising the alcohol purchase age:
its effects on fatal motor vehicle crashes in twenty-six states. Journal of Legal Studies,
16, pp. 249–66.
Foss, R.D.; Beirness, D.J.; Wells, J.K. and Williams, A.F., 1997, Effect of an intensive
sobriety checkpoint program on drinking-driving in North Carolina. Proceedings of the
14thInternational Conference on Alcohol, Drugs, and Traffic Safety (ed. C.Mercier-
Guyon), 2, pp. 943–48. (Annecy, France: Centre d’Etudes et de Recherches en
Médecine du Trafic, CERMT).
Grey, H.M., 1922, Lloyd’s: Yesterday and Today. (London: Syren).
Institute of Medicine, 1999, Reducing the Burden of Injury: Advancing Prevention and
Treatment. (Washington, DC: National Academy Press).
Insurance Institute for Highway Safety and Traffic Injury Research Foundation. 1999.
Graduated Licensing: A Blueprint for North America. (Arlington, VA: Insurance
Institute for Highway Safety).
Kraus, J.F., Franti, C.E., Johnson, S.L. and Riggins, R.S., 1975, Risk factors in
motorcycle collision injuries. Proceeding of the 19th Annual Conference of the
American Association for Automotve Medicine. American Association for Automotive
Medicine, Lake Bluff, IL. pp. 383–398
Kraus, J.F., Franti, C.E., Johnson, S.L., and Riggins, R.S., 1976, Trends in deaths due to
motorcycle crashes and risk factors in injury collisions. Accident Analysis and
Prevention, 8, pp. 247–255.
Kraus, J.F., Peek, C.; McArthur, D.L and Williams, A.F., 1994, The effect of the 1992
California motorcycle use helmet law on motorcycle crash fatalities and injuries.
Journal of the American Medical Association, 272, pp. 1506–1511.
Kraus, J.F., Riggins, R.S. and Franti, C.E., 1975, Some epidemiologic features of
motorcycle collision injuries. American Journal of Epidemiology, 102, pp. 74–109.
Lund, A.K., 1990. Helmet Use Laws: They Work. (Arlington, VA: Insurance Institute for
Highway Safety).
Lund, A.K. Williams, A.F. and Womack, K.N., 1991, Motorcycle helmet use in Texas.
Public Health Reports, 106, pp. 576–578.
Mayhew, D.R., Simpson, H.M.; Williams, A.F. and Ferguson, S.A. 1998, Effectiveness
and role of driver education and training in a graduated licensing system. Journal of
Public Health Policy, 19, pp. 51–67.
Muelleman, R.L., Mlinek, E.J. and Collicott, P.E., 1992, Motorcycle crash injuries and
costs: effect of a re-enacted comprehensive helmet use law. Annals of Emergency
Medicine, 21, pp. 266–272.
O’Neill, B., 1987, The Insurance Institute for Highway Safety: research arm of the U.S.
insurers and servant of the public. Transport Reviews, 7, pp. 83–94.
Opferkuch, R.C. and Frazier, D.O. 1987, Managing earthquake exposure. Risk
Management, (August), pp. 16–22.
Persaud, B., Hauer, E.J., Retting, R.A., Vallurupalli, R. and Mucsi, K., 1997, Crash
reductions related to traffic signal removal in Philadelphia. Accident Analysis and
Retting, R.A., 1996, Urban motor vehicle crashes and potential countermeasures.
Transportation Quarterly, 50, pp. 19–31.
Retting, R.A. and Greene, M.A., 1997, The influence of traffic signal timing on red-light
running and potential vehicle conflicts at urban intersections. Transportation Research
Record 1595, pp. 1–7. Washington, DC: Transportation Research Board.
Retting, R.A., Williams, A.F., Farmer, C.M. and Feldman, A.F. 1999a, Evaluation of red
light camera enforcement in Fairfax, Virginia. ITE Journal, 69, pp. 30–34.
Retting, R.A., Williams, A.F., Farmer, C.M. and Feldman, A.F., 1999b, Evaluation of red
light camera enforcement in Oxnard, California. Accident Analysis and Prevention, 31,
pp. 169–174.
Retting, R.A., Williams, A.F., Preusser, D.F., and Weinstein, H.B., 1995, Classifying
urban crashes for countermeasure development. Accident Analysis and Prevention, 27,
pp. 283–94.
Sakar, S., Peek, C. and Kraus, J.F., 1995, Fatal injuries in motorcycle riders according to
helmet use. The Journal of Trauma, 38, pp. 242–245.
Ulmer, R.G., Preusser, D.F., Williams, A.F., Ferguson, S.A. and Farmer, C.M., 1999,
Effect of Florida’s graduated licensing program on the crash rate of teenage drivers.
Accident Analysis and Prevention, in press.
Watson, G.S., Zador, P.L. and Wilks, A., 1980, The repeal of helmet use laws and
increased motorcyclist mortality in the United States, 1975–1978. American Journal of
Watson, G.S., Zador, P.L. and Wilks, A., 1981, Helmet use, helmet use laws, and
motorcyclist fatalities. American Journal of Public Health, 71, pp. 297–300.
Williams, A.F., 1986, Raising the legal purchase age in the United States: its effects on
fatal motor vehicle crashes. Alcohol, Drugs, and Driving, 2, pp. 1–12.
Williams, A.F., 1997, Earning a driver’s license. Public Health Reports, 112, pp. 453–
461.
Williams, A.F., 1999, Graduated licensing comes to the United States. Injury Prevention
5, pp. 133–135.
Williams, A.F., Zador, P.L., Harris, S.S. and Karpf, R.S., 1983, The effect of raising the
legal minimum drinking age on involvement in fatal crashes. Journal of Legal Studies,
12, pp. 169–179.
Zador, P.L., Lund, A.K., Fields, M. and Weinberg, K., 1989, Fatal crash involvement and
laws against alcohol-impaired driving. Journal of Public Health Policy, 10, pp 467–
485.
5
Methodological Considerations in the Analysis
of Injury Data: A Challenge for the Injury
Research Community
Shrikant I.Bangdiwala
INTRODUCTION
In most parts of the world, injuries are the leading cause of death and disability among
young adults, adolescents and children. As infectious and degenerative diseases are better
controlled, the importance of injuries in mortality and morbidity statistics has increased
in developing countries. The study of the determinants of injuries is receiving increasing
attention in most countries as the gravity of the problem is being faced.
The injury prevention and control research field can be considered to be a relatively
new area of research in public health, when compared to the study of chronic diseases
such as cardiovascular disease, cancer or infectious diseases such as tuberculosis,
malaria. Researchers in every field have had to struggle with many methodological
aspects in studying the particular disease, its aetiology and the impact of interventions.
Particular interests are the quantification of the burden of the disease, identification of the
causes of the disease, and evaluation of the impact of interventions to prevent and/or
control the disease.
It is important that the specific characteristics of the disease process be taken into
consideration when considering the appropriate methodology to use to address an issue in
any field of research. The injury epidemiology field has progressed considerably in the
last decade, by primarily ‘adopting increasingly sophisticated research methods from
more established branches of epidemiology’ (Cummings et al., 1995), but still has a
considerable way to go. It is time that the injury research community begins to address
particular needs of injury data by developing appropriate methodology that takes into
consideration the unique nature of the data. The study of diseases that are different in
nature requires different paradigms.
The purpose of this paper is to identify some of the unique characteristics of injury data
that must be addressed, to review some of the ways existing recent methodology has been
applied often in novel ways in the injury field, and to challenge researchers to tackle
unresolved problems. Only by doing so will the field of injury research mature
methodologically.
UNIQUE ASPECTS OF INJURY DATA
Every field of public health has had to tackle issues of how best to describe the
epidemiology of the disease, how to study the aetiology of the disease, and how to
understand ways to prevent and/or control the disease. These are common issues shared
across fields and many methodologies developed in one field are applicable in the study
of others. However, there are unique aspects of each disease that challenge researchers to
develop appropriately tailored methodologies if one wishes to address specific needs of
the research.
the injury epidemiology field has progressed considerably in the last decade… It is
time that the injury research community begins to address particular needs of
injury data by developing appropriate methodology that takes into consideration
the unique nature of the data
Injuries are characterized by the unique feature that they occur in the context of a sudden
and usually immediately apparent transfer of energy. This concept of energy transfer was
not well understood until proposed by Haddon (1972) as a conceptual framework for
motor vehicle injuries. This transfer of energy occurs during an event. Haddon divided
the factors related to injuries into three phases: pre-event, event, and post-event. Thus,
pre-event factors determine the occurrence of the event, event factors determine the
occurrence and/or severity of the injury, and post-event factors determine the severity
and/or outcome of the injury. Haddon created a nine-cell cross-classification matrix by
adding another dimension that considers where the factors are operating, whether in the
person, the injury producing vector, or in the environment. This matrix as a conceptual
model has proven to be a successful tool for analysing injury producing events and
recognizing factors important to their prevention. Recently, Runyan (1998) proposed a
third dimension that considers its application in decision making. Thus, one of the unique
features of studying the aetiology of injuries is the need to account for pre-event factors
as well as event factors.
Second, like infectious diseases and unlike chronic degenerative diseases, injuries are
characterized by the possibility of recurrence. Thus a given individual is at risk of re-
injury if prior events are non-fatal and also not substantially disabling so as to remove the
individual from exposures. A person can be involved in multiple motor vehicle crashes or
suffer from repeated falls. Unlike many infectious diseases, like influenza, where
immunity is not developed for all strains and one can argue that repeated events are
stochastically independent, these repeated injury events are clearly not independent, as
the likelihood for further events is related to the prior events and the injury outcome
suffered. Once injured, an individual is likely to alter their behaviour and thus the risk for
a second injury is no longer the same as it was for the first time. Thus, non-fatal injury
data analyses need to address the repeated correlated nature of the data.
A more complicated characteristic of injury data is the likelihood of multiplicity on
many fronts. Most diseases are affected by multiple factors and can be manifested in
A challenge for the injury research community 39
several ways, but this is more complex in injuries due to multiple factors operating in the
different cells of the Haddon matrix. An individual can be exposed to multiple competing
risks just like for any other disease, but the term ‘injury’ encompasses a large variety of
different times of trauma, from burns to fractures to asphyxiation, with a multiplicity of
factors to consider for events as well as for injuries from these events. In addition, some
injuries are manifested on multiple body sites, such as burns or fractures. Furthermore,
the severity of the injuries may vary by site and type of trauma. A single event such as a
motor vehicle crash can result in multiple types of injuries to multiple body sites with
different severities. If we then add that in one event there can be multiple individuals
involved, the complexity of the nature of the data is quite apparent. These complexities
are important in deciding the unit of analysis—event, individual or vector.
For many chronic diseases, the exposure period is prolonged and is usually the entire
lifespan of the individual. In most diseases, the individual is assumed to be exposed at all
times, and that exposure levels may vary as one modifies one’s behaviour. In injuries,
exposure to a particular injury is dependent on whether one is involved in a particular
activity that is conducive to an event in which the energy transfer can occur. Thus one is
not exposed to a motor vehicle injury if one is not near a motor vehicle and not exposed
to drowning if away for water sources. This more immediate or acute relationship of
exposure with outcome poses quantification problems that affects the design of studies, in
particular case-control studies where exposure is measured retrospectively.
A major feature of injury data is the role that knowledge, attitudes and behaviours play
in the aetiology of the disease. Clearly these considerations are really not unique to the
injury field, as they also affect the epidemiology of other diseases. However, the role of
these factors is often minimized in relation to more physiological factors when studying
other diseases. Many of the various causes of chronic and infectious diseases are
physical, chemical or biological factors that are distinct, identifiable and measurable,
while for injuries, the more difficult to measure and operationalize factors of knowledge,
beliefs and behaviour are more common (Petridou, 1997). The difficulty in assessing and
quantifying these factors may be one reason that they have traditionally not been
extensively incorporated in epidemiologic studies of other diseases, although recently
there has been a trend for considering these more subjective factors. Given that exposure
risks for injuries are more immediately dependent on the choice of engaging in a ‘risky’
behaviour, the more qualitative factors of knowledge, attitude and belief should be
incorporated in analyses of the epidemiology of injuries alongside the more traditional
demographic, physiologic, and environmental quantitative factors.
When considering the evaluation of the effectiveness of interventions to prevent or
control injuries, the experimental designs available for injury research are ethically
limited. For example, a randomized controlled trial (RCT), the accepted most rigorous
experimental design to test the effectiveness of an intervention, requires that individuals
be randomized to the interventions under consideration. It is not possible to use this
methodology to test the effectiveness of vector or environmental factors since these do
not operate at the individual level. It is often unethical or impractical to use an RCT to
test the effectiveness of an individual level intervention, as in the case of protective
devices such as seat belts in motor vehicles, or helmets or pads in sports activities. It is
unethical to randomize individuals to undertake a ‘risky’ behaviour, some with and some
without a protective device The experimental designs available are thus less rigorous,
using quasi-experimental designs such as community intervention trials or before-after
intervention studies.
Another unique feature of injury data is less a characteristic of the nature of the data,
but more so of the methodology for obtaining data, and can be thought as due to the
‘identity crisis’ of injuries. Injury prevention and control as a research discipline is a very
recent term, previously referred to as accidents. The term ‘accidents’ is still prevalent in
many countries, especially due to the difficulty in translating the term ‘injury’ to have a
similar connotation in other languages. The ‘random act of God’ implication of the term
‘accident’ relegated the documentation of injuries to legal and clinical requirements, not
to research needs for understanding how to prevent and/or control their occurrence. So,
while other diseases were viewed as preventable, thus information on relevant factors
was collected, injuries were not as well documented and relevant factors on the
circumstances that caused the injury were not routinely collected or considered relevant.
This is a battle that all injury researchers face in every country. This feature affects the
availability and quality of data and the feasibility of studies that rely on the use of
existing data sources. Another major implication of this lack of completeness in injury
data is the presence of missing data both cross-sectionally and longitudinally.
CURRENT STATE OF THE LITERATURE OF INJURY ANALYSIS
These unique features of injury data determine the possible methodologies that can be
used. The complexities of the data are daunting and for many years, injury researchers
focused on standard study designs well known in epidemiology and in standard
descriptive and inferential statistical methods for analyses well known in many fields of
application. The exception to this generalization is the field of motor vehicle injury
research, which has a relatively longer existence, and is therefore methodologically more
mature.
most articles on injury topics were mainly descriptive and focused on primarily
bivariate relationships with a few multivariate models
A review of publications in epidemiology and public health journals for the last ten years
showed that most articles on injury topics were mainly descriptive and focused on
primarily bivariate relationships with a few multivariate models. Most of the complexities
of the data arising from multiple dimensions, repeated measures, multiplicity of
factors/sites/severities, and missing data, are often ignored or the problem simplified to
satisfy requirements of less sophisticated methods, with the resulting loss of specificity
and loss of information.
The major speciality journals in the injury field are few, with Accident Analysis &
Prevention the oldest one (currently in volume 31) and Injury Prevention the relatively
new kid on the block (currently in volume 5). A review of the last 12 years of Accident
Analysis & Prevention shows the use of a variety of statistical analytic techniques, from
the relatively simple analysis of variance, multiple linear regression and logistic
regression, to complex multinomial logit models, empirical Bayes methods, and various
time-series models. More recently techniques developed in the biomedical fields for
analysing clinical trial data such as meta-analysis and survival analysis have been used.
Injury Prevention in its publications contains a section in each issue entitled
‘Methodological Issues’ which varies from short reviews of well-known statistical or
epidemiologic procedures such as logistic regression and odds ratios (Platt, 1997) to more
complex issues such as the methodological issues in case-control studies (Roberts, 1995).
State-of-the-art methodologies applied to the study of injuries

There are many recent articles that have taken methods developed for other disciplines
and applied them to the injury field. To the extent that these methodologies adequately
account for the unique characteristics of injury data, they are appropriate and their use
should be encouraged. However, many require an adaptation of the data to the method
that may not be appropriate, and these methods must be used with caution.
This section briefly illustrates some of the novel applications of methods developed in
other fields. These are grouped into methods for quantifying the burden of injury, for
identifying causal factors of injury, and for evaluating the effectiveness of interventions.
Quantification of the burden of injury

Every field of public health has had to struggle with how to quantify the burden of their
disease, with problems of definition, classification or typology, and case ascertainment.
In addition, many fields struggle with issues of long-term effects (disability) of the
disease, and how best to quantify the financial burden represented by those with the
disease. The injury field is no different in having to face many of these same issues.
Cummings et al. (1995) identified many aspects in what they called the ‘numerator
problem’ or the ability to properly count injuries. These include definitional problems,
problems of classification, and of ascertainment. The problem of clear and unambiguous
case definitions is not unique to injuries. However, the need to consider near-cases of
injury such as near-drowning or near-fall is different from other diseases where these
considerations are not relevant.
In quantifying the burden of injuries, the proper denominator for estimating rates is
often not clear (Cummings et al., 1995). In calculating an area-specific rate, the
numerator and denominator must properly account for residency and transient status in
order to avoid erroneous estimations. In motor-vehicle related injuries, the choice of
denominator—population, registered vehicles or vehicle-miles—affects the resulting
interpretation of the magnitude of the problem (Bangdiwala et al., 1985). This is related
to the proper measure of exposure risk.
Many of the measurement problems in injury stem from the fact that the collection of
data is dependent on existing data sources or registries or that primary data collection is
either collected retrospectively from individuals, with the usual limitations of such data
such as recall bias, or collected soon after the occurrence of the injury. The latter is
problematic as the acute care of the injured individual takes precedent over data
collection, the researcher is not present at the moment of data collection, and that clinical
and legal issues may inhibit the data that is obtainable.
Given the practical difficulty of primary data collection at the time of the injury, a very
common approach for data collection is to seek information at places where the injured
do tend to go to seek care or compensation, emergency rooms, hospitals, private
physician offices, police stations, insurance companies, rehabilitation centres. The
advantages and disadvantages of these sources of subjects are the subject of discussions
among researchers, with problems of undercount, access, quality of data, and
completeness of information. An important effort in this regard is the International
Collaborative Effort on Injury Statistics, which since 1994 has been directly addressing
the priorities and issues in injury statistics from an international perspective, including
international comparability issues on definitions, development and linkages of databases
and classification systems.
many of the measurement problems in injury stem from the fact that the collection
of data is dependent on existing data sources or registries, or that primary data
collection is either collected retrospectively from individuals, with the usual
limitations
In the injury field, a common approach is the use of surveillance systems. Surveillance,
according to Buehler (1998), is ‘the continuous and systematic process of collection,
analysis, interpretation, and dissemination of descriptive information for monitoring
health problems’. Modern concepts of surveillance were shaped by programmes to
combat infectious diseases, with the reporting of ‘notifiable’ diseases. However, the
injury research field has recognized its utility and has adopted its methodology to monitor
the occurrence of injury over time within specific populations.
The specific purposes of injury surveillance systems include:
• to provide descriptive epidemiology of injuries and possibly provide insight

into aetiology
• to monitor trends, search for outbreaks or clusters in time and geography
• to provide indication of the success of interventions
• to permit identification of groups at highest risk
• to guide prevention activities to problem areas and anticipate future trends.
Since surveillance systems are usually dependent on someone besides the
injury researcher as the primary data collector, they suffer from many potential
limitations:
• Continuity, since they are not necessarily permanent and depend on the
availability of continued funding
• Purpose, which usually is not research
• Completeness of information, since they are either retrospective or
concurrent with the care of the injured
• Limited detail, which limits the ability to investigate more in-depth
epidemiologic hypotheses
• Inadequate quality of the data, since usually out of the control of researchers
• Lack of relevance, since they seldom collect data useful for prevention, or
for fully understanding the circumstances of the injury or its severity.
This last point is illustrated by the recent trend in the injury research community to
stimulate the inclusion of narrative data in surveillance records, since the rich information
it contains is necessary for considering prevention; it is seldom collected, and if collected,
not analysed.
Surveillance systems usually include only those with the disease, the injured, or the
cases. They do not often provide a comparison group of individuals without the disease,
the non-injured, thus limiting the ability of researchers to investigate aetiology.
Surveillance systems as a source for useful data for research and public health policy
have been studied and evaluated for specific types of injuries or sub-populations (Lyons
et al., 1995), for the sources of the data (Garrison et al., 1994, Runyan et al., 1992), and
for their quality (Macarthur and Pless, 1999). They have many limitations including
inadequate ascertainment and coverage of cases, incompleteness of data, poor quality of
the data, and also that they often are designed as large investments of resources, are
labour intensive, and often they are not evaluated. One of the reasons for the large costs is
the effort to identify every case. Only recently has the wellknown statistical technique of
sampling been considered in injury surveillance systems. Another promising technique,
the use of capture-recapture statistical methods, has only since 1993 been applied in the
injury field (Chiu et al., 1993). It has been evaluated for ascertaining adolescent injuries
in a school district of Pittsburgh, Pennsylvania (LaPorte et al., 1995), as well as for
estimating deaths and injuries due to road traffic accidents in Karachi, Pakistan (Razzak
and Luby, 1998). These methods, originally developed in wildlife and fisheries to
estimate sizes of moving populations, utilize the information in the overlap of multiple
sources of cases to estimate the number of missed cases and thus adjust the incidence
rates appropriately. The benefit is that incidence rates can be calculated efficiently
without every case being identified. ‘Capture-recapture analysis not only provided an
approach to evaluate and adjust for undercount but also offered a formal means to
evaluate the most efficient combination of the sources to maximize completeness while
minimizing effort. The use of these techniques has the potential to evaluate and improve
injury surveillance…’ (LaPorte et al., 1995). However, many practical and theoretical
issues are yet to be adequately addressed.
Identification of the causes of injury

People who care directly for the injured and policy makers responsible for efforts to
prevent and control injuries, are the individuals more interested in quantifying the burden
of injuries. Researchers, on the other hand, are more involved in studying the causes of
injuries in order to develop and evaluate counter-measures. This may explain why more
sophisticated methodologies are applied to identifying the aetiology of injuries than to
other areas of study.
In studying the relationship between potential explanatory or causal factors of injuries,
the field draws from the existing designs in other fields of epidemiologic research. There
are different types of study designs, each providing information that is relevant to the
study of a particular disease and the factors associated with that disease. Studies can be
classified into observational or experimental, based on whether the investigator only
records data or intervenes in some fashion with the subjects.
Observational studies include case series, cross-sectional or prevalence studies,
retrospective studies such as case-control, and prospective studies such as cohort,
incidence, and surveillance. Experimental studies include the so-called ‘natural
experiments’, before-after studies, community and cluster trials, and randomized
controlled trials.
Case series are descriptions of exposure characteristics in injured individuals, and are
useful for detailed in-depth initial investigations of potential factors in injured subjects.
Recently, Cummings and Weiss (1998) argue for the utility of case series and exposure
series (descriptions of outcomes in exposed individuals) in providing information about
the aetiology of injury when viewed as ‘primitive forms’ of case-control and cohort
studies, respectively. Cross-sectional studies describe the burden of the disease in a given
area and time period, provide prevalence rates, and permit the study of relationships
between exposure factors and outcome measures that are evaluated simultaneously. They
are thus limited in their ability to study causal relationships.
Case-control studies compare the occurrence of exposure factors in a sample of
subjects with the disease or condition (cases) and the occurrence of exposure factors in a
sample of subjects without the disease or condition (controls). They thus are able to
quantify the strength of the relationship between the exposure factors and the outcome
measures, estimate percentages of exposure in cases and controls, but given their
retrospective design, also are not able to study causality. Nevertheless, they are very
strong methodologically and widely used.
an exciting new methodology for a research design is the case-crossover for

quantifying the importance of transient effects on the risk of acute events
Roberts (1995) addresses several methodological issues in case-control studies, from

defining the study base from which cases and controls are to be drawn, to the importance
of errors in the exposure measurement, and the potential for bias from combining
unrelated injury outcomes. For most disease processes, incidence rates are expressed as
the number of cases over a measure of person-time of exposure, which is person time
‘living’. For injuries, since these only can occur or be prevented when the person is
engaged in a relevant activity, the denominator is more appropriately ‘person time
exposed in a particular activity’. Marshall et al. (1998) illustrate potential biases from
using inappropriate controls in bicycle helmet studies, while Cummings et al. (1998)
study biases of selecting controls in an emergency department setting.
An exciting new methodology for a research design is the case-crossover for
quantifying the importance of transient effects on the risk of acute events (Maclure,
1991). The methodology was developed for cardiovascular diseases but has been applied
to child pedestrian injuries (Roberts et al., 1995), as it addresses the acute nature of the
injury event.
Cohort studies measure the occurrence over time of a condition or an outcome measure
in a sample of individuals initially free of the condition. They thus shed light on the
natural or clinical evolution of the disease, estimate incidence rates, and permit the study
of causal relationships since temporally exposure precedes outcome. Cohort studies of
injuries are not much different in methodological issues from cohort studies of other
diseases. Issues of incomplete follow-up, missing data, duration and cost are shared.
The advantage of longitudinal data comes with the potential disadvantages of
incomplete follow-up thus missing information. Methodologically, they also are more
complex as one usually has repeated measurements in an individual over time, and the
correlation among these measurements must be accounted for in the analysis.
More often than not, these issues from multiplicity are simply ignored, which tends to
overestimate significance as it reduces the true variability of estimates. Another common
approach is to simplify the problem, by analysing the data using cross-sectional looks of
the data, looking at changes between two time points, and recently, by looking at simple
survival analysis of the time to the occurrence of the first outcome.
There are appropriate methodologies that do exploit the richness of the information in
longitudinal data, such as
• Time series analyses

• Survival analysis with time varying co-variates
• Repeated measures analytic methods:
• Incidence density
• Mixed random and fixed effects linear models
• Latent class and hierarchical models
• Generalized estimating equation models (GEE)
Williamson et al. (1996) illustrated various methods for analysing repeated measures data
when the outcome is binary (injured or not). They applied GEE models and survey
sample methodology to a cohort study of injuries in rugby players. This application did
provide interesting insight into relationships between exposure factors and the likelihood
of injuries that are not possible if the repeated measures design of the longitudinal data is
ignored or simplified.
A major problem faced by injury studies, which rely on existing data collection
systems or are by design longitudinal, is the presence of missing data. The analysis of
incomplete multivariate data has traditionally involved such drastic measures as case-
deletion or completing the missing information using some imputation technique. The
reason was to force the incomplete data set into a rectangular complete data format and
thus be able to use standard analytic techniques. Even if not done consciously, many
statistical software packages automatically delete observations with incomplete data in
multivariate procedures such as linear regression. Imputation can be a simple replacement
of the missing value with some reasonable observed mean of non-missing values or a
predicted value from a regression model. When the missing data is a small fraction
relative to the rest of the data, case deletion may be adequate. Ad hoc imputation
techniques do not account for the fact that there is uncertainty in the imputed value.
Recently, Schafer (1997) proposed iterative algorithims for simulating multiple
imputations of missing values in incomplete data sets, to account for the uncertainty in
the imputation.
Evaluation of the impact of interventions to prevent and/or control injuries

Observational studies are useful in providing information on strength of relationships
among exposure factors and an outcome, but an experiment is necessary for evaluating
the efficacy of an intervention in a controlled situation. There are many different types of
experimental designs, which vary in the degree of control by the investigator over the
factors that potentially affect the outcome. They all must comply with ethical
considerations of intervening on human volunteers that have consented to participate in
such studies. An experiment is the definite design that establishes the causal relationship
of an exposure factor (the intervention) with the disease outcome.
The different types of experimental designs are
• Natural experiments where the factors occur differently in different communities, there
is no control and should not really be considered as an experiment.
• Before-after studies, where individuals are their own comparison group, there is little
control of factors and there is no randomization.
• Community and cluster randomized trials, where entire communities or clusters (groups
of individuals such as church groups, neighbourhoods, factories, schools) are assigned
randomly to interventions. The level of control of factors is at the group level, while
the outcome measure is usually at the individual level.
• Randomized con trolled trials where individuals are randomized to interventions,
offering the maximum control over factors that potentially can affect the outcome
measure.
Natural experiments arise from changes in a factor that is not controlled by the
investigator, as in a new law, counter-measure for preventing injuries from motor
vehicles or a comparison of two countries or communities with different policies. They
are sometimes confused with before-after studies, that arise when some treatment has
been deliberately implemented on a group or community and the task is to estimate its
safety effect. Although randomization is not involved, before-after studies have a
methodological advantage, which is that they have reduced variability since subjects are
their own comparison group. Before-after studies are also very common in the injury
field, as the political practice of evaluating policies is such a design. It is especially
common in the field of road safety (Hauer, 1997). Many complex time series models are
utilized in the analysis of trends in before-after studies.
thus randomized community trials or randomized controlled trials are the most
sound study designs for evaluating the efficacy of an intervention
The process of random assignment to intervention is considered as the most rigorous

procedure methodologically thus randomized community trials or randomized controlled
trials are the most sound study designs for evaluating the efficacy of an intervention.
They are both prospective studies comparing the effect and value of one of more
interventions against a control in human subjects with a given risk situation. They
measure causality in terms of the effect of an intervention: if one alters the risk factor,
does one alter the occurrence of the event/injury?
Trials allow comparison of alternative treatments in a controlled experimental setting,
to the extent that this is possible. The idea is to control the possible factors that may
affect the outcome, vary one (the intervention) and evaluate the outcome. This ideal
experimental setting is rarely achieved, since there are a multiplicity of factors that are
difficult to control. Randomization at least will make comparable the distributions of
these factors in the different groups, provided the study involves a large number of units
of randomization. However, despite such statistical efforts, controlling human behaviour
factors is often difficult and this is a primary concern in injury studies.
Cluster or group randomization occurs when we randomly allocate groups or
communities to interventions and we study outcomes in a sample of its members. This is
often called a quasi-experimental design. They arise from practical, ethical, and financial
considerations. Such studies are very commonly used in injury research. For example, it
may be difficult to randomly assign some individuals in a community to an intervention
and others to not receive that intervention, especially when the intervention is an
environmental countermeasure for road safety.
The analysis of community trials is somewhat more complex, since members within a
cluster are statistically correlated, so that the needed assumptions of statistical
independence of the data are not met. The effect of this correlation is to inflate the
variance of the estimates of outcome rates. There are many alternatives for handling such
correlated data. A common practice is to ignore the problem, thus overestimating the
significance of the results. Another common approach is to avoid the problem by
reducing the information to the cluster level, with the resulting loss of efficiency. The
best alternative is to confront the problem using standard statistical methods like the
variance inflation factor based on the intra-class correlation coefficient or using
appropriate regression models for clustered data (Graubard and Korn, 1994).
Challenge for the injury community

Like most disease processes, injuries are caused by a multivariate set of variables.
However, unlike the causes of cardiovascular or neoplastic diseases or of infectious
diseases, where the event and the disease are the same, the risk factors for the occurrence
of an event are not the same as those for the occurrence of an injury. This is exacerbated
by the common research practice of considering broad categories of injuries as a single
entity. However, since aetiologic mechanisms usually differ among the various types of
injuries, causal associations between potential risk factors and injuries may be obscured.
The measurement of exposure is problematic. Promising developments focus on more
reliable exposure ascertainment (Towner et al., 1994), or methods of quasi-induced
exposure (Stamatiadis and Deacon, 1997; Hakamies-Blomqvist, 1998). The road safety
field has been more active methodologically and many multivariate models have been
proposed or adapted, as in using empirical Bayes procedures to predict accidents on main
road with minor junctions (Mountain et al., 1996), and survival analysis to predict the
probability of being involved in an accident given a vehicle’s non-accident status
(Jovanis and Chang, 1989). The non-road safety field has used methodology from other
fields, as in Poisson regression and time-series analysis for examining trends over time in
rates of child injury (Kuhn et al., 1994). However, many challenges remain, as the unique
aspects of the field of injury need to be addressed. An important point of merging
qualitative data with quantitative data has been promisingly discussed by Oppe (1992).
Despite a considerable body of research on interventions to prevent and/or control
certain types of injuries, questions remain about the effectiveness of various intervention
strategies. Often similar studies are replicated in different settings for different sub-
populations. This effort is expensive, and often such studies are not completely
comparable due to different circumstances and changes over time, culture and other
factors. Many small studies are conducted due to limited financial resources, and the
results may even be different. It is often impractical to conduct one large study. The
recent methodological approach of systematic literature reviews and meta-analysis are
appropriate tools for synthesizing the existing evidence form evaluation studies and to
provide a basis for rational decision making. Such aggregate analyses benefit from the
increased statistical power provided by the large combined sample size, and may often
shed new light into the effectiveness of an intervention strategy. Sometimes they may
show that interventions believed to be ineffective may be quite effective or vice versa. By
demonstrating effectiveness with existing evidence, they can provide substantial savings
by avoiding future unnecessary studies. Recently, an entire supplemental issue of the
American Journal of Preventive Medicine illustrated the use of such methodology to
evaluate a variety of strategies to prevent motor vehicle injuries such as the proper use of
child occupant restraints for children, driver education programmes and alcohol laws and
regulations (Rivara et al., 1999).
As we direct ourselves to the new millennium, we must consider advancing the field of
injury prevention and injury control methodologically. In order to do this, as computer
intensive methods are becoming the norm in cutting-edge statistical methods, we must
familiarize ourselves with new techniques, such as:
systematic literature reviews and meta-analysis are appropriate tools for

synthesizing the existing evidence
• Simulated imputation for addressing missing data

• Generalized estimating equations for multiplicity arising from longitudinal
data
• Classification systems based on recursive partitioning
• Geographical positioning systems for spatial distribution.
There are also some well established statistical concepts that must be
considered in the field of injury research:
• Combining qualitative with quantitative
• Combining event factors with injury factors using conditional probability
• Clustering for dealing with multiplicities
• Sampling techniques
Petridou (1997) clearly argued that epidemiology, as the basic science for public health,
is the major ‘resource for identifying the underlying causes of injury, and eventually, for
controlling them’. Injury epidemiology is a young discipline and as such it is
understandable that it borrows methodology from more established fields. However, it is
imperative that it continues to explore alternatives, refinements and development of new
methods that address its needs more adequately. ‘What is now needed is to develop
epidemiologic methods that can effectively address the problems generated by the
peculiar circumstances that increase the long term likelihood of an accident or, just as
important, trigger it’ (Petridou 1997).
Methodology for analysing data should be appropriate to the needs of the subject
matter. Thus one should use appropriate methods that fit the processes in which the data
were generated (sampling), the nature of the data (continuous or discrete distribution) and
the frequency measured (repeatedly or not). If appropriate methods do not exist, these
need to be developed by competent researchers.
The potential for injury epidemiology is unrealized at present. The injury research
community—working closely with methodologists that fully comprehend the specific
issues in the nature of injury data—is challenged to continue tackling the important
methodological issues in order to advance the field and to be able to deal more efficiently
with limited resources with this grave problem affecting the entire world
REFERENCES
Bangdiwala, S.I., Anzola-Pérez, E. and Glizer, M., 1985, Statistical considerations for the
interpretation of commonly utilized road traffic accident indicators: Implications for
developing countries. Journal of Accident Analysis & Prevention, 17, pp. 419–427.
Buehler, J.W., 1998, Surveillance. In Modern Epidemiology, edited by Rothman, K.J.
Greenland, S., (Philadelphia: Lippincott-Raven), pp. 435–457.
Chiu, W., Dearwater, S.R., McCarty, D.J., et al., 1993, Establishment of accurate
incidence rates for head and spinal cord injuries in developing and developed
countries: A capture-recapture approach. Journal of Trauma, 35, pp. 206–211.
Cummings, P., Koepsell, T.D. and Mueller, B.A., 1995, Methodological challenges in
injury epidemiology and injury prevention research. Annual Review of Public Health,
16, pp. 381–400.
Cummings, P., Koepsell, T.D. and Weiss, M.S., 1998, Studying injuries with case-control
methods in the emergency department, Annals of Emergency Medicine. 31, pp. 99–
105.
Cummings, P. and Weiss, N.S., 1998, Case series and exposure series: The role of studies
without controls in providing information about the etiology of injury or disease.
Injury Prevention, 4, pp. 54–57.
Garrison, H.G., Runyan, C.W., Tintinalli, J.E., Barber, C.W., Bordley, W.C., Hargarten,
S.W., Pollock, D.A. and Weiss, H.B., 1994, Emergency department surveillance: An
examination of issues and a proposal for a national strategy. Annals of Emergency
Medicine, 24, pp. 849–856.
Graubard, B.I. and Korn, E.L., 1994, Regression analysis with clustered data. Statistics in
Medicine, 13, pp. 509–522.
Haddon, W., Jr., 1972, A logical framework for categorizing highway safety phenomena
and activity. Journal of Trauma, 12, pp. 193–207.
Hakamies-BIomqvist, L., 1998, Older driver’s accident risk: Conceptual and
methodological issues . Accident Analysis & Prevention, 30, pp. 293–297.
Hauer, E., 1997, Observational Before-After Studies in Road Safety: Estimating the Effect
of Highway and Traffic Engineering Measures on Road Safety, (New York: Pergamon
Press, Elsevier Science, Inc.).
Jovanis, P.P. and Chang, H.-L., 1989, Disaggregate model of highway accident
occurrence using survival theory. Accident Analysis & Prevention, 21, pp. 445–458.
Kuhn, L., Davidson, L.L. and Durkin, M.S., 1994, Use of Poisson regression and time
series analysis for detecting changes over time in rates of child injury following a
prevention program. American Journal of Epidemiology, 140, pp. 943–955.
LaPorte, R.E., Dearwater, S.R., Chang, Y.-F., Songer, T.J., Aaron, D.J., Anderson, R.L.
and Olsen, T., 1995, Efficiency and accuracy of disease monitoring systems:
Application of capture-recapture methods to injury monitoring. American Journal of
Epidemiology, 142, pp. 1069–1077.
Lyons, R.A., Lo, S.V., Heaven, M. and Littlepage, B.N.C., 1995, Injury surveillance in
children—usefulness of a centralised database of accident and emergency attendances.
Macarthur, C. and Pless, I.E., 1999, Evaluation of the quality of an injury surveillance
system. American Journal of Epidemiology, 149, pp. 586–592.
Maclure, M., 1991, The case-crossover design: A method for studying transient effects on
the risk of acute events. American Journal of Epidemiology, 133, pp. 144–153.
Marshall, S.W., Waller, A.E., Loomis, D.P. and Langlois, J.A., 1998, Selection of control
groups in injury case-control studies. In Abstracts of the 4th World Conference on
Injury Prevention and Control, Amsterdam.
Mountain, L., Fawaz, B. and Jarrett, D., 1996, Accident prediction models for roads with
minor junctions. Accident Analysis & Prevention, 28, pp. 695–707.
Oppe, S., 1992, A comparison of some statistical techniques for road accident analysis.
Accident Analysis or Prevention, 24, pp. 397–423.
Petridou, E., 1997, Epidemiology and injury prevention. Injury Prevention, 3, pp. 75–76.
Platt, R.W., 1997, Logistic regression and odds ratios. Injury Prevention, 3, p. 294.
Razzak, J.A. and Luby, S.P., 1998, Estimating deaths and injuries due to road traffic
accidents in Karachi, Pakistan, through the capture–recapture method. International
Journal of Epidemiology, 27, pp. 866–870.
Rivara, F.P., Thompson, D.C., Beahler, C. and MacKenzie, E.J., 1999, Systematic
reviews of strategies to prevent motor vehicle injuries. American Journal of Preventive
Medicine, 16(1S), pp. 1–5.
Roberts, I., 1995, Methodologic issues in injury case-control studies. Injury Prevention,
1, pp. 45–48.
Roberts, I., Marshall, R. and Lee-Joe, T., 1995, The urban traffic environment and the
risk of child pedestrian injury: A case-crossover approach. Epidemiology, 6, pp. 169–
171.
Runyan, C.W., 1998, Using the Haddon matrix: Introducing the third dimension. Injury
Runyan, C.W., Bowling, J.M. and Bangdiwala, S.I., 1992, Emergency department record
keeping and the potential for injury surveillance. The Journal of Trauma, 32, pp. 187–
189.
Schafer, J.L., 1997, Analysis of Incomplete Multivariate Data, (London: Chapman and
Hall).
Stamatiadis, N. and Deacon, J.A., 1997, Quasi-induced exposure: Methodology and
insight. Accident Analysis & Prevention, 29, pp. 37–52.
Towner, E.M., Jarvis, S.N., Walsh, S.S. and Aynsley-Green, A., 1994, Measuring
exposure to injury risk in schoolchildren aged 11–14. British Medical Journal, 308, pp.
449–452.
Williamson, D.S., Bangdiwala, S.I., Marshall, S.W. and Waller, A.E., 1996, Repeated
measures analysis of binary outcomes: Applications to injury research. Accident
Analysis & Prevention, 28, pp. 571–579.
6
Assessing the Burden of Injury: Progress and
Pitfalls
Ted R.Miller
INTRODUCTION
Only a systematic measurement of the burden that injury imposes on our society helps us
fully understand the critical need to cure this pervasive killer and crippler (Rice et al.,
1989). Cost is the Rosetta stone that makes burden estimates understandable. Cost
estimates offers a major advantage over the measurement of incidence by reducing
disparate outcomes—traumatic deaths, broken noses, burns, dog bites, even damaged
motorcycles—to a single compact metric. Compactness eases comprehension. That
makes cost data valuable for problem size and risk assessment, broad priority-setting,
resource allocation modelling, health and safety advocacy, regulatory analysis,
performance comparison, and programme evaluation. Cost data describe how injuries
affect society and drive analyses of the potential to reduce injuries cost effectively.
Increasingly this basic health services research tool has become the focal point of
debate and decision making. Cost is a powerful persuader. The press and politicians
understand a savings of $9 per taxpayer in medical and work loss costs far better than a
statistically significant 1 per cent reduction in injury deaths and hospital admissions.
From the perspective of a developing country, measuring injury burden can be a
critical guide to health sector resource allocation. By 2020, the World Health
Organization (Murray and Lopez, 1994) estimates that injury will rank ahead of all other
causes except mental health as a contributor to disability adjusted life year (DALYs) loss
in the developing world. Understanding burden size in a country and identifying the
injury causes that contribute the greatest burden is critical when formulating cost-
effective intervention projects. As donor organizations’ health sector interest shifts
toward injury, so must the effort devoted to characterizing the problems and evaluating
potential solutions.
COST CONCEPTS
Costs are estimated from a perspective. The US Panel on Cost-Effectiveness in Health

and Medicine (PCEHM) (Gold et al., 1996), recommends always including a cost
Assessing the burden of injury 53
BOX 1
USING INJURY COSTS TO ADVOCATE FOR CANADIAN GUN

REGISTRATION
To aid advocates, we estimated the costs of gunshot wounds in Canada and
made some US comparisons (Miller, 1995; Miller and Cohen, 1997).
Capping years of lobbying, Parliament was scheduled to debate gun
registration legislation with the outcome in serious doubt. The study was
deliberately published in the Canadian Medical Association Journal as the
formal debate began. It made headlines.The opposition leader responded
spontaneously, publicly claiming that the taxes on guns alone would pay the
entire firearm injury bill. The injury costs were an order of magnitude larger,
prompting a further, highly embarrassing front page story.The legislative
managers immediately called for a vote, making damage control impossible.
The law passed handily.
USING INJURY COSTS TO RAISE PREVENTION FUNDS IN FLORIDA

In DuVal County, Jacksonville Florida, the county health department’s
injury prevention specialist uses cost data when raising funds. In his initial
efforts, his proposals featured data about the economic costs of problems like
bicycle crash injuries and child safety seat non-use. He then evaluated the
impact funded programmes had on injury frequency and from that, estimated
the injury cost savings. Funders were impressed by the economic arguments
and the clear way they demonstrated the programme’s track record. They
gave the programme an edge.
when evaluating… laws that interfere with personal freedom, economists often
focus on external costs…high external costs justify public intervention
estimate from society’s perspective. That perspective embraces all costs associated with
injuries—costs to victims, families, government, insurers and taxpayers. Costs to
government, to insurers or to employers are frequently computed separately. When
evaluating mandatory helmet use laws, driver blood alcohol limits and other laws that
interfere with personal freedom, economists often focus on external costs—costs to
people other than the person whose behaviour is constrained. High external costs justify
public intervention.
Costs can be prevalence-or incidence-based. Prevalence-based costs measure all injury
related expenses during one year, regardless of when the injury occurred. For example,
the prevalence-based cost of head injuries in 1996 measures the total health care spending
on head injuries during 1996, including spending on victims injured many years earlier.
Prevalence-based costs are computed by summing all costs incurred during the year.
They are used to project health care spending and evaluate cost controls.
Medical costs sum the lifetime costs that are expected to result from injuries that occur
during a single year. For example, the incidence-based cost of head injuries in 1996
estimates present and future medical spending associated with all head injuries that
occurred in 1996. Incidence-based costs are computed by multiplying the number of
injury victims by lifetime cost per victim. They measure the savings that prevention can
yield.
Investments earn interest. In incidence-based costing, therefore, future costs must be
discounted to present value. This procedure ‘shows the amount that would be invested
today to pay future costs as they arise. The PCEHM (Gold et al., 1996) recommends that
all cost savings analyses include an estimate at a 3 per cent discount rate to accommodate
cross-study comparisons. Real rates of return on investment and discount rates that
individuals apply when making health decisions suggest this discount rate is a
conservative upper bound in the US (US Supreme Court, 1983; US Office of
Management and Budget, 1994; Viscusi, 1995) and elsewhere (Murray and Lopez, 1994).
Indeed, our group’s published work generally uses a 2.5 per cent discount rate.
Worldwide, governments often require analyses of proposed government investments at
discount rates of 7 per cent to 10 per cent. These high rates offset optimistic impact
estimates, lowering total expected benefits. For example, a $1 million cost saving 20
years hence has a present value of $625,000 at a 2.5 per cent discount rate but only
$275,000 at a 7 per cent discount rate.
BURDEN CATEGORIES AND MEASUREMENT METHODS
Injury burden falls into four categories: medical costs, other resource costs, work loss
cost and quality of life.
Medical costs include emergency transport, medical, hospital, rehabilitation, mental
health, pharmaceutical, ancillary, and related treatment costs, as well as funeral/coroner
expenses for fatalities and administrative costs of processing medical payments to
providers.
Other resource costs include police, fire, legal/court, and victim services (e.g., foster
care, child protective services), plus the costs of property damage or loss in injury
incidents.
Work loss costs value productivity losses. They include victims’ lost wages and the
replacement cost of lost household work, as well as fringe benefits and the administrative
costs of processing compensation for lost earnings through litigation, insurance, or public
welfare programmes like food stamps and Supplemental Security Income. As well as
victim work losses from death or permanent disability and from short-term disability, this
category includes work losses by family and friends who care for sick children, travel
delay for uninjured travellers that results from transportation crashes and the injuries they
cause, and employer productivity losses caused by temporary or permanent worker
absence (e.g. the cost of hiring and training replacement workers).
Quality of life includes the value of pain, suffering, and quality of life loss to victims
and their families.
Some aspects of injury burden are readily measured in monetary terms. These include
medical costs, other direct or resource costs, and work losses. Together, they are called
economic costs or human capital costs. Placing a monetary value on pain, suffering, and
lost quality of life, however, is challenging and controversial. For this reason, it often is
desirable to quantify this portion of burden with a non-monetary measure. Quality
adjusted life years ((QALYs) and DALYs are popular non-monetary measures. The best
approach may be to present both QALYs and monetized QALYs. Costs that include the
value of pain and lost quality of life are called comprehensive costs or willingness-to-pay
costs.
Medical Costs
Medical cost estimates are computed best bottom-up, by multiplying estimated medical
spending per case or visit by diagnosis times corresponding estimated case or visit
counts. Two coarser approaches are possible.
• Top-down. One can obtain total national medical spending, then apportion it
according to hospital days by diagnosis group. This prevalence-based approach
often is taken when comparing spending on injury and illness (Rice et al., 1985;
Moore et al., 1997). We applied it to cost gunshot wounds in Canada (Miller,
1996) and occupational injuries and employer costs in the US (Miller and
Galbraith, 1995a; Miller, 1997a).
• Factoring. One can adjust a national cost per case to local prices (and
preferably local length of hospital stay), then multiply times a local case count
to get local costs. We often use this method to make state or provincial cost
estimates. It is inexpensive, yet yields reasonably credible numbers.
Work Loss Costs
it is reasonable to assume parental work loss equals the loss that normally occurs
when an adult suffers a comparable injury
Work or productivity loss has two components: short-term losses during acute injury
recovery and lifetime losses due to death or permanent work-related disability. The value
of lost paid work includes both wages and fringe benefits. Employers or employees may
bear these costs. On average, US workers lose housework on 90 per cent of the work days
that they lose wage work. Thus household work days lost can be estimated from the days
of paid work lost (Miller, 1993). These days typically are valued at the wages paid for
comparable tasks (e.g., cooking, cleaning, child care).
Lifetime work loss costs value work losses in the current year plus the present value of
probable work losses in future years if the individual dies or is permanently disabled.
Children under age 15 will not lose work in the short-term. Thus, the lower bound on the
cost of short-term work loss due to a child’s injury is none. When injured children are
impaired sufficiently that they would not have been able to work if they had been
employed, someone else generally will lose work while serving as a care giver. As an
upper bound, then, it is reasonable to assume parental work loss equals the loss that
normally occurs when an adult suffers a comparable injury. For other age groups, the
value of lost work depends on the work that someone of the victim’s age and sex
normally would do and the amount they would earn.
Quality of life
A quality-adjusted life year or QALY is a health outcome measure that assigns a value of
1 to a year of perfect health and 0 to death (Gold et al., 1996). QALY loss is determined
by the duration and severity of the health problem. To compute it, one estimates the
fraction of perfect health lost during each year that a victim is recovering from a health
problem or living with a residual disability, then sums these fractions. People killed lose a
full QALY per life-year; this value may be adjusted for pre-existing conditions and the
general decline in health as people age.
The most practical way to assess health-related quality of life losses from a community
viewpoint is a two-step process. In the first step, one creates a set of scales for rating
health states, i.e., physical and emotional health status. The general public then is polled
to determine how they value the different health states relative to optimal health and to
death. A good measure should allow people to rate some fates as worse than death.
In the second step, either patient survey/observation or expert physician judgement is
used to estimate the temporal pattern of health status changes over time that result from a
medical problem. The rating scale then is used to estimate lost utility (an economist’s
measure of the relative value people place on different goods). What results is an estimate
of the QALYs lost to the medical problem. Over time, refinements to this method have
led to the recognition that death does not result in a full QALY of loss per year of life
lost. Because no one is perfectly healthy every day, estimated QALY losses normally
should be adjusted for major pre-existing health problems. US data now are available to
support adjustment (Krueger and Ward, 1998).
A good QALY scale is segmented by dimensions of functioning. The range of
functional dimensions varies in scope and detail. For example, the original Health Utility
Index (Torrance, 1982) considers four dimensions: social-emotional function, role
function, physical function, and health/sensory problems. Patrick and Erickson (1993)
summarize the scale dimensions used in a wide range of QALY rating studies.
Each of the existing QALY scales has drawbacks. Furthermore, in a case study (Gold
et al., 1996), the scales yielded widely varying estimates for the same disease. In part this
was caused by the different domains covered by the different measures, in part it was a
function of the underlying valuation task.
Primary considerations in selecting a baseline scale are: (1) how comprehensively the
scale covers the range of health status dimensions, (2) how finely health status is assessed
within dimension, (3) empirically, how credible the scale’s ratings are (measured by
consistency with ratings from other scales and by the range/clustering of possible scale
scores), (4) the representativeness of the population used in calibrating the ratings, (5) the
technical soundness of the rating method used in calibration, (6) the length of interview
required to rate current health status using the scale, and (7) importantly, what off-the-
shelf ratings are available for specific medical conditions.
Here are preliminary assessments of some of the leading scales.
• The Health Utility Index (Torrance, 1982; Drummond et al., 1987) was
calibrated primarily with a survey of 112 parents of school-age Canadian
children. A broader HUI scale recently was calibrated with particular reference
to disabling workplace injury (Torrance et al., 1992). A still more
comprehensive HUI scale then was calibrated through a 500-person sample
survey (Patrick et al., 1993). HUI scales are easily applied to a wide range of
diagnoses. Miller et al. (1989) finds HUI-based estimates compare reasonably
well with direct survey estimates of utility losses for selected conditions.
The original HUI has been mapped into various US National Center for
Health Statistics (NCHS) surveys including the National Medical Expenditure
Survey, the National Health Interview Survey, and the National Health and
Nutrition Examination Survey (Erickson et al., 1989, 1992), making it possible
to compare the health status of people with varying medical conditions.
Mapping yields a good measure of the population’s average quality of life.
Since most people have a variety of medical conditions, however, it can be
difficult to try to parse out the QALY impacts of a particular condition (Fryback
et al., 1993).
injury cost per gun is $840 in Canada and $630 in the US. This finding
strongly suggests controlling ready access to guns can reduce the toll
• The Quality of Well-Being (QWB) Scale (Bush et al., 1973; Kaplan et al.,
1976, Kaplan, 1982) is based on a San Diego survey in the early seventies. The
QWB’s performance has a low maximum loss level; no fate is scaled anywhere
near as bad as death. Although I am not overly fond of this scale, it has been
very widely applied. Like the HUI, NCHS has mapped the QWB into its data
sets (Erickson et al., 1989). Fryback et al. (1993) and those designing the
Oregon Medicaid rationing effort (Office of Technology Assessment, 1992)
also used it to calculate values for specific diagnoses. Oregon also recalibrated
the scale (Patrick and Erickson, 1993). Thus, the QWB can readily be used for
sensitivity analysis. I do not favour it, however, for base case analysis.
• The EuroQol scale (EuroQol Group, 1990; Brooks et al., 1991; Nord, 1991;
Williams, 1995) has been calibrated with national sample surveys in Britain,
Sweden, and the Netherlands, making its estimates more representative than any
other scale developed to date. The scale only offers a total of 245 health states,
meaning it is less detailed than the HUI. It may lack the sensitivity needed to
differentiate between alternatives that create moderate improvements along one
dimension of health status.
• The Injury Impairment Index (III) (Hirsch et al., 1983) was developed for
physicians to use in rating the time-phased consequences of injury on six
dimensions: cognitive, mobility, daily living, sensory, cosmetic, and pain.
Following procedures similar to Erickson et al. (1992), Miller et al. (1991,1995)
converted the functional ratings to utilities. Objective data were added on
permanent work-related disability frequency. Weights were synthesized for each
scale level within dimension and across dimensions from scores on HUI, QWB,
and other scales, then combined across dimensions. Validation has been limited,
however. This scale underlies virtually all of our group’s published estimates of
willingness to pay to reduce the risk of non-fatal injuries.
• The Functional Capacity Index (FCI) (MacKenzie et al., 1996, Luchter,
1998) is the result of an ongoing major US National Highway Traffic Safety
Administration investment. The work to date includes creating a 10-dimensional
Functional Capacity Index, convening physician panels to rate average adult and
paediatric functional losses by non-fatal injury diagnosis, validating the ratings,
and converting the scores to QALY estimates through limited population
surveys. The scales presently exclude impacts on pain and psychological
function and do not allow fates worse than death.
Another popular measure, the World Bank’s disability adjusted life years (DALYs), is
essentially QALYs where the importance of different aspects of functioning is based on
analytic judgement rather than a survey of public preferences. Furthermore, impact is
rated on a single 6-point disability scale with little supporting evidence for the ratings.
DALY estimates are little grounded in data. They have not been validated. Promisingly,
the Dutch are measuring DALYs with the EuroQol instrument, which will make them
more objective and allow validation.
In cumulating future years saved, like with any benefit, one needs to discount.
Numerous studies (Agee and Crocker, 1996; Cropper et al., 1991; 1992; 1994;
Johannesson et al., 1997; Kashner, 1990; Moore and Viscusi, 1990a; 1990b; Olsen, 1993;
Viscusi, 1995; Viscusi and Moore, 1989) find that discount rates for longevity are not
dissimilar to discount rates for monetary decisions. Indeed, the PCEHM and WHO
recommend discounting QALYs at the same rate as monetary losses (Gold et al., 1996;
Murray and Lopez, 1994).
Should QALYs be monetized?

QALYs can simply be reported as a complementary measure to economic costs.
Alternatively, they can be monetized. The strongest argument for monetizing is to permit
comparisons between health sector investments and investments in other sectors. Such
comparisons are especially important for multinational development banks and donor
organizations.
Pain, suffering, and lost quality of life for fatalities are best valued in dollars using an
approach economists call willingness-to-pay. This approach derives the value of pain and
suffering by asking people what they are willing to pay (called contingent value surveys)
or by studying what people actually pay for small changes in their chance of being killed
or injured. The value of fatal risk reduction, aggregated over many people, yields the
value of a statistical life. For example, suppose a study estimated that the average person
spends $300 on optional auto safety features that reduce the chance of dying prematurely
by 1 in 10,000. Dividing $300 by the 1 in 10,000 probability yields a $3 million value per
statistical life. That value has two components: (1) the value of the future work that will
be foregone and (2) the value of the pain, suffering, grief, lost companionship, and lost
quality of life. The value of lost future work is known. Subtracting it leaves the value of
the intangibles (Arthur, 1981; Miller et al., 1989). Importantly, when this subtraction-
based method is used to value lost quality of life, the total cost of an injury becomes
insensitive to the work loss estimate.
The values for lost quality of life in our published studies use the estimated mean value
of statistical life across 67 studies that Miller (1990, 1999) rated as technically sound. At
present, value estimates exist for 12 countries. The estimated mean value of statistical life
in the US, for example, is $3 million 1997 after-tax dollars with a coefficient of variation
of 28.5 per cent. Miller (1990) modified many published values to remove inappropriate
biases. Without these somewhat arbitrary adjustments, Viscusi (1993) recommends a
value of $3–5 million. These values are for people averaging age 38, two-thirds of them
males. Miller (1999) provides a formula for estimating values for other countries from
their gross domestic product, but its validity for different cultures and less developed
countries is uncertain.
For non-fatal injury, four methods can be used to place a dollar value on pain,
suffering, and lost quality of life: (1) monetization of estimated QALY loss, (2) analysis
of jury awards for non-economic damages in countries where juries make such awards
without restrictions, (3) survey methods, although these are costly and often fail to yield
credible estimates, and (4) multiplication of work losses by the average ratio of QALYs
to work loss across all non-fatal injuries to get a low-cost order-of-magnitude estimate.
valuing quality of life may be unfamiliar or disquieting. Given the wide range for
the value of a statistical life and the slim validation of the implied value, it also
QALY is difficult and controversial
Monetizing QALYs is the most popular approach. Working from a general equilibrium
model of the economy, Miller et al. (1989) show that the value of a statistical life times
the percentage utility loss associated with a non-fatal outcome equals the willingness to
pay to avoid that outcome. Most QALY systems are calibrated so that optimal health has
a value of 1, death has a value of 0, and fates worse than death are allowed. Under those
conditions, QALYs measure the desired utility loss. We have published more than a
dozen studies using this approach. Beyond our work, other studies that monetize QALYs
with willingness-to-pay values come from Australia (Fildes and Digges, 1998), Canada
(Bein et al., 1994; Newman et al., 1994), New Zealand (Guria, 1991), the United
Kingdom (Ball, 1998; Ives et al., 1993), and the US (French et al., 1996; Mauskopf and
French, 1989).
The value of reducing risk by one QALY can be estimated readily from the value of
statistical life by assuming the value per QALY does not vary with age or sex (Miller et
al., 1989). Specifically, one subtracts lifetime work loss from the value of statistical life
to avoid double counting, then divides the remainder by the expected years of life span
saved per life saved (discounted to present value). Multiplying the discounted years of
expected life lost for someone in a specific age group by the QALY value, and dividing
by the mean discounted all-victim lifespan yields a value tailored by age and sex.
With our value of statistical life, the value per QALY at a 2.5 per cent discount rate is
$85,600 (in 1997 US dollars). Validation of this QALY value has been minimal. Miller et
al. (1989) used it to accurately predict values of asthma risk reduction obtained in a
subsequent survey. For assaults (Miller et al., 1996), consumer product injuries (Miller et
al., 1998), and drunk driving (Smith, 1998), the pain and suffering component of jury
verdicts can be predicted well from III-based QALY losses (r2>0.5 in log-linear
regressions), with juries valuing QALYs consistent with a value of statistical life of $ 1.7
to $4 million. Thus QALY costs measure real and tangible losses. For example, when a
car strikes a bicyclist and a bicycle helmet prevents a severe brain injury, an auto insurer
may avoid paying millions of dollars to compensate the family’s lost quality of life.
BOX 2
SELLING QUALITY OF LIFE COSTS FOR TRANSPORT DECISION

MAKING IN AUSTRALIA
Transport infrastructure and regulatory decisions often trade safety to reduce
the time required to move people and goods. In Australia, we multiplied the
government’s official value of travel time by the number of hours in a life,
than compared this value with the AU$630,000 official value of life used in
time-safety tradeoffs (Cook, 1992; Miller, 1998). A lifetime of travel time
was worth AU$780,000. As Ezra Hauer (1994) says, it was better to be dead
than stuck in traffic. This imbalance was comprehensible. The comparison
offered a simple plausibility test for economic costs, one they failed. It
convinced policymakers that their benefit-cost analyses which failed to
account for quality of life losses were yielding incorrect answers.
The jury verdict method also has quietly gained popularity. Its theoretical framework
comes from Cohen (1988), Viscusi (1988) and Rodgers (1993). The basic notion is that
pain and suffering to a survivor can be approximated by the difference between the
amount of compensatory damages awarded by a jury minus the actual out-of-pocket
charges associated with the injury. Miller, Cohen, and Wiersema (1996) estimates pain
and suffering for physical assaults from log-linear jury verdict regressions, then compares
the results with III-based QALY estimates by ICD9 diagnosis group (Miller et al., 1995).
Some individual estimates vary fairly significantly, but the incidence weighted mean
estimates from the two methods vary by only 5 per cent. Thus, in large numbers, US jury
verdicts appear to be reasonably predictable.
Valuing quality of life may be unfamiliar or disquieting. Given the wide range for the
value of a statistical life and the slim validation of the implied QALY value, it also is
difficult and controversial. Nevertheless, the ease of dealing with a single burden measure
and the ability to compare between health and other sectors make it a valuable
supplement to economic cost and QALY loss estimates.
Recurring questions about costs

Audiences often ask eight questions about injury costs. This section answers them. In
some cases, it adds technical details.
Question 1. Preventing injuries means physicians lose business! Is that a cost to

society?
No. Physicians lose business, but jewellers and computer stores gain business, so it is a
wash. At the end of the day, however, you take home a diamond ring rather than a leg
cast. That means you benefit.
Question 2. Why are welfare payments not in the costs?

The issue here is perspective. Costs most often are computed from society’s
perspective. Welfare payments mean that we pay part of the wage loss for your injury
rather than you picking up the whole tab. If we counted both the work loss and the
welfare payment, we would double count.
From government’s perspective, however, welfare payments count. They also count
from an external cost perspective. This perspective is often used when analysing whether
the public should interfere with individual freedom. It evaluates the cost to people other
than an actor who bears some responsibility for the injury, for example, the costs to other
people when someone rides a motorcycle without a helmet for a year or the costs that
others pay when someone drives a kilometre drunk.
When an individual’s costs are estimated with willingness to pay, ignoring the value of
pride and self-sufficiency, a rational individual will not be willing to pay anything to
avoid costs reimbursed by the welfare system. To convert individual willingness to pay
into societal willingness to pay requires adding the welfare payments. From a practical
viewpoint, this means societal willingness to pay to prevent an injury equals the sum of
resource costs, work loss costs, and individual willingness to pay to avoid pain, suffering,
and lost quality of life (net of lost productivity).
costs most often are computed from society’s perspective. Welfare payments mean
that we pay part of the wage loss for your injury rather than you picking up the
whole tab
Question 3. Can we compare the economic costs to the Gross National Product
(GNP)? Do they describe injury’s impact on the GNP?
No, they are not comparable. The economic costs include household work loss. The
GNP does not value housework.
Comparing economic costs minus housework to the GNP is a valid yardstick for the
size of the injury problem. (For example, annual US highway crash costs are roughly
comparable to the GNP reduction that results from a recession.) However, economic
costs do not describe injury’s impact on the economy. For one thing, when you buy
domestic medical care rather than an imported South African diamond or SONY
television, that helps your country’s economy. Similarly, when someone dies, since we
are not at full employment, it means someone else gets a job. The victim’s wage loss is
not a GNP loss. The GNP loss in wages is just the friction cost (Koopmanschap et al.,
1995), essentially the costs of hiring and training plus the value of unique skills that are
lost. To compute the impact of injury on the economy, we would need to use an input-
output model to trace the effects of expenditure shifts.
Question 4. When people die, we no longer have to pay to feed and clothe them. Why
not subtract consumption from the wage loss?
The purpose of society is to let people consume. Since the deceased was a member of
society, his or her consumption losses are legitimate societal losses (Fein, 1958; Klarman,
1965; Rice, 1966; Rice and Cooper, 1967).
Question 5. For deaths, especially deaths of older people, are reduced future medical
costs a benefit?
From society’s perspective, they are not. Medical care is just a good like food which
people consume to maximize their quality of life. From a government or external cost
perspective, reduced medical care generally is a benefit.
Question 6. Why count earnings losses when unemployment is high? Someone else
needs the dead person’s job.
This is a difficult question. In a willingness to pay accounting framework, as discussed
earlier, the value of statistical life values both earnings loss and quality of life. For this
framework, economic theory dictates including earnings loss (Miller et al., 1989).
If one does not monetize QALY losses, the problem is more complex. QALYs
typically include the value the victim places on losing the ability to work. Adding them to
productivity losses would double-count. Friction costs (Koopmanschap et al., 1995) were
designed specifically to measure the productivity cost to people other than the victim.
They appear to be the appropriate productivity loss costs in a QALY accounting
framework where the QALY measure includes work impacts.
The economic cost framework is theoretically justifiable only as a lower bound
estimate of injury costs that excludes the impacts on quality of life. Miller et al. (1989)
show that work loss is included in this lower bound estimate.
Question 7. No dollars change hands when people lose quality of life. Are you adding
quality of life costs just to inflate the losses? Can users of these numbers omit the quality
of life costs?
The costs value real losses. US tort liability lawsuits against people or companies that
cause serious injuries yield pain and suffering awards comparable to our estimates,
although a wide gap might exist in less litigious countries (Cohen and Miller, 1997;
Miller et al., 1996; Smith, 1998). Suppose a big truck hits your car. Your baby is in a
child seat so she avoids a severe brain injury. That means an insurer will not be paying
millions of dollars to compensate your family for lost quality of life. We see those
savings.
Nevertheless, for many purposes and audiences, it is perfectly acceptable to report just
economic or medical costs. Problem size, for example, is best described to a hostile
audience with economic costs, possibly supplemented by non-monetized QALYs. Some
audiences are only interested in medical or resource costs, and costs to government are
especially relevant in political debates. Conversely, victims turned advocates often
choose to use costs that include monetized QALYs. Politicians feel uncomfortable
challenging the victims’ valuations for fear of seeming unsympathetic.
When analysing the cost savings of prevention, a cost per QALY saved or a benefit-
cost ratio based on comprehensive costs generally is desirable. If the resource cost
savings alone exceed intervention costs, however, it is reasonable to simply state that the
intervention yields net cost savings. A large benefit-cost ratio based on economic cost
savings also can be reasonable to report although it does not support sound comparison
with alternatives.
Question 8. When a habitual drunk driver or violent offender is saved, (s)he may harm
other people. Is that not a cost?
This is a difficult ethical question. It seems preferable to reduce benefits in such cases.
Computationally, we lack the data needed to adjust.
RECOMMENDED BURDEN TABLES AND THEIR PURPOSES
This section suggests a preliminary set of injury cost tables for international comparisons.
It also discusses additional tables that can be useful if the data is available. Miller and
Blewden (1999) explains the utility of more detailed measures around one injury-related
topic, impaired driving crashes and costs. Lack of incidence and outcome data constrains
costing efforts internationally.
by comparison, a kilometre driven sober costs only $ 0.10. A kilometre driven

drunk costs people other than the drinking driver $8.90, 1,400 times the $.006 cost
per sober kilometre
Practically speaking, it would be a major advance to have even a modest amount of

comparable data on non-fatal injury. A reasonable medium-term goal would be to
compare annual incidence, incidence rates and economic costs for all injuries
combined—road crashes; occupational injuries, and interpersonal violence by gender and
age group. Because different people often need different information, a cost breakdown is
desirable into at least three cost categories: (1) medical and other resource costs, because
these costs are of special interest to governments and insurers who typically pay them, (2)
work loss costs, which yield insight into income tax and productivity losses, and (3)
monetized or unmonetized quality of life losses, which provide the best measure of
human burden.
Increasingly, countries collect injury incidence data that include International
Classification of Diseases (ICD) diagnosis codes and external cause (e) codes. Ideally,
these countries should tabulate incidence, costs, and QALY losses: (1) by gender and age
group, accompanied by costs per resident, (2) by severity (fatal, hospital admitted, other
medically treated), (3) by age group and external cause/intent group (e.g., road crash,
self-inflicted) classified following the recommendations of the International
Collaborative Effort (ICE) on Injury Statistics, and (4) by age group and ICE diagnosis
group. If incidence data exist, it also is desirable to tabulate the costs of occupational,
consumer product, and sports injury. E codes do not identify these categories. Finally, a
comparative assessment of injury and illness costs by age group can be very helpful in
drawing attention to the major burden injury imposes.
Experience dictates not just reporting total annual costs, which tend to be billion dollar
numbers. Legislators deal with such large numbers daily and understand them. Extensive
survey and psychological research, however, shows most people cannot comprehend the
difference between $10 billion and $100 billion; both numbers are simply enormous
(Mitchell and Carson, 1989; Slovic et al., 1980; Soguel, 1995; Viscusi and Magat, 1987).
To aid public comprehension, billion-dollar annual costs should be supplemented with
costs divided by denominators that make sense from a policy viewpoint. For example,
firearm injuries cost $495 per US resident versus $180 per Canadian resident (in 12/93
US dollars) (Miller and Cohen, 1997). Does this mean Canadian handgun control works?
Perhaps, but perhaps it merely means the US is a more violent country. A different
divisor clarifies the picture. Injury cost per gun is $840 in Canada and $630 in the US.
This finding strongly suggests controlling ready access to guns can reduce the toll.
As another example, three divisors seem particularly appropriate for alcohol-related
crash costs. Cost per drink is relevant for debating and evaluating taxation and other
consumption reduction efforts such as enforcing laws against serving intoxicated patrons.
Cost per heavy drinker informs treatment, detention, ignition interlock, and other
sanctioning analyses. Cost per kilometre driven under the influence supports evaluation
of other harm reduction strategies. It also makes the riskiness of alcohol-involved driving
and the need for public intervention clear. Consider that the crash cost per kilometre
driven drunk in NZ approaches $18 (in 1997 NZ dollars) (Miller and Blewden, 1999). By
comparison, a kilometre driven sober costs only $ 0.10. A kilometre driven drunk costs
people other than the drinking driver $8.90, 1,400 times the $.006 cost per sober
kilometre. Strong public intervention clearly is justified.
PROGRESS IN MEASURING BURDEN
Without doing an exhaustive search of the literature on injury costing worldwide, this
section provides a partial report on costing practice and coverage by country.
Transportation safety costs have been studied in most developed countries and some less
developed countries. A partial overview of these studies is available in Elvik (1995),
Schwab-Christe (1995), and Ryan and Dyke (1998). Further information can be found in
the proceedings from the October 1994 conference Valuing the Consequences of Road
Accidents organized by the Institute for Regional and Economic Research at the
University of Neuchatel, Switzerland and the May 1998 conference on Measuring the
Burden of lnjury that. the European Consumer Safety Association (ECOSA) is editing.
In the US, our research group recently assessed economic costs and QALY losses
(monetized and unmonetized) for injuries by age group, cause group, diagnosis group,
and circumstance (e.g., occupational, consumer product). (See Miller, Romano and
Spicer, in press; Miller, Covington, and Jensen, in press; Miller et al., 1998; Miller et al.,
1999). This work updates less complete estimates for 1985 (Rice et al., 1989).
Increasingly, we are computing state-level cost breakdowns as well. Leigh et al. (1997)
also estimates the economic costs of occupational injury.
Injury costs elsewhere in North America are less complete. We have estimated
comprehensive costs for gunshot wounds by intent in Canada (Miller, 1995) and for road
crash costs in Ontario (Vodden et al., 1993) and British Columbia (Bein et al., 1994).
Nationally, economic costs of crashes are estimated periodically. Moore et al. (1997)
uses a top-down approach to estimate Canadian economic costs for injury as a whole and
compare them to illness costs. Angus et al. (1998) estimates Canadian economic costs for
unintentional injury, with separate breakdowns into 10 cause categories and severity.
(This report also reviews the world literature on injury costs, as does Koffijberg et al.,
1998.) Kerr and McLean (1996) and Day (1995) estimate the economic costs of violence
against women in Canada. Mexico began assessing its road crash costs at one time but I
am unsure if this effort was completed.
Most European countries have estimated the costs of road crashes periodically (OECD,
1988; Elvik, 1995). Several countries (Austria, Denmark, France, Norway, Sweden,
Switzerland, and the United Kingdom) use willingness-to-pay values of statistical life
from contingent valuation survey data. The remainder include only economic costs. The
United Kingdom also has made QALY-based estimates for crash injuries and has roughly
estimated comprehensive costs of occupational and consumer product injuries (Ball,
1998).
An on-going multinational effort coordinated by the European Consumer Safety
Association (ECOSA) is developing uniform methods for costing consumer product
injuries in Europe. The Netherlands has assessed its medical costs using the agreed
methods and has estimated the economic costs of injury (van Beeck et al., 1997). They
also have assessed the QALYs lost to injury versus illness as part of a EuroQol-based
assessment of national DALY loss by health condition. Austria has estimated economic
costs for injuries, as has a Swedish municipality (Lindqvist and Brodin, 1996). Norway
also has current injury medical cost estimates (Kopjar et al., 1996) and is estimating work
losses.
the lack of well-designed outcome evaluations for effective preventive measures

imposes a major barrier to filling the holes
Australia periodically evaluates the economic costs of transportation injury (Cook et al.,
1992). They are experimenting with ways to incorporate quality of life losses. Victoria
estimated economic costs of injury (Watson and Ozanne-Smith, 1997); Moller (1998)
used their cost factors to estimate national injury costs by age group and sex for 13 cause
groups.
New Zealand used a contingent valuation survey to cost injury deaths and hospital
admissions, then supplemented their estimates with medical cost data. Although their
work has focused primarily on road crashes, it could easily be applied to cost other
unintentional injury. They also have estimated all-cause injury medical costs (Phillips et
al., 1993).
In Central and South America, the Pan American Health Organization and
InterAmerican Development Bank are studying violence costs. Japan, Malaysia, Taiwan,
and Hong Kong all estimate road crash costs. With expert judgement, Japan also has
assessed QALY loss to crash injuries. Hsueh and Wang (1987) and Liu and Smith (1996)
have estimated the value of statistical life in Taiwan.
WHERE WE NEED TO GO
International collaborative efforts have been instrumental in developing a worldwide

picture of injury incidence. Outcome measurement scales like the Functional Capacity
Index (MacKenzie et al., 1994) are being developed and validated through multinational
efforts. These scales are valuable cost building blocks. It would be desirable for the
worldwide injury community to adopt one QALY scale as a standard for base case
analysis in cost-utility and benefit-cost analyses. Uniformity allows comparison of results
across studies. Sensitivity analysis with other scales or supplemental direct survey
assessments then can be added as appropriate.
It also is time to try stretching beyond national boundaries in developing other injury
cost factors. Especially if we standardize on a QALY scale, the obvious question is
whether countries need to collect their own QALY levels (i.e., utility loss estimates) by
functional capacity loss category. Surveys of 100 to 400 people in half a dozen countries
could answer questions like whether people trade off mobility against cognitive function
comparably. If they do, several modest survey investments could yield a large calibration
data set. I am particularly tempted by the idea of adopting Torrance’s latest HUI scale
and accompanying scaling, or possibly the FCI.
Other injury cost factors also may be comparable internationally. It may be possible to
develop temporary and permanent work-related disability probabilities and temporary
disability durations by diagnosis code that apply across developed countries. We need to
see if patterns between ICD diagnosis codes in length of stay, medical resource
intensity/cost, and percentage of medical spending in the acute care phase persist across
countries (e.g., how does the US ratio of 7.5:6.1 for lengths of stay for admitted
tibia/fibula fractures (ICD 823) versus humerus fractures (ICD 812) compare with other
countries’ ratios).
Another valuation issue where a worldwide approach seems credible is willingness to
pay for fatal risk reduction. Miller (1999) suggests values are reasonably transportable
across developed countries, but that finding needs validation, especially for different
cultures and less developed countries.
PROGRESS IN MEASURING COST SAVINGS OF PREVENTION
One particularly important area of progress in estimating injury burden is in application

for evaluation and resource allocation. We recently completed a meta-analysis of studies
examining the cost-savings of non-occupational injury prevention and treatment system
organization measures in the US (Miller and Levy, 1999). We identified 84 credible cost
savings estimates published between 1987 and 1999. The large majority of the estimates
addressed impaired driving, substance abuse, road crashes, or interpersonal violence.
None addressed suicide, drowning, poisoning, scald, cutting/piercing, childhood fall, or
unintentional firearm injury prevention. The lack of well-designed outcome evaluations
for effective preventive measures imposes a major barrier to filling the holes.
Many preventive measures, including more than half the road safety measures, offered
resource cost savings exceeding their implementation and operational costs. In health
care, interventions rarely are this cost-effective (Tengs et al., 1995). High benefit-cost
ratios and low costs per QALY saved suggest many injury prevention efforts should be
widely implemented. Cost effective road and vehicle safety efforts have been
implemented, as have child safety measures. Interventions that constrain adult behaviour,
however, have often met political resistance. Indeed, the US repealed two measures with
proven benefits—national interstate speed limits and Federal incentives for states to
mandate that motorcyclists wear helmets—in 1996. Promises of cost savings can help to
swing the political wheel. They are, however, only an advocacy aid. Political will is the
key to safety.
Each year brings greater knowledge of the worldwide burden of injury and ways to
reduce it cost-effectively. We still have much to learn. Costs and QALY losses are
critical public information and investment management tools. They can be assessed at
reasonably modest cost. It is time we accelerated the process.
REFERENCES
Agee, M.D. and Crocker, T.D., 1996, Parents’ discount rates for child quality. Southern
Economic Journal, 63, pp. 36–50.
Angus, D.E., Cloutier, E., Albert, T., Chénard, D., Shariatmadar, A., Pickett, W. and
Hartling, L., 1998, The economic burden of unintentional injury in Canada.
(Government of Ontario, Canada).
Arthur, W.B., 1981, The economics of risks to life. American Economic Review, 71, pp.
54–64.
Ball, D. J., 1998, Status of injury valuation in the United Kingdom. In Measuring the
Burden of Injury (February 15–16, 1996) edited by Ryan G.A. and Duke, P.M. 183–
196. (Nedlands: Road Accident Prevention Research Unit, University of Western
Australia, Fremantle, Western Australia).
Bein, P., Miller, T.R. and Waters, W., 1994, British Columbia road-user unit costs. In
Proceedings of the Canadian Transportation Research Forum Annual Conference, pp.
714–727. (Ottawa: University of Saskatchewan).
Brooks, R.G., Jedteg, S., Lindgren, B., Persson, U. and Bjork, S., 1991, EuroQol: Health-
related quality of life measurement: Results of the Swedish questionnaire exercise.
Health Policy, 18, pp. 37–48.
Bush, J. W., Chen, M.M. and Patrick, D.L., 1973, Health status index. In Cost-
effectiveness: Analysis of PKU program edited by Berg, R. pp. 172–209. (Chicago:
Hospital Research and Educational Trust).
Cohen, M.A., 1988, Pain, suffering, and jury awards: A study of the cost of crime to
victims. Law and Society Review, 22, pp. 537–555.
Cohen, M.A. and Miller, T.R., 1997, Willingness to award nonmonetary damages and the
implied value of life from jury awards. Working paper, Vanderbilt University, Nashville,
TN.
Cook, M., Motha, J., McKirgan, J. and O’HoIIoran, M., 1992, Social Cost of Transport
Accidents in Australia. (Report 79). (Canberra: Bureau of Transport and
Communication Economics).
Cropper, M. L., Aydede, S.K. and Portney, P.R., 1991, Discounting human lives.
American Journal of Agricultural Economics, 73.
Cropper, M. L., Aydede, S.K. and Portney, P.R., 1992,. Rates of time preference for
saving lives. American Economic Review, 82, pp.469–472.
Cropper, M.L., Aydede, S.K. and Portney, P.R., 1994, Preferences for life saving
programs: How the public discounts time and age. Journal of Risk and Uncertainty, 8,
pp. 243–265.
Day, T., 1995, The Health-related Costs of Violence against Women in Canada: The Tip
of the Iceberg. (Ontario, Canada: Centre for Research on Violence Against Women
and Children ).
Drummond, M.F., Stoddart, G.L. and Torrance, G.W., 1987, Methods for the Economic
Evaluation of Health Care Programs. (New York: Oxford University Press).
Elvik, R., 1995, An analysis of economic valuations of accident fatalities in 20 motorized
countries. Accident Analysis & Prevention, 27, pp. 237–247.
Erickson, P., Kendall, E.A., Anderson, J. P. and Kaplan, R.M., 1989, Using composite
health status measures to assess the nation’s health. Medical Care, 27, pp. S66-S76.
Erickson, P., Kendall, E.A., Odle, M.P. and Torrance, G.W., 1992, Assessing health-
related quality of life in the National Health and Nutrition Examination Survey.
(Hyattsville, MD: National Center for Health Statistics).
EuroQol Group., 1990, EuroQoI: A new facility for the measurement of health-related
quality of life. Health Policy, 16, 199.
Fein, R., 1958, Economics of Mental Illness. (New York: Basic Books).
Fildes, B. and Digges, K., 1998, Harm reduction for estimating countermeasures benefits.
In Measuring the Burden of Injury: Proceedings of a conference held at The Esplanade
Hotel, Fremantle, Western Australia, 15th and 16th February 1996 edited by G.A.
Ryan and P.M.Dyke. pp. 147–156. (Nedlands: Road Accident Prevention Research
Unit, Department of Public Health, The University of Western Australia, WA,
Australia).
French, M.T., Mauskopf, J.A., Teague, J.L. and Roland, J., 1996, Estimating the dollar
value of health outcomes from drug abuse interventions. Medical Care, 34, pp. 890–
910.
Fryback, D.G., Dashbach, E.J., Klein, R., Klein, B.E.K., Peterson, K. and Martin, P. A.,
1993, The Beaver Dam health outcomes study: Initial catalog of health-state quality
factors. Medical Decision Making, 13, pp. 89–102.
Gold M.R., Siegel, J.E., Russell, L.B. and Weinstein, M.C. (Eds.), 1996, Cost-
Effectiveness ‘in Health and Medicine. (New York, NY: Oxford University Press).
Guria, J. C., 1991, Estimates of Social Costs of Accidents and Injuries. (Wellington: New
Zealand Land Transport Safety Authority).
Hauer, E., 1994, Can one estimate the value of life or is it better to be dead than stuck in
traffic ? Transportation Research, A 28A, pp. 109–118.
Hirsch, A., Eppinger, R., Shame, T., Van Nguyen, T., Levine, R., Mackenzie, J., Marks,
M. and Ommaya, A., 1983, Impairment scaling from the abbreviated injury scale.
(Washington, DC: National Highway Traffic Safety Administration).
Hsueh, L. and Wang, S., 1987, The Implicit Value of Life in the Labor Market in Taiwan.
In Discussion Paper 8801. (Taiwan: Chung Hua Institution for Economic Research).
Ives, D., Soby, B., Ball, D. and Kemp, R., 1993, Evaluation of non-fatal casualty costs: a
report of the application of the relative utility loss approach. TRL Contractor Report
307. (Crowthorne, UK: Transport Research Laboratory).
Johannesson, M., Johansson, P.O. and Lofgren, K.G., 1997, On the value of changes in
life expectancy: Blips versus parametric changes. Journal of Risk and Uncertainty, 15,
pp. 221–239.
Kaplan, R.M., 1982, Human preference measurement for health decisions and the
evaluation of long-term care. In Values and Long-term Care edited by R.L.Kane and
R.M.Kane. pp. 157–188. (Lexington, MA: Lexington Books).
Kaplan, R.M., Bush, J.W. and Berry, C.C., 1976, Health status: Types of validity for an
index of well-being. Health Services Research, 11, pp. 478–507.
Kashner, T.M., 1990, Present-future gratification tradeoffs: Does economics validate
psychometric scales? Journal of Economic Psychology, pp. 247–268.
Kerr, R. and McLean, J., 1996, Paying for Violence: Some of the costs of violence against
women in B.C. (Victoria, B.C., Canada: Ministry of Women’s Equality,).
Klarman, E.H., 1965, Syphilis control programs. In Measuring the Benefits of
Government Investment edited by Dorfman, R. (Washington, DC: The Brookings
Institution)
Koffijberg, H., Meerding, W.J. and Mulder, S. (Eds.), 1998, Economic Analyses of
Accidents and Injuries: An Annotated Bibliography. (Amsterdam, Netherlands:
European Consumer Safety Association—ECOSA).
Koopmanschap, M.A., Rutten, F.F. H., van Ineveld, B.M. and van Roijen, L., 1995, The
friction cost method for estimating the indirect costs of disease. Journal of Health
Economics, 14, pp. 171–198.
Kopjar, B., Guldvog, B. and Wilk, J. (1996). Costs of medical treatment of injuries in
Norway Medisinske behandlingskostnader for skader i Norge. Tidsskr Nor Laegeforen,
116, pp. 512–516.
Krueger, K. and Ward, J.O., 1998, Healthy Life Expectancy, 1996 Tables. (Shawnee
Mission, KS: Expectancy Data).
Leigh, J.P., Markowitz, S.B., Fahs, M., Shin, C. and Landrigan, P.J., 1997, Occupational
injury and illness in the United States. Estimates of costs, morbidity, and mortality.
Archives of Internal Medicine, 157, pp. 1557–1568.
Lindqvist, K.S. and Brodin, H., 1996, One-year economic consequences of accidents in a
Swedish municipality. Accident Analysis and Prevention, 28, pp. 209–219.
Liu, J. and Smith, V.K., 1996, English Abstract of a Proceedings Paper in Chinese.
Lucher, S., 1998, A conceptual framework for measures of injury outcome. In Measuring
the Burden of Injury: Proceedings of a conference held at The Esplanade Hotel,
Fremantle, Western Australia, 15th and 16th February 1996 edited by G. A.Ryan and
P.M.Dyke. pp. 115–124. (Nedlands: Road Accident Prevention Research Unit,
Department of Public Health, The University of Western Australia, WA, Australia).
MacKenzie, E., Damiano, A., Miller, T.R. and Luchter, S., 1996, The development of the
Functional Capacity Index. Journal of Trauma, 41, pp. 799–807.
MacKenzie, E. and Miller, T.R., 1994, The Development of the Functional Capacity
Index. NHTSA.
Mauskopf, J.A. and French, M.T., 1989, Estimating the value of avoiding morbidity and
mortality from foodborne illnesses. In Estimating and Valuing Morbidity in a Policy
Context, 1989 AERE Workshop. EPA–230–08–89–065. (Washington, DC: US
Environmental Protection Agency).
Miller, T.R., 1990, The plausible range for the value of life: Red herrings among the
mackerels. Journal of Forensic Economics, 3, pp. 75–89.
Miller, T.R., 1993, Costs and functional consequences of United States roadway crashes.
Accident Analysis and Prevention, 25, pp. 593–607.
Miller, T.R., 1995, Incidence and Costs of Gunshot Wounds in Canada. Canadian
Medical Association Journal, 153, pp. 1261–1268.
Miller, T.R., 1996, Injury cost estimation: A pain in the neck. In Proceedings of the
February 1996 conference: Measuring the Burden of Injury edited by G.Ryan and
P.Dyke. pp. 163–182. (Nedlands: University of Western Australia, Australia).
Miller, T.R., 1997, Estimating the costs of injury to U.S. employers. Journal of Safety
Research, 28, pp. 1–13.
Miller, T.R., 1998, Injury cost estimation: A pain in the neck. In Measuring the Burden of
injury, Proceedings of February 1996 Conference edited by G.A.Ryan and P.M.Dyke.
pp. 163–182. (Nedlands: Road Accident Prevention Research Unit, University of
Western Australia).
Miller, T.R., 1999, Variations between countries in values of statistical life. Peer review
working paper JTEP.
Miller, T.R. and Blewden, M., 1999, Costs of alcohol-related crashes: New Zealand
estimates and suggested measures for use internationally. Peer review draft, (Auckland,
NZ: University of Auckland).
Miller, T.R., Calhoun, C. and Arthur, W.B., 1989, Utility-adjusted impairment years: A
low-cost approach to morbidity valuation in estimating and valuing morbidity in a
policy context. In Proceedings of Association of Environmental and Resource
Economists Workshop. EPA–23–08–89–05. (Washington, DC: U.S. Environmental
Protection Agency).
Miller, T.R. and Cohen, M.A., 1997, Costs of gunshot and cut/stab wounds in the United
States, with some Canadian comparisons. Accident Analysis and Prevention, 29, pp.
329–341.
Miller, T.R., Cohen, M.A. and Wiersema, B., 1996, Victim costs and consequences—A
new look. National Institute of Justice Research Report NCJ 155281 at U.S. GPO:
1996–495–037/20041. (Washington, DC: Government Printing Office).
Miller, T.R., Covington, K. and Jensen, A., 1999, Costs of injury by major cause, United
States, 1995: Cobbling together estimates, measuring the burden of injuries. In
Measuring the burden of injuries, Proceedings of a conference in Noordwijkerhout
edited by S.Mulder, Netherlands, May 13–15, 1998.
Miller, T.R. and Galbraith, M., 1995, The costs of occupational injury in the United
States. Accident Analysis and Prevention, Vol. 27, pp. 741–747.
Miller, T.R., Lawrence, B.A., Jensen, A.F., Waehrer, G., Spicer, R., Lestina, D. and
Cohen, M., 1998, The Consumer Product Safety Commission’s Revised Injury Cost
Model. (Final report).
Miller, T.R., Lestina, D.C. and Spicer, R.S., 1998, Highway crash costs in the United
States by driver age, blood, alcohol level, victim age, and restraint use. Accident
Analysis and Prevention, Vol. 30, pp. 137–150.
Miller, T.R. and Levy, D.T., 1999, Cost-outcome analysis in injury prevention and
control: 84 recent estimates for the United States. Peer review draft. (Landover, MD:
Pacific Institute for Research and Evaluation).
Miller, T.R., Pindus, N.M., Douglass, J.B. and Rossman, S.B., 1995, Databook on
Nonfatal Injury: Incidence, Costs, and Consequences. (Washington, DC: The Urban
Institute Press).
Miller, T.R., Romano, E.D. and Spicer, R.S., 1999, in press, The cost of childhood
unintentional injuries and savings from prevention. The Future of Children.
Miller, T.R., Viner, J.G., Rossman, S.B., Pindus, N.M., Gellert, W.G., Douglass, J. B.,
Dillingham, A.E. and Blomquist, G.C., 1991, The Costs of Highway Crashes.
(Washington, DC: The Urban Institute).
Mitchell, R.C. and Carson, R.T., 1989, Using Survey Methods to Value Public Goods:
The Contingent Valuation Method. (Washington, DC: Resources for the Future).
Moller, J., 1998, Australian Costs of Injury. (National Injury Surveillance Unit: NISU
website).
Moore, M.W. and Viscusi, W.K., 1990a, Models for estimating discount rates for long-
term health risks using labor market data . Journal of Risk and Uncertainty, Vol. 3, pp.
381–402.
Moore, M.W. and Viscusi, W.K., 1990b, Discounting environmental health risks: New
evidence and policy implications. Journal of Environmental Economics and
Management, Vol. 18, pp. S51–S62.
Moore, R., Mao, Y., Zhang, J. and Clarke, K., 1997, Economic Burden of Illness in
Canada, 1993. (Ottawa: Canadian Public Health Association).
Murray, C.J.L. and Lopez, A.D., 1994, Global Comparative Assessments in the Health
Sector: Disease Burden, Expenditures and Intervention Packages. (Geneva: World
Health Organization).
Newman, J. and Miller, T.R., 1994, Toward a biomechanical injury cost model. Accident
Analysis and Prevention, Vol. 26, pp. 305–314.
Nord, E., 1991, EuroQol: Health related quality of life measurements: Valuations of
health states by the general public in Norway. Health Policy, Vol. 18, pp. 25–36.
OECD, 1988, Route Guidance and In-car Communication Systems. (Paris, France:
Organization for Economic Co-Operation and Development).
Office of Technology Assessment, U.S. C., 1992, Evaluation of the Oregon Medicaid
Proposal. (OTA-H-531). (Washington, DC: US Government Printing Office).
Olsen, J. A., 1993, Time preferences for health gains: An empirical investigation. Health
Economics, Vol. 2.
Patrick, D.L. and Erickson, P., 1993, Health Status and Health Policy: Quality of Life in
Health Care Evaluation and Resource Allocation. (New York: Oxford University
Press).
Phillips, D.E., Langley, J. D. and Marshall, S.W., 1993, Injury: The medical and related
costs in New Zealand 1990. New Zealand Medical Journal, Vol. 106, pp. 215–217.
Rice, D.P., 1966, Estimating the cost of illness. In No. 6, U.S. Public Health Economics
Series. (Washington, DC: U.S.Public Health Service).
Rice, D.P. and Cooper, B.S., 1967, The economic value of human life. American Journal
of Public Health, Vol. 57, pp. 1954–1966.
Rice, D.P., Hodgson, T.A. and Kopstein, A.N., 1985, The economics of illness: A
replication and update. Health Care Financing Review, Vol. 7, pp. 61–80.
Rice, D.P., MacKenzie, E.J., Jones, A.S., Kaufman, S.R., deLissovoy, G.V., Max, W.,
McLoughlin, E., Miller, T.R., Robertson, L.S., Salkever, D.S. and Smith, G.S., 1989,
Cost of Injury in the United States: A Report to Congress. (San Francisco, CA:
Institute for Health and Aging, University of California; and Injury Prevention Center,
The Johns Hopkins University).
Rodgers, G.B., 1993, Estimating jury compensation for pain and suffering in product
liability cases involving nonfatal personal injury. Journal of Forensic Economics, Vol.
6, pp. 251–262.
Ryan, G.A. and Dyke, P.M. (Eds.), 1998, Measuring the Burden of Injury: Proceedings
of a Conference held at The Esplanade Hotel, Fremantle, Western Australia, 15th and
16th February 1996. (Nedlands WA, Australia: Road Accident Prevention Research
Unit, Department of Public Health, The University of Western Australia).
Schwab-Christe, N.G. and Soguel, N.C. (Eds.), 1995, Contingent Valuation, Transport
Safety, and the Value of Life. (Boston, MA: Kluwer Acdemic).
Slovic, P., Fischoff, B. and Lichtenstein, S., 1980, Facts versus fears: Understanding
perceived risk. In Societal Risk Assessment: How Safe is Safe Enough? edited by R.
Schwing and W.A.Albers, Jr. pp. 181–214. (New York: Plenum Press).
Smith, S.V., 1998, Why juries can be trusted. Voir Dire, Vol. 5, pp. 19–21.
Soguel, N.C., 1995, Introduction. In Contingent Valuation, Transport, Safety and the
Value of Life edited by N.G.S.Christe and N.C.Soguel. pp. 1–13. (Norwell, MA:
Kluwer Academic Publishers).
Tengs, T.O., Adams, M.E., Pliskin, J.S., Safran, D.G., Siegel, J.E., Weinstein, M.C. and
Graham, J.D., 1995, Five hundred life-saving interventions and their cost-
effectiveness. Risk Analysis, Vol. 15, pp. 369–390.
Torrance, G.W., 1982, Multiattribute utility theory as a method of measuring social
preferences for health states in long term care. In Values and Long-term Care edited by
R.L.Kane and R.M.Kane, pp 127–156. (Lexington: DCHeath).
Torrance, G.W., Zhang, Y., Feeny, D., Furlong, W.J. and Barr, R., 1992, Multi-attribute
Preference Functions for a Comprehensive Health Status Classification System.
(Working paper No. 92–18). (Ontario: Centre for Health Economics and Policy
Analysis, McMaster University Hamilton).
U.S.Office of Management and Budget, 1994, Guidelines and Discount Rates for Benefit-
cost Analysis of Federal Programs. (Circular A-94). (Washington, DC: Office of
Management and Budget).
U.S.Supreme Court., 1983, Jones and Laughlin Steel Corp. v. Pfeifer. pp. 2541–2558.
Van Beeck, E.F., van Roijen, L. and Mackenbach, J.P., 1997, Medical costs and
economic production losses due to injuries in the Netherlands. Journal of Trauma, Vol.
42, pp. 1116–1123.
Viscusi, W.K., 1988, Pain and suffering in product liability cases: Systematic
compensation or capricious awards? International Review of Law and Economics, Vol.
8, pp. 203–220.
Viscusi, W.K. (1993). The value of risks to life and health. Journal of Economic
Literature, Vol. 31, pp. 1912–1946.
Viscusi, W.K., 1995, Discounting health effects for medical decisions. In Valuing Health
Care: Costs, Benefits, and Effectiveness of Pharmaceuticals and Medical Technology.
Edited by F.A.Sloan. (New York: Cambridge University Press).
Viscusi, W.K. and Magat, W.A., 1987, Learning about Risk. (Cambridge, MA: Harvard
University Press).
Viscusi, W.K. and Moore, M.W., 1989,. Rates of time preference and valuations of the
duration of life. Journal of Public Economics, Vol. 38.
Vodden, K., Miller, T.R., 1993, The Social Cost of Motor Vehicle Crashes. (Final Report
to Ontario Ministry of Transport). Abt Associates of Canada Ottawa.
Watson, W.L. and Ozanne-Smith, J., 1997, The Cost of Injury. (Report No. 124).
(Melbourne: Monash University Accident Research Centre Victoria).
Williams, A., 1995, The Role of the EuroQol Instrument in QALY Calculations
(discussion paper 130). (York, UK: Centre for Health Economics, University of York).
7
Traffic Flow and Safety: Need for New Models
for Heterogeneous Traffic
Geetam Tiwari
INTRODUCTION
Knowledge of fundamental traffic flow characteristics—speed, volume and density—and

associated analytical techniques are essential requirements in planning, design, and
operation of transportation systems. The fundamental characteristics have been studied at
microscopic and macroscopic levels. Existing traffic flow models are based on time
headway, flow, time-space projectory, speed, distance headway and density. These have
lead to the development of a range of analytical techniques: demand-supply
analysis,capacity and level of service analysis, traffic stream modelling, shock wave
analysis, queuing analysis and simulation modelling. The focus of this research has been
homogenous traffic conditions—traffic which consists primarily of cars or motorized
vehicles of similar characteristics. We are now faced with a situation where more than
half of the world’s population is living in megacities in the developing region. Large
differences in income levels and social disparities have lead to the development of ‘cities
within cities’. Each level of the city, with its own level of technology and land-use
patterns, exists in close geographical proximity with other cities of different patterns.
This is reflected in the travel and traffic patterns existing in this region. The same road
space gets used by cars and buses, along with locally developed vehicles for public
transport such as three-wheeled scooter taxis, scooters and motorcycles, bicycles, tricycle
rickshaws, animal and human drawn carts. Infrastructure which is designed on the basis
of homogeneous traffic models has failed to fulfill the mobility and safety needs of this
traffic. Despite low levels of car ownership, accidents and congestion continue to plague
this region. This is the right time to question the applicability of models which have been
developed for a very different situation.
Traffic Flow and Safety

Two basic parameters of traffic flow which influence the safety of road users are traffic
volume or flow measured in terms of number of vehicles per unit time and traffic speed.
The amount of traffic, often called traffic flow or more general exposure, is often treated
as a matter of routine. Relationship between accidents and flow is, therefore, often
assumed to be a simple (linear) relationship. One major reason could be that it seems
obvious that an increase in flow leads to an increase in the number of accidents. If we
have a system without road users it is obvious that it cannot produce any traffic accidents.
And if the number of road users increases this will then obviously lead to increased
probability for an accident. Most modelling work has used flow as one major explanatory
variable. The most used model structure is the exponential (Ekman, 1996).
E(A)= k* exposure y (Kulmala, 1995).
This approach fulfils some basic ideas about traffic and traffic safety: it starts at the
origin, zero flow creates no accident. Furthermore, it is monotonically increasing with
increasing flow, as long as y is positive. However, this assumption is not completely
supported either by basic traffic theory or modern behaviour adaptation theories. The
results from estimations based on empirical data is often that y is slightly less than 1. This
means that accidents increase by increasing exposure less than a linear approach would
have given. This makes intuitive sense also considering the relationship between flow,
speed and fatalities.
Flow, speed and fatalities

The standard traffic flow model assumes a parabolic curve between speed and flow as
shown in Figure 1.
Figure 1 Speed flow relationship (May, 1990)
depending on the degree of heterogeniety in the traffic mix, fatalities and injuries
may decrease even if the exposure of vulnerable road users increases
As number of vehicles increase on a network, speed remains constant as long as the flow
is below a certain value. This condition is defined as level of service A. Speed reduces
Traffic flow and safety 75
gradually until the flow reaches capacity. Consider this relationship with the relationship
of speed and fatalities and injuries shown in Figure 2. The estimates for probability of
pedestrian deaths at different impact velocities are: 5–8 per cent at 30 kph, 25 per cent at
40 kph, 45–80 per cent at 50 kph, and more than 85 per cent at 60 kph (European
Transport Safety Council, 1995; University of Zurich and Swiss Federal Institute of
Technology, 1986.) For car occupants in crashes at 80 kph the likelihood of death is 20
times more than at 32 kph (IIHS, 1987). Clearly, increase in speed is associated with
disproportionate increase in number of fatalities. Also, the safe speed for car occupants is
much higher than for the pedestrian and bicyclists.
Speed influences energy consumption, pollution, noise, vehicle and road maintenance
costs, stress on road users and safety. In general, higher speeds have an adverse influence
on all these factors. The safety of road users is influenced both by the absolute speed of
vehicles and by the variation in speeds among vehicles on the road (Noguchi, 19**).
Other factors remaining constant, higher speeds increase the probability of a crash taking
place and the severity of injury in a crash, whereas a greater variations in speeds of
vehicles only increases the probability of the event. As illustrated in Figure 2, small
reductions in travelling speed result in large reductions in injuries and fatalities in both
urban and rural areas. This is because the stopping distance of a vehicle under braking is
proportional to the square of the original velocity and the damage to human beings is
related to the square of the impact velocity. Lower initial speeds means that the driver has
better control on the vehicle and the vehicle can stop much earlier and reduce the
probability of a crash. In the event of a crash the injuries are less severe at lower impact
velocities.
Figure 2 Speed and pedestrian fatalities
The relationship between flow, speed and fatalities requires further consideration in the
case of mixed traffic and vulnerable road users. A heterogenous traffic mix has an effect
on traffic safety, specially traffic fatalities. Figure 3 shows the distribution of the
percentage of non-motorized vehicles (NMV) fatalities versus the percentage of
motorized vehicles (MV) trips comprising the location’s modal split (Fazio and Tiwari,
1995). Theoretically, no NMV fatalities can result from a striking MV at the origin on the
graph because no MVs exist in the traffic stream at this point. When MVs account for
100 per cent of the trips, no NMV fatalities occur because of the absence of NMVs in the
traffic. Presence of NMVs also has a calming effect on traffic speed. Data from Delhi
also show that as NMV flow increases, the average speed difference between MVs and
NMVs decrease specially on roads where mixing of MV and NMV takes place (high
conflict between MV and NMV). As the speed difference or initial speed reduces,
number of fatalities and injuries reduce. Therefore, depending on the degree of
heterogeniety in the traffic mix, fatalities and injuries may decrease even if the exposure
of vulnerable road users increases.
Models which predict number of fatalities, injuries and accidents based on a linear
relationship with motorization or flow are inappropriate if they do not include speed
implications. Rate of fatalities would depend on how the increased flow affects mean
speed of the traffic stream as well variation of speed of the traffic stream. The modal
diversity present in mixed traffic (primarily in less motorized countries) makes the effect
of speed even more critical because the vulnerable road users (VRUs)—people outside
cars and buses constitute majority of the victims.
Figure 3 Degree of traffic homogeneity
Heterogeneous traffic Flow: present and future

Nearly 60 per cent of world population lives in developing regions-Africa, Asia and
South America (Figure 4).
Figure 4 Proportion of urban population in 1990 by region
models which predict number of fatalities, injuries and accidents based on a linear
relationship with motorization or flow are inappropriate if they do not include
speed implications
These regions are characterized by dominance of large cities—more than 60 per cent of
the urban population resides in million-plus cities. Of the 100 largest cities 62 are in this
region. A large proportion of the population residing in these cities lives below the
poverty line, 29–60 per cent (Figure 5). Therefore, the demand for non-motorized modes
and pedestrians on highways and urban areas is inevitable.
Urban areas in developing countries experience such extremes of wealth and poverty
that they can be characterized as having dual economies. One serving the needs of the
affluent and featuring modern technologies, formal markets, and outward appearance of
developed countries. The other serves disadvantaged groups and is marked by traditional
technologies, informal markets and moderate to severe levels of economic and political
deprivation (Dimitrou, 1990).
Urban poverty, characterized by unemployment, dependence on the informal sector,
low wages and insecure jobs, has a direct bearing on travel and transport demand of a
large segment of the population residing in urban areas. Their dependence on transport
which enables them access to job markets becomes essential for survival. This need is
more critical for them than for those with high income and secure jobs. However, this
segment of population is also transport poor. Even a subsidized public transport remains
cost prohibitive for them.
Figure 5 Population below poverty line
Cities in developing countries are characterized by heterogenous traffic (mix of non-

motorized and motorized modes) and mixed land-use patterns. Non-motorized vehicles
are owned and used by a large section of the population (Figure 6).
Figure 6 Comparison of vehicle composition

Car ownership rates in Asian countries are low compared to North America and OECD
countries. In 1993, it was 29 cars per 1000 residents in East Asian countries compared to
561 cars per 1000 residents in North America, 366 in OECD countries (AAMA, 1995).
Although the greatest growth rate in the number of motor vehicles is expected in many
Asian countries, most of these increases in absolute numbers of vehicle fleets will result
from increases in the numbers of motorized two-wheelers (MTWs) and three-wheelers
(World Resources Institute, 1996). In Thailand, Malaysia, Indonesia and Taiwan, two-
and three-wheelers account for more than 50 per cent of all motor vehicles.
30 per cent of the world population living in urban poverty in cities of developing
countries is transport poor… A sustainable transport system must meet the
demand of this captive ridership of non-motorized transport existing in the cities
of the South.
Non-motorized transport (NMT) constitutes a significant share of the total traffic in many
Asian cities. Shanghai, Hanoi, Kanpur and Tokyo all have a relatively high rate of
bicycle ownership and a high proportion of bicycle traffic (Figure 7). In Indian cities, the
share of NMT at peak hour varies from 30–70 per cent. The proportion of trips
undertaken by bicycles range between 15 and 35 per cent, the share tending to be higher
in medium and small sized cities. The patterns of NMT use changes with growth in city
size. In most NMT-dependent low-income cities, bicycles are used for entire trips (e.g.,
commuting, shopping). In Kenya, despite several constraints, the NMT including walking
are still the prevalent modes that provide more than 45 per cent of all the personal
transport in urban centres. In a high-income city like Tokyo, bicycles are increasingly
used as a feeder mode to rail stations as well as for shopping and other purposes (World
Bank, 1995). Every motorized public transport trip involves access trips by NMT at each
end. Thus, NMT including walking continues to play a very important role in meeting the
travel demand in cities in developing countries.
Table 1 gives selected indicators for a few Indian cities. Regardless of city size, it shows
that nearly 40–60 per cent of households have monthly incomes of approximately US
$50–60. In large cities like Mumbai, Delhi, Chennai, more than 60 per cent of people are
employed in the informal sector. For this population walking and bicycling to work is the
only mode of transport available. Assuming a minimum of 4 trips per household per day
at the cost of Rs 2 (US$ 0.05) per trip by public transport, a household would need to
spend Rs.320 (US$ 8) per month on transport. For low-income people living in the
outskirts of the city, the cost per trip may be two or three times this amount depending on
the number of transfers. On an average, low-income households cannot spend more than
10 per cent of its income on transport. This implies that a household’s income must be at
least Rs.3200 (US$ 80) to be able to use the public transport system at minimum rates.
According to a survey (ORG, 1994), approximately 28 per cent of households in Delhi
have a monthly household income of less than Rs.2000(US$ 50).
Figure 7 Comparison of vehicle composition among cities
Table 1
Indian city indicators
INDICATOR MUMBAI DELHI MADRAS BANGALORE LUCKNOW VARANSI
Population(million)
10.26 8.96 5.65 4.47 1.8 1.08
Household Income Distribution (Quintile Boundaries US $)
I (poorest 20%)
374 290 347 385 291 268
V (richest 20%)
2497 3292 2781 2487 2181 2084
Informal Employment(%)
68 66 60 32 48 49
Motorized Vehicles(per 1000 pop.)
51 205 102 130 130 85
Data from: Society for Development Studies, Delhi
Survey results show that nearly 60 per cent of respondents find the minimum cost of
work trips by public transport (less than Rs.2 per trip) unacceptable (CRRI, 1988). Even
at minimum costs, public transport trips account for 20 to 30 per cent of family income
for nearly 50 per cent of people living in unauthorized settlements. This segment is very
sensitive to the slightest variation in the cost of public transport trips.
The data above show that an estimated 30 per cent of the world population living in
urban poverty in cities of developing countries is transport poor. In this segment it is
harder for individuals and households to save and build up assets, and reduce their
vulnerability to sudden changes/loss in income. Low incomes also make it difficult for
households to ‘invest’ in social assets such as education that can help reduce their
vulnerability in the future. Therefore, access to affordable transport is necessary for
survival. A sustainable transport system must meet the demand of this captive ridership
of non-motorized transport existing in the cities of the South.
URBAN TRANSPORT AND LAND-USE PATTERNS
Heterogenous traffic is likely to continue in less motorized countries (LMCs) because of

the socio-economic realities and compulsions. Many cities in LMCs experience such
extremes of wealth and poverty that they can be characterized as having dual
economies—one for the affluent and the other for the disadvantaged. Very often formal
plans address the needs of one set of city residents, failing to recognize the existence of
the other set who very often form the majority.
this traffic is characterized by lack of any effective channelization, mode

segregation or control of speeds. To the formally trained planner it looks like
chaos moving towards total gridlock. Yet people and goods keep getting through.
And may, by some measures actually be doing better than in some controlled
conditions
The planning framework as adopted in the preparation of master plans in many Asian
cities has been completely divorced from resource assessment. The process also does not
invoke any procedures for involving community and bringing about consensus on
contentious issues. The net effect of the inadequacies of the planning process has been
that majority of urban growth has long taken place outside the formal planning tools.
Informal residential and business premises and developments increasingly dominate new
urban areas. Even in our megacities where half or more of a city’s population and many
of its economic activities are located in illegal or informal settlements, urban planners
still rely on traditional master-planning approaches with their role restricted to servicing
the minority, high income residents. Few weak attempts have been made to bring some
coordination of development and ‘services’ to informal areas through slum improvement
schemes.
Unlike traffic in cities in high income countries (HICs), bicycles, pedestrians and other
non-motorized modes are present in significant numbers on arterial roads and intercity
highways. Their presence persists despite the fact that engineers designed these highway
facilities for fast moving uninterrupted flow of motorized vehicle.
Increase in the level of congestion has been a major concern for planners and policy
makers in metropolitan cities. In Delhi average speeds during peak hour range from 10–
15 kph in central areas and 25–40 kph on arterial streets (CRRI, 1992). Compared to
Delhi, average speeds in other mega cities are less. In 1993, Delhi’s traffic fatalities were
more than double that of all other major Indian cities combined (Indian Express, 1994).
Clearly, criteria for recommending optimal speeds and congestion reduction does not
include desired level of safety, pollution and land-use patterns.
There is ample evidence to illustrate the mismatch between the careful planning and
growing transportation problems. Unless we understand the basic nature of problems
faced by our mega cities, the adverse impact of growing mobility on the environment and
safety would continue to multiply in future.
The existence of an active informal sector introduces a high degree of heterogeneity in
the socio-economic and land-use system. This is assumed to add to our problems of
congestion and pollution. However, the informal sector is an integral part of the urban
landscape providing a variety of services at low costs, at locations with high demand for
these services. Many view hawkers, pavement shops, cycle and motor vehicle repair and
spareparts shops as unauthorized developments along the road that reduce the capacity of
the planned network. However, since the market demands these services, they continue to
exist and grow along arterial roads as well. It is quite clear that long term land-use
transport plans must address the needs of the informal sector.
Understanding Heterogeneous traffic flow in developing country cities

Most developing country cities have been classified as ‘low cost strategy’ cities
(Thomson, 1977). In comparison with cities in the West, these cities consume less
transport energy. High densities, intensely mixed land-use, short trip distances, and high
share of walking and non-motorized transport characterize these urban centres (Newman
and Kenworthy,1989) .
Heterogenous traffic flow consists of modes of varying dynamic and static
characteristics sharing the same road space. Underlying concepts of traffic flow theory in
the US, Europe, and Australia are formed by motorized four-wheel road way traffic
dominating in those areas, i.e., homogenous traffic. All car following, lane changing logic
and system’s measure of effectiveness used in microscopic simulation programmes
ultimately use field data from these countries for caliberation.
In LMC cities, the road network is used by at least seven categories of motorized and
non-motorized vehicles. Vehicles ranging in width from 60 to 2.6 m, and capable of
maximum speeds ranging from 15 to 100 kph, share the same road space. All these
vehicles which have varied dynamic and static characteristics share the same
carriageway. This traffic is characterized by lack of any effective channelization, mode
segregation or control of speeds. To the formally trained planner it looks like chaos
moving towards total gridlock. Yet people and goods keep getting through. And may, by
some measures actually be doing better than in some controlled conditions.
In Delhi different traffic modes are not segregated and there is minimal enforcement of
speed limits. In this situation flow patterns result in a natural optimization of road use due
to self organization by road users. Though aggregate conflict data do not correlate with
fatalities, our data show that Delhi has high number of VRU fatalities. Therefore,
segregation and traffic calming techniques developed for Delhi conditions with special
reference to motorized two-wheelers are desirable. Techniques developed in HMCs do
not address the high volume of motorized two-wheelers and large variations in traffic
composition from site to site.
The peak hour motor vehicle flows in traffic with mix of MVs—cars, buses, two-
wheeler scooters, three-wheeler scooter—have been observed to be very high compared
to homogenous traffic sites of similar street width (Fazio et al., 1998). Having vehicles of
narrow widths in the traffic stream greatly increases the capacity of streets. Narrow
vehicles fill-in the lateral and longitudinal gaps between wide vehicles; heterogenous
traffic uses on-street space more efficiently than homogenous traffic (Figures 8, 9 and
10). Homogeneous traffic flow is modeled on the basis of lane discipline logic. An ideal
lane capacity is estimated of only passenger cars by using passenger car units or
equivalents.
For heterogenous traffic, having an ideal capacity by lane is mis-conceptual because
lane discipline is very loose. Vehicles have varying static and dynamic characteristics.
These share the same road space and move by sharing the lateral as well as the linear
gaps. For example, a motorcycle rider judges whether the lateral distance (width)
between a motor scooter and bus is acceptable to progress on the roadway. Another
motorcycle rider in the same situation would have a different critical width acceptance. If
the width is unacceptable, then an entity is constrained by preceding entities. Critical
width acceptance depends on three items. First, the travel speed of the vehicle/ entity
itself. The next item is the physical width of the vehicle. Distribution of the width
acceptances of specific entity groups is the third item, i.e., driver/rider/pedestrian
behaviour. Each vehicle/entity group has its own critical width acceptance.
Heterogeneous traffic can have many motorized two-wheelers, motorized three-
wheelers, bicycles, non-motorized three-wheelers, cars, buses, trucks, animal-drawn
carts, and human-powered push and pull carts. Additionally, if sidewalk facilities are
inadequate or lacking, this diverse mixture contains significant on-road pedestrian traffic.
In homogeneous traffic, traffic entities form one-dimensional queues (Figure 8); in
heterogeneous traffic, mass queues develop. These queues grow lengthwise as well as
laterally.
Lane discipline is deficient in heterogenous traffic not because driver behaviour is

significantly different, but because heterogeneous traffic consists of entities of
various widths and varying dynamic characteristics
The ‘car following’ notion used in homogeneous traffic flow models is not applicable in
heterogenous traffic (Figure 9). Since cars do not comprise most of the traffic mixture,
‘car following’ is an incorrect term for heterogenous traffic. Secondly, since width of
entities vary greatly in heterogenous traffic, figuring out which leading entity/vehicle it is
following is difficult. Leading entities may run parallel or staggered.
Figure 8 Queuing theory
Figure 9 Car following
Professionals have extensively derived models and algorithms from the ‘lane
changing’ notion of homogeneous traffic (Figure 10). Microscopic studies of this traffic
shows that the time headway between vehicles is an important flow characteristic that
affects safety, level of service, driver behaviour and capacity of a transportation system.
A minimum time headway must always be present to provide safety in the event that the
lead vehicle suddenly decelerates. The percentage of time that the following vehicle must
follow the vehicle ahead is one indication of level or quality of service. The distribution
of time headways determines the requirement and the oppurtunity for passing, merging,
and crossing. The capacity of the system is governed primarily by the minimum time
headway and the time headway distribution under capacity-flow conditions.
Clearly, underlying these concepts is the notion of lane discipline or lack of it. Lane
discipline is deficient in heterogenous traffic not because driver behaviour is significantly
different, but because heterogeneous traffic consists of entities of various widths and
varying dynamic characteristics. With homogeneous traffic, the width range is
approximately 2.1 m for cars to 2.6 m for trucks and buses. Homogeneous traffic drivers
find it optimal and advantageous to adopt lane discipline to traverse the roadway space
given the narrowness of the width range. For heterogeneous traffic, the width range is
approximately 6 m for pedestrians to 4.9 m for overburdened truck trailers. Drivers,
pedestrians, riders and animals find it optimal to advance by accepting lateral gaps
(widths) between preceding entities. Heterogeneous traffic uses road space more
efficiently than homogenous traffic. For this traffic, models based on width acceptance
can ultimately produce a good estimate of roadway capacity and assessments of
operations and safety of various facility designs.
Figure 10 Lane changing
This specific characteristic of heterogeneous traffic highlights the limitations of

conflict technique which has been developed for homogeneous traffic (Tiwari G et al.,
1998). There are very few detailed studies on traffic patterns and their influence on
accidents in LMCs. We use here a study conducted in Delhi which involved conflict
analysis for prediction of fatal crash locations. Peak hour traffic at fourteen selected
locations were videotaped. Trained observers recorded traffic compositions at mid-block,
average space, mean speeds by mode and conflicts by type, reactor mode and cause
mode. The study showed a weak crash-conflict association. Definitions of conflict
developed in HMCs where motorized vehicles are the dominant modes are inadequate in
heterogeneous traffic situations like Delhi. In heterogeneous traffic streams, bicyclists,
motorized two-wheelers and motorized three-wheelers have the freedom of moving
laterally as well. These manoeuvers at low speeds need not necessarily lead to fatal
crashes. Kouabenan (1996) and Salimen et al. (1992) have also reported that fatal
accidents generally differ in causation from non-fatal injuries. The 3.5 metre wide lanes
designed to fit four-wheeled motorized vehicles, such as cars, trucks, and buses can fit at
least two lanes of two wheelers. These space sharing manoeuvers of heterogeneous traffic
are also recorded as conflicts. These manoeuvers at low speeds need not necessarily lead
to fatal crashes. For mixed traffic, we need to develop better understanding of conflicts to
differentiate between conflicts that are a part of normal driving cycle and the ones that
are potential crashes.
These issues convince us that LMCs are experiencing a new phenomenon in road
traffic patterns and accidents for which there is little precedence. These patterns are new
and they need to be understood through careful scientific research.
Safety in Heterogeneous traffic

The proportion of road users killed in various modes of transport as a percentage of all
fatalities in different countries is shown in Figure 11. The data show that non-car
occupants constitute almost 75 per cent or more of all fatalities in most developing
countries. This flows logically from the fact that this class constitutes the majority of road
users however, roads and other transport infrastructure designs address the need of
motorized vehicles only. In addition, because VRUs are not protected by metallic or
energy absorbing materials, they sustain relatively more serious injuries even at low
velocity crashes.
Figure 11 Proportion of road users killed in different countries
In many LMCs highways run through rural areas with high density populations
where most people do not have access to motor vehicles.
It is possible that wherever the proportion of VRUs is high as a proportion of all road
users similar crash rates will be experienced as in India. However, some less motorized
countries (LMCs) do not have bicycle use rates as high as those of countries like India,
China and Vietnam, and these countries have lower involvement of bicycle fatalities
(Tiwari and Saraf, 1997). In these countries where bicycle use rates are lower, it appears
that MTW and pedestrian fatality rates are proportionately higher. Data show that in most
developing countries, the urban poor who are heavily dependent on NMVs, are the
victims of road traffic crashes.
Almost 80 per cent of all cars are owned by 15 per cent of the world’s population
residing in North America, western Europe and Japan. On the other hand, in LMCs like
India and China, less than one in a hundred families owns a car. Car ownership levels in
LMCs are so low that even at a reasonable economic growth rate (say 7–10 per cent per
year) for the next few decades, most families in LMCs are not likely to own a car in the
next quarter of a century. A comparison of per capita levels of vehicle registration and
proportions of cars and motorcycles in a few countries is given in Table 2. These
ownership patterns show that total vehicles registered per capita, and car and motorcycle
ownership levels differ greatly in HMCs and LMCs. These different ownership levels
influence traffic behaviour patterns and composition of traffic fatalities.
Table 2
Vehicle Registration and Proportions
COUNTRY GNP PER TOTAL NUMBER MTWsAS % CARS AS %

CAPITA OF OF TOTAL TOTAL
us$ VEHICLES /1,000 VEHICLES VEHICLES
PERSONS
Japan 34630 640 20 58
USA 24780 740 2 88
Germany 23980 570 9 89
France 23420 520 10 87
UK 18 340 410 3 86
Australia 18000 610 3 76
Republic of 8260 206 24 33
Korea
Malaysia 3 140 340 56 34
Thailand 2410 190 66 16
Philippines 950 32 26 28
Indonesia 810 58 69 15
Sri Lanka 600 50 60 13
China 530 21 40 24
India 320 30 67 14
Vietnam 210 27 91 9
These figures suggest two different phenomena that are relevant to road safety policies. It
appears that total vehicle registration levels remain below 100 per thousand persons in
countries that have per capita incomes of less than US $ 1,000 and that motorcycle
registrations decrease below 20 per cent of the total vehicle fleet only when per capita
incomes are much greater than US $ 8,000. The only exceptions are countries like China
where MTW production or availability has been controlled by government policy. Even
at per capita income levels of US $ 3,000 car ownership levels remain low and the
proportion of MTWs can be more than 50 per cent. Most LMCs, including India and
China, will not reach per capita income levels of US $ 3,000 in the next decade. As
incomes increase, the poorest people in countries like India and China will be able to own
bicycles and those who own bicycles today may opt to buy motorcycles when they
become richer. As the number of poor and lower middle class people in these countries is
larger than that belonging to the upper class, we are likely to witness greater increases in
absolute numbers of bicycles and motorcycles than cars in the next decade or so. Road
safety policies and countermeasures that are based on societies where car fleets constitute
about 80 per cent of the vehicle fleet will not be adequate for most LMCs where MTWs
comprise more than 40–50 per cent of the total number of vehicles.
future traffic models must account for the users of different transport modes
having conflicting requirements… Highway planning standards provide for
services needed by motorized vehicle users. However, there are no standards for
providing services needed by NMT
The prevalent high rates of pedestrian, bicycle and motorcycle traffic in LMCs
(proportions do differ from country to country) result in VRU fatalities constituting 60–
80 per cent of all traffic fatalities (Mohan, 1992). These patterns of traffic and accident in
LMCs are not only different from those that are prevalent in HMCs today, but are also
different from the experiences of HMCs in the past. The HMCs have never experienced
road traffic that includes such a high proportion of motorcycles, buses and trucks sharing
the same road space with pedestrians and bicyclists. In addition, in the earlier part of this
century when the present HMCs had low per capita incomes, motor vehicles (including
motorcycles) were relatively more expensive and not capable of high velocities and
accelerations. Therefore, speeds were lower and number of vehicles using roads was less
than that seen today. In a sense, motor-vehicle technology, roadway quality and social
systems were more compatible. On the other hand, LMCs now have to plan for use of
technologically advanced vehicles using relatively ‘less advanced’ roadways and
enforcement systems.
Because bicyclists and pedestrians continue to share road space in the absence of
infrastructure specifically designed for NMVs, they are exposed to higher risks of being
involved in a road traffic accident by sharing the road space with high speed modes.
Unlike cities in the West, pedestrians, bicyclists and MTWs constitute 75 per cent of total
fatalities in road traffic crashes. Buses and trucks are involved in more than 60 per cent of
fatal crashes. Buses are often very crowded inside and significant proportion of
passengers who die are those who fall from footboards of buses. In addition, many
indigenously designed vehicles (IDVs) such as tempos, jugar are present on roads of
Indian cities because of the absence of efficient and comfortable public transport
services. These IDVs operate as paratransit modes thus serving a useful role in the
context of existing social system (Tiwari, 1994).
In HMCs a very large proportion of the population owns motorized vehicles. In
addition, these countries can afford to have roads parallel to expressways to be used by
local traffic and vehicles not allowed on expressways. In many LMCs highways run
through rural areas with high density populations where most people do not have access
to motor vehicles. Also, many expressways in LMCs do not have parallel road links for
slow and non-motorized traffic. This forces slow and non-motorized traffic to use
expressways and to cross them illegally where that majority of the victims of road
accidents on intercity highways are the vulnerable road users.
FUTURE DIRECTIONS
Various road users have different and often, conflicting requirements. Motorized vehicles
need clear pavements and shoulders, while bicyclists and pedestrians need shaded trees
along the pavement to protect them from the summer sun. Owners of private transport
modes like MTW and automobiles prefer uninterrupted flow, fewer stops and minimum
delays at intersections, whereas public transport buses require frequent stops for picking
and discharging passengers. Motorized four-wheeled vehicles like cars, buses, etc.,
perform better if they move in queues with minimum braking and acceleration. Since our
infrastructure design does not account for the existing conflicting requirements of
different modes, all modes have to share the road space and operate in sub-optimal
conditions.
Experience of past decades of long-term integrated land-use transport plan exercise
suggests that the existence of informal sector and their travel needs must be recognized
for preparing effective plans. This should encourage mixed land-use patterns and
transport infrastructure especially designed for bicycles and other non-motorized modes.
Future traffic models must account for the users of different transport modes having
conflicting requirements. These models must account for the needs of motorized vehicles
for clear roads for uninterrupted traffic flow, at the same time they must address the
needs of bicyclists and pedestrians for shady trees, kiosks for drinks, food and bicycle
repair shops, etc., at shorter distances. Highway planning standards provide for services
needed by motorized vehicle users. However, there are no standards for providing
services needed by NMT. These services mushroom along urban or inter-city highways
to fulfill the demand of road users, however their existence is viewed as ‘illegal
encroachment’ on the designed road space.
Motorized vehicles are designed to operate at much higher speeds for better fuel
economy and emission levels. Roads are also designed to increase throughput of
motorised vehicles only. These measures decrease safety of NMV occupants and
pedestrians sharing the same road space. Therefore, safe facilities—segregated lanes,
convenient crossing opportunities from the point of view of NMV users should form an
integral part of the road designs. At present these facilities are viewed as cost increasing
measures which many developing countries cannot afford due to resource crunch.
Urban streets passing through the commercial development and highways passing
through small towns serve multiple purposes. They carry through traffic. However, the
adjacent land-use generates cross-traffic and demands multiple space usage, for example,
space for parking vehicles, space for hawkers and informal shopping, etc. The existing
design standards do not account for conflicting demand between local traffic and through
traffic resulting in sub-optimal conditions, i.e., long delays for through traffic and safety
hazards for local traffic specially at off peak hours, for both kinds of traffic.
We have to accept the fact that safety has to be promoted in most LMCs within
existing conditions. These include low per-capita incomes, presence of mixed traffic, low
capacity for capital intensive infrastructure and different law enforcement capabilities.
This approach will be important for most LMCs as they are not likely to experience
economic growth rates which puts them at par with HMCs within the next couple of
decades. This implies that pedestrians, bicyclists and motorized two-wheeler (MTW)
riders will remain dominant on LMC roads for many decades. This group of road users
will be called the vulnerable road users (VRUs).
Such traffic systems are very complex and will need new understanding. Therefore
traffic flow and safety models need to developed for this complex traffic to meet their
specific requirements not found in homogenous traffic conditions.
there is a need to accommodate the conflicting requirements ofNMV occupants

and pedestrians, and motorized traffic on our urban and inter-city highways
A major shift is required in design principles itself to promote safety in LMCs. If a large
number of users are pedestrians, bicyclists and other slow moving vehicles then road
designs have to address their needs in addition to the needs of motorized
vehicles.Motorized vehicles can use a longer route and over-bridges, however, a
pedestrian or bicyclist would prefer not to use an underpass or over-bridge just because it
is safer to do so. For this group of users convenience is an overriding priority.
There is a need to accommodate the conflicting requirements of NMV occupants and
pedestrians, and motorized traffic on our urban and inter-city highways. This includes
redesigning the road cross section setting more exclusive space for pedestrians and
NMVs ,and giving pedestrians and bicyclists priority over cars at certain places.
Speed control is perhaps the most important measure for reducing road traffic crashes
in LMCs. Methods to control speed in urban and residential areas should be given the
highest priority. Currently we do not have a good understanding of how this can be done
at low policing levels in the traffic mix seen in LMC urban areas. For rural roads there
are no good designs for non-expressway safety with high VRU participation.
The quality of roads issue has to be addressed in terms of providing better facilities to
non-motorized road users, developing suitable designs for heterogenous traffic and those
for slowing traffic in residential areas. In many LMCs even on national highways,
majority of people killed are pedestrians, bicyclists, those using two-wheelers or involved
in crashes with tractors/bullock carts. Therefore, unless these issues are addressed and
methods developed for area wide safety improvements we will not be gaining much by
concentrating on blackspot treatment.
REFERENCES
AAMA, World Motor Vehicle Data, 1995 Edition. American Automobile Manufactures’
Association: Detroit, USA. 1995.
Central Road Research Institute, Mobility Levels and Transport Problems of Various
Population Groups, CRRI, Mathura Road, New Delhi, India, 1988, p32.
CRRI, 1992, Development of Traffic and transport Flow Data Base for Road System in
Delhi Urban Area (New Delhi: Central Road Research Institute)
Dimitrou, H.T. Transport and Third World City Development. In Planning in Third
World Cities, (H.Dimitrou and G.Banjo editors), Routledge, London, 1990.
Ekman, L., 1996, On the treatment of flow in traffic safety analysis: A non-parametric
approach applied on vulnerable road users. Bulletin 136 (Sweden: University of Lund,
Lund Institute of Technology, Department of Traffic Planning and Engineering), p. 22
European Transport Safety Council, 1995, Reducing Traffic Injuries Resulting from
Excess and Inappropriate Speed. (Brussels: European Transport Safety Council).
Fazio J. and Tiwari, G., 1995, Nonmotorized-motorized traffic accidents and conflicts on
Delhi Streets. Transport Research Record, 1487, 1995, pp. 68–74.
Fazio J., Hoque, MD. M., Tiwari, G., 1998, Fatalities of Heterogenous Traffic in Large
South Asian Cities. In Proceedings of the Third International Symposium on Highway
Capacity, vol.1, edited by Rikke Rysgaard.(Denmark: Copenhagen), pp. 423–436.
Indian Express, 1994, Better Traffic policing urged, 26 February, Insurance Institute for
Highway Safety, 1987, IIHS Facts: 55 speed limit. (Arlington, Va, USA: Insurance
Institute for Highway).
Kulmala, R., 1995, Safety at rural three- and four-arm junctions: Development and
application of accident prediction models, (Espoo: VTT).
May, A.D., 1990, Fundamentals of Traffic Flow.
Mohan, Dinesh, 1992, Safety of vulnerable road users. Indian Highways, 20 (4), pp. 29–
36.
Newman, Peter and Kenworthy, Jeff, 1989, Cities and Automobile Dependence, An
International Source Book. (Brookfield, Vermont, USA: Grower Publising).
Noguchi K., 1990, In search of optimum speed: From the user’s viewpoint. Journal of the
International Association of Safety Sciences, Vol 14, No 1.
Operations Research Group, 1994, Household Travel Surveys in Delhi, Final Report ,
Sept., (New Delhi).
Tiwari, G., 1994, Safety aspect of public transort vehicles in developing countries.
Journal of Traffic Medicine.
Thomson, J.M., 1977, Great Cities and Their Traffic, (England: Penguin).
Tiwari, G, and R.Saraf, 1997, Bicycle Transport and Prevention of Theft—A Case Study
of Five Cities. (The Netherlands: Interface for cycling Expertise, I-CE).
Tiwari G., Mohan, Dinesh and Fazio, J, 1998, Conflict Analysis for prediction fatal crash
locations in mixed traffic streams, Accident Analysis and Prevention, Vol. 30, No.2,
pp. 207–215.
University of Zurich and Swiss Federal Institute of Technology, 1986, The Car
Pedestrian Collision. Interdisciplinary Working Group for Accident Mechanics
(Zurich: Switzerland).
World Bank, 1995, Non-Motorised Transport in Ten Asian Cities, Report TWU 20,
(Washington, DC, USA: The World Bank: Washington DC).
World Resource Institute, 1996, World Resource: A Guide to the Global Enviroment.
(Washington, DC, USA: The World Resources Institute).
8
Folkore and Science in Traffic Safety: Some
New Directions
Murray Mackay
INTRODUCTION
Today there are approximately 700 million passenger cars, trucks and buses and 300
million motorcycles and mopeds (motorized two-wheelers) worldwide. Nobody really
knows how many bicycles there are. Of passenger cars 80 per cent are owned by 15 per
cent of the world’s population living in North America, western Europe and Japan. Along
with food, shelter, security and health, individual mobility is a basic human need; one on
which our other basic needs are in part dependant. Indeed historically, development has
been almost synonymous with improvements in transport from the trans-Asian Silk Route
of Marco Polo to the farm-to-market roads in early United States to the trans-European
motorway network aiming to link west Europe more firmly to East Europe.
A major side effect of increased mobility is increasing exposure to injury in traffic
crashes. Using the measure of Disability adjusted life years (DALYs) Murray and Lopez
(1996) predicted that death and disability resulting from road accidents (in comparison
with other diseases) will rise from 9th to 3rd rank between 1990 and 2020. Road
accidents as cause of death and disability would rank below heart disease and clinical
depression, and ahead of stroke and all infectious diseases (see Figure 1).
What is clear from even the most superficial view of the epidemiology of road
accidents is that the characteristics of this form of injury vary enormously between
country and region. It is therefore wrong to describe these phenomena as an unified
subject with well defined chracteristics and a universally applicable array of counter-
measures. Conventional wisdom on the subject has evolved in western Europe and North
America in the context of fifty years of modest growth rates and currently very high
levels of car ownership and use in comparison to the rest of the world. At present, neither
conditions pertain in most countries. This paper argues that a fundamental review is
required of the nature of traffic injury in low income countries (LICs) and new strategies
need to be developed to diminish traffic injuries in LICs. For too long consultants have
been applying western counter-measures to traffic accident problems of LICs. New
knowledge, thus new strategies need to be developed within the political and cultural
framework of each country, which are more appropriate for the real traffic accident
problems faced by most of the world.
for too long consultants have been applying western counter-measures to traffic
accident problems of LICs. New knowledge and thus new strategies need to be
Folklore and science in traffic safety 93
developed within the political and cultural framework of each country, which are
more appropriate for the real traffic problems accident faced by most of the world
Figure 1 Change in rank order of disease burden for 15 leading causes 1990–
2020
BACKGROUND
At the macro level traffic-related deaths are the only variable which are recorded
worldwide, and even deaths are likely to be seriously under-reported in many LICs
(Mackay, 1984). Sample studies suggest that the following ratios for survivors occur in
most traffic environments:
Fatality 1
Major Permanent Disability 3
Hospital Admission 10
Injury Requiring Outpatient Treatment 30
These ratios are of course influenced by the quality of care available, reporting procedure
and precise definitions of the categories described. Table 1 indicates current trends of
fatalities in the 15 member countries of the European Unionand the United States.
Table 1
Traffic fatalites in industrial countries
YEAR
1990 2000 2010 2020
EU 55K 40K 30K 25K
USA 45K 42K 40K 38K
Data from: Various national and EU reports
Within the European Union there is a six-fold variation in fatality risk between the best
and worst countries, in terms of fatalities per motor vehicle kilometres travelled (FERSI,
1997). It is likely that these variations will diminish over the next two decades, the
general trend in terms of fatalities is likely to be downward in the industrialized countries
of Europe, North America, Japan and Australia.
This is not the case in low income and middle income countries. By 1990 these
countries (described, using the United Nations’ collective euphemism, as ‘developing’)
accounted for 70 per cent of the world’s traffic fatalities. Currently that proportion is 84
per cent and by 2010 it is likely be over 90 per cent, as shown in Table 2.
Table 2
Traffic fatalities worldwide
YEAR
1990 2000 2010 2020
Industrialized 150K About the same or little less
Developing 350K 800K 1.3M 1.9M
Total 500K 950K 1.4M 2M
Data from: Various national and UN reports
These projections are partly determined by growth in motor vehicle ownership. Table 3
gives a consensus of projection for a selection of motorized and developing countries.
Clearly a number of LICs are in the process of doubling their number of cars over a
period of about 10 years. Such projections however, fail to give a realistic picture of
either the nature of traffic in LICs, or the nature of traffic accidents or the trends in traffic
casualties for the future, because they do not include the most commonly used vehicles in
the country/region.
Table 3
Forecast of car populations 1995–2005
COUNTRY 1995 2005 PERCENTAGE CHANGE

(in thousands)
Canada 13,850 15,000 8.3
USA 148,200 1 56,000 5.4
Germany 40,500 44,000 9.8
Italy 30,000 32,000 6.7
UK 24,960 27,500 10.2
Japan 44,680 55,880 25.0
Mexico 8,060 10,900 35.2
Brazil 12,500 15,800 26.4
Poland 7,340 11,160 52.0
Malaysia 2,500 4,930 97.2
China 2,700 7,850 190.7
India 3,630 8,120 124.0
Data from : National reports, AAMA annual reports
TWO WHEELED VEHICLES
The number of cars, trucks and buses around the world is reasonably well documented,
although even with such vehicles there are serious uncertainties for some countries.
Vehicle registration data at the national level is often incomplete in some LICs, the age of
motor vehicles is undoubtedly much longer in LICs than in motorized countries, often
around 20 years, and thus projections based on annual new registrations, a technique used
in some countries for estimating the vehicle fleet are very uncertain.
When you attempt to quantity the population of two-wheeled vehicles even rough
estimates become very uncertain. No good figures exist as to the working life of a
motorcycle, registration data in many countries is fragmentary, under reporting being
commonplace. Contrary to popular belief the highest volume motor vehicle ever made is
not the Volkswagen Beetle, but the Honda Super Cub in its various national versions,
which has had a production volume of 24 million, over 50 per cent more than that of the
Beetle, and the Honda machine is still produced in significant quantities.
prediction of future injury trends in LICs based on car, truck and bus populations
are unlikely to be accurate, as their role, both in traffic and in accidents, is
relatively small
Table 4 illustrates the importance of the motorized two-wheeler in various countries.

These data are taken from Mohan and Tiwari (1998) and they illustrate the minor
importance of the passenger car in terms of the total vehicle fleet in LICs. Hence
prediction of future injury trends in LICs based on car, truck and bus populations are
unlikely to be accurate, as their role, both in traffic and in accidents, is relatively small.
Predictions of growth in two-wheeled vehicle use and particularly the transition from foot
to bicycle to motorcycle, as countries progress along the mobility spectrum, are
problematic. However as Mohan and Tiwari note, the data in Table 4 suggest that
motorcycle registration decrease to less than 20 per cent only when per capita income
exceeds US $8,000. China is an exception, but that is beacause motorcycle use is
specifically controlled by government policy. Even at income levels of US $33,000 per
capita the motorcycle part of the vehicle fleet can be greater than 50 per cent. Hence, it
seems likely that in many LICs over the next two decades we will see greater increase in
the use of bicycles and motorcycles than cars and trucks.
This is likely to be especially true in the growing megacities in low motorized
countries (LMCs). In such cities as Shanghai, Beijing, Delhi, Bangkok and Manila the
road infrastructure is already totally inadequate for the current traffic volumes and this
leads to a natural preference for two-wheeled vehicles as the only way of minimizing
delays from congestion. The development of a motorcycle taxi service in Bangkok is just
a sign of adaptation to those demands.
Table 4
Vehicle ownership in different countries
COUNTRY GNP PER TOTAL NUMBER MTWs AS % OF CARS AS % OF

CAPITA OF VEHICLES PER TOTAL TOTAL
(US $) 1000 PERSONS VEHICLES VEHICLES
Japan 34,630 640 20 8
USA 24,780 740 2 88
Germany 23,980 570 9 89
France 23,420 520 10 87
UK 18,340 410 3 86
Australia 1 8,000 610 3 76
Korea 8,260 206 24 33
Malaysia 3,140 340 56 34
Thailand 2,140 190 66 16
Philippines 950 32 26 28
Indonesia 810 58 69 15
Sri Lanka 600 50 60 13
China 530 21 40 24
India 320 30 67 14
Vietnam 210 27 91 9
Data from: Statistics are taken from country reports, international agency publications and
conference papers, year range 1992–1995, (Mohan and Tiwari, 1998)
Beyond the motorcycle, non-motorized traffic constitutes a very signifacant share of

urban travel in many Asian cities. Mohan and Tiwari note that in Indian cities the
proportion of trips undertaken by bicycles ranges from 15 to 35 per cent. Even in high
income cities like Tokyo the bicycle is significant as a travel mode for shopping and as
feeder at each end of a railway journey for commuters. This contrast with HICs are
reflected in the varying characteristics of road fatalities. Table 5 gives the relative
proportions of the various categories of road users for a selection of cities and countries.
Table 5
Road fatality categories as proportion of all road fatalities
PEDS CYCLISTS MTWs 4-WHEELERS OTHERS

Delhi(1994) 42% 14% 27% 12% 5%
Thailand(1987) 47% 6% 36% 12% —
Colombo (1991) 38% 8% 34% 14% 3%
Malaysia(1994) 15% 6% 57% 19% 3%
Ghana(1996) 15% 50% 4% 25% 4%
Barcelona(1996) 25% 3% 42% 30% —
Japan(1996) 25% 10% 19% 45% 1%
Netherlands(1990) 10% 22% 12% 55% —
Australia (1990) 18% 4% 11% 65% 2%
Great Britain(1997) 27% 5% 14% 53% 1%

USA(1996) 13% 2% 5% 79% 1%
Data from: National reports, UN annual reports
These examples reflect the great preponderance of that collective group, the vulnerable
road users, pedestrians, cyclists and motorcyclists—together being numerically far more
important in LMCs than vehicle occupants. Even in highly motorized countries however,
these vulnerable road users constitues 20 to 46 per cent of all road fatalities.
ACCIDENT RATES
Fatality and accident rates of various types are used to assess changes over time, make
comparison between diffrent countries and to described relative risks, often
inappropriately and drawing incorrect conclusions.
Two main rates are most useful in assessing traffic safety and personal safety. Traffic
safety is a measure of how safely the road transport system is performing. It is commonly
measured in terms of death per number of registered motor vehicles, or somewhat better,
using deaths per number of vehicle kilometers travelled. This latter indicator is a good
measure of how the road system is functioning, but in most parts of the world such
exposure data is not available. Even when it is available it is often based on sale of fuel
with consequent uncertainties as to how that variable relates to actual mileage covered.
the benefits of behavioural change programmes have often been over stated and
inadequately evaluated. This has resulted in wasted resources and neglect of other
more successful scientific strategies
The second rate often used is an assessment of individual risk on personal safety, and is
the number of road deaths per unit population. It follows that:
Personal Safety=Traffic safety x Motorization
Where motorization is the number of motor vehicles per head of population.

Table 6 lists these various rates for a selection of countries. Many intriguing
comparison can be drawn. Clearly as motorization increase, deaths per vehicle decline
and the traffic system as it develops becomes safer. Even on that measure however, there
are interesting differences; for example deaths per number of motor vehicles registered in
the USA is significantly higher than for a number of EU countries (e.g., Britain) and
Japan. Traffic safety politicians in the United States argue that a better parameter is to use
deaths per hundred million kilometres travelled, that fatality rate for the United States in
1996 being 1.06. That number compares very favourably with the EU average of 1.8 (see
Figure 2), with only Sweden and the UK having a lower rate.However, in terms of
personal safety, the United States has over twice the death rate of India, 50 per cent
higher than Kenya and on par with France. This arises because of the higher levels of
exposure in the USA because of the enormous amount of road travel which occurs. The
idea of exposure control, as a traffic safety strategy in North America, clearly finds little
favour.
Table 6
Fatality rates, traffic safety and personal safety
COUNTRY VEHICLES PER DEATHS PER DEATHS PER DEATHS

1000 PEOPLE 1000 MVs MILLION POP. PER YEAR
USA (1996) 760 0.21 158 41,907
Australia 647 0.13 85 10,805
(1998)
Great Britain ) 481 0.13 64 3,599
(1997
France (1994) 568 0.26 148 8,533
Thailand 219 1.17 257 15,176
(1994)
Saudi Arabia 167 1.44 241 4,077
(1994)
India (1993) 25 2.75 67 59,300
China (1993) 19 2.72 53 63,508
Kenya (1993) 15 7.13 108 2,516
Ethiopia (1990) 1 17.2 23 1,169
Mohan and Tiwari (1998) make the point that all such comparisons are questionable
when great differences in the mix of the vehicle fleet are present. To illustrate the point
they calculate road user specific fatality rates, shown in Table 7.
These rates show that the car occupant fatality rate is 30 per cent more in Delhi than in
the USA, but the motorcycle rate is 50 per cent less. The pedestrian death rate in Delhi is
3.5 times that of the United States, but if the exposure of the walking population is much
greater, which clearly is, then that high rate does not necessarily translate into a high
individual risk. Clearly without more accurate measures of exposure, comparisons
betwen countries with very disparate traffic paterns do little to illuminate genuine
differences and, particularly if deaths per number of motor vehicles is the sole measure of
comparison, can lead to profoundly wrong conclusions.
Figure 2 Fatality rate per 108 motor-vehicles km (1994)
Table 7
Road user specific fatality rates
FATALITIES CAR OCCUPANT MTWs PEDESTRIAN

PER 10,000 FATALITIES PER FATALITIES FATALITIES PER
MVs 10,000 CAR PER 106MTWS 106 POP.
USA 2.1 1.8 600 21
(1995)
Malaysia 5.5 3 741 40
(1994)
Delhi 10 2.3 402 75
(1991)
TRAFFIC INJURIES
The fact of gross under reporting of surviving casualties from road accidents is a
worldwide problem. In Britain about a third of slightly injures pedestrians and
motorcyclists, and about 50 per cent of cyclists who are treated by emergency room
personnel do not feature in police reported data. In LICs the problem of under reporting
of casualties in police data is likely to be profound. A comparison of ratios of fatal:
serious: slight casulties between Nigeria and Britain gave the following results.
NIGERIA BRITAIN
Pedestrians 1:2:3 1:8:22
Cyclists 1:4:3 1:16:61
Motorcyclists 1:3:2 1:17:42
Vehicle Occupants 1:2:2 1:13:50
Clearly the outcomes of road accidents in Nigeria cannot be as different as these ratios
imply, and hence major under reporting of surviving casualties must be the appropiate
conclusion.
Fatalities are different in kind from injury producing accidents and do not follow the
same trends. Figure 3 illustrates for Japan how fatalities have declined markedly during
the last 7 years, but in contrast, injuries have continued to climb. The same is true of
Great Britain over the last decade.
Figure 3 Traffic accident changes in Japan Data from: Traffic Accidents In

Japan, Ministry or Transport, Tokyo, 1999
The health sector has neglected traffic injury, although it is manifestly a major
public health problem
Thus in the context of LICs the absence of agreed definitions of injury severity and
pervasive under reporting of casualties in police-collected data sources mean that the the
true dimensions of the traffic injury problem are poorly documented and understood.
A FRAMEWORK FOR EFFECTIVE POLICIES
Fundamental to developing an effective response to road traffic death and injury is to

recognize that road accidents are capable of being studied and documented like any other
disease and that appropriate counter-measure can be successfully applied. The
epidemiology of traffic injury is very incomplete, and particularly because of the
profound differences between countries at various levels of motorization, the
characteristics of traffic crashes vary profoundly from country to country.
What is most apparent is that vulnerable road users, pedestrians, cyclists and
motorcyclists on a global basis are far more numerous as traffic casualties than vehicle
occupants. Particularly in LICs where traffic is much more mixed, where formal rules are
not applied, but informal rules develop between different road users, the counter-
measures traditionally applied in HICs may not be appropriate and may well not work.
The benefits of behavioural change programmes have often been over stated and
inadequately evaluated. This has resulted in wasted resources and neglect of other more
successful scientific strategies. This does not mean that behavioural programmes should
not be introduced. It does mean however, that they must be targeted, realistic and always
evaluated for effectiveness. In the long run behavioural change must be the way towards
greater safety on the world’s roads. Those behaviours which prove to be successful in one
country may well not be successful in other environments and cultures.
Historically traffic safety has been perceived largely as a public sector undertaking,
centred in departments of transport. The health sector has neglected traffic injury,
although it is manifestly a major public health problem, and generally the focus of
government, reflecting the perceptions of politicians and the community, has been to
leave traffic safety as a public sector activity, dominated by traffic engineering and
behavioural control counter-measures.
Until relatively recently the professions and the private sector have not been prominent
in developing initiatives in trafic safety. The training of doctors, automotive engineers
and indeed traffic engineers hardly addresses the issues of traffic accidents and injuries.
Few universities have any teaching or research activity in traffic safety. Such private
sector ventures as do occur are usually focused on specific local problems or campaigns.
Haddon in 1964 applied the host, agent, environment framework of classical
epidemiology to motor vehicles crashes acting on the pre-crash, crash and post-crash
aspects of an accident. Although that nine-phase structure provides amost useful way of
thinking about counter-measures, and forced a fundamental rethink of the nature of
causation, it lacks a specific recognition of the importance of exposure control as an overt
strategy. Perhaps a more inclusive framework is the five strategies outlined by the GTST
(1988):
• Exposure control
• Crash Prevention
• Behaviour Modification
• Injury Control
• Post-Crash Injury Management
Bounding these five strategies are the necessary accident and control data delineate and
quantify the problem. Furthermore, the introduction of any counter-measures should
always be accompanied by evaluation of any changes, which actually occur in deaths and
morbidity.
ultimately succesful programmes will be those which develop locally and at the
national level. They cannot be imported from outside
Because of the profoundly different characteristics of traffic injuries between low

motorized and highly motorized countries and the great preponderance of vulnerable road
user casualties in LMCs, the actual counter-measures developed, their relative priorities,
and the nature of infrastructure investment should be very different from the routes
followed historically in HMCs. The first stage is to develop a greater understanding of
traffic and acident epidemiology specifically in LMCs.
Necessary to that process is to encourage a more scientific approach by both national
goverments and national safety councils with political recognition of the importance of
traffic injury control. International agencies are limited in what they can offer in this area.
Long-term effective solutions are likely to be home grown and dependent more on
professional expertise developed at the national level. Global professional training in the
various dimensions of injury control is vital, but so is a recognition of the great difference
between countries in terms of the characteristics of crashes and the effectiveness of
various counter-measures. Ultimately succesful programmes will be those which develop
locally and at the national level. They cannot be imported from outside.
REFERENCES
FERSI, 1971, Road Safety Research and Policy in Europe. (Crowthorne, UK: Fourm of
Road Safety Research Institutes. TRL Inc.).
GTST, 1988, Reducing Traffic Injury: A Global Challenge. (Melbourne, Australia: Royal
Australasian College of Surgeons).
HaddonW., Suchman, E.A., and Klein D., 1964, Accident Research Methods and
Approaches. (London: Harper and Row Limited).
Mackay, M., 1984, The Collision Performance of Road Vehicles with Particular
Reference to Requirements of Developing Countries. (Rome, Italy: International Centre
for Transportation Studies Series).
Mohan D and Tiwari G., 1998,. Traffic safety in Low income countries. In Reflections on
the Transfer of Traffic safety Knowledge to Motorising Nations. (Melbourne, Australia:
GTST). pp. 27–56.
Murray, C.J.L. and Lopez, A.D. (eds), 1996, The Global Burden of Disease, Harvard
(Boston, USA: School of Public Health).
9
Demands to Vehicle Design and Test
Procedures for Injury Control of Vulnerable
Road Users in Traffic Accidents
Dietmar Otte
INTRODUCTION
The proportion of pedestrian fatalities relative to all road user fatalities has been recently
estimated by Mackay (1997) as being in the range of 16 to 36 per cent for motorized
countries in Europe and USA.
Each year in the European Union, approximately 7,000 pedestrians and 2,000 cyclists
are killed as a result of being struck by a motor vehicle, accounting for about 20 per cent
of total road traffic deaths (ETSC, 1999). In addition, each year an estimated 400,000
pedestrians and cyclists are injured by vehicle impact. The majority of pedestrian
accident casualties are the very young and the very old, and injuries are sustained as a
result of impact by a passenger car front. The European Commission is considering
recommendations for pedestrian test procedures that will define performance targets
aimed at reducing these pedestrian casualties in Europe. The total number of cars sold in
the European Community market is 12 million per year, so this legislation may result in
significant benefits.
In the past, much research has been done in Europe, USA and Japan investigating
causes of pedestrian injuries and the biomechanics of pedestrian impact. In Europe, a test
procedure was developed by the European Experimental Vehicles Committee (EEVC,
1994). This work has been co-ordinated by research institutes representing the
government and car industry in the EEVC Working Group 10 with a view to creating a
legislative standard for passenger cars and their commercial vehicle derivatives (Ref.
203). This standard, which consists of a number of component tests simulates three
common mechanisms of injury: a) Lower leg/knee joint impact against the bumper b)
Upper leg/hip joint impact against the bonnet leading edge, and c) Head impact (child
and adult) against the bonnet and wing tops.
These tests and the associated injury criteria, have been written into a draft European
directive, and are proposed as a requirement for new vehicles models. This draft
standardizes test procedures simulating a 40 km/h collision between a motor vehicle and
pedestrian.
Annual benefits that will be gained in terms of reduced fatal and serious injuries
experienced by pedestrian and cyclist accidents based on such test regulation has been
predicted by several institutes, i.e. MIRA (1997) has estimated a number between 3,795
and 14,914 injured persons. Total additional benefit gained across Europe for this 15 year
Demands to vehicle design and test procedures 105
period is predicted to be between ECU 5.7 billion and ECU 33.1 billion.
The European Transport Safety Council believes that the introduction of legislation
requiring new cars to pass EEVC tests is one of the most important actions that the EU
could take to improve road safety (ETSC, 1998). Using up-to-date casualty information
from International Road Traffic Accidents Database (IRTAD), ETSC estimates that if all
cars on the road today provided such protection then 655 to 2,226 fatalities and 21,548 to
24,944 serious pedestrian and cyclist casualties could be prevented annually, resulting in
an annual cost reduction of ECU 3.7 billion to ECU 5.1 billion, respectively
CURRENT DRAFT EUROPEAN PEDESTRIAN DIRECTIVE
The directive stipulates three types of component test for vehicles in M1 and NI
categories. These are outlined in Figure 1.
Figure 1 Pedestrian protection test methods proposed by EEVC WG 10
each year in the European Union, approximately 7000 pedestrians and 2000
cyclists are killed as a result of being struck by a motor vehicle
Legform to bumper impact

A minimum of three tests must be carried out at points judged most likely to cause injury,
on each test type. This is carried out by striking the front of the vehicle, or a
representative section of the vehicle, in a longitudinal direction with an impact device
representing the upper and lower leg two rigid tubular segments and the knee joint at a
fixed impact velocity of 11.1 m/s (40 km/h). These are connected by a replaceable,
deformable element which allows relative shear and bending displacement between
segments under impact conditions. The legform with a total weight of 13.4 kg is enclosed
in a soft foam covering with rubber skin.
The relative bending and shear displacements between the rigid segments are
measured, and an accelerometer near the top of the lower leg segment is measuring the
acceleration in the impact direction. During the test, the maximum dynamic angular
displacement measured between the upper and lower leg segments must not exceed 15°,
and their maximum dynamic shear displacement must not exceed 6 mm. At the same
time, the acceleration at the upper end of the lower leg must not exceed 150 g.
Upper legform to bonnet leading edge impact

The device consists of a rigid cylindrical segment carrying an impact element in the form
of a simply supported beam on its front face. The beam’s two supports incorporate
transducers to measure the force exerted at the interface, and the beam itself is strain
gauged to measure the bending moment exerted in it. The beam is covered by a layer of
foam with a rubber skin. The rear segment of the upper legform, reacting through the
force transducers to the forward cylindrical beam, incorporates attachments for ballast
weights by which the mass of the device can be varied.
The weight of the device and its impact velocity (up to 40 km/h) are determined
according to the geometry of the front of the vehicle, in accordance with a set of curves
published in the directive.
During the test, the sum of the forces exerted between the beam and the rigid segment
must not exceed 4 kN, and the bending moment in the beam must not exceed 200 Nm.
Headform to bonnet top impacts

These are carried out by striking points on the bonnet and top of the wings with either a
child headform (2.5 kg) or adult headform (4.8 kg) at a velocity of 40 km/h in free flight
at impact.
The two headforms are made of semi-rigid polymer material, covered with an
elastomeric skin. A metal insert in the centre of the sphere carries three accelerometers
measuring the load for the impact direction and also to the other two principal axes.
A minimum of nine tests must be carried out with each type of head impactor, within
areas marked out on the top of the bonnet defined according to a wrap-around distance,
starting from the ground plane below the bumper. This wrap-around distance is between
1000 mm and 1500 mm for the child headform and between 1500 and 2100 mm for the
adult. If the lower edge of the windscreen lies within the child or adult warp-around
distances, this will limit the extent of the head impact zone.
In order to pass the test, the Head Performance Criterion measured by the headform
must not exceed 1000 mm.
REVIEW OF CURRENT STATUS
Recent accident studies have been analysed, showing among other findings a decline in
the proportion of injuries caused by the bonnet leading edge of modern streamlined
passenger cars. Moreover, it is found that the windscreen and the A-pillars of cars are
also important injury areas, but these are not covered by EEVC test methods (Otte, 1999).
The need for more research in the future is recommended by many scientific authors.
EEVC working group 17 was created in 1998 for analysing the current situation on
new accident data, biomechanics and test specifications in the test method. Modifications
were made on the test procedure concerning optimised material conditions for decreasing
special vehicle designs, but the windscreen impact at higher speed levels has not been
included. A modified procedure based on the final version of the report will be available
by mid–1999 (EEVC WG 17, 1998).
In general it is difficult to estimate the exact cost benefit of such test procedures,
because changes in car design resulting from these test requirements will not be known in
advance. The spectrum of injury occurrence and injury patterns depend on the shape,
stiffness and localization of impacted area on the vehicle front and are also influenced by
the energy of impact, and biomechanical limits of the bony structures of the human body.
These are also likely to change with the introduction of the standard.
Another aspect of the proposed regulation is that it is based on analyses of pedestrian
accidents only. There is no mention of bicycles with their specific impact points on the
car front in the reported statements for the regulation.
In this paper an overview of the current injury situation is given on the basis of in-
depth investigations carried out by Accident Research Unit at the Medical University
Hannover ARU-MUH, Germany. This work (Otte, 1994) has been financed by the
German Federal Highway Research Institute (BAST) for more than 25 years. The results
of pedestrian accidents analyses have also been integrated into the studies by EEVC.
With such an ongoing project a comprehensive investigation of the accident situation can
be conducted by comparing different time histories, including old and new car designs.
proposed regulation is …based on analyses of pedestrian accidents only. There is

no mention of bicycles
The accident documentation was compiled according to a statistical random sampling

procedure, using a weighting procedure, that can be considered as representative for the
area under investigation, namely the city and county of Hanover.
In order to gain an initial overview of the situation of pedestrians, the relationship
between the collision speed of the cars and the total degree of severity of injury—
Maximum Abbreviated Injury Scale, MAIS—to the pedestrian was recorded( American
Association for Automotive Medicine, 1990).
Importance of car shape areas with bumper and bonnet leading edge
In order to demonstrate those injuries to pedestrians which are influenced by the design
of the front edge of the car bonnet, the latter were investigated in the context of its
frequency as cause of injury; the team had in each case linked the injury to the impacting
part of the vehicle respectively the road.
For comparison of the temporal development of the vehicles, cars involved in accidents
were differentiated by model of car, i.e. the year when the model was introduced in the
market, the actual year of construction of the car was not taken into consideration (Otte,
1998). The vehicle groups were defined as follows:
cars released in the market before 1990, so called ‘old vehicle’ n=615
cars released in the market in 1990 or later, so called ‘new vehicle’ n=80
It can be established, that there were fewer impacts with the front edge of car bonnets in
the case of new vehicles. Injuries due to impact with the bonnet in ‘new vehicles’ were
recorded at 4.4 per cent, while in older vehicles the figure was 19.7 per cent. It should,
however, be borne in mind at this point that an impact was only recorded if it led to
injury.
The frequency of this kind of injury proved to be relatively independent of the height
of the pedestrian. Impacts with the bonnet mainly causes injuries to lower extremities of
the body of persons above the height of 120 cm, while shorter persons frequently sustain
injuries to the thorax and arms. The influence of the speed of impact can be clearly
ascertained, 30.3 per cent of collisions at speeds of over 40 km/h involving older vehicles
and 13.9 per cent involving modern vehicles resulted in injury.
The relationship between the collision speed and the severity of injury sustained by the
pedestrian according to the Abbreviated Injury Scale is shown in Figure 2.
Figure 2 Injury severity grade MAIS for different groups of collision speed
It was established that at impact speeds of less than 40 km/h there was hardly any risk of
the pedestrian sustaining extremely serious injuries (MAIS 5/6). At an impact speed
between 41 and 60 km/h there was a spread of 30 per cent minor injuries (MAIS 1), 64.7
per cent severe injuries (MAIS 2–4) and 5.3 per cent extremely severe or fatal injuries
(MAIS 516). At a speed above 60 km/h, 24.2 per cent of the pedestrians sustained
extremely serious injuries or were killed.
In a traffic scene 82.6 per cent of impact speeds can be established below 40 km/h and
only 17.4 per cent above 40 km/ h (Figure 3).
Figure 3 Cumulative frequency of collision speed (100 per cent all vehicles)
Accident Research Unit Hanover
However, 66.3 per cent of pedestrians with severe head injuries AIS 2 and above (AIS
2+) suffered these injuries at impact speeds of up to 40 km/h (Figure 4); 33.7 per cent
received these injuries at impact speeds above that level.
it was established that at impact speeds of less than 40 km/h there was hardly any
risk of the pedestrian sustaining extremely serious injuries
There proved to be a lower incidence of injury to the head, thorax and legs in post-1990
models (Table 1); the incidence of injury to the pelvis was slightly lower at 15.3 per cent
compared to 19 per cent in pre-1990 models. However, the percentage of injuries to the
neck, 3.5 per cent, was remarkably higher for new cars compared to 2.6 per cent for old
cars.
Figure 4 Cumulative frequency of collision speed (1 00 per cent all vehicles

collided with pedestrians AIShead 2 and above) Accident Research Unit
Hanover
Table 1
Frequencies of injured body regions for new and old cars (100% all pedestrians each
vehicle group)
INCIDENCE OF INJURY ACCORDING INTRODUCTION OF CAR MODEL IN

TO AREA OF BODY THE MARKET
Pre-1990 Post-1990
Total number 615 80
Head 60.7% 58.6%
Neck 2.6% 3.5%
Thorax 23.6% 18%
Arms 44.8% 45.4%
Abdomen 8.6% 7.5%
Pelvis 19% 15.3%
Legs 77% 65.7%
When looking to accidents at collision speed of up to 40 km/h (n=386 pre-1990 models
and n=49 post-1990 models), similar situation can be recognized (Table 2), injuries to the
thorax and legs occur less frequently and neck injuries occur with a higher frequency in
post-1990 models.
Table 2
Frequencies of injured body regions for new and old cars (1 00% all pedestrians each
vehicle group) for collision speed of up to 40 km/h

Pre-1990 Post-1990
Total number 386 49
Head 55.2% 57.7%
Neck 1.9% 3.2%
Thorax 19.6% 18.2%
Arms 42.6% 43.4%
Abdomen 5.9% 6.5%
Pelvis 1 7.9% 16.2%
Legs 75.3% 65 %
Bonnet leading edge
as cause of injury 16.8% 4.5%
Accidents at speeds of more than 40 km/h show a clearly higher incidence of injuries for
all body regions (Table 3). At speeds of more than 40 km/h there were head injuries over
80 per cent of cases, compared with accidents at lower impact speeds in nearly 60 per
cent (Table 2). The head has the highest injury frequency, especially for newer car
models. Injuries to the pelvis are, however, less frequent in post-1990 cars. Proportion of
injuries caused by the front edge of the bonnet for new cars was 4.5 per cent, clearly
lower than that for older vehicles (16.8 per cent) (see Table 2).
Table 3
Frequencies of injured body regions for new and old cars (100 per cent all pedestrians
each vehicle group) for collision speeds of above 40 km/h

Pre-1990 Post-1990
Total number 134 15
Head 81.9% 100%
Neck 7.7% 12.4%
Thorax 40.7% 34%
Arms 63.1% 70%
Abdomen 17.8% 12.8%
Pelvis 27.5% 1 8.6%
Legs 85.9% 69.3%
Bonnet leading edge 30.3% 13.9%
as cause of injury
The incidence of injury through impact with the bonnet leading edge increases at higher
speeds (Table 3). In the case of pre-1990 cars 30.3 per cent of pedestrians were injured on
impact with the bonnet at speeds above 40 km/h.
Figure 5 is shows the frequency of impacts with the vehicle front and the road and the
resultant injuries. This was analysed within each case at different speed ranges.
In general it can be seen that higher impact speeds result in more severe injuries due to
impact with the car. For collision speeds of up to 40 km/h nearly in 40 per cent of most
severe injuries were due to impact with car, and for speeds ranging from 41 to 60 km/h
this proportion was 80 per cent
Impact speed of car (km/h)

it is remarkable that in newer cars in collision speeds up to 20 km/h almost no

head injuries were registered
Figure 5 Portion of highest injury severity of impact distinguished for vehicle

impact only, road impact only and some assessment of injury
causation
Impacts to the windscreen region

Results of 762 car to pedestrian collisions (Otte, 1999) were used to analyse for the
detailed injury situation of the head impacted the area of the windscreen. The
characteristic deformation pattern of the glass and frame was studied compared to the
injury pattern of the head. Within the framework of the study the investigated vehicles
were categorized as ‘old cars’ (if production was pre-1990) and ‘new cars’ (cars
produced after 1990). It can be stated that in newer vehicles 11.6 per cent of head injuries
occur by impact with the windscreen pane and 9.3 per cent are caused by front hood
impact.
Head injuries and collision speed

Two-thirds of all head injuries were soft-tissue lesions, including abrasions, kceration-
contusion injuries as well as cuts of the skin (severity degree AIS 1) for 91.5 per cent of
all pedestrians. Of them, 51.1 per cent suffered injuries by road impact, 16.9 per cent of
soft-tissue injuries on the car were caused by the hood and 19.8 per cent by the
windscreen. Fractures of the skull are frequent to the vault of the cranium and the middle
face. Fractures of the skull base are rare. Another frequent injury type is concussion. This
type of injury is received by 7.9 per cent pedestrians by an impact to the pane. It was
observed that the windscreen quite often causes injuries and is also responsible for
fractures of the vault of the cranium, the skull base, the middle face and the skull-brain
trauma (Table 4)
Table 4
Different types of head injuries and their impact location (100% all pedestrians)
PEDESTRIAN WITH HEAD TOTAL STREET LOCALIZATION OF

INJURIES IMPACT
n=543 (100%) car
front hood screen pillar other
soft tissue 91.5 51.1 2.7 16.9 19.8 2.4 5.8
fracture of face 5.7 2.1 0.2 1.5 1.5 0.5 —
skull fracture 5.2 2.0 0.1 0.7 2.0 0.4 0.2
fracture of skull base 1.9 0.3 — 0.3 0.8 0.5 0.1
concussion 28.8 14 0.7 5.0 7.9 0.1 0.5
brain injury 9.0 2.3 — 2.1 3.3 1.2 0.8
The probalitity of occurrence of diffrent injuries definitly depends on the collision

speed of the car (Figure 6).
Figure 6: Cumulative frequency of different types of injuries in relation to the

impact speed of car
Of all soft-tissue injuries 80 per cent were sustained in collision speeds up to 40 km/h
(Figure 7), 60 per cent of the fractures and 40 per cent of serious skull-brain traumas, the
so-called contusion were registered. The cumulative frequency reveal that for old as well
as for new cars approximately 75 per cent of all established collision speeds were
exceeding 25 km/h, while in comparison for all pedestrian collisions 80 per cent were
registered in speeds up to 45 km/h (Figure 8). The correlation between head injury
severity and impact speed is pointed out in Figure 6, 74 per cent of all head injuries
occurred at speed ranges of up to 40 km/h, while 34 per cent happened at speeds above
60 per cent. Here 28 per cent were severe injuries of the brain.
Figure 7 Frequency of types of head injuries in head impacts to the windscreen

(100% head injuries of pedestrians per car impact speed range)
the probability for the head to hit the front pane or the lateral cross-beam the so
called A-pillar is visibly increasing with higher collision speeds
It is remarkable that in newer cars in collision speeds up to 20 km/h almost no head

injuries were registered. The severity of head injuries also determines the injury severity
of the entire body.
Figure 8 Cumulative frequency of impact speed of cars distinguished between

all pedestrian and those with AIS head 2+ for older and newer cars
Head impact points on car shape

The probability for the head to hit the front pane or the lateral cross-beam the so called
A-pillar is visibly increasing with higher collision speeds (Figure 9). In collision speeds
up to 30 km/h a frequency of windscreen contacts below 10 per cent is established. In
speeds of 31 to 40 km/h already 38.4 per cent of pedestrian accidents with head impact to
the front pane or the A-pillar occurred. Injuries were incurred to 26 per cent. With speeds
of 51 to 60 km/h already 60.4 per cent of pedestrian accidents occurred and with speeds
of more than 70 km/h 90 per cent were registered with head impact to the front pane
region or the A-pillar. Not all contacts to the front pane region lead to injuries. This is
apparent when comparing the different frequency curves for contacts and injuries of the
head and serious head injuries of AIS 3+ as illustrated in Figure 9. It also demonstrates
that with increasing collision speed serious head injuries are more frequent. Head injuries
of severity degree AIS 3+ occurred only in speeds above 30 km/h and those severity
grades significantly started more frequently above 50 kmlh.
Figure 9 Frequency of injuries of head contact to the windscreen or A-pillar

compared to contacts at all to this region in relation to impact speed
(100 per cent all persons per speed range)
All documented and measured impact points on the glass of the windscreen pane or the
lateral windscreen frames and the A-pillar respectively were analysed in each case and
transferred to a standardised drawing in Figure 8.
the results from these investigations lead to the conclusion that real world
conditions have not been considered for EEVC test regulations for
pedestrian impacts
As far as the frequency of all impact points is concerned Figure 10 illustrates impact
patterns showing remarkably many points on the lower half part of the windscreen pane.
This reveals a rotation of the entire body within the impact kinematic. During the ladling-
up phase the head experiences a downward directed movement. An impact to the lateral
A-pillar is with 5.1 per cent quite rare, and it doesn’t show a high injury severity in most
cases. Children quite often, and in this study especially, contact of the front hood,
compared to this adults, however contact the hood surface is quite rarely.
The detailed analysis of the head impact points shows that serious head injuries do not
only occurring on the border region of the pane to frame but very frequently on the pane
surface. The dominating influential parameter for the windscreen impact is the collision
speed of the car. Only in speeds of more than 40 km/h are impacts of the hood often
found on the pane surface and here regions also found in the centre of the pane and,
therefore, are the cause for serious head injuries.
It could be established, that up to impact speeds of 40 km/h exclusively soft tissue
lesions (74.1 per cent) and the concussion (20.5 per cent) and fracture of the skull
happened in 2.8 per cent only. Compared to this in the speed range of more than 40 km/h
fractures could be registered with higher frequency. With speeds above 60 km/h 7.9 per
cent of the head injuries were fractures of the skull, 7.3 per cent skull base fractures and
very often a severe head brain trauma could be observed.
Figure 10 True to scale head impact location on the front shape by detailed case
analysis on standardized front shape
QUANTIFICATION OF THE INFLUENCE OF DIFFERENT IMPACT

AREAS FOR INJURY PATTERN
For showing the influences of injuries of different body regions to the resultant severity
of the body MAIS in the frame of an ongoing study a virtual MAIS was calculated. It was
asked for the avoidance of injuries of the specific body region caused by a forgiven
impact area at the car (Table 5) and assessed for the resulted injury pattern of the whole
body. Therefore in the data file the AIS of each injury of the specific body region and
impact point of car, i. e. the head to windscreen, was set to zero and for that person the
new so called virtual MAIS was evaluated as maximum of all AIS grades of body region
except the specific region and impact area, i.e. the head at windscreen. Only those cases
were selected for this calculation with a maximum of 40 km/h for the impact speed of the
car, which is the most often observed impact speed in pedestrian accidents and is the
defined speed range for EEVC test procedure.
Table 5
Pedestrians colliding with car front of car up to 40 km/h (as percentage)
AGE OF PEDESTRIAN EXISTING VIRTUAL SITUATION

UP TO 12 YEARS CASE AVOIDING INJURIES OF
n=155 head upper leg lower all
leg
windscreen pelvis/front knee measures
bonnet w+b
maximum injury severity — — 1.9 1.9 — 7.6 12
grade not injured
MAIS 1 76.8 76.8 78 78 78.1 72.3 72.7
MAIS 2–4 23.2 23.2 20.1 20.1 21.9 20.1 15
MAIS 5/6 — — — — — — —
AGE OF PEDESTRIAN EXISTING VIRTUAL SITUATION
>12 YEARS CASE AVOIDING INJURIES OF
n=333 head upper leg lower all
leg
windscreen pelvis/front knee measures
bonnet w+b
maximum injury severity — 2.9 3.5 4.5 2.6 10.5 18.5
grade not injured
MAIS 1 60.5 60 58 59 59.7 57.3 55.3

MAIS 2–4 39 36.6 38 36 37.2 31.7 25.7
MAIS 5/6 0.5 0.5 0.5 0.5 0.5 0.5 0.5
For selected impact speed range of up to 40 km/h very severe injuries of MAIS 516 are
very rare. For pedestrians up to 12 years no child suffered such severity and 0.5 per cent
of the adults. Avoiding the impact of the head to the bonnet the best effectiveness can be
found (real world 23.2 per cent MAIS 2–4 compare to virtual situation of 20.1 per cent
MAIS 2–4). For pedestrians older than 12 years the reduction of the impact of the lower
leg and knee region to the bumper area is found as a high effectiveness (39 per cent
MAIS 2–4 in existing cases to 31.7 per cent MAIS 2–4 in virtual situation). There will be
an increase of uninjured and minor injury severity grades if all measures are taken into
account together, resulting in 12.3 per cent uninjured children and 18.5 per cent adults.
safety for pedestrians in impacts with cars has clearly improved within the past
few years
CONCLUSION
Safety for pedestrians in impacts with cars has clearly improved within the past few
years. More than one third of pedestrians, however, suffered serious injuries. The head is
regarded as the most seriously injured body region—60 per cent of all pedestrians suffer
head injuries and 70 per cent of all collision occur at speeds up to 40 km/h. Up to this
speed level 60 per cent of pedestrians are affected with head injuries. Most of the serious
injuries are registered in this speed range. It could be established, that with higher speed
levels, above 40 km/h, a still higher injury frequency with a probability for head injuries
of more than 80 per cent occur, linked with often severe head injuries.
For cars introduced to the market since 1990, so-called new cars, 93.6 per cent of
pedestrians were registered with head injuries if the impact speed was above 40 km/h.
With impact speeds up to 40 km/h injuries caused by the windscreen region are
relatively rare (4.7 per cent). With speeds exceeding 40 km/h for newer cars, the
windscreen region is very frequently associated with (63.8 per cent) injury causation.
This may be due to the low number of cases in this speed range. On the other hand, it
could be due to a shorter front hood, which is often found in newer compact cars.
The study clearly shows that the head impact often occurs to the lower half of the
windscreen. Serious injuries of A1S 2+ are caused just as frequently in the lower and
lateral edge region of the windscreen. The head of an adult pedestrian is mostly
impacting the lower half of the windscreen and that this region is therefore responsible
for severe head fractures and brain trauma (AIS 2+ injuries). The lateral frame of the
windscreen, the so called ‘upper A-pillar’ is an area seldom impacted by the head and no
higher injury severity could be established than for the glass pane surface. Severe head
injuries AIS 2+ exclusively happen with speeds of more than 40 km/h, while up to 40
km/h minor injuries such as soft tissue lesions and concussions are frequent.
The bonnet leading edge is responsible for 4.5 per cent of injuries up to 40 km/h for
new cars compared to 16.8 per cent for old cars. But it is not quiet right to say that
injuries caused by the front leading edge are very rare. For the impact speed range above
40 km/h a higher incidence of injury causation was observed—13.9 per cent.
The results from these investigations lead to the conclusion that real world conditions
have not been considered for EEVC test regulations for pedestrian impacts. Impact
speeds of up to 40 km/h are not responsible for the severity of the head impact to the
windscreen. The collision speed is the main indicator for the injury severity. The
causation for serious injuries must be seen in view of the fact that severity of head
injuries often determine the overall grade of the trauma. Patients with poly-traumatic
symptoms, who suffered injuries to at least three different parts of the body with injury
severity degree MAIS 2+ nearly always suffer head injuries. The proportion of
pedestrians with poly-traumatic symptoms amounts to only 4.7 per cent of all injured
pedestrians, but 61.4 per cent of pedestrians suffered an impact to the windscreen. Beside
the fact that chances of survival for these patients are quite limited, there exists, on the
other hand, a high probability of long-term consequences. From the medical-traumatic
point of view, measures are required for modification of the windscreen region,
especially the pane itself. The head impact at least should be integrated in the EEVC test
rules.
It must be taken into account, that 12.4 per cent of all bicycles colliding with cars
suffered head injuries from impact against the windscreen. The study shows, that most
impact points are located on the lower half of the pane. Concerning to this result an area
of approximately 20 cm beneath the lateral frame and the border to lower metal parts of
the bonnet should be used as point for a head impactor. The head impactor should impact
with his lateral forehead region to the pane surface under an angle of 45 degrees of the
body length axis related to a horizontal line. An accident-conform impact speed,
diverging from defined test criteria does not seem to be convincingly necessary.
On the other hand, the analysis of different studies, discussed in this paper,
acknowledged the existing impact areas on the car, used as test points by EEVC.
Furthermore the study clearly shows the following demands for future developments:
• An energy absorbing attachment of the pane in the anchorage system of the

frame should be developed, a more elastically material of the glass compound
system could be used for such a purpose.
• An optimizing of the angle of the windscreen pane could help to reduce the
head impact speed in connection with the kinematics of the ladling-up
movement of the pedestrian body.
• An impact within the glass region of the windscreen there is not high risk
for severe injuries to the head, admittedly the impact to the frame region is very
severe.
The results of this study should help for further development of car designs and finding
measures. Some more in-depth research activities are needed for this purpose.
REFERENCES
American Association forautomotive Medicine, 1990, The Abbreviated Injury Scale AIS;
Revision. (lllinois: USA).
EEVC, 1994, Working Group 10 Report, European Experimental Vehicles Committee,
November.
EEVC, 1998, EEVC WG 17: Improved test methods to evaluate pedestrian protection
afforded by passenger cars, EEVC Working Group 7 Report, December.
ETSC briefing, January 1998.
ETSC, 1999, Crash-ETSC’s newsletter on European Vehicle Crash Protection, Spring.
Mackay, M., 1997, A review of biomechanics of impacts in road accidents,
Crashwothiness of transportation systems, structural impact and occupant protection,
(Kluver, London).
MIRA: Motor Industry Research Association, 1997, Study of Research into Pedestrian
Protection, Report No 97 forEurop. Commission. DC III: 456502–01, August, United
Kingdom
Otte, D., 1994, Road accident research by the Medizinische Hochschule Hannover as an
example of the significance and usefulness of a scientific team working at the scenes of
accidents. Brandschutz, Deutsche Feuerwehr Zeitung, 6, pp 370–377.
Otte, D., 1998, Pedestrian impact at front end of car, Paper for EEVC WG 1 7 doc. 10
rev., Accident Research Unit; Medical University Hannover.
Otte, D. 1999, Severity and mechanisms of head impacts in car to pedestrian accidents,
proc. IRCOBI, (Barcelona, Spain).
10
Risky Business: Safety Regulations, Risk
Compensation and Individual Behaviour
James Hedlund
INTRODUCTION
You are the proud owner of a new car. Perhaps it just came from the factory and you are
the first owner; perhaps you acquired it second-hand or third-hand. But it is new to you.
And one of its unfamiliar features is an air bag-a device in the steering column that will
inflate during a crash and reduce the forces that may injure you. You may be well aware
of the air bag-you may like it or you may fear it. Or you may not even know that it is
there.
Question: Will your driving change because of the air bag? Will you drive faster or
more recklessly, because you believe you will be safer in a crash?
You are a logger. You earn your living cutting down trees with a chain saw. It is a
dangerous occupation: injuries from falling trees or from the chain saw are common. To
reduce these injuries, your employer or the government requires that you use safety
equipment: a hard hat, safety goggles, gloves, steel-toed shoes. You are very aware of
this equipment every time you work: it is hotter with a hat, the goggles can steam up, the
gloves protect your hands from snapping branches.
Question: Will your work practices change because of the safety equipment? Will you
work more quickly and less carefully because you believe you are safer?
These are typical risk compensation questions. Stated more broadly: will your
behaviour change in response to features or rules intended to improve safety? Do
individuals compensate for their increased safety; do they in some way trade this safety
for increased performance?
This paper provides an overview of risk compensation, reviews its history, discusses its
theoretical foundations, outlines evidence for and against its claims, provides the author’s
own views, and gives a brief annotated bibliography for further reading. It concludes with
practical advice: what relevance, if any, does risk compensation have for injury
prevention workers who seek to reduce unintentional injuries and their consequences?
THE SETTING: INJURY PREVENTION STRATEGIES AND RISK

COMPENSATION
The discipline of injury prevention began when injuries were understood to be both
predictable and preventable rather than unavoidable accidents or behavioural problems.
Injuries are the unintended consequences of individual behaviour within a risky
environment. This understanding led to three fundamental injury strategies to prevent
injuries, as described in the comprehensive report Injury in America (Committee on
Trauma Research, 1985):
• persuade persons at risk to change their behaviour
• require behaviour change by law or administrative rule
• provide automatic protection through product and environmental design.
Injury prevention policymakers and workers agreed that it usually was easier to change
objects than people. Consequently, as Injury in America reports: ‘Each of these general
strategies has a role in any comprehensive injury-control program; however a basic
finding from research is that the second strategy—requiring behaviour change—will
generally be more effective than the first, and that the third—providing automatic
protection—will be the most effective’.
In the United States, injury prevention and control methods were first used on a
national scale in 1966 when Congress established the organization that soon became the
National Highway Traffic Safety Administration (NHTSA) and appointed William
Haddon as its first administrator. The NHTSA was directed it to reduce traffic injuries
and fatalities. The Consumer Product Safety Commission (CPSC) followed in 1972.
While NHTSA employed all three strategies, it emphasized the latter two: safer
behaviour through laws (such as mandatory motorcycle helmet use) and safer vehicles
through required safety features (such as shatterproof windshields). CPSC relies almost
exclusively on the third strategy, safe product design.
injuries are the unintended consequences of individual behaviour within a risky

environment
The favoured strategies implicitly assume that people will not react to safety laws or safer
products in unexpected ways. But this assumption may not be correct. If you are required
to wear a seat belt or cycle helmet while on the road or to wear padding and a helmet
while playing ice hockey, it is possible that you may change your driving or skating
behaviour. If your ski boots now have quick-release bindings to prevent ankle injury, you
may possibly ski more aggressively. If your medicine bottles now have ‘child-resistant’
caps, you may possibly be more careless in leaving them near children. If your behaviour
does change, the safety gains expected from the safety laws or safer products may not be
realized.
This potential behaviour change in response to safety rules, safer products or safer
environments has been called many things. In this paper, ‘behavioural adaptation’
Risky business 125
describes all behavioural change in response to our perceived changes in risk and ‘risk
compensation’ describes the special case of behaviour change in response to laws and
regulations. The distinction becomes murky at times. For example, if a new safety feature
appears on all chain saws, our behavioural reaction will not depend on whether the
feature is required by government regulation or adopted voluntarily by all manufacturers.
But our risk compensation definition focuses on the injury prevention strategies of
greatest interest, where the government attempts to require us to be safer.
These strategies have also generated the greatest controversy regarding risk
compensation. Its adherents maintain that safety measures imposed on us cannot succeed.
The debate has produced several books and over 200 journal articles over the past 25
years, many of which contain rather strong views from risk compensation proponents and
opponents.
Scientific risk compensation studies are based in economics, psychology, decision
theory, and mathematical statistics. But on a practical level we all are risk compensation
experts as we live our everyday lives. I invite each reader to think about risk
compensation from at least two perspectives: as an injury prevention and control
specialist whose goal is to reduce injuries across society, and as an individual subject to
safety laws and using products designed to reduce injury.
A BRIEF HISTORY OF RISK COMPENSATION
It is perfectly obvious that we change our behaviour in response to changes in perceived

injury risk. Most obviously, we take additional precautions if we believe our risk has
increased. For example, when roads and sidewalks are icy, we walk more carefully for
fear of falling and we drive more slowly to be sure that we can stop safely.
Note that we can react only to what we perceive or know. There is an experience
shared by almost everyone who has driven a car in snowy conditions. You ‘know’ that
the road may be slippery; you may or may not drive more slowly or allow more distance
to brake. But one time when you apply the brakes, the car skids. If you are lucky, you
correct the skid or come to a stop without consequence. If you are not, your skid may
produce a damaged fender or worse. And afterwards, at least for a time, you drive even
more slowly and apply the brakes more cautiously. Actual risk has not changed, but your
perception of risk has, and your actions change as a result.
Early behavioural adaptation thoughts in road safety

Some observers raised the possibility of behavioural adaptation and risk compensation
shortly after features or regulations were implemented to reduce injury. The early
literature deals with road safety, for several reasons. In motorized countries, traffic
crashes are the largest cause of accidental injury and death. As a result, extensive safety
regulations were implemented—traffic laws, traffic police to enforce these laws, required
safety features for vehicles and roadways—which affect almost everyone in these
countries every day. Consequently, road safety provided the best opportunity to examine
any behavioural reaction to these safety regulations. More recently, risk compensation
has been discussed in response to regulations affecting consumer products and workplace
safety.
A letter from the early days of motoring notes quite colourfully that road safety
measures may produce behavioural adaptation (Adams, 1995):
Letter to The Times, 13 July 1908
Dear Sir,
Before any of your readers may be induced to cut their hedges as suggested by the
secretary of the Motor Union they may like to know my experience of having done so.
Four years ago I cut down the hedges and shrubs to a height of 4ft for 30 yards back
from the dangerous crossing in this hamlet. The results were twofold: the following
summer my garden was smothered with dust caused by fast-driven cars, and the average
pace of the passing cars was considerably increased. This was bad enough, but when the
culprits secured by the police pleaded that ‘it was perfectly safe to go fast’ because ‘they
could see well at the corner’, I realized that I had made a mistake. Since then I have let
my hedges and shrubs grow, and by planting roses and hops have raised a screen 8ft to
10ft high, by which means the garden is sheltered to some degree from the dust and the
speed of many passing cars sensibly diminished. For it is perfectly plain that there are
many motorists who can only be induced to go at a reasonable speed at crossroads by
consideration for their own personal safety.
Hence the advantage to the public of automatically fostering this spirit as I am now
doing. To cut hedges is a direct encouragement to reckless driving.
Your obedient servant,

Willoughby Verner
behavioural adaptation describes all behavioural change in response to our

perceived changes in risk and risk compensation describes the special case of
behaviour change in response to laws and regulations
Forty years later, Smeed (1949) noted in a widely-quoted observation: ‘It is frequently
argued that it is a waste of energy to take many of these steps to reduce accidents. There
is a body of opinion that holds that the provision of better roads, for example, or the
increase in sight lines merely enables the motorist to drive faster, and the result is the
same number of accidents as previously. I think there will nearly always be a tendency of
this sort, but I see no reason why this regressive tendency should always result in exactly
the same number of accidents as would have occurred in the absence of active measures
for accident reduction. Some measures are likely to cause more accidents and others less,
and we should always choose the measures that cause less.’
Academics also began thinking and writing about these ideas. But they attracted little
public attention and certainly had little influence on injury prevention programmes and
policy.
Risky business 127
Risk compensation and economics: Sam Peltzman’s evaluation of motor

vehicle safety standards
Risk compensation in response to government actions became a public issue with Sam
Peltzman’s 1975 paper—‘The effects of automobile safety regulation’—in the Journal of
Political Economy, Nine years earlier, in 1966, the United States Congress established
NHTSA’s predecessor and directed it to set motor vehicle safety standards—to make
vehicles safer by regulation. Prior to this, road safety concentrated on regulating
individual behaviour, through traffic laws and developing a safe road environment. These
strategies affected individual drivers, cyclists, and pedestrians on the one hand and
involved public expenditures for developing infrastructure on the other but did not
directly affect private industry. NHTSA moved quickly to comply with its directive,
issuing 20 Federal Motor Vehicle Safety Standards (FMVSS) in January 1967 that
applied to all new motor vehicles. Nine more standards followed by the end of 1968.
Some FMVSS sought to prevent crashes by setting standards for brakes, tires, and
mirrors; the majority sought to reduce crash injuries by requiring features such as seat
belts, shatterproof windshields and energy-absorbing steering columns.
As product regulations affected business, the FMVSS attracted the attention of
economists. Some economists believe that markets for goods and services usually
function best when left alone; that government regulation or control, even with the best of
intentions, distorts prices and quantities of goods and influences both producers and
consumers to act in ways not in society’s best interests. The FMVSS ran directly counter
to this point of view.
Peltzman evaluated the effects of the FMVSS: had they in fact improved safety as their
advocates had predicted? He began with the assumption that individuals are rational
economic consumers who will act in their own best interests. If you have more of
something than you want, you will (if you can) exchange it for something else that you
lack or desire. Each day involves many decisions to trade your time, your money, or your
goods for other goods or money or time, to improve your overall well-being.
Peltzman considers safety (or risk) an economic good that you will trade with other
goods in the same way. If you do not have as much as you wish, you will acquire more.
You will spend more time and walk more slowly if icy sidewalks are too risky. You will
buy gloves if pulling thistles without them is likely to injure your fingers. And if you
have more safety than you want, and are able to trade some of it for something else you
desire, you will. On the road, if your car is safer than you want it to be, then Peltzman
presumes you will drive faster, trading safety for time. In his words, you will trade safety
for ‘driving intensity’ to achieve the balance that you find best.
Peltzman then proceeded to test his idea with an econometric (statistical) analysis of
FMVSS effects. He concluded that the standards were ineffective (Peltzman, 1975): ‘The
main conclusion is that safety regulation has had no effect on the highway death toll.
There is some evidence that regulation may have increased the share of this toll borne by
pedestrians and increased the total number of accidents.’
This conclusion, from a University of Chicago economist, based on an analysis using
standard econometric methods, startled the road safety community. Peltzman not only
claimed that the FMVSS had not performed as predicted, he also said that they may have
reduced safety for pedestrians. His analysis was based on classical economic theory. And,
of course, it challenged the role of government in attempting to improve safety through
regulation. In Peltzman’s view, NHTSA was useless and perhaps even counter-
productive.
Peltzman’s paper prompted a lively debate over the next 20 years, with criticism and
support from Joksch (1976a, 1976b), Robertson (1977a, 1977b, 1977c, 1981, 1984,
1996), Crandall and Graham (1984), Graham and Garber (1984), Orr (1984), Zlatoper
(1984), Garbacz (1990), Chirinko and Harper (1993), and Peltzman himself (1976, 1977),
among others. The debate concentrated on Peltzman’s statistical analysis. Typical issues
included the variable used to measure FMVSS effects (traffic fatalities, fatalities per mile
of travel or fatalities per crash), the time period analysed (Peltzman used data from 1947–
1972), the regression equation’s functional form (linear or logarithmic), what other
factors should be included in the model as controls and how they are best measured
(changes in drunk driving, rural-urban travel mix, speed limits and gas shortages, the
number of young drivers, etc.), and how to account for motorcycles, trucks and other
vehicles not regulated in the same way as passenger vehicles. Each choice may affect the
results substantially. As just one example, Graham and Garber (1984) chose to‘…
recalculate Peltzman’s regression estimates, changing the functional form. We did this
merely using data on the absolute levels of death rates, average speeds, and so on, rather
than their logarithms. This change alone casts serious doubt on Peltzman’s conclusions:
The resulting estimates suggest that regulation averted roughly 5,000 fatalities between
1966 and 1972, rather than causing about 10,000 deaths as Peltzman’s estimates suggest.’
Blomquist (1988) reviewed 11 studies from 1975 to 1986, some with multiple
analyses. He summarized the key conclusions reported by all 11 as follows: due to the
FMVSS, passenger car occupants are safer, non-occupants are less safe, overall road
travel is safer, and occupants probably are less safe than engineering studies had
predicted. In a recent critical review, Levy and Miller (1999) agree that the FMVSS
improved occupant safety but find only weak support for any effect on non-occupants.
behavioural adaptation generally does not eliminate the safety gains from
programmes, but tends to reduce the size of the expected effects
Peltzman moved on to other economic issues. After replying to his initial critics he did
not write on risk compensation again. The FMVSS remained in force, without serious
challenge (indeed, manufacturers had adopted many of the FMVSS voluntarily before
they were required by regulation). But Peltzman had introduced risk compensation into
the road safety debate. His 1975 paper probably is cited more frequently than any other
risk compensation study.
Risk compensation and psychology: G.J.S.Wilde’s risk homeostasis theory

G.J.S.Wilde, Professor of Psychology at Queen’s University in Ontario, had been
thinking and writing about behavioural adaptation and risk compensation since the early
seventies. He set forth the fundamentals of a theory in several papers leading to The Risk
Compensation Theory of Accident Causation and Its Practical Consequences in 1976.
Risky business 129
With his 1982 Risk Analysis paper, ‘The theory of risk homeostasis: implications for
safety and health’, its four accompanying commentaries and Wilde’s response, his ideas
attracted substantial attention.
In Wilde’s view, risk is an inherent part of our human psychological makeup. Not only
can we not escape risk, we need risk. Indeed, Wilde hypothesizes that we each have a
‘target level of risk’. We each measure risk on our own ‘risk thermostat’. If the perceived
risk of a situation exceeds our target level, we will act to reduce it. And if the perceived
risk is less than our target level, we will increase our risk back to our target level through
more dangerous actions.
Wilde’s name for this process is risk homeostasis, by analogy with the self-regulatory
and unconscious manner in which we maintain our body’s temperature. Risk homeostasis
is then an extreme form of behavioural adaptation: not only do we modify our behaviour
in response to external changes designed to make us more or less safe, but we seek to
counteract these changes completely and return to our desired risk level. Because the risk
in our environment constantly changes, we are constantly forced away from our target
risk level, but we always move back toward it, always countering external influences
(such as injury prevention measures) that attempt to decrease or increase our risk.
Wilde (1982) developed a mathematical formulation of risk homeostasis in road safety:
the ‘accident rate per time unity of driver exposure is invariant regardless of road
geometry’ (or, for that matter, regardless of anything else). In other words, your accident
rate per hour of travel on a high-speed divided motorway is the same as your rate per
hour on a low-speed neighborhood street. Wilde also described his theory in catchy
language: ‘the sum of the sins is constant’. Others have called it even more picturesquely
‘the law of conservation of misery’. Wilde (1994) also extended risk homeostasis beyond
road safety to other injury prevention (and in fact disease prevention) arenas. ‘Risk
homeostasis may thus apply not only to road use, but also to industrial safety, sports,
making love, smoking, drinking, doing home repairs, climbing ladders, physical exercise,
investing money, gambling, and who knows how many other activities.’
Risk homeostasis challenges the very foundations of injury prevention strategy. It
holds that the only effective safety measures are those that alter your desired risk level.
Anything that merely modifies the environment or that regulates your behaviour without
affecting your target risk level is useless. It is as far removed as possible from Haddon’s
recommended strategies.
Wilde’s risk homeostasis theory was challenged by Leonard Evans, Frank McKenna,
and others. Wilde published extensively over the next 10 years, engaging his critics in
running debates (see Evans, 1986a; 1986b and Wilde, 1986; McKenna 1982; 1985a;
1985b; 1987; 1988; 1990 and Wilde, 1984; 1985; 1989. Wilde summarized his views in
popular form in his 1994 book Target Risk.
Risk compensation and public policy: John Adams’ campaign against seat
belt use laws
John Adams, a geographer at University College, London, began investigating the effects
of seat belt use laws in his work on transportation planning, where he was strongly
critical of policies that favoured continual increases in the number of private cars. He
concluded that seat belt laws were not effective. In fact, as Peltzman concluded for the
FMVSS, Adams believed that belt laws reduced risk for passenger car occupants but
increased risk for pedestrians and cyclists.
Adams generalized his observations across most standard road safety measures: when
they have any effect at all, they merely redistribute risk from vehicle occupants to cyclists
and pedestrians. He adopted much of Wilde’s risk homeostasis as the behavioural basis
for his findings.
Adams’ motivation was to influence public policy. He took on the issue of seat belt
laws while it was being debated in the British Parliament. He opposed belt use laws in
newspapers and in letters to the government as well as in formal papers. His primary
analytic method for estimating belt law effects was to compare overall road fatality trends
in countries with and without belt laws. His results were easy to show on a chart and
easily understood by newspaper readers and politicians alike. He emphasized the idea of
individual liberties, that belt use laws infringe on our freedom to act as we wish as long
as we do not interfere with others. And he also could state his thoughts clearly, succinctly
and controversially: ‘protecting motorists from the consequences of bad driving
encourages bad driving’ (Adams, 1985b). His 1995 book Risk, John Adams summarizes
and extends his ideas to a broad discussion of risk in society.
they typically find no effects for measures to protect occupants in the

event of a crash (such as seat belts) but may find effects for measures that
attempt to prevent crashes by improving vehicle performance (such as
better brakes or tires)
Adams’ views and analyses on belt law effectiveness were countered forcefully by many,
notably Murray Mackay. His critics argued that Adams’ methods were suspect. In
particular, overall road fatality trends are affected by many factors and a detailed
statistical analysis with good data and appropriate controls is needed to evaluate belt law
effects.
After extensive debate, Parliament voted to adopt a belt use law effective in January
1983 which would lapse after three years unless Parliament voted again to continue it.
This provided the opportunity for a careful and detailed evaluation under favourable
conditions: a rapid increase in belt use from about 40 per cent to over 90 per cent, a large
population affected at the same time (over 16 million cars in 1983), and a good records
system for recording crashes, injuries and fatalities. The results were quite consistent. A
simple comparison of casualties for the 11 months before and 11 months after the law
showed fatalities dropped by 23 per cent and serious injuries by 26 per cent (Mackay,
1985a). A more sophisticated time series analysis by British Department of Transport
statisticians reported fatality and serious injury reductions of about 20 per cent for drivers
and 30 per cent for front-seat passengers (Department of Transport, 1985). Finally, an
independent evaluation by Durbin and Harvey, distinguished statisticians from the
London School of Economics, reported that serious injuries and fatalities had dropped 23
per cent for car drivers and 30 per cent for front-seat passengers but had risen 3 per cent
for rear-seat passengers, dropped 0.5 per cent for pedestrians, and risen 5 per cent for
cyclists, with the last three results not statistically significant (Department of Transport,
Risky business 131
1985). Adams (1995) countered by claiming that the observed decrease was due to a
well-established downward trend in casualties and to a simultaneous campaign to reduce
drunk driving.
In 1986, Parliament voted to retain the seat belt use law. Irwin (1987) provides a
thoughtful view of scientific and political issues throughout the British seat belt law
debate. Adams (1985a) and Mackay (1985) give contemporary views from both sides.
Evans (1991) and Adams (1995) analyse the issues and evaluate the results from a longer
perspective.
Risk compensation in 2000

Peltzman’s, Wilde’s, and Adams’ papers prompted debate in journals and at conferences.
A 1984 Symposium on Human Behaviour and Traffic Safety, sponsored by General
Motors, contained substantial contributions by Wilde, Adams, O’Neill, Mackay and
others (Evans and Schwing, 1985). Ergonomics published a special issue on ‘Risky
Decision-making in Transport Operations’ in 1988 with 21 papers representing all sides
of the issue. The First and Second European Congresses of Psychology, in 1989 and
1991, respectively, devoted considerable attention to the issue (papers are published in
Trimpop and Wilde, 1994). More recently, Safety Science published a special issue in
1996 on ‘Risk Homeostasis and Risk Assessment’. Managerial and Decision Economics
published a special issue on ‘Product Safety and Managerial Decisions’ in 1996 in which
three papers explicitly discuss the possibility of risk compensation in the use of cigarette
lighters, lawn mowers and bicycles. As noted earlier, Wilde and Adams summarized their
views in books: Target Risk (1994) and Risk/John Adams (1995), respectively. Finally,
Leonard Evans captured the views of many in his pithy quote ‘…there are no epicycles
and there is no phlogiston…similarly, there is no risk homeostasis’ (1986a) and in
Chapter 11 of his book Traffic Safety and the Driver (1991).
In terms of ideas and controversy, though, little has changed in the past decade. An
excellent, thorough, and dispassionate summary of risk compensation theories and
evidence is found in the OECD (1990) report Behavioural Adaptations to Changes in the
Road Transport System, prepared by a working group with representatives from 16
European and North American countries. The report’s key conclusion is that:‘…
behavioural adaptation to road safety programmes does occur although not consistently.
…behavioural adaptation generally does not eliminate the safety gains from programmes,
but tends to reduce the size of the expected effects’.
The extreme views of risk homeostasis have attracted little support. For example, the
editors of a safety journal (Stanton and Glendon, 1996) noted that: ‘One of our reviewers
requested that we send him no more papers on RHT [risk homeostasis], adding that
“when an enterprise is built on sand there is not much likelihood of a future and repeating
old arguments is no substitute for progress”.’
EVIDENCE FOR AND AGAINST RISK COMPENSATION
To summa rize so far: risk compensation occurs if we change our behaviour in response
to a law or regulation designed to increase our safety. Risk homeostasis theory predicts
that we will attempt to compensate exactly, over the long run, to restore the prior level of
risk. The two broad reasons why we might change our behaviour are economic and
psychological. As economic beings we can exchange safety for other things in order to
maximize our overall well-being. As psychological beings we may have an inherent
desire for risk that leads us to compensate for unwanted external attempts to reduce it.
If we change our behaviour, we can do so in several ways. We may change how we
perform a specific task, we may change the task, or we may add or eliminate the task
altogether. If my car is required to have antilock brakes, I may drive faster on some roads
and I may drive instead of taking public transit when roads are icy. If my local children’s
playground has been required to replace its concrete with softer surfaces, I may now
allow my children to perform more dangerous tricks, I may allow them to go to the
playground unsupervised, or I may allow them to use the previously ‘off-limits’
playground.
Our behavioural changes may have system effects: consequences beyond the current
act or to others. If we drive faster because of better brakes, we may endanger others. If
we now permit our children to play at the safer playground, they may have to cross a
dangerous street to get there.
With this broad framework, let us examine the evidence supporting and opposing risk
compensation. The evidence falls into three broad categories: evaluations, experiments,
and theory. We consider each in turn.
Evaluations
risk compensation is relevant only if a safety measure produces behaviour change

which in turn increases risk
Risk compensation occurs if people react to a safety law or regulation by acting less
safely. To see if this has occurred, we either can examine individuals to see if their
actions have changed, or we can examine aggregate data to see if the law or regulation
has had its intended effect.
Both evaluation types must deal with the ‘noticeable difference’ issue. A change so
small that it cannot be detected may be interesting theoretically but has no practical
value. The question is not whether risk compensation occurs or not, but rather whether
we can detect a noticeable difference in behaviour or aggregate effects that can be
attributed to it.
Individual action: Was explicit compensating behaviour observed? This is a logical
way to evaluate risk compensation, but for two reasons it is virtually impossible to carry
out satisfactorily in practice. First, risk compensation predicts that behaviour will change
but does not predict how it will change, so we do not know what to observe. Behaviour
may change in a way that is not at all obvious. Wilde (1994) suggests that measures to
reduce drunk driving may in fact have an overall road safety benefit but may cause those
who would have driven drunk to act in more risky ways when not on the road. Or,
compensating behaviour may take place well after the fact. Adams (1988) suggests that
Risky business 133
laws requiring traffic to stop when children are boarding or getting off a school bus,
reinforced by flashing lights and signals on the buses themselves, may teach children not
to be careful when getting off any bus, so that in later years they are in danger when
boarding or exiting a transit bus. No study can examine all possible ways in which
compensating behaviour might occur.
Second, behaviour change is difficult to measure. We may be able to measure large
changes such as performing the task more quickly but usually cannot measure more
subtle changes such as increased carelessness. If we attempt to observe changes in
specific individuals we probably will not have enough observations to draw meaningful
conclusions. If we instead observe groups, either before and after or with and without a
safety measure, we must control for all other factors that may influence behaviour.
A few studies have looked for driving behaviour changes following various road safety
measures. They typically find no effects for measures to protect occupants in the event of
a crash (such as seat belts) but may find effects for measures that attempt to prevent
crashes by improving vehicle performance (such as better brakes or tires). For example,
O’Neill et al. (1984) report on studies in Canada and England after seat belt use laws
were implemented. They examined travel speeds and following headways and reported
no evidence of riskier behaviour due to the belt use laws. Sagberg et al. (1997) observed
travel speeds and following headways for Oslo taxi drivers with and without air bags and
anti-lock brakes. They reported shorter headways for cars with anti-lock brakes but no
significant difference for cars with air bags.
Most important, even if evaluations show that behaviour has changed, the changes may
have no effect on accidents or injuries. Faster driving may not necessarily lead to
increased crash risk; different methods of sawing wood may not lead to increased
injuries. Risk compensation is relevant only if a safety measure produces behaviour
change which in turn increases risk. This must be evaluated with aggregate data.
Aggregate data: Did the measure reduce injuries as intended? Did the motor vehicle
safety standards reduce traffic injuries and fatalities? Did child-resistant medicine caps
reduce accidental poisoning? These questions require all the standards of a good
evaluation: a sound experimental design, good data, good controls for other factors, and
appropriate statistical analyses. Even competent researchers often have difficulty meeting
these standards. While overall injury counts are high, individual injuries typically are rare
events. Data on which to base an evaluation frequently are inaccurate or imprecise.
Effects may be small: most traffic measures do well to reduce casualties by 10 per cent.
Injury control measures seldom are implemented in controlled experimental conditions
but are put in place in the real world, where many other changing factors can affect the
results.
The literature contains tens of thousands of studies evaluating injury prevention
measures. For example, a literature search produced 54,078 abstracts or titles that might
be relevant to nine motor vehicle injury prevention strategies (Rivara et al., 1999). I
suspect that most evaluations concluded that injuries were reduced. But good studies are
rare. For instance, a meta-analysis of drunk driving prevention and control literature from
1960 through 1991 identified 6,500 documents, of which only 125 passed minimal
standards of scientific rigour and quality (Wagenaar, 1999). To test for risk
compensation, though, a high-quality study must do two additional things. First, it should
compare the effect with what was predicted, with what would have happened had there
been no behavioural change as a result of the measure. Predictions are of course
imprecise, and predictions made by advocates may well oversell the expected benefits.
However, results falling far short of predictions may suggest risk compensation. Second,
the study should examine system effects. Safety measure evaluations frequently fail to
look beyond the population directly affected by the measure: for example, seat belt
evaluations often consider consequences only to vehicle occupants.
Evans (1985, 1991) provides a good sampling of actual and predicted effects. He
examined 24 studies from the road safety literature and compared the effects predicted
and actually realized. For measures designed to increase safety he found examples where
safety increased even more than expected, about as expected, less than expected, where
the measure was ineffective (safety did not change at all), and where the measure actually
decreased safety—a perverse effect. Similarly, for measures expected to decrease safety,
he found the same range of effects, from a decrease greater than expected to an actual
increase in safety—an equally perverse effect. Evans concluded that behavioural
adaptation to traffic safety measures is widespread, that the effects can vary widely, and
that there is no evidence for the complete compensation predicted by risk homeostasis.
However, many of Evans’ measures are not laws or regulations, so his conclusions on
behavioural adaptation cannot be applied immediately to risk compensation. His studies
also do not report on any system effects. To examine both issues, let us consider three
road safety areas where risk compensation issues have been raised frequently: vehicle
safety standards, seat belt laws, and motorcycle helmet laws.
a meta-analysis of drunk driving prevention and control literature from 1960

through 1991 identified 6,500 documents, of which only 125 passed minimal
standards of scientific rigor and quality
• Vehicle safety standards: The evaluations discussed previously address both

issues. They typically examine data on all road fatalities, thus including all
plausible system effects. Several evaluations also compare results with
predictions. Blomquist’s summary (1988) suggests that the FMVSS did make
occupants safer, but less than expected; that they may have increased risk for
non-occupants; and that they increased overall road safety: in short, they may
have produced some risk compensation, but their overall effect was positive.
Note, though, that these FMVSS analyses were far from consistent, as they
reported effects on total road fatalities ranging from a decrease of 60 per cent to
an increase of 5 per cent.
• Seat belt use laws: The British seat belt law studies discussed previously
consider system effects on all road users. The results also are consistent with the
well-established protective benefits of belts in a crash and with the observed
increases in belt use resulting from the law (Evans, 1991). As noted earlier,
Adams (1995) challenges these findings. Evans (1991) compares United States
seat belt law results with predictions. He finds observed fatality reductions close
to, but typically less than, predicted reductions. He attributes this to two factors:
Risky business 135
that belted drivers are generally safer drivers than unbelted, and that belt use
rates are lower at night than during the day (when belt use observations typically
are taken).
Many other belt use law evaluations have been published with far less
conclusive results. Typically they attempt to measure fatality reductions in
relatively small populations resulting from laws that have not increased belt use
substantially. Evans (1991) maintains that these situations do not permit fatality
reductions to be measured directly. The studies usually report fatality reductions
lower than had been predicted and usually do not consider effects on other road
users (Levy and Miller, 1999, review a few studies that do examine possible risk
compensation).
• Motorcycle helmet use laws. These laws in the United States provide an
excellent test of risk compensation for several reasons. Helmets clearly reduce
head injury in a crash, and head injury is the leading cause of death for
motorcyclists.
Motorcyclists are very aware of the protection offered by a helmet, but some
motorcyclists dislike helmets for various reasons. The potential risk
compensation mechanism is obvious: a motorcyclist who would not choose to
wear a helmet may drive more recklessly if helmet use is required by law.
Helmet use depends strongly on the presence of a law: about half of all
motorcyclists wear helmets if there is no law but almost all wear helmets if there
is a law. States have enacted, repealed, and re-enacted helmet laws many times
over the past 35 years, providing many opportunities to measure law effects.
Finally, system effects of helmet use laws are negligible: motorcycle crashes
very rarely cause serious injury to anyone other than motorcyclists themselves.
At the request of Congress, in 1991 the United States General Accounting
Office reviewed all available studies of helmet law effectiveness (GAO, 1991).
It summarized information from 46 studies and concluded that helmet laws
reduced motorcyclist fatalities by 20 to 40 per cent, a result in reasonable
agreement with helmet effectiveness in a crash. Evans (1991) compares
predicted and observed reductions and reaches similar conclusions. Motorcycle
helmet laws have produced no detectable risk compensation.
Summary. This brief review indicates that risk compensation may have occurred in
response to some safety measures but not in response to others. The review also illustrates
the difficulty of conclusively establishing or refuting risk compensation, as its proponents
acknowledge. Adams (1995): ‘…the multi-dimensionality of risk and all the problems of
measuring it discussed earlier, preclude the possibility of devising any conclusive
statistical tests of the [risk compensation] hypothesis.’
Experiments
Controlled experiments in laboratory settings eliminate much of the messiness and
variability of real world evaluation. These experimental studies of risk compensation take
a variety of forms. Subjects typically perform some task for a reward that depends on
their performance. They also face penalties for an ‘accident’. The experimenter varies the
reward and the accident risk and observes changes in the subject’s performance.
Two examples give a flavour of these laboratory experiments. For a summary of many
experimental studies on risk compensation, see Glendon et al. (1996).
Jackson and Blackman (1994) report on a study in which subjects ‘drove’ a driving
simulator through a predetermined route of city blocks with traffic lights and pedestrian
traffic. Subjects received a monetary reward for completing the route more quickly than
the average subject. They were penalized monetarily for ‘accidents’ (striking solid
objects, crossing the curb line, running red lights, etc.) or for being caught speeding
(exceeding the speed limit while passing through a randomly-located speed check).
Various speed and vehicle control measures were recorded. The speed limit, speeding
costs, and accident costs were varied. The authors found that, ‘consistent with risk
homeostasis theory, increased speed limit and reduced speeding fine significantly
increased driving speed but had no effect on accident frequency. Moreover, increased
accident cost caused large and significant reductions in accident frequency but no change
in speed choice.’
G.J.S.Wilde reports on several experiments in Target Risk (1994). Wilde’s experiments
seek to estimate whether a person takes too much or too little risk compared to his or her
skill level when the goal is to maximize net benefit. In the basic experiment, a subject is
seated at a computer screen. At random times, a large square appears in the center of the
screen. The subject’s objective is to press a button as close to but no sooner than 1.5
seconds after the square appears. The subject’s reward increases the closer the response is
to 1.5 seconds. Responses sooner than 1.5 seconds receive no reward and may impose a
penalty. Wilde calls these ‘brinksmanship’ experiments: your reward increases the closer
you come to the brink of disaster. In many replications of these and similar experiments
Wilde finds, for example, that as the penalty for responding too quickly increases,
response times also increase so that the number of responses drawing penalties decreases.
behavioural adaptation to traffic safety measures is widespread, that the effects

can vary widely, and that there is no evidence for the complete compensation
predicted by risk homeostasis
These experiments show clearly that people will modify their behaviour in response to
the reward and penalty structure of their environment. This is hardly news: behavioural
change in response to reward and risk has been observed in decisions to invest, to buy
insurance, to gamble at the casino, and in hosts of other ways. But this is far from
relevant to injury prevention. Laboratory experiments carry no risk of injury or death.
Performance in the laboratory likely has little or no relation to risk compensation. Even
risk compensation proponents recognize this: ‘Many of the central questions of risk
homeostasis theory, such as the effect of legislation on seat belts or crash helmets, are not
well suited to study within a simulated environment’ (Glendon et al., 1996).
‘Resorting to laboratory and simulation studies may be methodologically pleasing (and
morally innocuous), but it is doubtful that the theory in question [risk compensation] can
ever be cogently tested under such contrived conditions…. In other words: simulation of
risk, like a sham duplicating the real thing, is a contradiction in terms’(Wilde, 1982b).
Risky business 137
‘Risk, by definition, cannot be simulated’ (Wilde, 1994).
Theory
Theory provides the only alternative to the difficulties of testing risk compensation in
either experimental or real-world settings. We are, or we are not, economic beings who
constantly balance costs and benefits in deciding whether to speed up a bit on this road
(knowing that our air bag will protect us if worst comes to worst). We are, or we are not,
motivated by an inherent desire for physical risk. Both common sense and the evidence
briefly reviewed earlier suggest kernels of truth in both propositions, but something far
short of complete acceptance. Further, theory is useful when it can make predictions
about real-world behaviour that are testable. This leads right back to evaluation.
On a personal level, we all know that we sometimes change our behaviour in response
to changes in our environment that appear to increase or decrease our risk. We also know
of and welcome many changes in our environment that have made us safer. On a
professional level, we know that many safety measures have not produced the benefits we
had expected. We also know of many that have indeed reduced injuries substantially.
With these thoughts as incentive to read further, I will give you my views.
A PERSONAL VIEW OF RISK COMPENSATION
First, my disclaimer (or, to understand where I stand, you should know where 1 sit). I am
not a professional economist or psychologist or decision theory professional. I am not an
academic (though I was one once). Instead, I have worked in road safety for 24 years,
developing and implementing many measures that risk compensation proponents find
useless or worse (such as belt use laws, air bags, and various drunk driving counter-
measures).
I believe it is quite clear that from both common sense and formal studies that
behavioural adaptation and risk compensation can occur in some situations. We are
humans, not machines; we react to changed conditions; we are famous for not always
doing as we are told or as is expected of us. On the other hand, the evidence is
overwhelming that risk compensation does not always occur and that risk homeostasis
cannot be justified on any basis short of metaphysics.
Thus the issue becomes not yes or no, but when and how much. When may
compensation occur in response to a safety measure? How likely is it to occur? What are
the possible consequences, both specific and on the system at large?
I suggest that four factors influence compensation in response to a safety measure.
Each factor has several aspects, and the factors interact with each other. None is original.
Many authors have proposed some of them. OECD (1990) suggests that five similar
factors influence risk compensation in response to changes to motor vehicles. These four
factors lead to overall guidance and to principles for action.
Visibility
How obvious is the change produced by the safety measure? Do I even know there has
been a change?
Some changes are very obvious, especially if they affect performance through direct
feedback: brakes, studded tires, and other vehicle handling characteristics, child-resistant
caps on medicine bottles, protective equipment for athletes or workers. Other changes are
apparent if I look for them but easily can be overlooked. I ‘know’ there’s a smoke
detector outside my study, but I think about it only when I replace its battery. Finally,
some changes may be completely or psychologically invisible. The only way I can tell it
is there is from information in the media or in product information. Many features to
reduce or prevent injuries to vehicle occupants, such as side door beams or penetration-
resistant windshields, are invisible for all practical purposes.
Laws and regulations restricting my behaviour can be very obvious, if advertised and
enforced vigorously. On the other hand, they too can be invisible, as we all know.
Evans (1991) takes a strong position:‘Technical changes that are readily apparent to
the driver are very likely to induce user responses’—brakes, handling, tires, etc. On the
other hand, ‘… there is no case of a safety change invisible to road users which has
generated a measurable user response’.
Rule 1: If you do not know it is there, you will not compensate for a safety measure.
Effect
How does the change affect me, both physically and mentally?
This factor has several dimensions. First, how does the change affect my physical
performance of the task, through direct sensory feedback or otherwise? Is it annoying,
like child-resistant medicine caps that too often are adult-resistant as well? Is it physically
uncomfortable, as helmets are for some motorcyclists or seat belts for some vehicle
occupants? Does it make the task easier, like improved vehicle handling or brakes? Or
more difficult, like the lawnmower deadman switch that requires me to hold the handle
constantly?
if it does not affect you, you will not compensate for a safety measure
Second, how does the change affect my attitude? Does it annoy me, like a requirement to
wear seat belts may for a libertarian? Or do I welcome it, like a guardrail added to a
dangerous curve? These two dimensions clearly interact, as changes affecting my
performance may also affect my attitude.
Finally, how does the change affect my perception of risk? Do I feel safer because I am
wearing a bicycle helmet? Do I feel that it has eliminated all risk, as the Titanic’s
passengers and crew may have believed? Or do I think it has little or no effect on my risk,
because I felt there was no risk in the first place, because I believe the measure is
ineffective, or because I don’t know that anything has changed?
Rule 2: If it does not affect you, you will not compensate for a safety measure.
Risky business 139
Motivation
What influences my behaviour? What is my motivation in doing the task? What is my
economic utility function? What are my psychological needs?
This factor is key in most risk compensation theory discussions. Economists hold that
my only influence is my pursuit of fairly well understood economic goals. If I am
driving, they believe my goal is to be transported in the shortest time and to avoid the
economic costs of crashes and injuries. So if my car becomes safer, either because it is
less likely to crash or less likely to injure me if it does, and if the additional safety is not
useful to me, then I will drive faster. The same reasoning applies if 1 am required to use
safety equipment such as seat belts or cycle helmets. In the workplace, especially if my
salary depends on my output, my goal is to maximize my production while keeping my
injury risk below an acceptable level.
Risk homeostasis theorists, on the other hand, hold that my basic goal is to maintain
my desired risk level: ‘It is primarily risk to self that governs behaviour on the
road’ (Adams, 1995).
Both views are simplistic. We are motivated by many factors, both economic and
behavioural. On the road we want to get from here to there while avoiding both personal
injury and crashes. We may or may not care about saving time: while late for a business
appointment, we may cherish every second; on a casual trip, we may decide to take a
longer and slower route because the drive is prettier.
We also are motivated by habit and by our desire to simplify decisions. While we may
make more or less rational decisions in an unfamiliar situation, we quickly fall into habits
and put many daily operations on ‘automatic pilot’. So we do not think each time about
how fast we drive down our neighborhood street, we do it just as we always have done.
Once accustomed to wearing a seat belt or a bicycle helmet, many of us do it every time
without thinking about it. In fact, after seat belt wearing has become a habit, we feel
uncomfortable if a belt is not available. The first few times we wear a hard hat on a
construction site we notice it, but it soon becomes second nature; again, we may feel
uncomfortable without it.
These factors all influence our motivation to compensate for safety changes. If I am
motivated to change behaviour, and if the safety change allows me to do so and still
maintain my previous risk level, I may well compensate. But if there is no motivation for
behaviour change, I will not. As Fuller (1994) states quite vividly in the setting of road
safety: ‘Except where speed increases are rewarding, only very special road users, such
as homicidal maniacs, putative suicides and demolition engineers ever intentionally opt
for a greater chance of collision with obstacles in front of them. The rest of humanity sets
out by and large with a distinct preference for self-preservation and a marked dislike of
visits to the albeit friendly but expensive hospital and vehicle repair shop’.
Rule 3: If you have no reason to change your behaviour, you will not compensate for a
safety measure.
Control
How much do I control the situation? Can I change my actions even if I want to?
Workplace situations frequently are tightly controlled by rules, supervisors, and the
physical environment, so may allow little opportunity for compensation. Piecework
settings provide more flexibility and freedom. Driving allows considerable freedom:
traffic laws provide nominal control, but since most laws are not enforced rigorously,
individual drivers have considerable latitude for their actions. Household settings allow
virtually complete control.
Sports provides interesting examples of the interplay between injury prevention,
compensation, and control. In many sports, players are required to wear protective
equipment. Ice hockey and American football are prime examples. Some players have
compensated by acting more violently within the confines of the rules. In some instances
this has led to rules changes to control player actions more tightly and prohibit some
violent actions.
Rule 4: If your behaviour is tightly controlled, you will not compensate for a safety
measure.
A compensation index
Each of these four factors—visibility, effect, motivation, and control—is far more
complex than this simple discussion suggests. But together they provide a useful
framework for considering potential risk compensation in response to a safety measure: a
highly imprecise Compensation Index. Assess each factor subjectively, from ‘no, not at
all, zero’ to ‘maybe, moderate, some’ to ‘yes, strong, a lot’. As a first approximation:
prefer measures that are invisible to people, or that do not affect their actions or
attitudes, or for which they have no motivation or freedom to change behaviour
• Compensation is unlikely if any of the four is zero: the measure is invisible,

does not affect me, or I have no motivation or no freedom to change my
behaviour.
• As all factors reach moderate levels, the likelihood of compensation
increases.
• If each is high, behavioural compensation is likely: the measure is highly
visible, affects me substantially, I have good reasons to change behaviour and I
have the freedom to do so. Then I will consume at least some of my increased
safety as performance. And even then it’s unclear whether the overall effect will
be to compensate partially, completely, or more than completely for the safety
measure.
Let us apply the Index to a few of the examples considered previously.
• FMVSS: The crash-prevention standards such as better brakes and tires that
improve vehicle performance rate moderate to high on each factor:
compensation is likely (although, as noted previously, many crash-prevention
FMVSS merely codified existing industry practices). The injury-prevention
standards such as improved side door structure are essentially invisible: no
Risky business 141
compensation is expected.
• Seat belt use laws: Belt use is quite visible and affects belt users (though the
effect may be ignored as belt use becomes habitual). However, there is little or
no motivation for more risky driving because belts do not affect the risk of a
crash. Thus compensation is unlikely.
• Safety equipment for chain saw operators: The equipment is very visible and
allows me to work faster with lower injury risk. If I am paid by the quantity
produced, I am highly motivated to increase my output and likely will
compensate substantially. If I am paid by the hour and my working practices are
constrained by rules, I likely will not compensate.
These predictions are in reasonable agreement with the best evidence on risk
compensation in these situations. As a mental experiment, consider several injury
prevention measures that you know well. Does the Index predict compensating behaviour
or not? Is this prediction accurate?
WHY SHOULD YOU CARE ABOUT RISK COMPENSATION?
Injury prevention and control professionals seek to do something, right now if possible, to
reduce unintentional injuries. Academic discussions matter only if they have practical
consequences. With this in mind, it is appropriate to summarize this discussion of risk
compensation with a short list of Things to Think About and Actions to Take in planning
and implementing injury prevention and control measures.
Risk compensation can occur—people are not machines

We all change our behaviour in response to changes in our environment. Our behaviour is
determined by a complex set of rational and behavioural factors. Safety measures change
our environment, so we may change our behaviour in response to them. Never assume
blindly that behaviour will not change.
Four factors influence risk compensation—visibility, effect, motivation, and

control
Risk compensation occurs only in certain circumstances. The four factors and the
Compensation Index help you think through the circumstances of an injury prevention
measure to estimate whether risk compensation is likely or not.
To reduce or eliminate risk compensation, use measures rating low on at least

one factor
Prefer measures that are invisible to people, or that do not affect their actions or attitudes,
or for which they have no motivation or freedom to change behaviour.
Consider system effects
Injury prevention measures may have effects beyond the individual

actions they influence directly. These effects may be harmful or
helpful. Think about them before acting.
Do not over-predict benefits
Many injury prevention measures promise more benefits than they

deliver, due to bad science, political pressures, or failure to consider
risk compensation or system effects. While calm and realistic benefit
estimates are difficult to produce in the heat of battle, unduly
optimistic predictions will hamper injury prevention efforts in the long
run.
Trading safety for performance is not all bad
Safety is not our only goal. All action produces risk; there is no risk-
free life. As society and as individuals we constantly change our
balance of performance and risk (in many dimensions of each). If some
safety benefits predicted for an injury prevention measure become
performance improvements instead, we should be pleased with
contributions to society in both areas.
FOR FURTHER INFORMATION
Key papers
• Sam Peltzman’s 1976 Journal of Political Economy paper first brought the
subject to popular attention. For a sampling of commentaries, rebuttal, and
support see Joksch (1976a and 1976b), Robertson (1977a, 1977b, 1981, 1984,
and 1996), Orr (1984), Graham and Garber (1984), and Peltzman (1976 and
1977). Blomquist (1988) summarizes 13 of these studies.
• Gerald Wilde’s Risk Analysis paper (1982a), four commentaries (Slovic and
Fischhoff, Graham, Orr, and Cole and Withey), and Wilde’s response (1982b)
begin the debate on risk homeostasis. To follow it further, see McKenna (1982),
Wilde (1984), McKenna (1985a), Wilde (1985), McKenna (1985b), McKenna
(1987), Wilde (1988), McKenna (1988), Wilde (1989), and McKenna (1990);
also Evans (1986a), Wilde (1986), and Evans (1986b).
• John Adams’ early work is not in easily-accessible literature, but his views
Risky business 143
are rebuttal. Leonard Evans’ 1985 Human Factors paper discusses 26 traffic
safety studies presented clearly and at some length in his 1985 paper. Mackay
(1985) offers a that illustrate all varieties of risk compensation.
Books
• The OECD report Behavioural adaptations to Changes in the Road Transport

System (1990) gives an excellent, thorough, and dispassionate summary of risk
compensation theories and evidence in road safety.
• John Adams’ Risk and Freedom (1985) treats road safety regulation in
general and seat belt and motorc ycle helmet use laws in particular. His
Risk/John Adams (1995) is a broad discussion of risk and risk compensation
throughout society.
• Gerald Wilde’s Target Risk (1996) summarizes his views, again
concentrating on road safety, provides numerous examples from the literature,
and discusses some of his experimental work.
• Leonard Evans’ Traffic Safety and the Driver (1991) is the definitive work
on the behavioural aspects of road safety. Chapter 11, User Responses to
Changes in Traffic Systems, discusses risk compensation.
• Edward Tenner’s Why Things Bite Back: Technology and the Revenge of
Unintended Consequences (1996) takes a broad view of why things don’t
necessarily work out the way we had expected. The sections on sport speak
directly to behaviour change in response to equipment change intended to
improve safety and performance.
Collections
The literature on risk compensation is spread across many disciplines and journals (in
preparing this paper I reviewed papers in over 60 different journals). For quick access to
several points of view on risk compensation issues, I recommend two books and five
special journal issues (in chronological order).
• Risk Analysis vol. 2, no. 4 (1982) contains Wilde’s fundamental paper, four
commentaries, and Wilde’s response.
• Human Behaviour and Traffic Safety, edited by Leonard Evans and
R.C.Schwing (1985), contains 16 papers from a 1984 symposium together with
both formal and informal discussion. It includes risk compensation papers from
Adams, Mackay, O’Neill et al., and Wilde.
• Accident Analysis and Prevention vol. 18, no. 5 (1986) is a special issue on
risk, with seven papers.
• Ergonomics vol. 31 no. 4 (1988) is a special issue on ‘Risky decision-
making in transport operations’ containing 21 papers from a 1986 workshop on
risk compensation on the road and in the workplace.
• Challenges to Accident Prevention: The Issue of Risk compensation
Behaviour, edited by R.M.Trimpop and G.J.S.Wilde (1994) contains 12 papers
presented at the first and second European Congresses on Psychology
(Amsterdam 1989 and Budapest 1991).
• Safety Science vol. 22, no. 1–3 (1996) is a special issue on risk homeostasis
and risk assessment, with 13 papers.
• Managerial and Decision Economics vol. 17, no. 5 (1996) is a special issue
on product safety and managerial decisions with three papers on risk
compensation considerations regarding cigarette lighters, bicycle helmets, and
lawn mowers.
REFERENCES
Adams J.G.U., 1985a, Smeed’s Law, seat belts and the emperor’s new clothes. In Human
Behaviour and Traffic Safety, edited by L.Evans and R.C.Schwing (New York, NY:
Plenum), pp. 193–248.
Adams J.G.U., 1985b, Risk and Freedom: the Record of Road Safety Regulation. (Cardiff,
Wales: Transport Publishing Projects).
Adams J.G.U., 1988, Risk homeostasis and the purpose of safety regulation. Ergonomics,
31, pp. 407–428.
Adams, J.G.U., 1995, Risk/John Adams (London: UCL Press).
Blomquist G.C., 1988, The Regulation of Motor Vehicle and Traffic Safety (Boston, MA:
Klewer Academic Publishers).
Chirinko, R.S. and Harper, E.P.Jr., 1993, Buckle up or slow down? New estimates of
offsetting behavior and their implications for automobile safety regulation. Journal of
Policy Analysis and Management, 12, pp. 270–296.
Cole, G.A. and Withey, S.B., 1982, The risk of aggregation. Risk Analysis, 2, pp. 243–
247.
Committee on Trauma Research, Commission on Life Sciences, National Research
Council, and the Institute of Medicine, 1985, Injury in America: A Continuing Public
Health Problem. (Washington DC: National Academy Press).
Crandall R.W. and Graham J.D., 1984, Automobile safety regulation and offsetting
behavior: some new empirical estimates. Am. Econ. Rev., 74, pp. 328–331.
Department of Transport, 1985, Compulsory Seat Belt Wearing Report by the Department
of Transport. (London: Her Majesty’s Stationery Office).
Evans L., 1985, Human behavior feedback and traffic safety. Human Factors, 27, pp.
555–576.
Evans L., 1986a, Risk homeostasis theory and traffic accident data. Risk Analysis, 6, pp.
81–94.
Evans L., 1986b, Comments on Wilde’s notes on “Risk homeostasis theory and traffic
accident data.” Risk Analysis, 6, pp. 103–107.
Evans L., 1991, Traffic Safety and the Driver. (New York, NY: Van Nostrand Reinhold).
Evans L. and Schwing R.C., eds., 1985, Human Behavior and Traffic Safety. (New York,
NY: Plenum).
Fuller R.G.C., 1994, To be safely thus? Conditions for risk-compensation and its
modification. In Challenges to Accident Prevention: The Issue of Risk compensation
Behavior, edited by R.M.Trimpop and G.J.S.Wilde. (Groningen, the Netherlands: Styx
Publications), pp. 75–80.
GAO, 1991, Highway Safety: Motorcycle Helmet Laws Save lives and Reduce Costs to
Risky business 145
Society. (Washington, DC: US General Accounting Office).
Garbarcz C., 1990, How effective is automobile safety regulation? Appl Econ, 22, pp.
1705–1714.
Glendon A.I., Hoyes T.W., Haigney D.E. et al., 1996, A review of risk homeostasis
theory in simulated environments. Safety Science, 22, pp. 15–25.
Graham J.D., 1982, On Wilde’s theory of risk homeostasis. Risk Analysis, 2, pp. 235–
237.
Graham J.D. and Garber S., 1984, Evaluating the effects of automobile safety regulation.
J.Policy Analysis and Management, 3, pp. 206–224.
Hoyes T.W. and Glendon A.I., 1993, Risk homeostasis—issues for future research. Safety
Science, 16, pp. 19–33.
Irwin A., 1987, Technical expertise and risk conflict—an institutional study of the British
compulsory seat-belt debate. Policy Sciences, 20, pp. 339–364.
Jackson J.S.H. and Blackman R., 1994, A driving-simulator test of Wilde’s risk
homeostasis theory. Journal of Applied Psychology, 79, pp. 950–958.
Joksch H.C., 1976a, Critique of Sam Peltzman’s study: The effects of automobile safety
regulation. Accident Analysis and Prevention, 8, pp. 129–137.
Joksch H.C., 1976b, The effects of automobile safety regulation: comments on
Peltzman’s reply. Accident Analysis and Prevention, 8, pp. 213–214.
Klen T, 1997, Personal protectors and working behavior of loggers. Safety Science, 25,
pp. 89–103.
Levy D.T. and Miller T., 1999, Review: risk compensation literature, the theory and
evidence. Journal of Crash Prevention and Injury Control, to appear.
Mackay M., 1985, Seat belt use under voluntary and mandatory conditions and its effect
on casualties. In Human Behavior and Traffic Safety, edited by Evans, L. and Schwing,
R.C., pp. 259–278. (New York, NY: Plenum).
McKenna F.P., 1982, The human factor in driving accidents: an overview of approaches
and problems. Ergonomics, 25, pp. 867–877.
McKenna F.P., 1985a, Do safety measures really work? An examination of risk
homeostasis theory. Ergonomics, 28, pp. 489–498.
McKenna F.P., 1985b, Evidence and assumptions relevant to risk homeostasis.
Ergonomics, 28, pp. 1539–1541.
McKenna F.P., 1987, Behavioural compensation and safety. Journal of Occupational
Accidents 9, 107–121.
McKenna F.P., 1988, What role should the concept of risk play in theories of accident
involvement? Ergonomics 31, 469–484.
McKenna F.P. , 1990, In defense of conventional safety measures: a reply to G.J.S.
Wilde. Journal of Occupational Accidents 11, 171–181.
OECD, 1990, Behavioural adaptations to Changes in the Road Transport System. Paris:
Organization for Economic Co-operation and Development.
O’Neill B., Lund A.K., Zador P. and Ashton P., 1984, Mandatory belt use and driver risk
taking: an empirical evaluation of the risk compensation hypothesis. In Human
Behavior and Traffic Safety, eds. L.Evans and R.C.Schwing, pp. 93–107. (New York,
NY: Plenum).
Orr L.D., 1982, Goals, risks, and choices. Risk Analysis 2, 2239–242.
Orr L.D., 1984, The effectiveness of automobile safety regulation: Evidence from the
FARS data. Am. J.Public Health 74, 1384–1389.
Peltzman S., 1975, The effects of automobile safety regulation. J.Political Economy 83,
677–725.
Peltzman S., 1976, The effects of automobile safety regulation: a reply. Accid. Anal, or
Prev. 8, 139–142.
Peltzman S., 1977, A reply to Robertson. J.Economic Issues 11, 672–678.
Rivara R.P. et al., 1999, Systematic reviews of strategies to prevent motor vehicle
injuries. Am. J.Preventive Medicine 16, 1–5.
Robertson L.S., 1977a, A critical analysis of Peltzman’s “The effects of automobile
safety regulation.” J.Economic Issues 11, 586–600.
Robertson L.S., 1977b, Rejoinder to Peltzman. J.Economic Issues 11, 679–683.
Robertson L.S., 1977c, State and federal new-car safety regulation: effects on fatality
rates. Accid. Anal, or Prev. 9, 151–156.
Robertson L.S., 1981, Automobile safety regulations and death reductions in the United
States. Am. J.Public Health 71, 818–822.
Robertson L.S., 1984, Automobile safety regulation: rebuttal and new data. Am. J.Public
Health 74, 1390–1394
Robertson L.S., 1996, Reducing death on the road: the effects of minimum safety
standards, publicized crash tests, seat belts, and alcohol. Am. J. Public Health 86, 31–
34.
Sagberg R., Fosser S. and Saetermo I-A.F., 1997, An investigation of behavioural
adaptation to airbags and antilock brakes among taxi drivers. Accident Analysis and
Slovic P. and Fischhoff B., 1982, Targeting risks. Risk Analysis, 2, pp. 227–234.
Smeed R., 1949, Some statistical aspects of road safety research. Journal of the Royal
Statistical Society, Series A 112, pp. 1–34.
Stanton N. and Glendon I., 1996, Risk homeostasis and risk assessment. Safety Science,
22, pp. 1–13.
Tenner E., 1996, Why Things Bite Back: Technology and the Revenge of Unintended
Consequences. (New York, NY: Random House).
Trimpop R.M. and Wilde G.J.S., eds., 1994, Challenges to Accident Prevention: The
Issue of Risk compensation Behavior. (Groningen, the Netherlands: Styx Publications).
Wagenaar, A.C., 1999, Importance of systematic reviews and meta-analyses for research
and practice. Am. J.Preventive Medicine, 16, pp. 9–11.
Wilde G.J.S., 1976, The Risk compensation Theory of Accident Causation and Its
Practical Consequences. (Kingston, Ontario: Queen’s University).
Wilde G.J.S., 1982a, The theory of risk homeostasis: Implication for safety and health.
Risk Analysis, 2, pp. 209–225.
Wilde G.J.S., 1982b, Critical issues in risk homeostasis theory. Risk Analysis, 2, pp. 249–
258.
Wilde G.J.S., 1984, Evidence refuting the theory of risk homeostasis? A rejoinder to
Frank P.McKenna. Ergonomics, 27, pp. 297–304.
Wilde G.J.S., 1985, Assumptions necessary and unnecessary to risk homeostasis.
Ergonomics, 28, pp. 1531–1538.
Wilde G.J.S., 1986, Notes on the interpretation of traffic accident data and of risk
homeostasis theory: A reply to L.Evans. Risk Analysis, 6, pp. 95–101.
Wilde G.J.S., 1988, Risk homeostasis theory and traffic accidents: Propositions,
deductions and discussion of dissension in recent reactions. Ergonomics, 31, pp. 441–
468.
Wilde G.J.S., 1989, Accident countermeasures and behavioral compensation: The
position of risk homeostasis theory. Journal of Occupational Accidents, 10, pp. 267–
292.
Risky business 147
Wilde G.J.S., 1994, Target Risk: Dealing with the Danger of Death, Disease and
Damage in Everyday Decisions. (Toronto: Castor and Columba).
Zlatoper T.J., 1984, Regression analysis of time series data on motor vehicle deaths in the
United States. Journal of Transport Economics and Policy, 18, pp. 263–274.
11
Aging and Transportation: Mobility or Safety?
Liisa Hakamies-Blomqvist
INTRODUCTION
Research on ageing and traffic safety has a history dating to the thirties. In an early paper
(DeSilva, 1938), concern was expressed at the steadily increasing proportion of drivers
over 40 years of age. During the beginnings of traffic gerontological research, which was
dominated by research teams in the US, ageing of drivers was considered the main issue.
Since late sixties, empirical evidence on different aspects of ageing and driving has
accumulated, and mainstream older driver research has shifted focus a few times. The
scope of research also has widened from drivers to other road users, and from treating
older road users mainly as a safety problem to also encompass mobility and quality of life
related issues.
This paper describes the main trends in gerontological traffic safety research during the
last three decades, with focus on research on private car accidents, since this has been the
most active area of traffic gerontology. The scope of the report makes it impossible to
present an exhaustive review of all relevant research; rather, the aim is to illustrate certain
paradigmatic shifts with a few examples of representative studies. For sake of clarity, this
report is organized chronologically by decade; the author apologises for the unavoidable
geometrical oversimplification.
THE SEVENTIES: IS THERE AN OLDER DRIVER PROBLEM?
The ‘older driver problem’ first became established as a result of a wave of intensive
research activities on older drivers by the end of the sixties, mainly, in the US. As a result
of these efforts (for an overview, see Grow, 1972), general trends in accident rates and
exposure characteristics were established. Thus, it was found that older drivers had less
accidents per capita but more per mileage and that they in general drove less than did
middle-aged drivers (Finesilver, 1969; McFarland, 1964; Planek, 1972). The importance
of different aspects of vision for safe driving was thoroughly studied by Hills and Burg
(1977)
At the North Carolina Symposium on Highway Safety in 1972, Planek (1972)
summarized the status quo in a critical review the focus of which he defined in the
following manner: ‘In studying the effect of the ageing process on driving, we shall be
primarily interested in drivers over 55 years of age, although some age-associated change
in driving activity itself may start as early as 50. This discussion will focus on defining
the deficiencies of ageing drivers and examining them in relation to driving performance
Aging and transportation 149
research. Hopefully, from such a review, we can begin to assess the impact of the ageing
driver in today’s traffic both quantitatively and qualitatively.’
According to Planek, deficiencies of the summarizing the findings available, ageing
driver seemed to fall into the two overlapping categories: those due to the ageing process
itself and those due to some medical disability. The age-related deficiencies were
delineated along three general areas: sensory reception, neural processing and
transmission, and motor response.
Planke’s views reflected the Zeitgeist in research on ageing drivers in two ways. First,
in harmony with the concurrent interest in accident proneness (Echterhoff, 1990), the
driver’s ‘deficiencies’ were seen exclusively as the cause of all problems, and the
characteristics of the traffic system were taken largely for granted. Second, the concepts
used for the driver’s internal faculties clearly date from a period before the ‘cognitive
revolution’; the lack of an information-processing viewpoint and cognitive terminology is
flagrant for a modern reader. Another analysis presented by Mann (1972) at the same
meeting reflects similar thinking. Although discussion about the importance of decision
making processes and about possible safety-related changes in the traffic system soon
emerged (Planek and Overend, 1973), the general recommendations made by the US
National Conference on the Ageing Driver (National Conference on the Ageing Driver
(NCAD, 1974) were dominantly oriented towards screening drivers and eliminating those
with higher risk from the driver population.
It can only be speculated whether the impact of this first wave of interest on future
older-driver research and traffic planning might have been different had a more system-
oriented view of the human operator already prevailed. As it was, the problem became
established in terms of decreased safety of deficient drivers and the need to do something
about them.
older drivers are ‘good insurance risk’ but have more accidents per distance
driven
In Europe, interest in ageing and driving seems to have emerged later, although some
valuable work was done as early as the fifties (Häkkinen, 1954). In some European
countries, age-related controls or restrictions to driver licensing were stipulated in early
seventies, but generally speaking the issue did not get much attention before the eighties
(Hakamies-Blomqvist, 1996).
THE EIGHTIES: WHAT IS THE NATURE OF THE OLDER DRIVER

PROBLEM?
Once it had been settled that there was, or would be, an ‘older driver problem’, research
in the eighties was directed toward a more thorough understanding of the general traits of
this problem. Early findings on accidents were supported and complemented by later
studies. Thus, it was shown that older drivers are ‘good insurance risk’ (Cooper, 1990;
Wiener, 1972) but have more accidents per distance driven (Brorsson, 1989; Cerelli,
1989; Evans, 1987; Evans, 1988; Graca, 1986). When involved in an accident, older
drivers were found to be more often than not the legally responsible party, i.e., the party
at fault (Knoflacher, 1979; McKelvey and Stamatiadies, 1989; Partyka, 1983; Verhaegen
et al., 1988; Viano et al., 1990). Older drivers were shown to be over-represented in
accidents occurring at intersections and other complex traffic situations (Broughton,
1988; Cerelli, 1989; Hauer, 1988; OECD, 1985; Partyka, 1983; Stamatiadis et al., 1991;
Viano et al., 1990; Yanik, 1985). Older drivers were also found to be convicted more
often for corresponding violations in traffic (McKelvey and Stamatiadis, 1989; Rothe,
1990). On the other hand, they turned out to have less single-vehicle accidents
(Campbell, 1966; Cerelli, 1989; Moore et al., 1982). It was also established that accidents
of older drivers occurred mostly in the daytime (Broughton, 1988; Campbell, 1966;
Cerelli, 1989; Hauer, 1988) and that accident-involved older drivers were less often
alcohol-intoxicated than accident involved young or middle-aged drivers (Berghaus et al.,
1983).
While the results of the research papers ofthis period usually are neat and
straightforward, the discussion in these papers tend to be disappointing. Attempts to
explain the emerging accident picture of older drivers were mostly based on an ad hoc
combination of elements belonging to knowledge bases: (1) a task analysis of car driving,
now mostly described in information processing terms (e.g., flow charts), (2)
gerontological data about age-related changes in different aspects of human performance,
and (3) accident statistics. From these three database, hypothetical explanations were
generated by matching the information about which faculties are necessary for safe
driving, which functions change with age, and which kinds of accidents occur. The
elements chosen to explain the accident picture were mostly picked on the basis of their
face value. Thus, since (1) vision certainly is one of the necessary faculties for safe
driving, and (2) several visual functions deteriorate with age, and (3) older drivers were
over-represented in intersection accidents where they failed to see their collision partners
in time, age-related changes in vision often were blamed for their accidents.
A central weakness of such an approach was that though empirical evidence of the
nature of the accidents accumulated, insight into their causation did not, since the same
speculations were expressed again and again with little or no effort towards deepening
the understanding of the behavioural mechanisms mediating the effects of functional
deficits via driving behaviour to accident statistics.
A definite improvement, however, compared to earlier research, was a shift of focus
towards higher cognitive functions in search of factors explaining the age-related
accident risk. During the seventies, sensory and motor phenomena were the focus; during
the eighties, critical deficiencies were described in referring to speed. Neural conduction
time was mercifully forgotten as a major factor explaining accidents.
The success story of the UFOV research (Useful Field of View), a dynamic measure of
the functionally available field of view, is a good example of the power of a more
cognitive approach. In an early study, Burg and Hills had tried to establish statistical
relations between, basic visual functions accident risk (Burg, 1967; Burg, 1968; Hills and
Burg, 1977). Despite a sample of over 17,500 drivers, their findings were modest.
Discussing their failure to achieve stronger predicting power, Hills and Burg claimed that
perceptual rather than sensory measures would be useful in predicting accidents; for
instance ‘useful visual field’ instead of total visual field was proposed as a tool for future
researchers. Ten years later, such a measure was developed. In a retrospective accident
study, in combination with a measure of cognitive performance, UFOV explained 29 per
cent of the variance in older drivers’ prior 5-year rates of accidents at intersections
(Owsley et al., 1991).
While the adaptation of a cognitive frame of thinking was a major improvement, the
main weakness of ‘modern’ flow chart models of driver behaviour was their ahistorical
nature. Such models only presented a snapshot and failed to take into account the nature
of driving as a skilled performance with a long learning history. Conceptualizing driving
as skilled behaviour rather than, or in addition to its being a complex information
processing task has become increasingly popular in the nineties; in the beginning of the
nineties, some important advances also were made in the study of ageing of cognitive
skills (Bosman, 1993). This approach certainly shows promise toward a deeper
understanding of the weaknesses, strengths and compensatory strategies of older drivers.
One of the corner stones of the definition of the older driver problem, the claim
concerning their higher accident rates, was contested in the eighties when it was pointed
out that a higher risk of injury leads to sampling bias in accident data bases. As
demonstrated by Evans (1991) on the basis of several earlier studies, the greater physical
frailty of older individuals explains an important part of their higher rates of injury and
fatal accidents. Challenging the traditional problem definition, Evans concluded that for
older drivers, accident risk in fact was a minor issue, and limitations in mobility, due to
self-imposed compensatory restrictions in driving exposure, were the real problem. In the
eighties, many European countries felt the need of conducting a national state-of-the-art
review on older road users. Depending on the actors involved, these different national
papers have somewhat different focuses and levels of discussion.
older drivers were shown to be over-represented in accidents occurring in

intersections and other complex traffic situation
In 1989, a report by the European Conference of Ministers of Transport summarized the

situation in Europe regarding population and licensing trends, accidents, and
implementation of safety measures focusing on older road users. The general findings
and conclusions were largely in harmony with an earlier OECD report (OECD, 1985).
From the beginning, the European discourse on ageing and driving has had a somewhat
different orientation than the North American one. Characteristic for the European
approach is a system perspective: issues like factors affecting modal choice and safety as
unprotected road user have higher priority in Europe than in the US where the focus is
more on private car use. In Europe there is also an explicit attempt to design traffic
facilities ‘for all’; groups with special needs are not primarily seen as targets for specific
solutions but rather as criterion users guiding the design toward solutions beneficial for
all users.
THE NINETIES: WHOSE PROBLEM IS THE ‘OLDER DRIVER

PROBLEM’?
Arguably the most important change of direction in older driver accident research is the
recent shift of focus from a general approach toward a differential one. Research efforts
in the eighties had attempted to describe the general nature of the older driver problem.
However, it was, at the same time, increasingly recognized that the problem may not be a
genetal one. Gerontological research had long since shown that inter-individual variance
in increases with age. Clinical experience pointed out certain sub-groups of older patients
having illnesses that could affect abilities essential for safe driving as a major source of
safety concern, more than ‘normal ageing’. Thus, while earlier research mostly was
guided by the question ‘Why do older drivers have higher accident risk?’ the alternative
question ‘Which older drivers have higher accident risk?’ has gained momentum in the
nineties (Hakamies-Blomqvist, 1998).
Of risk-increasing illnesses, dementia of Alzheimer type has received most attention.
The scientific and political discussion about dementia as risk factor for car driving started
with a study by Friedland et al. (1988) claiming that older drivers with dementia had 4.7
times the accident risk of healthy older drivers. The authors recommended on the basis of
these findings that a diagnosis of dementia (in their case Dementia of Alzheimer Type,
DAT) should always lead to a revocation of driver licence. Similar findings and
conclusions were presented by Lucas-Blaustein later the same year (Lucas-Blaustein et
al., 1988). Serious protests were, however, immediately presented against this
recommendation which was claimed to be premature and to have negative consequences
(Drachman, 1998). While most studies showed a substantial risk increase due to dementia
(Carr et al., 1990; Drachman et al. 1993; Dubinsky et al., 1992; Fitten et al., 1995; Gilley
et al., 1991; Logsdon and Larson, 1992 ; Lucas-Blaustein et al., 1988; O’Neill et al.,
1992; Trobe et al., 1996; Tuokko et al., 1995), it was also shown that the risk does not
necessary increase in the beginning of the illness, and that many patients with DAT
diagnosis have intact driving ability (Hunt et al., 1997). In 1994, an international
consensus conference was organized (documented in Lundberg et al., 1997) concerning
dementia and driving. The conclusion was that patients with beginning or mild dementia
should be allowed to drive but there should be a periodical follow-up in order to monitor
changes in performance.
Although it has been convincingly demonstrated that dementia increases a driver’s
accident risk it does not necessarily follow that dementia is a major factor in older
drivers’ accident statistics; demented drivers may limit their exposure or stop driving
entirely. Several studies have shown that between 20 and 30 per cent of drivers affected
by dementia continue driving (Gilley et al., 1991, Carr et al., 1990, Logsdon and Larson,
1992). In a Swedish-Finnish collaborative study, Johansson et al. (1997) addressed the
question of how big a share all older drivers’ accidents could actually be attributed to
drivers suffering from dementia. They used micropathological methods to study certain
parts of the brain of aged drivers killed in traffic accidents, and concluded that those with
neuropathological changes in their brains, indicating possible or probable DAT (dementia
of Alzheimer type), were clearly over-represented, around 50 per cent of the killed
drivers.
If demented drivers turn out to be heavily over-represented among accident-involved
older drivers, as indicated by Johansson’s study, then it follows that non-demented
drivers have a smaller share of the accidents than has been generally believed. The
distribution of individual accident risk among the elderly may indeed be bimodal rather
than normal: in virtue of their cautious driving style and self critical attitude, normally, or
‘successfully’ ageing older drivers may be extremely safe drivers whereas illnesses
causing dementia may lead to a dramatic increase of risk—which brings up the old
concept of ‘accident proneness’ again in a fresh shape. In harmony with this thinking,
periodic medical controls for older drivers are often suggested as safety measure when
discussing illness-related risk increase. Those studies in which existing systems have
been evaluated have, however, failed to show beneficial effects of medical screening of
older drivers (Hakamies-Blomqvist, 1996; Hull, 1991), except for small safety benefits
for regular vision controls (Levy et al., 1995; Shipp, 1998).
while earlier research mostly was guided by the question ‘Why do older drivers
have higher accident risk?’ the alternative question ‘which older drivers have
higher accident risk?’ has gained momentum in the nineties
The realization of the importance of cognitive decline in accident causation has

stimulated interest in driver diagnostics and related research. Generally speaking,
research on diagnostic testing of older drivers, while flourishing in the US, is still young
in Europe, and there are several problematic issues. First, there is the ‘eternal’ problem of
accidents as outcome measure. Since accidents are rare multi-determined events it is
difficult to achieve sufficient statistical power using accident data as outcome variable.
However, the validity of any surrogate measures remains to be demonstrated. Second, it
is not always clear whether the aim of the diagnostic studies is in reality to find methods
sensitive for safety decrements, or rather, to find methods sensitive for early dementia,
since signs of beginning dementia or other cognitive decline tend to correlate strongly
with any surrogate measures of increased risk. In the latter case, the research done in a
traffic context risks to duplicate more powerful research addressing the same issues
within a medical context.
The concept of risk has continued to intrigue traffic gerontologists in the nineties.
Janke (1991) demonstrated in a pertinent review paper how comparisons based on
accidents per mileage exaggerate the risk of driver groups with small yearly mileage as
compared to those driving more. Maycock (1997) analysed European accident statistics
with respect to the frailty effect and concluded that half of the increased fatality risk of
drivers aged 75 years or more, compared to drivers aged 30 years, might be due to the
enhanced susceptibility of the older drivers to be killed in the accidents in which they are
involved, rather than to their higher accident rates. Hakamies-Blomqvist (1994, 1998)
pointed out that the finding that older drivers tend to be legally responsible parties in their
collision accidents can be equally well stated as that older drivers are very hard to hit as
innocent parties because of their cautious and defensive driving style. Hence, surrogate
measures of induced exposure relying on the ratio of at-fault to not-at-fault tend to
exaggerate the risk of older drivers. Finally, in a number of studies an attempt has been
made to dissect the effects of age, cohort, and period on older drivers’ accident picture.
The overall findings indicate that typical age-related findings emerge at a later age in
cohorts born later as compared to those born earlier (Evans, 1993; Stutts and Martell,
1992; Hakamies-Blomqvist and Henriksson, in press).
There has been an impressive amount of research on different aspects of driving
behaviour during the nineties, which it is not possible to review here. Suffice it to
mention a few topical areas: trends and changes in travel behaviour, actual driving
behaviour such as merging decisions at intersections, compensatory strategies such as
limiting one’s driving exposure, and reasons of driving cessation.
FUTURE DIRECTIONS
In harmony with the redefinition of the older driver problem, recent traffic gerontological
research seems to have adopted a wider scope than earlier. Travel behaviour is more
often studied in the whole context of everyday life. Aspects other than safety are often in
focus, and an increasing interest in a phenomenological point of view is evident. Another
recent trend is the emergence of older female drivers as a special group. Integrative
research efforts containing a ‘user perspective’ can be expected in future, as well as a
growing interest in ‘life after licence’.
REFERENCES
Berghaus, G., Pieper, W. and Staak, M., 1983, Der alkoholisierte ältere
Verkehrsteilnehmer—Typologie und Trends anhand der polizeilichen Erfassung.
Report 42. (Köln: Unfallund Sicherheitsforschung Strassenverkehr).
Bosman, E.A., 1993, Age-related differences in the motoric aspects of transcription
typing skill. Psychology and Ageing, 8, pp. 87–102.
Brorsson, B., 1989, The Risk of Accidents Among Older Drivers. Scandinavian Journal
of Social Medicine, 17, pp. 253–256.
Broughton, B.S., 1988, The variation of car drivers accident risk with age. Research
Report 135. (Crowthorne, UK: Transport and Road Research Laboratory).
Burg, A., 1967, The relationship between vision scores and driving record: general
findings. Report 67–24. (California: Department of Engineering, University of
California).
Burg, A., 1968, Vision test scores and driving record: additional findings. Report 68–27.
(California: Department of Engineering, University of California).
Campbell, B.J., 1966, Driver age and sex related to accident time and type. Traffic Safety,
10, pp. 36–42.
Carr, D., Jackson, T. and Alquire, P., 1990, Characteristics of an elderly driving
population referred to a geriatric assessment center. Journal of the American Geriatrics
Society, 38, pp. 1145–1150.
Cerrelli, E., 1989, Older Drivers: The Age Factor in Traffic Safety. NHTSA Technical
Report DOT HS 807 402. (Washington, DC: National Highway Traffic Safety
Administration).
Cooper, P.J., 1990, Differences in accidents characteristics among elderly drivers and
between elderly and middle-age drivers. Accident Analysis and Prevention, 22, pp. 499–
508.
DeSilva, H.R., 1938, Age and highway accidents. Scientific Monthly, 47, pp. 536–545.
Drachman, D.A., 1988, Who may drive? Who may not? Who shall decide? Annals of
Neurology, 24, pp. 787–788.
Drachman, D.S., J.M., 1993, Driving and Alzheimer’s disease: The risk of crashes.
Neurology, 43, pp. 2448–2456.
Dubinsky, R.M., Williamson, A., Gray, C.S. and Glatt, S.L., 1992, Driving in
Alzheimer’s Disease. Journal of American Geriatrics Society, 40, pp. 1112–1116.
Echterhoff, W., 1990, Geschichte der Verkehrspsychologie. Zeltschrift fur
Verkehrssicherheit, 36, pp. 50–70.
Evans, L., 1987, Fatal and severe crash involvement versus driver age and sex. In 31st
Conference of the American Association for Automotive Medicine, New Orleans, pp.
59–78.
Evans, L., 1988, Older driver involvement in fatal and severe traffic crashes. Journal of
Gerontology: Social Sciences, 43, pp. S186–S193.
Evans, L., 1991, Traffic Safety and the Driver, (New York: Van Nostrand Reinhold).
Evans, L., 1993, How safe were today’s older drivers when they were younger? In
Transportation Research Board, 72nd Annual Meeting. Washington, DC.
Finesilver, S.G., 1969, The Older Driver: A Statistical Evaluation of Licensing and
Accident Involvement in 30 States. (Denver: University of Denver College of Law).
Fitten, L.J., Perryman, K.M., Wilkinson, C.J., Little, R.J., Burns, M.M., Pachana, N.,
Mervis, J.R., Malmgren, R., Siembida, D.W. and Ganzell, S., 1995, Alzheimer and
vascular dementia and driving. A prospective and laboratory study. Journal of the
American Medical Association, 273, pp. 1360–1365.
Friedland, R.P., Koss, E., Kumar, A., Gaine, S., Metzler, D., Haxby, J.V. and Moore, A.,
1988, Motor vehicle crashes in dementia of the Alzheimer Type. Annals of Neurology,
24, pp. 782–786.
Gilley, D.W., Wilson, R.S., Bennet, D.A., Stebbins, G.T., Bernard, B.A., Whalen, M.E.
and Fox, J.H., 1991, Cessation of driving and unsafe motor vehicle operation by
dementia patients. Arch Intern Med, 151, pp. 941–946.
Graca, J.L., 1986, Driving and ageing. Clinics in Geriatric Medicine, 2, pp. 577–589.
Grow, N.L., 1972, The literature of ageing pedestrians and drivers: a bibliography 1962–
1972. In Ageing and Highway Safety: The Elderly in a Mobile Society, Vol. 7 edited by
Planek, T.W., Mann, W.A. and Wiener, E.L. (Chapel Hill, NC: North Carolina
Symposium on Highway Safety).
Hakamies-BIomqvist, L., 1994, Ageing and fatal accidents in male and female drivers.
Journal of Gerontology, Social Sciences, 49, pp. S286–S290.
Hakamies-BIomqvist, L., 1996, Research on older drivers: A review. IATSS Research,
20, pp. 91–101.
Hakamies-BIomqvist, L., 1998, Older drivers’ accident risk: conceptual and
methodological issues. Accident Analysis and Prevention, 30, pp. 293–297.
Hakamies-BIomqvist, L. and Hensriksson, P., in press, Cohort effects in older drivers,
accident type distribution: Are older drivers as old as they used to be? Transportation
Research Part F, Traffic Psychology and Behavior.
Hakamies-BIomqvist, L., Johansson, K. and Lundberg, C., 1996, Medical screening of
older drivers as a traffic safety measure—A comparative Finnish-Swedish evaluation
study. Journal of the American Geriatrics Society, 44, pp. 650–653.
Häkkinen, S., 1954, Sambandet mellan alder och trafikolyckor (Relation between age and
traffic accidents). Nordisk Psykologi, 6, pp. 77–92.
Hauer, E., 1988, The safety of older persons at intersections. In Transportation in an
Ageing Society: Improving Mobility and Safety for Older Persons. Special report 218,
Vol. 2. (Washington, DC: Transportation Research Board, National Research Council).
Hills, B.L. and Burg, A., 1977, A re-analysis of California driver vision data: General
findings. TRRL Laboratory Report 768. (Crowthorne, UK: Transport and Road
Research Laboratory).
Hull, M., 1991, Driver license review: Functionally impaired and older drivers.
(Hawthorn: VIC Roads Road Safety Division).
Hunt, L.A., Murphy, C.F., Carr, D., Duchek, J.M., Buckles, V. and Morris, J.C., 1997,
Environmental cueing may affect performance on a road test for drivers with dementia
of the Alzheimer type. Alzheimer Disease and Associated Disorders, 11, pp. 13–16.
Janke, M.K., 1991, Accidents, mileage, and the exaggeration of risk. Accident Analysis
and Prevention, 23, pp. 183–188.
Johansson, K., Bogdanovic, N., Kalimo, H., Winblad, B. and Viitanen, M., 1997,
Alzheimer’s disease and apolipoprotein E e4 allele in older drivers who died in
automobile accidents. The Lancet, 349, p. 1143.
Knoflacher, H., 1979, Altersspezifische Unfallumstände. Zeltschrift für
Verkehrssicherheit, 25, pp. 131–134.
Levy, D.T., Vernick, J.S. and Kim, A.H., 1995, Relationship between driver’s license
renewal policies and fatal crashes involving drivers 70 years or older. Journal of the
American Medical Association, 274, pp. 1026–1030.
Logsdon, R.G., Teri, L. and Larson, E.B., 1992, Driving and Alzheimer’s disease.
Journal of General Internal Medicine, 7, pp. 583–5 88.
Lucas-Blaustein, M.J., Filipp, C.L., Dungan, C. and Tune, L., 1988, Driving in patients
with dementia. Journal of the American Geriatrics Society, 36, pp. 1087–1091.
Lundberg, C., Johansson, K., Ball, K., Bjerre, B., Blomqvist C., Braekhus, A., Brouwer,
W., Bylsma, F., Carr, D.B., Englund, L., Friedland, R., Hakamies-Blomqvist, L.,
Klemetz, G., O’Neill, D., Odenheimer, G.L., Rizzo, M., Schelin, M., Seideman, M.,
Tallman, K., Viitanen, M., Waller, P.F. and Winblad, B., 1997, Dementia and
driving—An attempt at consensus. Alzheimer Disease and Associated Disorders, 11,
pp. 28–37.
Mann, W.A., 1972, Problems of the ageing driver. In Ageing and Highway Safety: The
Elderly in a Mobile Society edited by Planek, T.W., Mann, W.A. and Wiener, E.L.
(Chapel Hill, NC: North Carolina Symposium on Highway Safety)
Maycock, G., 1997, The Safety of Older Car-drivers in the European Union.
(Basingstoke: AA Foundation for Road Safety Research).
McFarland, R.A., 1964, On the driving of automobiles by older people. Journal of
Gerontology, 19, pp. 190–197.
McKelvey, F.X. and Stamatiadis, N., 1989, Highway Accident Patterns in Michigan
Related to Older Drivers. In Transportation Research Record 1210. (Washington, DC:
Transportation Research Board, National Research Council).
Moore, R.L., Sedgley, I.P. and Sabey, B.E., 1982, Ages of car drivers involved in
accidents, with special reference to junctions. Supplementary Report 718. (Crowthorne,
UK: Transport and Road Research Laboratory).
NCAD, 1974, U.S. National Conference on the Ageing Driver, May 2–4, 1974. Journal
of Traffic Medicine, 2, pp. 45–46.
O’Neill, D., Neubauer, K., Boyle, M., Gerrard, I., Surmon, D. and Wilcock, G.K., 1992,
Dementia and driving. Journal of the Royal Society of Medicine, 85, pp. 199–201.
OECD, 1985, La Sécurité des Personnes Âgéess dans la Circulation Routière. (Paris:
OECD).
Owsley, C., Ball, K., Sloane, M.E., Roenker, D.L. and Bruni, J.R., 1991, Visual/
cognitive correlates of vehicle accidents in older drivers. Psychology and Ageing, 6,
pp. 403–415.
Partyka, S.C., 1983, Comparison by Age of Drivers in Two-car Fatal Crashes.
(Washington, DC: US Dept. of Transportation, National Highway Traffic Safety
Administration).
Planek, T.W., 1972, The ageing driver in today’s traffic: a critical review. In Ageing and
Highway Safety: The Elderly in a Mobile Society, Vol. 7 edited by Planek, T.W.,
Mann, W.A. and Wiener, E.L. (Chapel Hill, NC: North Carolina Symposium on
Highway Safety).
Planek, T.W. and Overend, R.B., 1973, How Ageing Affects the Driver. Traffic Safety,
17, pp. 13–39.
Rothe, J.P., 1990, The Safety of Elderly Drivers. (London: Transaction Publishers).
Shipp, M., 1998, Potential human and economic cost-savings attributable to vision testing
policies for driver license renewal, 1989–1991. Optometry and Vision Science, 75, pp.
103–118.
Stamatiadis, N., Taylor, W. and McKelvey, F.X., 1991, Older drivers and intersection
traffic control devices. Journal of Transportation Engineering, 117, pp. 311–319.
Stutts, J.C. and Martell, C., 1992, Older driver population and crash involvement trends,
1974–1988. Accident Analysis and Prevention, 24, pp. 317–327.
Trobe, J.D., Waller, P.P. Cook-Flannagan, C.A., Teshima, S.M. and Bieliauskas. L.A.
(1996). Crashes and violations among drivers with Alzheimer’s disease. Archives of
Neurology, 53, pp. 411–416.
Tuokko, H., Tallman, K., Beattie B.L., Cooper, P. and Weir, J., 1995, An examination of
driving records in a dementia clinic. Journal of Gerontology: Social Sciences, 50B, pp.
S173-S181.
Verhaegen, P.K., Toebat, K.L. and Delbeke, L.L., 1988, Safety of older drivers—A study
of their over-involvement ratio. In 32nd Annual Meeting of the Human Factors
Society, Vol. 1. (Anaheim, California: The Human Factors Society)
Viano, D.C., Culver, C.C., Evans, L. and Frick, M. ,1990, involvement of older drivers in
multivehicle side-impact crashes. Accident Analysis and Prevention, 22, pp.177–188.
Wiener, E.L., 1972, Elderly pedestrians and drivers: the problem that refuses to go away.
In Ageing and Highway Safety: The Elderly in a Mobile Society edited by Planek,
T.W., Mann, W.A. and Wiener, E.L. (Chapel Hill, NC: North Carolina Symposium on
Highway Safety).
Yanik, A.J., 1985, What accident data reveal about elderly drivers. SAE Technical Paper
Series 851688. The Engineering Society For Advancing Mobility Land Sea Air and
Space.
12
Adolescents’ Risk-taking Behaviour, Myth or
Reality: Evidence from International Data
Anne Tursz
INTRODUCTION
Professionals working in the area of adolescent health generally agree that adolescence is
a period characterized by a high frequency of risk-taking behaviour. This consensus has
been possible especially since Jessor and Jessor (1977) defined the psychological and
social utility of risk for this age group, Zuckerman (1971, 1979) showed that the
‘sensation-seeking’ curve reaches its apogee at the end of adolescence (16–19 years of
age), and Holinger (1981) compared adolescent risk-taking behaviour with a tendency for
self-destructiveness. Recently, several articles and books (Tonkin, 1987; Special issue of
the Journal of Adolescent Health, 1991; Tursz et al., 1993; Jonah, 1997) and publications
from surveillance systems (Kann et al., 1998) have reviewed the state of knowledge
about these behaviours, and the possibility of preventing them (Dryfoos, 1991).
Agreement within clinical and public health circles on the issue of risk as a
characteristic of adolescence is based primarily on conclusions of a statistical nature.
Mortality and morbidity linked to certain risk-taking behaviours seem particularly high at
that age. Epidemiological studies on risk-taking during adolescence investigate risk-
taking behaviours per se such as, consumption of alcohol, tobacco, illegal drugs,
dangerous driving, unprotected sex, suicide attempts, violence as well as their negative
consequences-intentional or unintentional violent deaths, accidents, mortality related to
drug abuse, unwanted pregnancies, sexually transmitted diseases, including AIDS.
As a matter of fact, in most countries, definitely in industrialized countries (IC), no
peak in frequency is observed among adolescents for these causes of morbidity and
mortality, except in the case of accidents (Tursz, 1997). Accidents are the primary cause
of death among children and adolescents, with a peak in mortality for the 15 to 24 age
group in most IC (WHO, 1995).
The age limits of adolescence vary from one study to another. Data presented in this
article cover different age groups but most are between 11 and 19 years of age. However,
mortality figures concern subjects between 15 and 24 years (adolescents and young
adults).
Bibliographical research of recent studies on adolescents’ risk-taking behaviour
produce many more references on sexual behaviour, STDs and AIDS, than on
behavioural causes of unintentional injuries. Questioning scientific data bases on the
specific issue of adolescents’ accidents leads to the identification of a large number of
articles reporting the results of studies conducted in ICs (mainly on traffic accidents in
Adolescents' risk-taking behaviour, myth or reality 159
USA), but very few on accidents in developing countries (DC), and almost none on risk-
taking behaviour in DCs. Nevertheless, it was recently estimated that 4 out of 5 young
people live in DCs (Friedman, 1989), unintentional injuries (mainly road traffic
accidents) are a growing problem in DCs (Smith and Barss, 1991), with nearly three-
quarters of road deaths in the world occurring in DCs (Odero et al., 1997), and increasing
population density being associated with a proportionately greater number of traffic
related deaths among the youth (Söderlund and Zwi, 1995).
In this paper, the intention therefore is to present data on adolescents’ accidents, when
possible, from both ICs and DCs. Epidemiological data analysed in this article deal with
both, mortality and morbidity related to risk-taking behaviours and with the behaviours
themselves. Mortality and morbidity figures relate only to unintentional injuries and
accidents, since they constitute a health problem internationally recognized as specific to
adolescence, whereas suicide rates, for example, increase with progression of age in most
countries (WHO, yearly publication). Regarding risk-taking behaviour and protective
behaviour, examples will be presented and discussed from various areas, such as drug
consumption or sexual behaviour. This review of epidemiological data does not presume
to be exhaustive but rather aims at presenting and analysing relevant examples. It also
aims at identifying some methodological problems in study design and data collection
which sometimes make the interpretation of figures difficult, especially at an
international level.
some risks may even be rewarded. This is the case with risk in sports, sports being
one of the few ways violence may repeatedly be expressed with the agreement of
Western society
Analysis of epidemiological data leads to a consideration of what really is specific about

adolescence. Indeed, looking at figures, rates and percentages of the consumption of toxic
substances, of sexual partners, raises numerous questions. Are these figures high, average
or low? In comparison to what or to what standard? The standard for figures found
among adults, for example? Or are they being compared to some theoretical standard
linked to the level of social acceptance of the negative consequences of risk-taking by an
adolescent (the enormous economic, social and psychological cost of traffic accidents, for
example)? The level of social acceptance of risk varies, with different societies and
cultures selecting different acceptable and unacceptable risks. Some risks may even be
rewarded. This is the case with risk in sports, sports being one of the few ways violence
may repeatedly be expressed with the agreement of Western society.
A social definition of risk, including ideas of negotiation in terms of social norms, of
consensus, is far from the epidemiological definition used in public health (probability of
an event occurring, that an individual will die or fall sick during a given period). This
definition, with its very negative connotation and statistical data gathered in
epidemiological studies, serve as the basis for planning prevention strategies. One may
very well question the narrowness of this definition as well as its implications for
rejecting any notion of benefit associated with risk. Taking into account the utility of
risk-taking, and the cultural and social value of certain risks is necessary for planning
prevention programmes with credibility among those we wish to communicate with—
adolescents.
MORTALITY AND MORBIDITY FROM UNINTENTIONAL INJURIES
Accidental mortality is principally caused by traffic accidents, 77 per cent of all

accidental deaths in the EU (WHO, 1995). Accidental mortality shows a peak for
adolescents and young adults between 15 and 24 years of age in most ICs (see Table 1,
rates are higher for males in the age group 25–34 years only in Finland and Spain). This
high mortality is attested to by the place occupied by this pathology among all causes of
death in the age group 15–24 years in EU countries—38 to 70 per cent of all causes of
death according to country and gender (WHO, 1995). However, in countries where
driving licenses can be obtained at 18, the rates are much higher among young adults
(especially males) from 20 to 24 years than among adolescents from 15 to 19 years. In
France, in 1996, rates of fatal accidents among males were 35.3 per 100,000 among 15 to
19 year olds and 58.4 per 100,000 among 20 to 24 year olds, while for females it was 11
and 13.5 respectively (INSERM, 1996). Mortality rate for this age group is higher than
that for adults older than 25 years of age, and in the United States, the curve of the
number of fatal traffic accidents in relation to the distance travelled annually shows an
evident peak for 16 to 19 year olds, with rates four times higher than those for adults 30
to 34 years old (Williams and Carsten, 1989). In USA, 73 per cent of all deaths among
the 10–24 year olds result from three causes: unintentional injuries, homicides and
suicides (Kann et al., 1998). There is a very constant pattern for adolescent and youths
accidental mortality in ICs. Table 1 shows a dramatic increase in rates after the age of 15
and the considerable male overmortality (sex ratio from 3.0 to 5.2).
Table 1
Accidental mortality among 5 to 34 year olds, according to sex and age, in the European
Union 15 member states*
COUNTRY YEAR 5–14 YEARS 15–24 YEARS 25–34 YEARS

M F M F M F
Austria 1994 9.7 3.1 75.4 16.8 48.7 8.1
Belgium 1989 9.0 8.8 69.9 16.7 46.3 12.0
Denmark 1993 8.4 6.9 32.9 8.5 32.7 8.8
Finland 1994 10.0 5.1 35.3 6.8 36.0 6.9
France 1993 7.3 4.6 55.1 16.0 46.3 11.3
Germany 1994 7.4 4.3 47.0 12.4 31.8 6.9
Greece 1994 7.3 5.8 70.5 16.1 68.3 13.5
Ireland 1992 10.3 2.8 51.0 11.8 36.1 6.4
Italy 1992 8.1 3.7 59.9 12.4 40.9 8.7

Netherlands 1994 6.3 4.0 24.1 8.0 19.1 4.9
Portugal 1994 14.8 8.7 69.7 13.7 64.7 9.8
Spain 1992 9.4 6.1 68.3 16.7 70.6 13.2
Sweden 1993 6.3 2.8 25.4 8.5 19.4 5.7
United Kingdom 1992 7.2 3.4 31.3 8.9 23.9 5.3
*: Rates from Luxembourg are not calculated due to too small numbers of deaths
Data from: World Health Statistics Annual, 1995
Accidental morbidity also shows a peak between 15 and 24 years of age in ICs (Fife et
al., 1984; Williams and Carsten, 1989; Tiret et al., 1989).
In DCs, the mortality peak at 15–24 year olds is not observed in all countries (Smith
and Barss, 1991), and in many of them mortality rates (especially traffic death rates)
increase after the age of 25. Nevertheless some recent studies show the importance of
accidents among adolescents. A study conducted in two districts of Algiers in 1986
indicates that higher traffic accident rates were observed in the 20–24 years age group for
both sexes and in the 15–24 age group for males (Bezzaoucha, 1988). In 1991 in a South
African township, the highest injury rate was found among adolescents aged 15–19 years,
with two major causes: traffic accidents and violence (Zwi et al., 1995). In this study,
adolescents’ injuries were found to be more severe than those of subjects of other ages.
Adolescents’ injuries in DCs are often related to causes which are rare in ICs, such as
occupational accidents, as noted in India, where in 1983, it was estimated that 44 million
12–18 year olds were industrial or rural workers (Chaudhuri, 1990). If it has been
estimated that middle-income countries appear to have, on average, the largest road
traffic mortality burden, after adjusting for motor vehicle numbers, the poorest countries
show the highest road traffic-related mortality rates (Söderlund and Zwi, 1995). This is to
be compared with the relationship observed between socio-economic status and the type
of accident in adolescents in ICs (Williams et al., 1996).
taking into account the utility of risk-taking, and the cultural and social value of
certain risks is necessary for planning prevention programmes with credibility
among those we wish to communicate with—adolescents
Important differences observed in all countries between males and females for mortality
as well as morbidity rates favour an explanation in terms of the role of behaviour, since
these differences cannot be fully explained by differences in risk exposure and as risk and
protective behaviours varies notably in traffic according to gender.
RISK BEHAVIOURS
Risk-taking
Some nation-wide surveys address all types of risk-taking behaviours, such as the
national school-based American Youth Risk Behaviour Surveillance System (YRBSS)
(Kann et al., 1998), or the French national study Adolescents, conducted in 1993, in
schools, among subjects aged 11 to 19 years (Choquet and Ledoux, 1994). Other studies
are more specific and we shall emphasize those addressing the problem of behaviours
leading to injuries, mainly traffic accident related injuries. Research on risk behaviour
come almost exclusively from ICs.
The data from the 1997 YRBSS and the French ‘Adolescents’ study show marked
differences as concerns the level of risk-taking. In the American study, 36.4 per cent of
high school students had smoked cigarettes during the 30 days preceding the survey; 50.8
per cent had drunk alcohol, and 26.2 per cent had used marijuana; 2.1 per cent had ever
injected an illegal drug; 7.7 per cent had attempted suicide during the previous 12
months; 48.4 per cent had ever had sexual intercourse (Kann et al., 1998). In the 1993
French study, 14.5 per cent of subjects smoked regularly; 39.8 per cent reported drinking
alcohol occasionally and 3.3 per cent reported drinking regularly and had been inebriated
3 or more times in the past year; 12 per cent had smoked hashish but only 0.9 per cent
had tried heroin at least once (0.2 per cent had tried it 10 or more times); 6.5 per cent had
already attempted suicide; 31.3 per cent of the boys and 22.4 per cent of the girls had
ever had sexual intercourse (Choquet and Ledoux, 1994).
These different risk-taking behaviours are by and large more frequent among males
than females, whether this concerns the consumption of alcohol or drugs. On the other
hand, studies carried out in France on tobacco consumption show tobacco use among
girls identical to that among boys (Choquet. and Ledoux, 1994), and even higher among
girls aged 16 years and older (Sasco et al., 1993). In fact, not all types of risks are the
subject of studies and it is probable that females may take more risks than young males
when the risk is associated with values judged relevant for their sex, which is doubtless
the case for tobacco consumption and may be becoming so in the case of operating two-
and four-wheeled motor vehicles. Perhaps conforming to a masculine model is socially
worthwhile in the eyes of some girls. In all studies, changes with age are in the direction
of a regular increase in the frequency of risk-taking behaviours.
In DCs risk-taking behaviour by adolescents and young people is seldom studied and
may be addressed in a quite theoretical manner, risk-taking behaviour being described as
a component of the ‘rebellion’ specific to this age (Chaudhuri, 1990). One explanation
might be that, in DCs, environmental hazards as a cause of injuries (poor road
infrastructures, occupational hazards) may appear as much more problematic than that of
inappropriate behaviour.
In the area of traffic violations, young drivers have certain specific risk behaviours
(Assailly, 1992) associated with sensation seeking (Jonah, 1997) and possibly leading to
single vehicle collision (Zhang et al., 1998) in particular, speeding and following too
closely (Jonah, 1997; Zhang et al., 1998) especially when driving at night (Zhang et al.,
1998), and risky decision making at intersections (Chiron et al., 1994). The latter
problem has been identified in a recent study carried out among 980 high school students
who use motorbikes in the Rhone region of France (Chiron et al., 1994), this study
showing that 20 per cent of boys and 8.5 per cent of girls never signal for turns and only
25 per cent of boys and 38 per cent of girls always do so.
Indeed, in the area of dangerous driving and traffic violations, gender differences are
quite significant. A study of risk for driving while intoxicated (DWI), conducted among
an American white suburban population, addressed the perception of car and driving
according to gender (Farrow and Brissing, 1990). Males scored higher on a sensation-
seeking scale, used more alcohol, perceived (for themselves) greater driving skills in
risky situations and tended to use automobile to enhance self efficacy. Females seemed
more realistic and more responsible in the assessment of a dangerous drinking situation.
Nevertheless, when young women were driving while intoxicated, they were, more often
than males, given warnings rather than cited for traffic violation; this, of course,
introduces a bias in the reliability on the statistics of DWI by gender.
Dangerous driving tends to increase with age, especially DWI, as shown in a study of
2250 Iowa high school students (Schootman et al., 1993).
The accumulation of risks

Several studies show the frequency of association of different risk behaviours in the same
subject (Friedman, 1989). Thus, in the French national study on adolescents (Choquet
and Ledoux, 1994), the percentage of those having used drugs 10 or more times during
their lifetime is 25 per cent among daily smokers and 1 per cent among non-smokers, 28
per cent among regular drinkers and 1 per cent among non-consumers of alcohol. Ninety-
four percent of non-smokers and 96 per cent of non-drinkers have never used drugs. A
longitudinal study carried out recently among 2,480 students in the county of San Diego
(USA) showed that the use of marijuana and the consumption of alcohol were predictors
of later tobacco use (Eckardt et al., 1994). In a study of a large sample of Dutch
secondary school students aged 16 to 19 years, strong statistical associations were found
between suicide attempts and, hard drug abuse and use of sedatives among girls, cigarette
smoking and use of sedatives among boys (Garnefski and de Wilde, 1998). These results
suggest that suicidal behaviour and addiction-risk behaviours should both be regarded as
part of a complex interaction of multiple behavioural problems.
Protective behaviours
A number of converging phenomena are apparent from a comparison of the previously
cited studies which concern risk behaviours and the rates of utilization of the different
systems of protection in risky situations.
In most of the studies, the rate for use of seat belts is low (Litt and Steinerman, 1981;
Preusser, 1987; Centers for Disease Control, 1994); it was 32.5 per cent (‘safety belts
used “always” when riding in a car or a truck as a passenger’) in the 1992 Youth Risk
Behaviour Survey(Centers for Disease Control, 1994). In the same study, regular wearing
of a motorcycle helmet was more frequent, at 43.6 per cent.
there is a very constant pattern for adolescent and youths accidental mortality in
ICs
Comparisons of percentages for condom use in various studies from ICs are often made
somewhat difficult because of age group differences and the definitions used for variables
associated with condom use. The percentage of utilization was 58.3 per cent in the 1992
Youth Risk Behaviour Survey (Centers for Disease Control, 1994) and 56.8 per cent in the
1997 YRBSS (Kann et al., 1998) reported use during most recent sexual intercourse,
subjects aged 14 to 19 years; in the French national study on adolescents (Choquet and
Ledoux, 1994) subjects aged 11 to 19 years), percentages were 41 per cent during
established heterosexual relationships and 71 per cent during occasional heterosexual
relationships; and, in the 1991–1992 national study on the sexual behaviour of the French
(Spira, Bajos et le groupe ACSF, 1993), 74.8 per cent in the case of relationships with a
single partner and 90.9 per cent in the case of relationships with multiple partners
(reported condom use by men during heterosexual relationships during the past 12
months, subjects aged 18–19 years). In the French study carried out in 1994 among 15–
18 year olds (Lagrange and Lhomond, 1995), the percentage use of a condom during the
first sexual relationship was 78.9 per cent for boys and 74.4 per cent for girls, and 72.5
per cent and 51.1 per cent, respectively, during the most recent intercourse.
As concerns sex differences in the use of safety devices, boys are less likely than girls
to wear back seat belts and moped helmets, and seat belt and helmet use decreases
dramatically with age (Schootman et al., 1993).
The analysis of changes in preventive behaviours with age is complex. As concerns
condom use, study data converge: condom use uniformly diminishes with age. One might
attribute this phenomenon to the establishment of stable and faithful sexual relationships,
but this is an insufficient explanation as attested to by the decrease in use, not only
among people with a single partner, but also among those citing several (Spira, Bajos et
le groupe ACSF, 1993).
Factors influencing risk-taking and protective behaviours

Studies have been carried out on the relationship between socio-economic status and
behaviours that can compromise health. A Finnish study on alcohol and tobacco
consumption and nutritional habits of 16–18 year olds indicates that social mobility and
achieved social position (in terms of schooling) are more important than original social
class (Karvonen et al., 1999). Two studies, one on factors in childhood and youth
predicting alcohol abuse in Swedish women (Spak et al., 1997) and one on drinking
behaviour of adolescents in Scotland (Schucksmith et al., 1997) found no relationship
with social class. In the area of injuries, a survey of health behaviour of 4,710 school
children aged 11, 13, and 15 years (Williams et al., 1996), identified no relationship
between the occupation of the father and a score of family welfare on the one hand, and
the incidence and severity of injuries on the other. However differences were found as
concerns circumstances of accidents and behaviours: children from low economic status
groups had more road accidents and engaged with friends in risky activities (crossing
busy streets, skateboarding, drinking), whereas children from high economic status
groups had more sports injuries, and presented more protective behaviours (notably use
of bicycle helmets).
Psychological factors seem to be more important than socio-economic factors: in the
two studies previously quoted (Spak et al., 1997; Schucksmith et al., 1997) on drinking
behaviour, family factors were strongly associated with alcohol consumption. In the
Scottish study, a supportive family environment was associated with lowered prevalence
of alcohol use (Schucksmith et al., 1997) and sexual abuse in childhood was the strongest
predictor of alcohol abuse in young Swedish women (Spak et al., 1997). Suicide and
suicide attempts are also known to be more frequent among adolescents with prior
physical or mental abuse (Alvin, 1993; Choquet. and Ledoux, 1994) and violent
adolescents have frequently had a personal experience with violence in their childhood
(Choquet and Ledoux, 1994). Personal factors such as high self-esteem have been shown
to be associated with few dangerous behaviours (Friedman, 1989); conversely, school
drop-outs are over-represented among those with suicidal thoughts (Berg Kelly, 1991).
Time trends in risk taking and protective behaviours

Adolescents’ knowledge about situations of risk changes, as does their preventive
behaviour. Thus in France, a recent summary of research carried out between 1985 and
1992 on youths and AIDS showed an improvement in the state of knowledge about
protective measures and a moderate increase in the percentage of condom users (Beltzer
et al., 1994), but more recently this percentage decreased due to an unjustified
overconfidence in new treatments for AIDS.
Both the 1997 YRBSS (Kann et al., 1998) and data from the Fatality Analysis
Reporting System, from the CDC in Atlanta, show an increase in the use of seat belts by
young people. In 1988, 39.6 per cent of young drivers involved in fatal motor vehicle
crashes were using restraints, but by 1995, this percentage reached 54 per cent (Phebo
and Dellinger, 1998). In the 1997 YRBSS, 19.3 per cent of high school students had
rarely or never worn a seat belt, in 1992 (Centers for Disease Control, 1994), 67.5 per
cent did not use ‘always’ a seat belt (data which are indeed not easy to compare). On the
other hand, comparisons between YRBSS 1992 and 1997 data show an increase in
consumption of alcohol, tobacco and marijuana (Centers for Disease Control, 1994; Kann
et al., 1998).
METHODOLOGICAL PROBLEMS
The data on risk behaviours and their consequences are essentially of two kinds: 1)
statistical data furnished by health services (e.g. accidental morbidity) or by
administrative services (national mortality statistics, police accident reports, data from the
legal system on drug use); 2) data on behaviour, usually self-reported during answers to
questionnaires (generally closed-ended questions filled out by the respondents), or,
rarely, recorded during observations.
studies on risk behaviours carried out among adolescents in health services cannot
make a claim to being representative
These data are gathered in often non-representative populations. Only data gathering
directly in the homes of adolescents guarantees representativeness (for non-
institutionalized populations) as long as the parents are not present during the interview.
Many studies, including those at the national level, are carried out in school settings
(Choquet and Ledoux, 1994; Kann et al., 1998). Even though the vast majority of
adolescents attend schools, this approach may introduce serious selection bias by
excluding those young people who may be most at risk because they are school dropouts,
unemployed, or frankly marginal. Special methods should be used for including these
young people in studies, as was done in the 1992–1993 Swiss national study (Narring et
al., 1994) where access was had to adolescents through social workers, firms who employ
adolescents, care or helping institutions, and detention centres for minors. Analysis of the
results in this study shows clearly the important differences which exist between youths
from these settings and those attending schools, especially as concerns consumption of
tobacco and illegal drugs.
Studies on risk behaviours carried out among adolescents in health services cannot
make a claim to being representative. In addition, they often have the disadvantage of
giving health professionals a vision of the general adolescent population which is biased
by observations made among their clientele in clinical settings.
The quality of information sources and statistical data is uneven and interpretations
should be made with care, especially when crude routine data are concerned. Thus in
France (Tursz, 1995), accidental mortality studies generally use national statistics based
on death certificate data. These do not allow the determination of whether an adolescent
who died in a traffic accident was a driver or a passenger, whether responsible or not for
the accident, or whether the adolescent died in an accident caused by an adult. Data
gathered in police departments and courts are usually deficient to a greater or lesser
degree because of under-reporting. In France, traffic accidents are recorded by police
departments and classified according to the category of the user (driver, passenger,
pedestrian), type of vehicle (two or four-wheeled), age, the circumstances of the accident,
the seriousness of the injuries. While under-reporting is very low for fatal cases, it
becomes proportionally higher as the seriousness of the injury decreases.
Self-reported behaviours on questionnaires also pose problems of reliability. A study
recently carried out in Maryland on the consistency of answers about sexual behaviour
(Alexander et al., 1993) and which followed a cohort during several years, shows that
this consistency is sometimes weak and varies according to the ethnic origins of the
subjects, the type of behaviour studied, and the questions asked (variables used for
measuring sexual activities included: the existence or not of sexual relations, number of
times, age at first experience). For example, by repeating questions to subjects each year,
it was observed that 67 per cent of answers were inconsistent concerning age at the time
of first sexual experience. In addition, self-administered questionnaires may be
inappropriate because they are too rigid or incomprehensible for young marginal subjects.
One solution is to replace them with interviews, as in the Swiss study (Delbos et al.,
1995).
Some retrospective studies cover a long recall period and may therefore be biased. For
example, this could be the case for the study of predictive factors for alcohol dependence
and abuse in young Swedish women (Spak et al., 1997); alcohol-dependent women may
have given more precise answers on their childhood and adolescence than the non-
dependent ones.
Some types of behaviour may be systematically underdeclared, notably by specific
groups (e.g. ethnic groups), as shown in a study of the declaration of alcohol
consumption by Hispanic and Black drivers in the USA (Ross et al., 1991).
As has already been noted, the comparison of several studies often leads to identifying
different phenomena, and these differences between studies may be difficult to explain,
particularly because of differing definitions of risks, behaviours and age groups, but also
because detailed data are rarely available on the cultural context. Numerous studies
compare risk behaviours of adolescents from different ethnic backgrounds. These are
usually epidemiological studies which are unable to deal in any but a superficial manner
with the problem of the relative importance of ethnic and socio-economic factors in the
observed differences. Epidemiological studies identify target groups for prevention
programmes (African-American adolescents for example) without elucidating the basic
mechanisms associated with risk-taking which would clarify the most appropriate
prevention strategies. Only an anthropological approach would permit an evaluation of
the true role of cultural factors and elucidate the fact that answers to questionnaires have
social and cultural value which varies according to the study population.
The importance of comparing research results, particularly at an international level, lies
in the role these comparisons have in suggesting guidelines for prevention. They
illuminate the varied cultural, socio-economic, legislative and regulatory frameworks
(legislation on firearms in Europe and United States, for example) within which risk-
taking behaviours and their consequences are imbedded. For example, the considerable
differences in the rates of adolescent pregnancies noted between Europe and the US-
American figures are more than 10 times higher (Tursz, 1997) suggest the need for
carrying out in-depth research among pregnant young girls on the local context in which
they live, their motivations and the meaning of their pregnancy for them. It is only with
this kind of knowledge in hand that it will be possible to identify those groups to which
information and prevention messages should be directed and to adapt the nature of
prevention strategies (contraception, among others).
RISK-TAKING BEHAVIOURS: SPECIFIC TO ADOLESCENTS?
self-administered questionnaires may be inappropriate because they are too rigid

or incomprehensible for young marginal subjects
Are these risk-taking behaviours rare or even non-existent, among adults? Are they of
comparable frequency regardless of age, the difference being mainly their negative
consequences for health, which are more frequent during adolescence? The answer to
these questions is not clear, except for accidental pathologies which show characteristic
peaks of mortality, morbidity, seriousness and lethality at that age. It has not been clearly
demonstrated that subjects less than 20 years old have greater frequency of chronic
alcoholism, STDs or severe consequences related to illegal drug use (Tursz, 1997).
When demonstrating the specificity of adolescent risk-taking behaviour, one comes up
against a methodological problem: the rarity of studies permitting the comparison of two
age groups (adults and adolescents). And this is the case whether the studies include
subjects of all ages, or are done on adults and adolescents separately but using
comparable methodologies.
National studies in USA (Department of Health and Human Services, 1992) and
France (Haut Comité d’Etude et d’Information sur l’Alcoolisme, 1985), and measuring
blood alcohol levels of all subjects involved in traffic accidents, show clearly that levels
are lower among subjects under 21 years of age than among their elders; highest levels
being found among subjects aged 21 to 34 years in the American study (Table 2), and
among those 41 to 60 years old in the French study.
Table 2
Blood alcohol concentration (g/L), by driver’s age, in drivers involved in fatal traffic
accidents (USA, January-March 1991)
AGE IN YEARS NUMBER OF DRIVERS BLOOD ALCOHOL CONCENTRATION

0.0 % 0.10–0.90 % >1.0 %
15–20 1,639 71.4 9.7 18.9
21–24 1,358 57.3 9.7 33.0
25–34 3,014 60.6 7.0 32.4
35–64 3,958 74.0 5.4 20.6
>65 1,175 88.7 3.8 7.5
Data from: US Department of Health and Human Services
As for illegal drug use, one may note its beginnings among teenagers and increase among
young adults. The National Household Survey on Drug Abuse showed that in 1984 in
USA, 18–25 year-olds were most involved, with percentages for hashish and marijuana
consumption 2 times higher than among 12–17 year-olds, and 3.5 times higher for heroin
(NIDA, 1986).
A study of fatal acute reactions to opiates or cocaine in 6 large Spanish cities between
1983 and 1991 (Sanchez et al., 1995) indicates the rarity of the phenomenon before the
age of 19, the highest frequency among subjects aged 25–29 years, and an increase in the
average age of death from this cause from 25.1 in 1983 to 28 years in 1991 (Table 3).
Table 3
Mortality by acute reaction to opiates or cocaine in six major Spanish cities, by age
group and year of death (per 100,000 persons)
YEAR AGE GROUP (IN YEARS)

<15 15–19 20–24 25–29 30–34 All ages
1983 0.0 2.0 5.4 4.9 1.8 1.2
1987 0.2 3.7 12.5 14.4 7.5 3.3
1991 0.2 2.6 22.6 35.3 27.7 8.2
Annual average rate 0.0 2.8 14.1 19.6 11.5 4.1
Rates ratio
1991/1983 2.6* 1.3 4.2 7.2 15.2 7.1
Rates difference
1991/1983 0.2 0.6 17.2 30.4 25.9 7.0
* Rates ratio 1991/1984, because the mortality rate was 0 in 1983.
Data from: Sanchez et al. and the State Information System on Drug Abuse Working Group. J
Epidemiol Community Health 1995.
This study did not depend on routine mortality statistics, rather used autopsy reports and
a system of information for recording drug addiction. It is therefore unlikely there were
differences in the identification and recording of cases as a function of age.
Some preventive behaviours are more frequent among adolescents than among adults,
in particular, the use of the condom. In the study on the sexual behaviour of the French
(Spira, Bajos et le groupe ACS F, 1993), the percentage of condom users among men
with only one partner falls from 74.8 per cent among 18 to 19 year olds to 42 per cent
among 20 to 24 year olds, and from 90.9 per cent to 74.5 per cent respectively among
those with several partners (Table 4). Changes in the percentage of users in the case of a
single partner may be related to the establishment of a permanent relationship, but it is
interesting to note that, within the context of risk behaviour (multiple partners), the
youngest subjects are more careful than their elders.
Table 4
Condom utilization by men in heterosexual intercourse during the 12 last months,

according to age (National survey on sexual behaviour in France, 1991–1992)
AGE TYPE OF SEXUAL ACTIVITY

One partner Multiple partners

18–19 74.8 90.9
20–24 42.0 74.5
25–34 25.8 69.0
35–44 26.6 59.7
45–54 13.4 44.2
55–59 15.5 40.7
Data from: Spira and Bajos, La Documentation Française, 1993
Epidemiological studies identify target groups for prevention programmes …

without elucidating the basic mechanisms associated with risk-taking which would
clarify the most appropriate prevention strategies
In France, all of the recent studies on the attitudes of young people vis-à-vis AIDS
(Beltzer et al., 1994) show they are quite knowledgeable about the mechanisms of
transmission and the modes of prevention of the disease. Adolescents also take care of
their health and, in some countries, consult physicians, as shown in the French
‘Adolescents’ study (Choquet and Ledoux, 1994) and in a study of self-reported health
status of 3500 Swedish adolescents aged 13 to 16 years who reported an average of 5.5
medical appointments per year (Berg Kelly et al., 1991). A national study on ‘young
people and their health’ (Health Barometer, 1992) indicates that 69.3 per cent of subjects
questioned about their fears concerning their health cited traffic accidents as the most
important problem (Baudier et al., 1994). Furthermore, the French national study on
‘Adolescents’ (Choquet and Ledoux, 1994) shows that they want information
(particularly on AIDS, for 62.8 per cent of them). A recent American study, comparing
three groups of subjects on their perception of risks (adults, their adolescent children and
‘high risk’ institutionalized adolescents) found that the feeling of ‘invulnerability’ is no
more developed among adolescents than among adults. In addition, one finds the same
tendency to deny personal risk in the three populations, social distance playing a more
important role in perception of vulnerability than does age—more socially distant people
are perceived as being more at risk (Quadrel et al., 1993).
These adolescents, who seem to be quite knowledgeable about health problems which
may affect them, who have an appropriate perception of risk and who engage perhaps in
no more risky behaviours than do adults, are, nevertheless, more often victims of serious
accidents than are adults. Explanations for this apparently paradoxical situation are to be
looked for in the conditions in which they take risks, for example in the area of risks
related to driving, conditions which are different from those encountered by adults, as
noted previously (Assailly, 1992; Chiron et al., 1994; Jonah, 1997; Zhang et al., 1998).
Thus, in spite of a general exposure to risk which is lower than for adults (fewer number
of kilometres travelled yearly, lower average alcohol consumption), adolescent
automobile drivers or passengers are exposed to particularly dangerous conditions:
driving at night after having spent the evening with friends, automobiles overloaded with
teenage passengers (Preusser et al., 1998), automobiles in poor operating condition, acute
alcohol use. Furthermore, it has been widely demonstrated that sensitivity to alcohol is
higher at that age and that, with equivalent levels of alcohol, adolescents have a much
higher probability of having an accident that do adults (Zylman, 1973)].
Finally, the role of experience is important. Adults often react out of habit, and,
through trial and error, they have acquired effective responses to problems which arise in
risky situations. Adolescents suffer more from inexperience than from ignorance. It
should be noted, however, that opinions diverge on the respective roles of age and
experience. For some authors, research indicates that youth plays a much greater role
than inexperience and accidents of young motorcyclists are associated with a particular
pattern of behaviour (a willingness to violate the rules of safe riding) (Rutter et al., 1996),
driving experience appearing to have very minor, if any, influence (Levy et al., 1990).
This comes out clearly in the debate over the ideal age for gaining access to driving an
automobile. For some, it should be advanced because young drivers have a higher rate of
accidents than do adults with the same experience (first year of driving). For others, the
age should be lowered since the frequency of accidents, traffic violations and dangerous
behaviour diminishes with experience, and therefore learning to drive should begin early
(with a long period of driving accompanied by an adult before the examination for the
driver’s license) (Assailly, 1992).
In the final analysis, prospective longitudinal studies are lacking which would allow
follow-up into adulthood of risk behaviours begun in adolescence, or for identifying
behaviours which begin in adulthood. Due to the methodological issues and the cost of
these studies, there are very few of them, but they tend to show that risky behaviours of
adults are often initiated in adolescence, if not in early childhood, and that predicting
factors can be identified. According to a study conducted in Quebec, aggressiveness and
anti-social behaviours in kindergarten children are associated with violence in
adolescence and delinquency at adulthood (Tremblay, 1996). Based on the Dunedin
(New Zealand) birth cohort, a study showed that motorcycle riding by young males,
which is a significant cause of serious injuries, had strong predictors, such as below
average reading skills and fighting in a public place at age 15 (Reeder et al., 1997).
FUTURE DIRECTIONS: SOME REFLECTIONS ON PREVENTION
Disseminating information on risks is the appropriate response if we assume that

adolescents do not know or do not understand risks. However, as has been abundantly
noted above, it seems rather that adolescents know about risks and decide to take them
because the benefits are greater than the negative consequences. Several solutions are
therefore possible. One could limit the freedom of adolescents who, by taking risks, do
not subscribe to adult values (for example, by raising the age for drivers’ permits or by
curfews which prevent night driving). Above all, a larger range of choices could be made
available to adolescents so that they do not find themselves inextricably confronted with
risks. Indeed, external constraints, principally socio-economic ones, frequently force
adolescents to take risks (poor vehicle maintenance or defective motorcycle helmet, both
due to financial reasons; poor safety of the traffic environment in DCs). To the extent
possible, it is important to offer protective systems to adolescents which are financially
accessible. However, one should not try to hide the fact that adolescents from the most
disadvantaged social classes will remain most at risk.
The roles played by the modification of the environment, in the sense of increasing its
safety, and by safety standards for products, need no further demonstration (the situation
of DCs and the low mortality rates observed in the countries known to have developed
traffic safety, such as Sweden or the Netherlands [Table 1] speak for themselves). In the
area of traffic accidents, there are numerous examples, including recent safety measures
such as helmets for cyclists. Thus, in Australia, it was possible to demonstrate on the one
hand that wearing a helmet considerably reduced the number of head injuries, and on the
other hand, that the passing of a law could bring about a massive increase in the rate of
utilization of helmets (Rogmans, 1993).
the feeling of ‘invulnerability’ is no more developed among adolescents than

among adults
While working to improve the material environment for adolescents, it is nonetheless

necessary to continue to provide them with pertinent information, especially since this is
their wish, as noted above. It is therefore necessary to reflect on what are the best
channels for communicating this information. In some countries, health personnel can
continue to play an important role, as noted in the French national study on adolescents
(Choquet and Ledoux, 1994). School physicians can notably contribute to the
identification of predictive factors such as child abuse or school drop out. The family can
also play a very important role, since, according to several studies, adolescents consider
their parents as the persons they discuss things with most. Thus, in the French study
(Choquet and Ledoux, 1994), 64.6 per cent of adolescents prefer to talk to their parents
about health matters, and 68 per cent about school difficulties.
Society can play an important role by way of publicity. Adolescent beauty, youth and
innocence have been used, and are still used in some countries, for vaunting the merits of
speed in automobiles or the consumption of alcoholic beverages and cigarettes. A study
carried out in 1986 among 602 high school students in the United States showed that
adolescents who were exposed to advertising for cigarettes were more likely to smoke
than others (Botvin et al., 1993), and a 1993–4994 British study indicates that onset of
smoking is largely influenced by cigarette advertising among girls (While et al., 1996).
Focusing on adolescent risk behaviours appears to be justified because adolescence is
the time when these behaviours are initiated, and therefore potentially the time for
modifying or even preventing them, since they continue into adulthood. There is a risk,
however, of constructing programmes which give adolescents a negative image of
themselves, which deny similarities with adulthood, and deny a positive aspect of risk
behaviours at this age: an attempt to gain independence. Such programmes may be totally
counterproductive in the same way as would programmes conceived for particular
cultural or ethnic groups which fail to recognize the meaning and utility of certain risks
for these groups. More generally, preventive strategies should take into account the social
value of risk (Douglas, 1986), the place of adolescents in society, and the role played by
adults and society in adolescent risk behaviour.
REFERENCES
Adolescents at risk. Special Issue, 1991, Journal of Adolescent Health, 8, pp. 585–647.
Alexander, C.S., Somerfield, M.R., Ensminger, M.E., Johnson, K.E. and Young, J.K.,
1993, Consistency of adolescents’ self-report of sexual behaviour in a longitudinal
study. Journal of Youth Adolescence, 22, pp. 455–471.
Alvin, P., 1993, Suicidal adolescents: lessons to be learned from early intervention,
Journal of Paediatric Child Health. 29, Suppl 1, pp. S20-S24.
Assailly, J.P., 1992, Les jeunes et le risque. Une approche psychologique de l’accident,
Vigot, Paris.
Baudier, F., Janvrin, M.P. and Dressen, C., 1994, Les jeunes français et leur santé.
Opinions, attitudes et comportements, Promotion et Education, 1, pp. 29–35.
Beltzer, N., Moatti, J.P. and Souteyrand, Y. Ed., 1994, Les jeunes face au SIDA: dela
recherche à l’action. Une synthèse des enquêtes et des recherches françaises, 2ème
édition, (Paris: ANRS).
Berg Kelly, K., 1991, Self-reported health status and use of medical care by 3500
adolescents in Western Sweden. II. Could clustering of symptoms and certain
background factors help identify troubled young people? Acta Paediatr Scand, 80, pp.
844–851.
Berg Kelly, K., Ehrver, M., Erneholm, T., Gundevall, C., Wennerberg, I. and Wettergren,
L., 1991, Self-reported health status and use of medical care by 3500 adolescents in
Western Sweden. I, Acta Paediatr Scand, 80, pp. 837–843.
Bezzaoucha, A., 1988, Etude épidémiologique d’accidents de la route survenus chez des
habitants d’Alger, Rev Epidemiol Sante Publique, 36, pp. 109–119.
Botvin, G.J., Goldberg, C.J., Botvin, E.M. and Dusenbury, L., 1993, Smoking behaviour
of adolescents exposed to cigarette advertising, Public Health Rep, 108, pp. 217–224.
Centers for Disease Control, Atlanta, 1994, Health risk behaviours among adolescents
who do and do not attend school. United States 1992, MMWR , 43, pp. 129–132.
Chaudhuri, N., 1990, Adolescent injuries. Indian Pediatr, 27, pp. 1261–1267.
Chiron, M., Le Breton, B., Alauzet, A. and Weber, D., 1994, La prise de risque chez les
élèves de seconde. Usage des transports et habitudes de vie: opinions et pratiques de
980 élèves du département du Rhône. Rapport INRETS n°189, Lyon.
Choquet, M. and Ledoux, S., 1994, Adolescents. Enquête nationale. (Paris: INSERM,
Analyses et prospective).
Delbos Piot, I., Narring, F. and Michaud, P.A., 1995, La santé des jeunes hors du système
de formation: comparaison entre jeunes hors formation et en formation dans le cadre de
l’enquête sur la santé et les styles de vie des 15–20 ans en Suisse romande. Sante
Publique, 1, pp. 59–72.
Douglas, M., 1986, Risk Acceptability According to the Social Sciences. (London:
Routledge/Kegan Paul).
Dryfoos, J.G., 1991, Adolescents at risk: a summation of work in the field. Programs and
policies. Journal of Adolescent Health, 12, pp. 630–637.
Eckardt, L., Woodruff, S.I. and Elder, J.P. ,1994, A longitudinal analysis of adolescent
smoking and its correlates. J Sch Health, 64, pp. 67–72.
Farrow, J.A., and Brissing, P., 1990, Risk for DWI: a new look at gender differences in
drinking and driving influences, experiences, and attitudes among new adolescent
drivers. Health Education Quarterly, 17, pp. 213–221.
Fife, D., Barancik, J.I. and Chatterjee, B.F., 1984, Northeastern Ohio trauma study: II.
Injury rates by age, sex and cause. American Journal of Public Health, 74, pp. 473–
478.
Friedman, H.L., 1989, The health of adolescents: beliefs and behaviour. Social Science
Medicine, 29, pp. 309–315.
Garnefski, N. and de Wilde, E.J., 1998, Addiction-risk behaviours and suicide attempts in
adolescents. Journal of Adolescence, 21, pp. 135–142.
Haut Comité d’Etude et d’Information sur l’Alcoolisme., 1985, Alcool et accidents.
Etude de 4796 cas d’accidents admis dans 21 hôpitaux français. (Paris: La
Documentation Française).
Holinger, P.C., 1981, Self destructiveness among the youth: an epidemiological study of
violent death. International Journal of Social Psychiatry, 27, pp. 277–282.
INSERM , 1996, Statistiques des causes médicales de décès. (Paris: INSERM).
Jessor, S.J. and Jessor, R., 1977, Problem Behaviour and Psychosocial Development: A
Longitudinal Study of Youth. (New York: Academic Press).
Jonah, B.A., 1997, Sensation seeking and risky driving: a review and synthesis of the
literature. Accid Anal Prev, 29, pp. 651–665.
Kann, L., Kinchen, S.A., Williams, B.I. et al., 1998, Youth Risk Behaviour
Surveillance—United States, 1997. State and local YRBSS co-ordinators. Journal of
Sch Health, 68, pp. 355–369.
Karvonen, S., Rimpelä, A.H. and Rimpelä, M.K., 1999, Social mobility and health
related behaviours in young people. Journal of Epidemiol Community Health, 53, pp.
211–217.
Levy, D., 1990, Youth and traffic safety: the effects of driving age, experience and
education. Accid Anal Prev, 22, pp. 327–334.
Litt, I.F. and Steinerman, P.R., 1981, Compliance with automotive safety devices among
adolescents, Journal of Pediatrics, 99, pp. 484–486.
Narring, F., Tschumper, A., Michaud, P.A. et al., 1994, La santé des adolescents en
Suisse. Rapport d’une enquête nationale sur la santé et les styles de vie. (Lausanne:
Institut Universitaire de Médecine Sociale et Préventive, Cahiers de Recherche et de
documentation n°113a).
NIDA, 1986, Drug use among Amercican high school students, college students and
other young adults. National trends through 1985, NIDA.
Odero, W., Garner, P. and Zwi, A., 1997, Road traffic injuries in developing countries: A
comprehensive review of epidemiological studies. Trop Med Int Health, 2, pp. 445–
460.
Phebo, L. and Dellinger, A.M., 1998, Young driver involvement in fatal motor vehicle
crashes and trends in risk behaviours, United States, 1988–1995. Injury Prevention, 4,
pp. 284–287.
Preusser, D.F., 1987, The effect of New York’s seat belt use law on teenage drivers.
Accid Anal Prev, 19, pp. 73–80.
Preusser, D.F., Ferguson, S.A. and Williams, A.F., 1998, The effect of teenage
passengers on the fatal crash risk of teenage drivers. Accid Anal Prev, 30, pp. 217–222.
Quadrel, M.J., Fischhoff, B. and Davis, W., 1993,. Adolescent (In) vulnerability.
American Psychology, 48, pp. 102–116.
Reeder, A.I., Chalmers, D.J., Marshall, S.W. and Langiey, J.D., 1997, Psychological and
social predictors of motorcycle use by youn adult males in New Zealand, Soc Science
Medicine, 45, pp. 1357–1376.
Rogmans, W.H.J., 1993, Helmets for All. (Amsterdam: ECOSA).
Ross, L.H., Howard, J.M., Ganikos, M.L. and Taylor, E.D., 1991, Drunk driving among
American blacks and Hispanics. Accid Anal Prev, 23, pp. 1–11.
Rutter, DR. and Quine, L., 1996, Age and experience in motorcycling safety. Accid Anal
Prev, 28, pp. 15–21.
Sanchez, J., Rodriguez, B., de la Fuente, L. et al and the State Information System on
Drug Abuse ,SEIT, Working group, 1995, Opiates or cocaine: mortality from acute
reactions in six major Spanish cities. J Epidemiol Community Health, 49, pp. 54–60.
Sasco, A.J., Pobel, D., Benhaim, V., de Bruin, K., Stiggelbout, A. and Tuyns, A., 1993,
Smoking habits in French adolescents. Rev Epidemiol Sante Publique, 41, pp. 461–
472.
Schootman, M., Fuortes, L.J., Zwerling, C, Albanese, M.A. and Watson, C.A., 1993,
Safety behaviour among Iowa Junior High and High school students, American
Journal of Public Health, 108, Suppl 1, pp. 11–14.
Schucksmith, J., Glendinning, A. and Hendry, L. ,1997, Adolescent drinking behaviour
and the role of family life: a Scottish perspective. J.Adolesc, 20 (1), pp. 85–101.
Smith, G. and Barss, P., 1991, Unintentional injuries in developing countries: the
epidemiology of a neglected problem. Epidemiology Review, 13, 228–266.
Söderlund, N. and Zwi, A.B., 1995,. Traffic-related mortality in industrialised and less
developed countries. Bulletin of the World Health Organization, 73, pp. 175–182.
Spak, L., Spak, F. and Allebeck, P., 1997, Factors in childhood and youth predicting
alcohol dependence and abuse in Swedish women: findings from a general population
study. Alcohol Alcohol, 32, pp. 267–274.
Spira, A., Bajos, N. et le groupe ACSF., 1993, Les comportements sexuels en France.
Rapport au Ministre de la Recherche et de l’Espace. (Paris: La Documentation
Française).
Tiret, L., Garros, B., Maurette, P. et al., 1989, Incidence, causes and severity of injuries
in Aquitaine, France: a community based study of hospital admissions and deaths.
American Journal of Public Health, 79, pp. 316–321.
Tonkin, R.S., 1987, Adolescent risk-taking behaviour. Journal of Adolescent Health
Care, 8, pp. 213–220.
Tremblay, RE., 1996, Prédire et prévenir la violence des adolescents, In : Les adolescents
face à la violence, ed : Rey, C., Syros, Paris.
Tursz, A., Souteyrand, Y. and Salmi, R., 1993,. Adolescence et risque. (Syros, Paris).
Tursz, A., 1995, Injury mortality and morbidity reporting systems in France,
Unintentional injuries in children and adolescents. In Proceedings of the International
Collaborative Effort on in jury statistics symposium. Bethesda, United States, eds :
Fingerhut, L. and Hartford, R., pp 15–1/15–16, CDC/NCHS, DHHS Publication No,
PHS, 95–1252, Hyattsville, Maryland.
Tursz, A., 1997, Problems in conceptualising adolescent risk behaviours: international
comparisons. Journal of Adolescent Health, 21, pp. 116–127.
US Department of Health and Human Services, 1992, Trends in alcohol-related traffic
fatalities by sex, United States, 1990. MMWR, 41, pp. 189, 195–197.
While, D., Kelly, S., Huang, W. and Charlton, A., 1996, Cigarette advertising and onset
of smoking in children: questionnaire survey. British Medical Journal, 313, pp. 398–
399.
Williams, A.F. and Carsten, O., 1989, Driver age and crash involvement. American
Journal of Public Health, 79, pp. 326–327.
Williams, J.M., Currie, C.E., Wright, P., Elton, R.A. and Beattie, T.F., 1996,
Socioeconomic status and adolescent injuries. Soc Sci Med, 44, pp. 1881–1891.
World Health Statistics Annual. Yearly publication. (Geneva: WHO).
World Heath Statistics Annual, 1995. (Geneva: WHO).
Zhang, J., Fraser, S., Lindsay, J. et al., 1998, Age-specific patterns of factors related to
fatal motor vehicle traffic crashes: focus on young and elderly drivers. Public Health,
112, pp. 289–296.
Zuckerman, M., 1971, Dimensions of sensation seeking. Journal of Consulting Clinical
Psychologists, 36, pp. 45–52.
Zuckerman, M., 1979, Sensation Seeking: Beyond the Optimal Level of Arousal. (New
York: Elbaum).
Zwi, K.J., Zwi, A.B., Smettanikov, E., Söderlund, N. and Logan, S., 1995, Patterns of
injury in children and adolescents presenting to a South African township health
centre. Injury Prevention, 1, pp. 26–30.
Zylman, R., 1973, Youth, alcohol and collision involvement. Journal of Safety Research,
5, pp. 55–72.
13
Injury Prevention and Children’s Rights
James Nixon
INTRODUCTION
It is in relatively recent times that the community has afforded children some formal
protection. It is again relatively recently that Western societies have recognized children
as being different from adults. There is no cause to be complacent about their safety and
welfare, as children continue to be injured, intentionally or unintentionally, across all
cultures.
What is not universal, however, are perceptions of childhood, injury and the potential
to prevent injury. There is also limited political or economic capacity to foster change in
these areas. The purpose of this paper is to examine injury prevention from the
perspective of the child’s right to safety.
I argue that advocacy for a safer environment for children may lead to a recognition
that safety measures are enforceable, thus ascribing a measure of right. This paper
examines the development of rights to safety for children through the view of my
Western eyes. I trace developments in children’s rights in a Western tradition, based for
the most part on English practices. Some may see this as so narrow an approach it is
bound to give a jaundiced view. My purpose is to give an outline of what has worked in
some Western communities and to view that from a rights perspective. This approach is
not universally applicable. However, if there are lessons from history and experience that
might benefit children anywhere, then there would be progress towards universal rights
for children.
Advances for children have followed changes from agrarian to industrial and post-
industrial forms of society. Many of these had their origins in England, and in that
context there are some lessons to be learned. Children exist universally but approaches to
childhood are not universal. Children live in a social, cultural and economic context.
Western interpretations of childhood or rights cannot be claimed to be universal nor even
agreeable to all Western eyes. Again, my purpose is to give recognition to the view that
children require protection in particular areas where adults do not, and, therefore, should
be treated as a special case.
CHILDREN ARE A SPECIAL CASE
There has been a move over the past century towards protecting children from harm and
exploitation. Indeed, there is now recognition in some quarters that children have rights
independently of parents and adults. Nations are now moving to modify laws to ensure
that children’s rights are adequately incorporated therein. Change in the way we perceive
children has been slow. The challenge for the future is to maintain this pressure for
change. Children continue to be hurt both seriously and unnecessarily in an adult oriented
world. Advocacy is one of the best means to stimulate change.
Throughout history, children were killed, abandoned, terrorized and sexually abused.
Society afforded children a low priority and paid no attention to their special needs.
WHAT ARE RIGHTS?
This paper does not argue the theories of how rights might develop or be justified, or
about the moral or ethical fundamentals that drive and shape rights. The paper recognizes
that certain rights for children have already been specified in some jurisdictions. Further,
an ideal specification for children’s rights has been proposed and accepted by most
countries of the world. Two have not yet acceded. Rights for children are no longer
speculative, they are possible.
The questions to be explored are: ‘How have these rights come about?’ and, ‘How can
the United Nations Convention on the Rights of the Child, which by its existence and
widespread support has merit, be used to advance children’s safety and maximize its
potential for good?’
Fundamentals of rights
children require protection in particular areas where adults do not, and,

therefore, should be treated as a special case
Alston et al. (1992) edited the work of 15 authors who have expressed disparate views of
children’s rights. The following short summary serves not to condense their positions but
to highlight different approaches to justification of children’s rights.
Rights may be based on a moral obligation to children or conversely on fundamental
moral rights of children. Each approach has implications for the consequences of any
regulations framed from those positions. Another position is to argue that rights exist to
protect the interests of children. Yet another argument is that they exist to give the subject
of the rights, the child, the freedom to waive or enforce their will. However, if rights are
recognized for children, how can they practically enforce their will when they do not
have the resources, experience or the power to do so?
A further approach to rights argues that the moral justification for rights lies in equality
and autonomy. If this is so, how can young children exercise the autonomy required to
benefit from the right. On the other hand if rights have roots in autonomy a young child
or a toddler might require that certain conditions be maintained in order to grow and
develop to adulthood, thus benefit from the right.
The question of similarities and differences between adults and children will colour
how, and whether, a society will protect children’s interests specifically. It also
determines what adult obligations are to children or whether adults make decisions for
Injury prevention and children's rights 179
children.
These questions overlay the fundamental approach to children’s rights, embodied in
the United Nations Convention—that matters affecting children, should comply with the
best interests of the child. Interpretation of the best interests of the child will, of course,
differ from one person, court or nation to another and has the potential, as has recognition
of rights for children, to create difficulties between the state, the family and the child.
The complexities of the philosophical arguments have been summarized by Parker
(1992). He suggested that differing approaches to children’s rights frequently require a
measure of assertion in order to proceed.
Rights cannot be discussed without consideration of obligations or of claims to rights.
Parker points out that a claim to rights can be ascribed by others and not necessarily by
the claimant who, as a child, may not be able to do so. Thus, advancing injury prevention
measures, through advocacy for change in laws or community norms, specifically for
children, is to recognize children’s rights. One function of children’s rights is to protect
the interests of children. Children do not vote or control much of the world’s resources. It
requires individuals or organized advocates to speak out for children’s safety issues and
to be advocates on children’s behalf.
Recognition of rights for children

The Magna Carta provided the first documentation of rights in the English speaking
world when King John traded rights for lords in exchange for peace. Not universal rights,
but it was a beginning. Magna Carta, however, did not directly advance the cause for
children’s rights. The theme of children not being considered until later is a theme which
recurs in the development of rights and the development of a safer environment. A recent
example of this time lag was exhibited in the development of the airbag for motor
vehicles. It was apparently designed, tested, and implemented with adults as focus. The
unsuitability for children was discovered only after fatalities occurred with its use on the
roads.
Other documents have defined rights for citizens. Habeas Corpus enshrines the right
for detainees to have the legality of their detention tested in court. Another document to
define rights was The Bill of Rights in the United States. It was introduced as ten original
amendments to the USA Constitution in force on the 15th December, 1791. The
amendments preserve and define freedom of religion, speech, the press, peaceable
assembly, and the right to petition the government; the right to bear arms; limited use of
military by the state; controls over policing powers such as search and seizure; right to
due process of law, the courts and trial by jury; freedom from excessive, cruel or unusual
punishments; and rights assigned to the individual states. As no specific prerogatives for
children are mentioned, the question is whether children are adequately protected.
An early attempt at defining acceptable standards for children occurred in the United
Kingdom in 1908. The Children’s Act had the aim of protecting children from harm,
tobacco and alcohol among other things. One purpose of the act was to prevent
exploitation. However, ironically it limited freedom in order to achieve that.
Eglantyne Jebb, founder of the Save the Children Fund, drafted the five item document
known as The Declaration of Geneva in 1922. This document was recognition that
‘mankind owes the child the best that it has to give…above all considerations of race,
nationality or creed.’
This declaration was adopted by the General Council of the Save the Children
International Union in February 1923. It was adopted by the Assembly of the League of
Nations in 1924 and reaffirmed in 1934. This document was the forerunner of the
Declaration of the Rights of the Child, proclaimed by the General Assembly of the
United Nations on 20 November, 1959. The Declaration of the Rights of the Child was
accepted by 78 member states of the United Nations. Table 1 shows a chronology of
some of the milestones in the development of rights for children.
Table 1
Chronology of milestones towards codification of rights for children
1922 The Declaration of Geneva

1924 The Geneva Convention
1948 Universal Declaration of Human Rights
1959 Declaration of the Rights of the Child
1978 Move to prepare a Convention on the Rights of the Child
1979 The International Year of the Child
1989 Consensus in the General Assembly for the Convention
1992 The Convention ratified by 1 6 nations on the first day if its release
1999 Only 2 nations yet to ratify the Convention
children have not been recognized as having particular rights in the drafting of
any of the documents defining rights for adults. It was only some 200 years ago
that children were recognized as being different and having different needs from
adults
Children have been afforded some rights in other resolutions, declarations and
conventions including special protection for children in foster placement and adoption,
(resolution 41/25 of the General Assembly of 3 December 1985). However, the
Convention on the Rights of the Child has gained the most support in less time than any
other United Nations convention.
The Universal Declaration of Human Rights, adopted by the United Nations in 1948
agreed that everyone is entitled to the rights and freedom set out in the Declaration
without distinction of any kind. These include race, colour, sex, language, religion,
political or other opinion of national or social origin, property, birth or other status. The
Universal Declaration of Human Rights also proclaims that childhood is entitled to
special care and assistance. The groundwork was set for the further development and
codification of rights for children in the Declaration of the Rights of the Child and the
acceptance of these through the United Nations General Assembly and the Convention on
the Rights of the Child.
As stated, The United Nations Convention on the Rights of the Child has been
supported like no previous United Nations convention. The Convention was ratified by a
record number of countries on the first day. One hundred and seven countries ratified and
a further 35 countries signed the convention in just over one year. It has been ratified by
all but two countries (more than any other convention), and more quickly than any other
United Nations convention. Such a display may indicate widespread support, or the
alternative cynical view, that being seen to do good things for children is good politics.
The United Nations Convention does not confer a solution to the question of children’s
right to safety, but rather presents a strong advocacy that the vulnerabilities of children
should be considered in decision making.
In summary children have not been recognized as having particular rights in the
drafting of any of the documents defining rights for adults. It was only some 200 years
ago that children were recognized as being different and having different needs from
adults. Codification of rights for children has developed relatively recently. The
development of rights for children has lagged many years behind the development of
human rights generally.
Rights and protection of children

Provision for the physical protection of children has been a slow process. It is one issue
that individuals have defended strongly, despite derision and opposition from colleagues.
This advocacy to recognize and deal with ‘child abuse’ has been strongest over the past
30 years. The pace of change in community attitudes in England highlights some of the
difficulties in recognition of children in that community.
In sixteenth century England some interest in children occurred. This was a time when
it was thought children might be useful to the future of Cromwell’s Commonwealth. The
first Poor Law, enacted in 1531, gave some protection to the care of children. However,
rights of children were limited and not widely recognized. Thirty-three years later, in
1564, the first English slave trader, Sir John Hawkins, began plying his trade. He did not
discriminate against children. He captured and traded them as easily as he did adults.
Legislation regarding children over the next few centuries was mainly involved with
law and order, rather than individual safety or protection of children. Some
apprenticeship laws protecting against ill treatment at work were passed.
Recognition of children’s right to protection

In England in 1839 the Infants’ Custody Act (2&3 Vict c54) was the first legislation in
the United Kingdom to acknowledge that children had a right to protection. This act gave
the court the power to determine custody of children. It was the first statutory
intervention into the common law rights of the father. Previously illegitimate children
were in the custody of the mother while legitimate children were in the custody of the
father. A mother could not petition for custody of her own legitimate children. Today,
decisions about custody of children, take into account the child’s development, in some
cases the child’s wishes, and, in general, consider the interests of the child to be
paramount.
Child welfare acts have always recognized the need to protect children from alcohol
and tobacco. However, legislative prohibition is as ineffective a means of social control
for children as it is for adults.
At the turn of the century England was one of very few nations that
• allowed children of any age to buy beer and spirits for off premises
consumption
• allowed children of 13 years and older to drink beer on the premises
• allowed children of 16 years to drink spirits on the premises
• provided a penalty of only 20 shillings on the licensee
The ‘Anti-sipping act’ was introduced in England in 1901. The age limit of
13 years was changed to 14 years and liquor was required to be sold in sealed
vessels. It had been called the anti sipping act as children, sent to collect liquor
for parents, would sip on the way home.
The 1908 Children Act in England was the first recognition of an obligation by the state
to provide general protection for children. This legislation drew together piecemeal laws
and amendments of previous decades. At the one time it controlled the sale of tobacco
and prohibited children in brothels, bars and street corners from begging. The attempt
was to provide a minimum level of care for children. This act prohibited children under
14 years entering licensed premises and prohibited giving liquor to a child under five
years of age.
Rights to protection from abuse and maltreatment have been developing since the
beginning of the twentieth century. However, it is only since the sixties and early
seventies that these rights have been made more explicit, and mechanisms have been put
in place to detect abuse and enforce protection.
Punishment and rights

It is a commonly held belief that physical punishment is an acceptable form of
discipline—‘spare the rod and spoil the child’. Attitudes toward physical punishment, by
both the state and individuals, have been slow to change in countries with a western
tradition.
many psychological punishments such as threatening, scaring or ostracising are as

detrimental to a child as beating
In the early eighteenth century (1718) the English courts were authorized by parliament
to order transportation as a direct punishment. Juveniles still were not recognized as
being different from adults, and were therefore also liable for transportation, and indeed,
were transported.
Later in the second half of the eighteenth century, there seemed to be some softening
towards punishment of offenders. In 1768 English judges were empowered to substitute
transportation for capital convictions.
The nineteenth century has been called ‘the century of the child’. Real inroads began to
be made into changing attitudes towards children’s care, welfare and health needs. More
and more books were written about children and for children. Some, such as Charles
Dickens’s works, exposed the social injustices toward children and no doubt prompted
philanthropic movements on children’s behalf.
Pickpocketing was a common crime among destitute boys in the early 1800s and in
1808 the English Parliament agreed to transportation instead of death for pickpocketing.
This show of soft-heartedness towards pickpockets was not extended to children who
stole. Four years later, in 1814, on one February day at the Old Bailey, five children were
condemned to death for stealing and burglary. Their ages 8, 9, 11, 12, and 12 years,
respectively.
Corporal punishment has been an accepted means of discipline in western tradition. It
has been progressively abandoned in state run institutions. In many countries it has
remained part of private school culture. Abandonment of corporal punishment has been
taken one step further in Sweden where in 1979 the Swedish Parliament gave support
(almost without opposition) to legislation, which reads:
‘A child may not be subject to physical punishment or other injurious or humiliating
treatment.’ Assault has always been a punishable offence. The effect of this law was to
sweep aside any confusion or ambiguity which was present in Swedish law (and is still so
in Australian law) about the difference between physical chastisement and assault. The
law went a little further. It recognized that many psychological punishments such as
threatening, scaring or ostracising are as detrimental to a child as beating. Under this law
these could be considered to be injurious and humiliating treatment and were forbidden.
The physical abuse of children has been recognized in our community for many years.
In the forties, spiral fractures of long bones in children were linked to abuse. It took until
the sixties for this to have widespread acceptance and to be consistently acted upon.
However, child abuse is still not accepted as a possibility in some countries. In addition,
it is still not accepted by some responsible adults, that parents are denied the right to deal
with their children as they please.
The Convention on the Rights of the Child is clear that children should be protected
from such intentional injury.
SAFETY IN THE WORKPLACE
Galen, in his treatise ‘de Sanitate Tuenda’ epitomized the attitudes of AD 157 when he
wrote, ‘the life of many men is involved in the business of their occupation and it is
inevitable that they should be harmed by what they do…and it is impossible to change it.’
In the case of workplace health and safety in factories the needs of children have been
addressed ahead of those of adults. Exclusion of children from factories in England was
not popular and took years many to accomplish.
The view that injury could not be prevented has been a widely held view. This
sentiment prevailed for the 100 years that it took to introduce the ‘Plimsoll line’ or safe
load line on ships. It prevailed for the 100 years that it took the English Parliament to
prevent sending children into chimneys to clean them. It also prevailed well into the
twentieth century until workers’ unions fought for safer conditions on the job, and when
Ralph Nader began to advocate for safer vehicles on the roads and for safer consumer
products in the home. Rights do not come easily and cannot be assumed. For every
advocate for a right there will be an individual or group with an interest in opposing
change and the possibility of loss in the face of gain by others.
Associated with the industrial revolution in England (c.1750–1870) were the inequities
of the workhouse system namely indifference, neglect and cruelty. The exploitation of
child labour in the 18th Century led to reform in the early 19th century. One example of
this was the Health and Morals of Apprentices Act passed in 1802 which led to the
limitation of children’s working hours to 12 hours a day.
By 1815 no child under ten years of age was to work in mills. In spite of this
prohibition, in the same year, the parliamentary transaction report, Hansard, reported a
speech where a Lancashire mill owner agreed with a London parish to take one idiot with
every 20 sound children supplied from the poor houses—so children of non-specified
ages were still being recruited.
The greatest changes in the workplace have taken place over the past three decades.
The International Covenant on Economic Social and Cultural Rights, ratified by more
than 90 countries, provides in Article 7(b) for ‘safe and healthy working conditions’.
While the level of safety might vary between workplaces, standards exist. The level of
safety required continues to rise and does so at a quicker rate than it took the English
Parliament to recognize the plight of sailors going to sea in dangerously overloaded
vessels, or to recognize the level of injury and death to children used in the chimney
cleaning trade.
Safety for workers on the job is widely recognized in the West, as are their rights to a
safe working environment. Ironically children in the workplace are often less protected
than their adult counterparts because of the vulnerability associated with their age.
Western societies may like to think that they are enlightened about children in the
workplace. Consider this report from a factory inspector in the United Kingdom, reported
in 1984 of a boy of 14 years of age found assembling electronic boards. He was working
in a room 8 feet by 5 feet by 9 feet. There was a diesel motor in the room giving off
copious amounts of fumes. The boy had a tube to breathe through one end of which was
passed through the window.
CHILDREN AND SAFETY
ironically children in the workplace are often less protected than their adult
counterparts
Safety matters, like so many matters in this world, are considered for children after they
are considered for adults. Advocacy has the potential to change attitudes and practices
with long term improvement where protection of children is concerned. The advocacy,
which led to the abandonment of the use of boy chimney sweeps in England, is especially
important as an injury prevention strategy.
Reports from the 1750s indicated that the practice of using young, small children to
sweep chimneys was widespread in England. The effects of this practice on the children
were injury from falls, burns from having fires lit under them to force them up the
chimney, Sweepers Cancer. This was a pioneering work in identifying an occupational
hazard leading to disease.
In 1788, David Porter petitioned parliament to restrict the use of boys under eight years
of age as sweeps. No change occurred and it has been recorded that in 1795 Thomas
Allen aged three and a half years was indentured as a chimney sweep. The first legislative
change occurred in 1834 (49 years later) when parliament forbade the binding of any boy
less than 10 years of age to a sweeper.
Europe had not used children in chimneys for many years. Design changes had been
incorporated into European chimneys to give access from the outside rather than by
travelling the length of the chimney. Europe also made use of alternative mechanical
devices to do the job which continued to be undertaken by children in England until
1875. This was 100 years after Dr Potts reported on the chimney sweepers cancer, and 87
years after David Porter first petitioned parliament about the use of children as sweeps.
WHAT THE UNITED NATIONS CONVENTION SAYS ABOUT

CHILDREN AND SAFETY
Following are some of the articles from the United Nations Convention which have some
bearing on developing rights for children to a safe environment.
Article 3.3
States parties shall ensure that the institutions, services and facilities responsible for the
care and protection of children shall conform with the standards established by competent
authorities, particularly in the areas of safety, health, in the number and suitability of their
staff, as well as competent supervision.
Article 4
States parties shall undertake all appropriate legislative, administrative, and other
measures for the implementation of the rights recognized in the present convention. With
regard to economic, social and cultural rights, states parties shall undertake such
measures to the maximum extent of their available resources and, where needed, within
the framework of international co-operation.
Article 5
States parties shall respect the responsibilities, rights and duties of parents or, where
applicable, the members of the extended family or community as provided for by local
custom, legal guardians or other persons legally responsible for the child, to provide, in a
manner consistent with the evolving capacities of the child, appropriate direction and
guidance in the exercise by the child of the rights recognized in the present convention.
Article 6
1. States parties recognize that every child has the inherent right to life.
2. States parties shall ensure to the maximum extent possible the survival and
development of the child.
Article 24
1. States parties recognize the right of the child to the enjoyment of the highest attainable
standard of health and to facilities for the treatment of illness and rehabilitation of health.
States parties shall strive to ensure that no child is deprived of his or her right of access to
such health care services.
2. States parties shall pursue full implementation of this right and, in particular shall
take appropriate measures.
A to diminish child and infant mortality; B to ensure the provision of all necessary
medical assistance and health care with emphasis on the development of primary health
care; E to ensure that all segments of society, in particular parents and children, are
informed, have access to education and are supported in the use of basic knowledge of
child health and nutrition, the advantages of breast feeding, hygiene and environmental
sanitation and the prevention of accidents.
3. States parties shall take all effective and appropriate measures with a view to
abolishing traditional practices prejudicial to the health of children.
UNITED NATIONS CONVENTION AND INJURY PREVENTION
Article 6 recognizes that children have an inherent right to life and that states should, to
the maximum extent possible ensure the survival and development of the child. This
provides a rationale to set priorities to deal with life threatening injury. In most parts of
the world motor vehicle injury will be a top priority.
The framework provides specifically for a safe environment for children in the care of
institutions and other agencies (Article 3.3) and in Article 24.2 identifies a right to the
highest attainable standard of health care, including the provision of information about
primary health care and the prevention of accidents.
To achieve this states should work to diminish child and infant mortality and ensure
health care is available with an emphasis on primary health care. The question of a right
to an injury free environment is not canvassed as an issue of rights. The question is
whether an injury free world is attainable or desirable.
the question of a right to an injury free environment is not canvassed as an issue of

rights. The question is whether an injury free world is attainable or desirable
There are many areas of injury prevention activity which cause unresolved debate among
the injury prevention community. Some examples where there is not consensus include
approaches to home use of trampolines, safety of playground equipment in public parks
and schools and use of baby walkers. A rights approach does nothing to inform that
debate. In fact a rights framework may lead to a dilemma about conflicting rights, for
example, the right to safety versus the right to developmental experiences, or risk taking
or a civil right.
THREE CASES OF INJURY PREVENTION: IS A RIGHTS

FRAMEWORK USEFUL?
Three cases are presented for consideration of the potential ambiguities in a rights
approach to injury prevention.
Three cases are presented for consideration of the potential ambiguities in a rights
approach to injury prevention.
Swimming pool fencing

Safety legislation was introduced in Queensland, Australia in 1990. This legislation
required new domestic swimming pools to be fenced on all four sides and for the fence to
have a self-closing, self-latching gate. In spite of numerous letters to the editor in
newspapers about children drowning in fenced pools, on investigation all of these
children had gained access through a gate which had been left propped open or was not
functional.
The achievement of pool fencing legislation in Queensland came about almost 20 years
after the problem was first recognized. The drowning rate for children 0–3 years, was
among the highest recorded in the world. An ordinance was first introduced into the
Queensland Capital City in 1978. This ordinance required all pools to be fenced on four
sides. The ordinance was vetoed by the State Government following political
intervention. The power of veto over local government had been used by the state only on
one previous occasion. This time the veto affected the lives of children.
The current legislation is more far reaching than the original local government city
wide legislation in that it is state legislation and the safety measure is now in place across
all local government areas.
The process of change was not without vigorous debate and persuasion of public
opinion. The public debate throughout the fencing campaign included phases in which
the threat was defined, a solution was publicized, the arguments were debated by interest
groups, opposition was identified, public opinion was surveyed and the legislation was
introduced and evaluated.
An addendum to the fencing legislation has been reported recently in the press in
Queensland. A set of apartments, many of which were rented out, had a swimming pool
with a faulty or missing lock on the access gate. The local authority served a notice on the
‘body corporate’, the body of owners responsible for the property, giving a period in
which the lock should be replaced. In the mean, time a child of three years of age entered
the pool enclosure and drowned.
This case involves elements of law (the pool was required to be fenced with a self
latching gate), ethics (the latch had not been repaired or replaced when it was known to
be faulty or missing by the ‘body corporate’ and rights (the child’s right to safety, defined
by the existence of the law). The resolution of this case could serve to establish the
child’s right to protection from this recognized community hazard.
In the case of pool fencing the target of the prevention strategy is children aged three
years and under. These children do not have the developmental proficiency to
discriminate, or negotiate with confidence, hazards such as unguarded pools in their close
environment. The problem was also at a level where child deaths were involved at a rate
higher than anywhere in the world. Advocacy for the rights to life of this group were part
of the community debate. Arguments against fencing included the cost to the home owner
and to the property value, that pools were on private property, that parents should
supervise children, and the aesthetic of having a pool with no fence. When the debate was
lost in 1978 and the state vetoed the local government ordinance, those arguments held
more weight than those of protecting life. By 1990 community perception had changed
and the right was established. Continued energetic advocacy will be required to keep the
protection in place. hand over hand, have been identified as producing the majority of
injuries. Of these injuries fractures to the upper limb predominate. Some of these
fractures are serious enough to affect long term function. Observation studies of children
in playgrounds suggest that children are five times more likely to play on a horizontal
ladder than the next most popular piece of equipment.
Removal of monkey bars will undoubtedly reduce injuries to children in playgrounds,
but at what cost. Children will be denied the most popular piece of equipment, and one of
the few pieces of equipment available to enhance upper body development.
What is the obligation of the injury prevention community? Is it to advocate for
removal and thereby injury reduction? Should injury prevention advocates entertain the
alternative arguments of enhancing child development or arguments of preferences for
particular equipment? Should injury research inform on the risks and avoid the public
debate? Public parks administrators in some places have removed play equipment to
avoid legal liability. Information about improving the safety of the equipment through
lowering or improving the surface is not yet available, and therefore not likely to be used
as an alternative to removal. The point here is that action may ensue instead of
community debate.
Does the responsibility of the injury prevention community go beyond identifying the
problem and educating parents? Should alternative solutions come from them or from
play leaders, or child development specialists who have an alternate interest to injury
prevention advocates. What of the preferences of the children? Should they be taken into
account?
Consumer safety
Children live in a world surrounded by an environment and products built by, and to a
scale for, adults. Consumer safety has been a relatively recent movement. The idea of
producing products to standards based on function and performance rather than on
engineering principals is still not widespread.
does the responsibility of the injury prevention community go beyond identifying

the problem and educating parents?
Safety of trampolines for home use has been discussed recently among subscribers to the
International Society for Child and Adolescent Injury Prevention. Home use trampolines
have been identified in a number of studies as posing significant threat to children. That
is, of the numerous injuries reported 15 per cent were assessed as severe while 12 per
cent were spinal injuries including fractures and paraplegia (Furnival et al., 1999). The
American Paediatric Association has recognized the hazard since 1981 and in 1999,
reconfirmed the view that they be banned.
Proponents of the trampoline cite the many hours of fun had by children using
trampolines, the aspect of development of coordination and the need for children to learn
to handle risk. Is the level of risk a consideration from a rights perspective? If the risk
Paediatric Association has recognized the hazard since 1981 and in 1999, reconfirmed the
view that they be banned.
Proponents of the trampoline cite the many hours of fun had by children using
trampolines, the aspect of development of coordination and the need for children to learn
to handle risk. Is the level of risk a consideration from a rights perspective? If the risk is
for death, then certainly rights are a consideration. The insurance industry had an
influence over objections to pool fencing in Queensland, when it declared that with
legislation in place, a pool owner who did not comply would void their public safety
insurance. The insurance industry has influenced the ban on trampoline use in some
schools through increased premiums, also reflecting the risk.
FUTURE DIRECTIONS
The future of children’s rights to safety is unassailable. The language of rights is

currently being used for adult safety in the workplace and on roads. It is only a matter of
time before children are included more generally in the rubric.
Because children constitute a ‘special case’ it will be some time before there is clarity
about the theoretical basis of rights for children. This debate will continue and progress
the cause of children’s rights.
On a practical front, it is clear from the United Nations Convention that the child’s
right to life is paramount consideration and this will be used successfully to promote
some aspects of injury prevention. It seems to be stating the obvious to suggest that in
setting priorities for injury prevention, fatal injury should be considered prior to non-life
threatening injury. However, fatal injury does not reduce health costs, but reducing the
impact of non-fatal injury is of greater economic benefit to a community. The focus in
some programmes seems to be towards reducing injuries as a means of reducing health
costs. Taking a rights perspective injury prevention professionals should advocate to
address fatal injury and for community resources be placed to address the most pressing
fatal injury causes.
Such advocacy can lead to changes in public policy backed by the strength of
legislation. However, legislation can be changed at the stroke of a pen. Helmet wearing
legislation for motor cyclists has been enacted and rescinded in some states of the United
States of America. Pedal cycle helmet laws have been threatened in some States of
Australia and pool fencing legislation was introduced and rescinded in New South Wales,
Australia. In a future where children’s rights are established parallel to those of adults,
children will have the same right to organize and challenge safety legislation, say, to ban
cycle helmet wearing or to challenge bans on trampolines or baby walkers. Alternatively,
children might recognize the damage to their own class, caused through the interests of
adults, and move to enforce bans where the interests of adults have thwarted them.
Advocacy in Australia has been effective in establishing that toddler aged children
have a right to protection from unfenced swimming pools. Twenty years of advocacy by
scores of people led to the introduction of legislation requiring domestic swimming pools
to be fenced. A rights framework for non-fatal injury prevention is more of a problem.
Arguments for reducing non-fatal injury to children using a rights framework are likely to
lead to debate about conflicting rights within the injury prevention community and the
broader community. The injury prevention community must focus on reducing injury
effectively, and not be diverted by the interests of others. The community will decide, but
injury prevention must advocate its position as actively as any other interest group.
In injury prevention there is a constant balance between opposing vested interests—
political, economic or social. Thus for many issues, especially those involving children, it
has been possible to stonewall change. The lesson for us is that injury prevention or
public health practitioners have a vested interest and must learn the skills associated with
winning in the political, economic and social ventures of life.
in a future where children’s rights are established parallel to those of adults,

children will have the same right to organize and challenge safety legislation
It took almost 100 years in England from the time opposition was mounted to stop the
practice of putting small children up chimneys as sweeps. The introduction of
compulsory fencing of swimming pools in the State of Queensland, Australia took 18
years of campaigning. While the time taken to implement change is long, the best
measure of the delays is in the number of deaths and suffering that might have been
prevented.
The frequently seen example of an adult cycling without a helmet followed by a child
cycling with a helmet, in spite of legislation requiring helmet use, illustrates the dilemma
of considering rights in the context of safety. While neither has the right to flout the law,
the adult considers the freedom to express his will and not use a helmet as a right. Should
community provided health services be withheld or a premium paid to provide services in
the case of injury to the adult? Does the position change if the child is injured? There are
currently no correct answers as far as rights for children are concerned. Advocacy for
injury prevention and children’s rights must take account of the balance between social,
political and economic forces to provide the most effective preventive solutions to injury
problems.
Laws confirming rights can and do change. The future of the debate on rights and
injury prevention depends very much on continued advocacy and involvement in that
debate. Through public debate of safety measures support and opposition for safety
measures may come from unexpected quarters. Advocates for the kidney foundation in
California have been heard to argue strongly in favour of rescinding helmet laws in that
state.
The future will see advocates liaising with odd partners in order to progress the cause
of preventing injury to children.
FURTHER READING
Alston, P.Parker and S.Seymour, J., 1992, Children, Rights and the Law. (Oxford:
Clarendon Paperbacks).
American Academy of Paediatrics Committee on Injury and Poison Prevention and
Committee on Sports Medicine and Fitness, 1999, Trampolines at home, school, and
recreation centres, Paediatrics, 103(1), pp. 1053–1056.
Furnival, R.A., Street, K.A. and Schunk, J.E., 1999, Too many trampoline injuries.
Paediatrics, 103, p. 57
Mott, A.Rolfe, James, K., Evans, R., Kemp, R., Dunstan, A., Kemp, K. and Sibert, J.,
1997, Safety of surfaces and equipment for children in playgrounds. The Lancet, 349,
pp. 1874–1876.
Parker, S., 1992, Child support: rights and consequences. In Children, Rights and the
Law. Edited by Alston, P.Parker and S.Seymour, J. (Oxford: Clarendon Paperbacks).
14
Evidence-based Injury Prevention and Safety
Promotion: State-of-the-art
Leif Svanström
INTRODUCTION
Concepts and definitions

Generally speaking there is the same contradiction between the concepts of injury
prevention and safety promotion as between disease prevention and health promotion.
The starting point of injury prevention is an outcome of a process where the medical view
decides the prevention activities in the same way as in disease and disease prevention.
However, safety as well as health is a much broader concept and safety promotion is
based in the society and community and how the individual or populations understand the
concept of safety and what measures need to be taken. Such actions might lead to
prevention of injuries or might not.
Because health and safety is a fundamental right of human beings, safety is also a
prerequisite to the maintenance and improvement of the health and welfare of a
population. It is a fundamental need of human beings.
Safety can be defined as a state or situation characterized by an adequate control of
physical, material or moral threats and which contributes to a perception of being
sheltered from danger (Andersson and Svanström, 1998).
Thus safety cannot be defined in absolute terms. Safety is a dynamic state. Safety is not
merely the absence of injuries or threats. Safety promotion should not be narrowed down
to injury prevention.
Safety promotion can be defined as a process that aims to provide populations the
means to ensure the presence and maintain the conditions that are necessary to reach and
sustain an optimal level of safety.
Safety promotion is all organized efforts by individuals, organisations and
communities to achieve that ultimate goal. Structural as well as attitudinal and
behavioural changes all aim at creating sustained supportive environments for safety.
By environment we mean not only the physical but also the social, cultural,
technological, politic, economic and organisational environment. Changing attitudes and
behaviours of man is not a goal in itself but merely a means to achieve an environment
and social structure where safety is built in.
In practice, on the road to achieve that goal, man has used a combination of efforts like
safety promotion, injury prevention, treatment and rehabilitation. The choice of one
approach over the other is important. Structural and environmental change demands a
Evidence-based injury prevention and safety promotion 193
democratic work, both participatory and representative, and can be more demanding than
the choice of organising treatment and rehabilitation after lack of real safety leading to
injury or environmental or machine damages.
The idea of an evidence-based safety promotion is based on the closeness between the
concepts of injury prevention and safety promotion. Evidence-based safety is always
related to a defined outcome. We have not been able to identify any studies of scientific
character where the outcome is a feeling of safety among individuals or populations.
Most outcomes are defined as injuries or lack of injuries or change of injury incidents.
Some examples of evidence-based findings are given here. The majority is from the areas
of safety promotion for children and elderly.
CHILD SAFETY PROMOTION
Safety promotion in the traffic environment
safety can be defined as a state or situation characterized by an adequate control

of physical, material or moral threats and which contributes to a perception of
being sheltered from danger
“There is good evidence that the use of cycle helmets and child car seat restraints can
reduce serious injury to children involved in road traffic accidents. Urban road safety
measures such as a provision of crossing patrollers, measures to re-distribute traffic and
improve the safety of individuals can reduce the rate and severity of childhood
accidents.” (Effective Health Care, 1996).
In an urban safety project the effect of measures to redistribute traffic and improve the
safety of individual roads was assessed in five English towns compared to match control
areas (Lynam et al., 1988). There was an overall accident reduction of 13 per cent
attributable to the schemes but there were great variations between schemes. Slight
injuries declined proportionately more than serious ones. Measures that were particularly
successful were those which protected two-wheeled vehicles (such as right turn
prohibition and central road dividers) and there was a general reduction in child cyclist
casualties. Each scheme cost about 250,000 Pounds and first year rates of return indicated
considerable accident costs savings.
The speed at which a car is driven affects the severity of pedestrian injuries (20 mph
leads to 5 per cent death; 30 mph-45 per cent; 40 mph-85 per cent). Therefore transport
policies aiming at reducing excessive car speeds maybe effective. However, there is very
little evaluation of such interventions.
‘There is little reliable evidence to suggest that children can be successfully trained to
avoid injury on the roads’ (Effective Health Care, 1996). Controlled trials indicate that
teaching children road crossing skills, however, can change reported behaviour and that
instruction in the classroom can be as effective as the road sign. “Educational programs
by themselves appear to have little effect. However, a number of community programs,
which involve local participation, and use of broad range of interventions have been
effective at reducing childhood injuries from a wide variety of causes. These need to be
based on accurate data derived from surveillance systems.” (Effective Health Care 1996).
Some studies show no evidence that children’s knowledge of road safety (e.g. Traffic
Clubs) had been improved. The Streetwise Kids Club’ was introduced in London, but
membership was low, particularly in lower social class groups (Downing, 1987). An
evaluation of the ‘Eastern Region Traffic Club’ showed increased participation and a
positive effect on aspects of behaviour and a 20 per cent reduction in casualties involving
children emerging from behind a vehicle (Bryan-Brown, 1995).
Renaud et al. (1989) has evaluated simulation games promoting traffic safety for
children. Using a simulation game designed to teach children to obey certain traffic safety
rules, an experimental study was conducted with 136 five-year-old children in four
Quebec schools. Within each classroom, subjects were randomly divided into four
groups: three intervention groups and one control group. Each of the experimental groups
was subjected to a different intervention with outcome measured using three instruments
related to attitudes, behaviour and transfer of learning of pedestrian traffic safety. Results
suggest that simulation games including role-playing/ group dynamics and
modelling/training can change attitudes and modify behaviour in the area of pedestrian
traffic safety and children of this age.
Several surveys and epidemiological studies have reported that cyclists who wear
helmets have a reduced risk of severe head injuries (Graitcer et al., 1995). In the past
decade there have been a variety of educational approaches used to promote the use of
bicycle helmets. These programs have included classroom curricula, programs that
subsidize the purchase of helmets, and programs that provide helmets in a health care
setting. None of these individual educational strategies has been shown to have any
significant impact in increasing helmet use. Only helmet promotion programs that are
organized by community-wide coalitions and use a variety of educational and publicity
strategies have been shown to be effective. The most successful of these programmes—
the Seattle Children’s Bicycle Helmet Campaign organized by Harborview Injury
Prevention and Research Center—used multiple strategies. These included classroom
education, discount purchase programmes, bike rodeos, distribution of printed material
through a variety of venues, and intensive promotional efforts by sports leaders, bicycle
clubs, and the media to increase children’s helmet use (Bergman et al., 1990). Using this
broad based approach, the Harborview program has been able to increase helmet wearing
rates among children to more than 40 per cent (Rivara et al., 1994). No information is
available on the direct and indirect cost of these promotional efforts. The relative
difficulty in implementing educational programs to promote helmet use, their potentially
grate costs, and their success in greatly increasing helmet use, have led to the introduction
of mandatory helmet wearing laws as a principle strategy of many governmental
jurisdictions. Studies from the introduction of legislation on helmets in Victoria,
Australia, showed that during the period 1975–1980 less than 5 per cent of all Victorian
bicyclists wore helmets. In March 1991, approximately 9 months after implementation of
the law, the average use rate for Victoria bicyclist was 75.2 per cent. A special survey,
conducted in May 1992, indicated that this rate increased further to 83 per cent (Cameron
et al., 1994). The number of head injuries decreased by 48 per cent during the first year
and an additional decline to 70 per cent of the pre-law levels was noted in 1991/2.
One major evaluation of the effect of community wide programs to promote the
wearing of cycle helmets showed a significant reduction in the rate and severity of
casualties. In 1990, following ten years of cycle helmet promotion campaigns, the state of
Victoria in Australia introduced the first law in the world requiring cyclists to wear
helmets. The increase of helmet wearing rates from 31 per cent immediately before to 75
per cent in the year following legislation was associated with a 48 per cent reduction in
head injury admissions or death between 1989/90 and 1990/91 and a reduction of 70 per
cent over the two year period 1989/90–1991/92. As with seat belt legislation, the
experience in Victoria has shown that legislation following education campaigns can
increase use. In a recently published study (Ekman et al., 1997) shows for some
intervention areas of Sweden, for children under 15, an average annual decrease in all
bicycle-related injuries of 3.1 per cent, equivalent to a decrease of 48 per cent over the
study period, 1978–93 (for head injuries, 59 per cent). Sweden as a whole showed a
reduction of 32 per cent in bicycle-related injuries (head injuries, 43 per cent). In
Skaraborg, children have been the targets of helmet-wearing programs at local and
regional levels since 1982 and at national level since 1987.
There is considerable evidence that child car seat restraints (for young children) when
properly used, reduce car occupant injuries (Agran et al., 1989).
In the United States Child passenger restraint use and motor vehicle related fatalities
among children (MMWR, 1991). In 1990, child safety seats were used for an estimated
83 per cent of infants and 84 per cent of toddlers, compared with 60 per cent and 38 per
cent, respectively, in 1983. Use of child safety seats reduced the likelihood of fatal injury
by an estimated 69 per cent for infants and 47 per cent for toddlers. Adult safety belts
used for toddlers reduced the likelihood of fatal injury by 36 per cent.
Safety promotion in the home environment
there is little reliable evidence to suggest that children can be successfully trained
to avoid injury on the roads
General home injuries are more common in households with poor social circumstances.
Rather than focusing on individual parenting behaviour it has been suggested that
increasing financial and social support to deprived households with young children would
have a beneficial effect on injury rates. However, no relevant evaluation has been
identified. A programme targeting poor, unmarried or teenage mothers of pre-school
children in the USA indicated that homes which had several visits from a nurse home
visitor had fewer home hazards than those which had not been visited (Olds et al., 1994).
The use of safety devices in the home such as smoke detectors, child resistant containers
and thermostat control for tap water can reduce the risks of home injuries. Targeting of
households at higher risk combined with home visits, education and the free distribution
of devices is likely to make the most impact.
A variety of protective safety devices have been tested under experimental and field
conditions and have been shown to reduce the risks of home injuries.
These include smoke detectors and child resistant container closures. Others are also
associated with reduced risk such as fireguards, stair-gates, safety catches for cupboards,
coiled kettle flexes, safety harnesses, safety film for interior glazing and thermostat
control of tap water (Department of Trade and Industry, 1991).
Programs aimed at raising awareness of home hazards encouraging parents and
children to reduce or avoid these risks have met with varying success. Home visits to
people in poorer areas with specific advice on hazards, combined with health education
and media campaigns resulted in around 50 per cent more households making changes to
the home environment (Colver et al., 1982).
Burns, scalds and smoke detectors.

There have been a series of evaluations of programs designed to increase the use of
smoke detectors. In one program, smoke detectors were given away free and 81 per cent
were operational 8–12 months after the campaign (German et al., 1985). One study
showed a small reduction in fatalities due to fires in a community where smoke detectors
were required by law in all homes. However, compliance was low (McLoughlin et al.,
1985).
Tap water temperature reduction.

One study showed that provision of a free thermometer when combined with physician
counselling was more effective than counselling by itself at reducing scalds (Katcher et
al., 1989).
Erdmann et al. (1991) point to the importance of legislation. Five years after a 1983
Washington State law required new water heaters to be pre-set at 49 °C, 77 per cent of
homes had tap water temperatures of less than 54 °C. In 1977, 80 per cent of homes had
tap water temperatures grater than 54 °C. The period of July 1979 through May 1988
showed an average admission rate of 2.4 per year, compared with 5.5 per year in the
seventies.
Poisoning
Clarke et al. (1979) points to the effectiveness of child-resistant closures, required under
the Poison Prevention Packaging Act of 1970, in reducing the incidence of accidental
ingestion of aspirin and aspirin-containing products among children less than 5 years of
age has been investigated. For baby aspirin, it is estimated that safety packaging has
reduced the incidence of ingestion 45 per cent to 55 per cent. For non-baby aspirin
products, the reduction has been 40 per cent to 45 per cent.
A controlled trial of children resistant containers for paraffin showed a 47 per cent
drop in paraffin ingestion compared to no change in the control area. Walton (1982)
reported on a study involving regulated substances like aspirin, acetaminophen,
prescription drugs and household chemicals. The ingestion rates for all substances that
require child resistant closures has declined from 5.7 per 1000 children 1973 to 3.4 per
1,000 children in 1978. It is estimated that child resistant closures have prevented nearly
200,000 accidental ingestion since 1973 in the USA. The death rate due to poisoning of
children has declined from 2.0 per 100 000 children to 0.5 per 100,000.
Safety promotion in the play and sport environment.

In the review over child injury prevention (Effective Health Care 1996) no good quality
studies were identified which evaluated the effectiveness of injury interventions
associated with sports. Similarly there has been little evaluation of playground layout,
equipment and surfacing in terms of achieving injury reduction; no evaluated studies of
training schemes for adults or children were identified.
A recent review suggests that a number of safety measures such as rule changes and
use of safety equipment may be effective at reducing injuries as a result of organized
sports in the 15–24 year age group (Coleman et al., 1996).
Organisation-based interventions
One study showed that provision of a free thermometer when combined with
physician counselling was more effective than counselling by itself at reducing
scalds
Especially the role of primary health care organisations has been investigated (Bass et al.,
1993). Twenty articles met the criteria for inclusion. Of these, 18 showed positive effects
of injury prevention counselling including 5 randomized/controlled, 10 non-
randomized/controlled, 2 multiple-time series and 1 descriptive study. In 15 of the
positive studies, physicians performed the counselling. Positive outcomes as measured by
increased knowledge, improved behaviour, or decreased injury occurrence were reported
for both motor vehicle and non-motor vehicle injuries. The literature review supports the
recommendation of the AAP to include injury prevention counselling as part of routine
health supervision.
Miller et al. (1995) has evaluated the role of the paediatrician. The authors have
estimated the savings achievable with comprehensive childhood injury prevention
counselling organized around the three Framingham Safety Surveys used in The Injury
Prevention Program (TIPP) developed by the American Academy of Paediatrics. TIPP
paediatrician injury counselling sessions between the ages of 0–4 years can achieve
estimated savings of US$ 880 per child or US$ 80 per visit. If all 19.2 million children
ages 0–4 years completed TIPP, they estimated that US$ 230 million would be saved
annually in medical spending, and injury costs would decline US$ 3.4 billion. Each dollar
spent on TIPP childhood injury prevention targeting children 0–4 years returns nearly
US$ 13. TIPP encompasses up to 11 visits between the ages 0–4 years. Topics covered
include child safety seat and smoke detector use, crib safety, water safety, firearm safety,
pedestrian safety, play equipment safety, fall prevention, burn prevention, choking and
suffocation prevention and poisoning prevention.
Bablouzian et al. (1997) has evaluated how high risk pregnant women, who were
enrolled in home visiting program that augments existing health and human services
received initial home safety assessments. Clients received education about injury
prevention practices, in addition to receiving selected home safety supplies.
Results showed a significantly larger proportion of homes were assessed as safe at
discharge compared with the initial assessment. That was true for the following hazards:
children riding unbuckled in all auto travel, Massachusetts Poison Centre sticker on the
telephone, outlet plugs in all unused electrical outlets, safety latches on cabinets and
drawers, and syrup of ipecac in the home. Thus four home hazards were significantly
reduced for which safety supplies were provided. Education and promotion of the proper
use of child restraint systems in automobiles significantly reduced a fifth hazard children
riding unbuckled in auto travel.
Community-based interventions
The Falköping Program in Sweden included establishment of an extensive network of
people interested in injury prevention, education of policy makers and health workers,
raising a public awareness, and provision of a local shop selling child safety products.
The intervention area experienced a reduction of 27 per cent in home accidents and 28
per cent in occupational accidents (Schelp, 1987).
In the ‘State-wide’ Child Injury Prevention Program (SCIPP) USA (Guyer et al., 1989)
nine intervention communities and five control communities were selected in
Massachusetts. Interventions targeted burns, poisoning, falls, suffocation and passenger
traffic accidents. Households in the intervention communities had greater safety
knowledge and higher behaviour scores than controls. There was a significant reduction
of motor vehicle passenger injuries in the intervention communities. No evidence was
found for the reduction of other target injuries.
The ‘Safe Block Project’, in Philadelphia, U S (Schwarz et al., 1993) targeted a poor
inner-city African-American community, using community workers and recruiting black
representatives from the local community. This method of “cascade training” was
successful in getting households involved. The intervention included an educational
program, home visits and the provision of safety equipment. The intervention was
partially effective for those home hazards requiring minimal or moderate effort to correct.
No information was provided on baseline comparability of the areas and no data were
collected on accident rates.
In a community intervention study Svanström et al. (1995) report on a programme
focusing childhood safety. The Lidköping Accident Prevention Program was compared
with four bordering municipalities and to the whole of Skaraborg County and included
five elements: surveillance, provision of information, training, supervision and
environmental improvements. In Lidköping there was an on average annual decrease in
injuries leading to hospital admissions from 1983 to 1991 of 2.4 per cent for boys and 2.1
per cent for girls compared with an smaller increase in one comparison area and a decline
in the other. (Four border municipalities: girls +2.2 per cent, boys +0.6 per cent;
Skaraborg county −0.3 per cent for girls and −1.0 for boys.)
Conclusion on prevention of childhood unintentional injuries

In a review made by Dowswell et al. (1996) it is concluded that examples of
interventions that have been effective in reducing injury include are bicycle helmet
legislation, area wide traffic calming measures, child safety restraint legislation, child
resistant containers to prevent poisoning, and window bars to prevent falls. Interventions
effective in changing behaviour include bicycle helmet education and legislation, child
restraint legislation, child restraint loan schemes, child restraint educational campaigns,
pedestrian education aimed at the child/parent, provision of smoke detectors, and parent
education on home hazard reduction. For the community based campaigns, the key to
success has been the sustained use of surveillance systems, the commitment of inter-
agency of co-operation and the time needed to develop networks and implement a range
of interventions. Education, environmental modification and legislation all have a part to
play and their effect in combination is important.
Educational programmes by themselves appear to have little effect. However, a

number of community programmes, which involve local participation, and use of
a broad range of interventions have been effective at reducing childhood injuries
from a wide variety of causes
Educational programmes by themselves appear to have little effect. However, a number

of community programmes, which involve local participation, and use of a broad range of
interventions have been effective at reducing childhood injuries from a wide variety of
causes.
PROMOTING SAFETY FOR ADOLESCENTS
Literature has been reviewed under the standard setting headings of road, work, home,
and sports and leisure (Munro et al., 1995). The most effective measures appear to be
legislative and regulatory controls in road, sport and workplace settings. Environmental
engineering measures on the road and in sports have relatively low implementation costs
and result in fewer injuries at all ages. There is little evidence that purely educational
measures reduced injuries in the short term. Community based approaches may be
effective in all age groups, and incentives to encourage safer behaviour hold promise but
require further evaluation. The potential of multi-factorial approaches seems greater than
narrowly based linear approaches. The conclusion is however that few interventions to
reduce injury in adolescents have been rigorously evaluated. There were no studies
identified relating to 15–24 year olds in domestic settings. Studies, however
demonstrating the effectiveness of smoke detector programs are clearly relevant to this
age group.
Alcohol being a major risk problem has led to a number of studies on prevention, as in
a by Foxcroft et al. (1992). 33 studies of which 24 were randomized controlled trials or
had well matched controls were evaluated. Assessment of the quality of the studies
showed that only 10 of the 33 studies included met four core methodological criteria,
indicating the poor quality of the studies. Overall, no prevention program was
convincingly effective. Of the 29 studies of prevention programs with short-term follow-
up, 16 were partially effective, 11 were ineffective and 5 had negative effects (increased
alcohol consumption). There were 12 prevention programs with medium-term follow-up.
Of these five were partially effective, five ineffective and two had negative effects. Only
two prevention programs had long-term follow-up, one was effective and one ineffective.
SAFETY PROMOTION IN THE WORK ENVIRONMENT
Prevention work designed to reduce the number of occupational injuries can be

undertaken at a national or regional level and can also take place locally at the level of
the organisation and the individual (Menckel, 1990). Near accident reporting has been
used. A review of available literature produced 24 research reports in the of near accident
reporting. This had been completed between 1968 and 1982 and had been conducted at
the local level. The review demonstrated that large quantities of information of near
accidents, which otherwise would not have been recorded, were gathered by verbal
means while written reporting provided relatively little information. No clear-cut
conclusions with respect to the effects on near accident reporting could be drawn, above
all because of limitations to the documentation of the near accident reporting that had
taken place. Some of the studies pointed to an increase in the level of safety activities
following a period of near accident reporting while such an increase could not be detected
with certainties in others (Carter and Menckel, 1985). Field experiments involved
occupational health nurses and safety engineers from occupational health services and
those with local responsibility for safety and production. Improvements were detected in
accident and prevention work with quicker accident investigation, an increase in the
number of counter measures proposed and a certain reduction in the number of accidents
in the number of days per accident.
An example of a very successful intervention has been reported by Springfeldt (1996)
Serious occupational injuries cause by rollover tractors can be prevented by fitting
Rollover Protection Structures (ROPS). In some countries authorities have required such
protection on tractors for many years. Others have recommendations or no rules at all. In
a survey of tractor rollover injuries and protective measures, the developments in twelve
countries are described. Sweden and New Zealand were first to develop ROPS in the
fifties Mandatory regulations for new tractors were introduced in Sweden 1959, Denmark
1967, Finland 1969, West Germany, Great Britain and New Zealand in 1970, the United
States 1972, Spain 1975, Norway 1977 and in Switzerland 1978. In Canada and Australia
federal authorities have no regulations at all concerning ROPS and tractors.
In Sweden the frequency of fatal rollovers by 100,000 tractors per year has been
reduced from 17 to 0.3 since mandatory regulations were introduced. In Norway the
frequency has decreased from 24 to 4 between the periods 1961–1969 and 1979–1986
and in Finland from 16 to 9 from 1980–1987. From 1961–1986 a West German rate has
been reduced from 6.7 to 1.3. In New Zealand the risk has decreased from 37 to 30 from
the period 1949–1958 to the period 1969–1974. In countries that introduced mandatory
regulations long ago an evident decrease in the number of rollover injuries is reported,
especially when retrofitting of ROPS on old tractors is required (Springfeldt, 1993).
PREVENTING FALLS AND SUBSEQUENT INJURY IN OLDER

PEOPLE
The risk of falls increases with age. Falls in older people often result in injury and death.
Such injuries, frequently fractures, are a common and costly cause of hospital admission.
General overview of interventions preventing injuries to the elderly

There is a recently published review on the evidence-base of this area (Effective Health
Care, 1996). They conclude:
• There is some evidence to suggest that exercise, such as balance training, is

effective in reducing the risk of falls in older people. Access to such
interventions should be offered and ways of promoting uptake should be
investigated. New programs should be part of controlled evaluations.
• Home visits and surveillance to assess and where appropriate, modify
environmental and personal risk factors can be effective in reducing falls. This
can be carried out by nurses, health visitors, occupational therapists or trained
volunteers.
• Soft hip protector pads have been shown to dramatically reduce hip
fractures in frail older people in residential care. Their effect and acceptability
in the community needs further research.
• High dose Vitamin D supplementation with or without calcium appears to
be effective in reducing fractures. Research is needed to identify the most cost-
effective strategy.”
in Sweden the frequency of fatal rollovers by 100,000 tractors per year has been
reduced from 17 to 0.3 since mandatory regulations were introduced
Sackett et al. (1991) identified thirty-six trials which evaluated interventions to prevent
falls (exercise (23), home assessment (9), type of shoe (1), interventions in institutional
settings (3), nutritional supplementation (1) and hip protectors (1), two studies examined
interventions covering more than one area. Pooling the results of these studies showed
that people assigned to an exercise group had an estimated 10 per cent lower risk of
falling than controls. Pooling the results of studies evaluating a balance training only
intervention showed a reduction in the risk of falling of 25 per cent. In one trial people
offered the balancing exercise Tai Chi, had 37 per cent lower risk of falling than the non-
intervention group (Wolf et al., 1993)
The review also covered studies involved visiting older people at home, an assessment
of the safety of the home environment, and a range of interventions such as safety checks,
safety modifications, referral to care, and recommendations for exercise. In a study of
over 2,000 people, Hornbrook et al. (1994) found that those offered a home intervention
to remove and repair safety hazards showed a reduction in falls compared with controls.
Similarly Carpenter and Demopoulos (1990) found that older people visited at home by
trained volunteers reported one-third of the number of falls as did controls.
Six trials were identified which evaluated the effectiveness of interventions other than
exercise and assessment in people who live in institutions. Frail people are at particularly
high risk of falling when getting out of bed. The use of a bed alarm, which alerts an
assistant when a person to get out of the bed, was evaluated in a small trial. This showed
a reduction in falls, which was not statistically significant (Tideiksaai et al., 1993).
Significant protection against falling was apparent from interventions which targeted
multiple, identified, risk factors in individual patients, and from interventions which
focused on behavioural interventions targeting environmental hazards plus other risk
factors.
Medical interventions preventing injuries for elderly

Recently a systematic review of RCTS (Randomized Clinical Trials), completed within
the Cochran Collaboration, looked at the effectiveness of Vitamin D and calcium
supplementation in reducing fractures (Gillespie et al., 1996). However, the so far most
extensive review has been made by INAHTA (Hailey et al., 1996) Bone Density
Measuring is poor at determining present fracture risk since it cannot distinguish clearly
between patients with fracture and those without. ‘It is estimated, under realistic
assumption, that a bone screening program would lead to the prevention of only 2 per
cent of fractures in menopausal women (screening uptake of 50 per cent, sensitivity of 38
per cent, reduced fracture risk of 30 per cent with Hormone Replacement Therapy
(HRT), lifetime compliance with HRT of 30 per cent.) fracture risk with ever and never
users of HRT’. Consequently 393 women would need to be invited for screening in order
to avoid one fracture.
Technology interventions preventing injuries for elderly

In a study from Denmark on the effect of hip protectors (Lauritzen et al., 1993) 10 of the
28 wards in the nursing home were randomized to receive external hip protectors; thus
167 women and 80 men were given protectors and 277 and 141 men no protectors. A fall
register was set up for 2 treatment wards and 2 control wards. There were 8 hip and 15
non-hip fractures in the hip-protector group and 31 hip and 27 non-hip fractures in the
control group. The relative risk of hip fractures among women and men in the
intervention group was 0.44. None of the 8 residents in the intervention group who had a
hip fracture was wearing the device at the time of the fracture. The study indicates that
external hip protectors can prevent hip fractures in nursing home residents.
Organizational interventions preventing injuries to the elderly

A randomized controlled trial (Rubenstein et al., 1990) from a long-term residential care
facilities for elderly persons included an intervention group of 79 people and a control
group of 81. The assessment included a detailed physical examination and environmental
assessment by a nurse practitioner, laboratory tests, electrocardiogram and 24-hour Holter
monitoring. At the end of the two-year follow-up period, the intervention group had 26
per cent fewer hospitalizations and a 52 per cent reduction in hospital days compared
with controls. Patients in the intervention group had 9 per cent fewer falls and 17 per cent
fewer deaths than controls by two years.
Community interventions preventing injuries to the elderly

Tinetti et al. (1994) studied 301 men and women living in the community who were at
least 70 years of age were given either a combination of adjustment in their medications,
behavioural instructions, and exercise programmes aimed at modifying the risk factors or
usual health care plus social visits. During one year of follow-up, 35 per cent of the
intervention group fell as compared with 47 per cent on the control group—a significant
reduction.
Svanström et al. (1996) have shown similar results for a whole community in a quasi-
experimental design. The incidence of femoral fractures in Lidköping—the study area-
shows a significantly declining trend for females (6.6 per cent per year) and (a 5.4 per
cent decline for males during the study period of 1987–1992. The control area of
Skaraborg County showed a minor decrease and Sweden as a whole showed an
increasing trend during the same period
people assigned to an exercise group had an estimated 10 per cent lower risk of
falling than controls
Ray et al. (1997) give support through a randomized controlled trial with randomization
of nursing homes. Seven pairs of middle Tennessee nursing homes with one facility in
each pair randomly assigned to the intervention. Facilities had 482 (261 control, 221
intervention) residents who qualified for the study. The mean proportion of recurrent
fallers in intervention facilities was 19.1 per cent lower than that in control facilities.
Intervention targeted four specific safety domains: environmental and personal safety,
wheelchairs, psychotropic drugs and transferring and ambulation.
Plautz et al. (1996) has reported on the trial established by the Department of Public
health, City and County of San Francisco—the Community and Home Injury Prevention
Program for Seniors (CHIPPS) to reduce the rate of unintentional injuries among elderly
residents. The objective was to reduce rates of falls, scalds and burns. Reported falls were
reduced by 60 per cent after the intervention, from 0.81 to 0.33 falls per person year.
Scalds were reduced from 9 to 0 and burns from 7 to 0 during the six-month periods
before and after the intervention.
SAFETY PROMOTION AT THE LOCAL LEVEL—SAFE

COMMUNITY INTERVENTIONS
Community interventions may be distinguished by their shift away from the focus on
individual responsibility and towards multi-faceted community wide interventions, which
ensure that everyone in a community is aware or involved.
As far as we know The Falköping Accident Prevention Program (FAPP) is the first
evaluated comprehensive programme aiming at promoting safety and preventing injuries
at the local community level. The idea behind is to address all kinds of safety and prevent
injuries in all areas, addressing all ages, environments and situation and involving non-
governmental as well as governmental community sectors. FAPP is based in Skaraborg
County, Sweden. An injury register was started in 1978 and intervention began in 1979.
Three years later the total rate of injuries had fallen by 23 per cent. In Falköping home
injuries decreased by 26.7 per cent, occupational injuries by 27.6 per cent, traffic injuries
by 27.7 per cent and other injuries by 0.8 per cent (Schelp, 1987). A corresponding
decrease in the number of other emergency visits, ie non-accident related, has not been
noticeable during the same period of time. The incidence of home injuries in the study
area decreased from 26 per 1,000 per year in 1978 to 17 per 1,000 per year in 1981/2
(Schelp L & Svanström L 1986). In 1978 there were 49 occupational injuries per 1,000
gainfully employed persons/ per year whereas the corresponding figure for 1981/2 was 34
occupational accidents per 1,000 gainfully employed persons/year (Schelp and
Svanström, 1986).
The program has then been followed over the period from 1978–1991 (Svanström et
al., 1996), using indicators of processes as well as outcome. Since 1983 the outpatient
injury rate has levelled off and the inpatient rate shows an average annual increase of 8.7
per cent for females and 4.9 per cent for males, which is significantly higher than the
increase for Sweden, which is 2.3 per cent for females and 0.5 per cent for males.
For injuries seen in either ambulatory or hospital settings, it seems that the effect of the
early phase of the program was the lasting one, but for injuries admitted to hospital the
effect was temporary. The onset of the increase by the end of 1982 coincides with the
break-up of the cross-sectoral organization originally set up to run FAPP.
In the beginning of eighties the programme was followed by others in Norway like
Vaeroy and Harstad and Sweden, like Lidköping and Motala.
In the program implemented in Motala municipality in the western part of Ostergötland
County in Sweden. The incidence on non-trivial injuries treated in health care decreased
by 41 per cent, while the trivial injuries increased by 16 per cent (Timpka et al., 1998).
The incidence of health care treated injuries had decreased by 13 per cent from 119 (per
1,000 population years) to 104. In the control area, corresponding injury incidences were
104 and 106. The hospital-treated injuries decreased by 15 per cent from 19 per 1 000
population years to 16, while in the control area the incidences remained at 13 per 1 000
population years (Lindqvist et al., 1998). The larger decrease of non-trivial injuries was
observed in all ages and injury event environments.
The best-evaluated programme in Norway is in Harstad (Ytterstad B 1995).
In a quasi-experimental study (Ytterstad and Wasmuth, 1995), hospital-treated traffic
accident injuries were recorded prospectively for seven and a half years in two
Norwegian cities, Harstad and Trondheim. Traffic safety was promoted in a
comprehensive community program in Harstad, with Trondheim as a control. A 27 per
cent overall reduction of traffic injury rates was found in Harstad from period 1 to period
3 (each period 30 months duration), whereas a correspondent significant increase was
found in the comparison city. Promotion of bicyclist helmet use and pedestrian safe
behaviour was implemented by activating public and voluntary organisations and media
(Ytterstad, 1995). Significant rate reductions were observed below the age of 16 for both
bicyclists (37 per cent) and pedestrians (54 per cent). For bicyclists this reduction was
larger among males (43 per cent) than females (23 per cent). Head injury rates decreased
for bicyclist children below 10 years of age but increased for those 10–15 years old.
Accident analyses based on the local database revealed coffee to be the most frequent
liquid causing scalds, which mostly occurred in the kitchen (Ytterstad and Sögaard,
1995). Sixty-six per cent of the injured were boys and two-thirds were below two years of
age. From the first to the second of the three periods the mean burn injury rate decreased
53 per cent, from 53 to 25 per 10,000 person years. In the Control City located 1,000 km
away, the rates increased from 62 per cent to 68 per 10,000 person years.
Sports injuries account for considerable morbidity and expenditure of resources
(Ytterstad, 1995). It accounted for 17 per cent of recorded unintentional injuries in
Harstad. Post-intervention injury rates for downhill skiing was reduced by 15 per cent
when adjusting for exposure.
older people visited at home by trained volunteers reported one-third of the

number of falls as did controls
Outside the Nordic countries the Safe Community Model has been especially popular in
Australia. Jeffs et al. (1993) reports on the Illawarra area of New South Wales.
Reductions of 17 per cent in attendance by children for injuries and a 14 per cent fall in
accident-related hospital admissions of children have been observed over the course of
the four year period, 1987–1991 before and after the intervention.
The evaluation of the Latrobe Valley Better Health Injury Prevention Program, a
community-based intervention in south east Victoria, Australia (Day et al., 1997) showed
that the age standardized rate per 100,000 persons for emergency department
presentations for all targeted injury fell from 6594 in the first program year to 4821 in
1995/96. There were significant decreases in the presentation rates for home injuries
among all age groups except for those 65 years and over, playground injuries among 5–
14, 15–24 and 25–64 years old and sport injury among 15–24 year olds only. The direct
program cost per injury prevented was $ 272. Significant reductions were observed for
assaults among 10–24 year olds compared to those over 25 years. The conclusion is that
the reductions were associated to some extent with changes in injury risk and protective
factors and were greatest for the injury issues subjected to the most intense activity.
FUTURE DIRECTIONS
This review of some of the evaluated interventions that exist in literature was only meant
to show that extensive information exist of value for program managers and decision-
makers. We look forward to further diving into the literature database and your
contribution.
REFERENCES
Agran, P., Dunkle, D., Winn, D., 1987, Effects of legislation on motor vehicle injuries to
children. Am J Dis Child, 141, pp. 959–964.
Andersson, R. and Svanström, L., 1998, Critical Factors Required for the Successful
Mobilisation of Communities to Enhance Safety. Manuscript prepared for Seminar on
Safety and Safety Promotion: Conceptual and Operational Aspects. Château
Frontenac, City of Quebec, Canada. February 5 and 6, 1998. Karolinska Institutet,
Department of Public Health Sciences, Division of Social Medicine.
Bablouzian, L., Freedman, E.S., Wolski, K.E., Fried, L.E., 1997, Evaluation of a
community based childhood injury prevention program. Injury Prevention, 3, pp. 14–
16.
Bass, J.L., Christoffel, K.K., Widome, M., Boyle, W., Scheidt, P., Stanwick, R., Roberts,
K., 1993, Childhood injury prevention counselling in primary care settings: A critical
review of the literature. Paediatrics, 92, pp. 544–550.
Bergman, A.B., Rivara, F.P., Richards, D.D. et al., 1990, The Seattle Children’s Bicycle
Helmet Campaign. Am J Dis Child, 144, pp. 727–731.
Bryan-Brown, K., 1995, The effects of children’s traffic club. In Road accidents In Great
Britain (Department of Transport).
Cameron, M.H., Vulcan, A.P., Finch, C.F. et al., 1994, Mandatory bicycle helmet use
following a decade of helmet promotion in Victoria, Australia: An evaluation. Accident
Analysis and Prevention, 26, pp. 325–337.
Carter, N. and Menckel, E., 1985, Near accident reporting: A review of Swedish
Research. Journal of Occupational Accidents, 7, pp. 41–64.
Christian, M.S. and Bullimore, D.W., 1989, Reduction in accident injuries severity in rear
seat passengers using restraints. Injury, 20, pp. 262–264.
Clarke, A. and Walton, W.W., 1979, Effect of safety packaging on aspirin ingestion by
children. Paediatrics, 63, pp. 687–693.
Coleman, P., Munro, J., Nicholl, J., Harper, R., Kent, G. and Wild, D., 1996, The
Effectiveness of Interventions to Prevent Accidental Injury to Young Persons Aged 15–
24 Years: A Review of the Evidence. (Medical Care Research Unit, Sheffield Centre for
Health and Related Research, University of Sheffield).
Colver, A., Hutchinson, P. and Judson, E., 1982, Promoting children’s home safety.
British Medical Journal, 285, pp. 1177–1180.
Day, L.M., Ozanne-Smith, J., Cassell, E. and McGrath, A., 1997, Latrobe Valley Better
Health Project: Evaluation of the injury prevention program 1992–1996. VicHealth.
Accident Research Centre. Report No 114. July.
Department of Trade and Industry, 1991, Child Safety Equipment for Use in the Home.
(London: DTI, Home and Leisure Accident Research).
Downing, C., 1987, Evaluation of the impact and penetration of a children’s traffic club.
Second International Conference on Road Safety, Groningen.
Dowswell, T., Towner, E.M. L., Simpson, G. and Jarvis, S.N., 1996, Preventing
childhood unintentional injuries-what works: A literature review. Injury Prevention, 2,
PP. 140–149.
Effective Health Care. Nuffield Institute for Health, University of Leeds. NHS Centre for
Review and Dissemination, University of York Effective Health Care. Preventing falls
and subsequent Injury in older people. Effective Health Care. April 1996 Vol. 2 No. 4.
Effective Health Care. June 1996, Vol 2, No 5. Nuffield Institute for Health, University
of Leeds, NHS Centre for Reviews and Dissemination, Universtiy of York
Ekman, R., Schelp, L., Welander, G. and Svanström, L., 1997, Can a combination of
local, regional and national information substantially increase bicycle-helmet wearing
and reduce injuries? Experiences from Sweden. Accident Analysis and Prevention, 29,
PP. 321–328.
Erdmann, T.C., Feldman, K.W., Rivara, F.P., Heimbach, D.M., Wall, H.A., 1991, Tap
water burn prevention: The effect of legislation. Paediatrics, 88, pp. 572–577.
Foxcroft, D.R., Lister-Sharp, D. and Lowe, G., 1992, Alcohol misuse prevention for
young people: a systematic review reveals methodological concerns and lack of
reliable evidence of effectiveness. Addiction, 5, pp. 531–537.
Gillespie, W.J., Henry, D.A., O’Connell, D.L. and Robertson, J., 1996, Vitamin D,
Vitamin D analogues and calcium in prevention of fractures involutional and post-
menopausal osteoporosis. Cochrane Database of Systematic Reviews, Issue 3.
Gillespie, L.D., Gillespie, W.J., Cuming, R., Lamb, S.E. and Rowe, B.H., 1997,
Interventions to reduce the incidence of falling in the elderly. (University of York:
NHS Centre for Review and Dissemination).
Gorman, R., Charney, E., Holtzman, N. and Roberts, K., 1985, A successful citywide
smoke detector giveaway program. Paediatrics, 75, pp. 14–18
Graitcer, P L, Kellerman, A.L. and Christoffel T. A review of educational and legislative
strategies to promote bicycle helmets. Injury Prevention 1995; 1:122–9.
Guyer B, Gallagher S, Chang B, Azzara C, Cupples L, Colton T. Prevention of childhood
injuries: Evaluation of the Statewide Childhood Injury Prevention Program (SCIPP)
Am J Public Health 1989; 79:5121–27.
Hailey D, Sampietro-Colom L, Marshall D, Rico R, Granados A, Asua J,SheIdon T.
INAHTA project on the effectiveness of bone density measurement and associated
treatments for prevention of fractures. Statement of findings. Published on behalf of
the International Network of Agencies for Health Technology Assessment by: Alberta
Heritage Foundation for Medical Research, Canada. 1996.
Hornbrook M C, Stevens V J, Wingfield D J, Hollis J F, Greenlick M R, Ory M G.
Preventing falls among community dwelling older persons: Results from a randomised
trial, The Gerontologist 1994;34:16–23.
Jeffs D, Booth D, Calvert E. Local injury information, community participation and
injury reduction. Australian Journal of Public Health 1993; 17:365–72.
Katcher M, Landry G, Shapiro M. Liquid crystal thermometer use in paediatric office
counselling about tap water burn prevention. Paediatrics 1989;83:766–71.
Lauritzen J B, Petersen M M, Lund B. Effect of external hip protectors on hip fractures.
The Lancet 1993;341:11–3.
Lindqvist K, Timpka T, Schelp L, Åhlgren M. The WHO Safe Community program for
injury prevention: evaluation of the impact on injury severity. Accepted. Public Health.
1998).
Lynam D, Mackie A, Davies C. Urban Safety Project: 1. Design and Implementation of
Schemes. Department of Transport, Transport and Road Research Laboratory, 1988.
McLoughlin E, Marchone M, Hanger L, German P, Baker S. Smoke detector legislation:
its effect on owner occupied homes. Am J Public Health 1985;75:852–62.
Menckel E. Intervention and Cooperation. Occupational Health Services and Prevention
of Occupational Injuries in Sweden. Arbete och hälsa. Vetenskaplig skriftserie.
1990:31.
Menckel E, Carter N. The development and evaluation of accident prevention routines: A
case study. Journal of Safety Research 1985; 16:73–82.
Miller T R, Galbraith M. Injury prevention counselling by paediatricians: A benefit-cost
comparison. Paediatrics 1995;96:1–4.
MMWR. United States, 1982–1990. MMWR 1991; vol 40:sid 600–2.
Munro J, Coleman P, Nicholl J, Harper R, Kent G, Wild D. Can we prevent accidental
injury to adolescents: A systematic review of the evidence. Injury Prevention 1995;
1:249–55.
Olds D I, Henderson C R, Kitzman H.Does prenatal and infancy nurse home visitation
have enduring effects on qualities of parental caregiving and child health at 25 to 50
months of life? Paediatrics 1994;93:89–98.
Plautz B, Beck D E, Selmar C, Radetsky M. Modifying the environment: A community
based injury reduction program for elderly residents. American Journal of Preventive
Medicine 1996; 12:33–8.
Ray W A, Taylor J A, Meador K G, Thapa P B, Brown A K, Kajihara H K, Davie C,
Gideon P, Griffin M R. A randomised trial of a consultation service to reduce falls in
nursing homes. JAMA 1997;278:557–62.
Renaud L, Suissa S. Evaluation of the efficacy of simulation games in traffic safety
education of kindergarten children. American Journal of Public Health 1989; 79:307–
9.
Rivara F P, Thompson D C, Thompson R S, et al. The Seattle children’s bicycle helmet
campaign; changes in helmet use and head injury admissions. Paediatrics
1994;93:567–9.
Rubenstein L Z, Robbins A S, Josephson K R, Schulman B L, Osterweil D. The value of
assessing falls in an elderly population. A randomised clinical trial. Annals of Internal
Medicine 1990;113:308–16.
Sackett D L, Haynes R B, Guyatt G H, Tugwell P. Clinical Epidemiology: a basic science
for clinical medicine. 2nd Edition. Boston: Little Brown & Co, 1991.
Schelp L. Svanström L. One year incidence of home accidents in a rural municipality.
Scandinavian Journal of Social Medicine 1986; 14:75–82.
Schelp L, Svanström L.One year incidence of occupational accidents in a rural
municipality. Scandinavian Journal of Social Medicine 1986; 14:197–204.
Schelp L. Community intervention and changes in accident pattern in a rural
municipality. Health Promotion 1987;2:109–25.
Schwarz D F, Grisso J A, Miles C, Holmes J H, Sutton R L. An injury prevention
program in an urban African-American community. Am J Public Health 1993;83:675–
680.
Schwarz D F, Grisso J A, Miles C, Holmes J H, Sutton R L. An injury prevention
program in an urban African-American community. Am J Public Health 1993;83:675–
680.
Springfeldt B. Effects of occupational and safety rules and measures with special regards
to injuries. Advantages of automatically working solutions. The Royal Institute of
Technology . Department of Work Science. Doctoral dissertation 1993.
Springfeldt B.Rollover of tractors—international experiences. Safety Science.
1996;24:95–110.
Svanström L, Ekman R, Schelp L, Lindström A. The Lidköping Accident Prevention
Program—A community approach to preventing childhood injuries in Sweden. Injury
Prevention 1995; 1:169–72.
Svanström L, Ader M, Schelp L, Lindström Å. Preventing femoral fractures among
elderly: The community safety approach. Safety Science 1996;21:239–46.
Svanström L, Schelp L, Ekman R, Lindström Å. Falköping, Sweden, ten years after: still
a safe community? International Journal for Consumer Safety 1996;3: l-7.
Tideiksaai R, Feiner C F, Maby J. Falls prevention: The efficacy of a bed alarm system in
an acute-care setting. Mount Sinai journal of Medicine 1993;60:522–7.
Timpka T, Lindqvist K, Schelp L, Åhlgren M. Community-based injury prevention:
effects on health care utilisation. Submitted 1998. Journal of Epidemiology.
Tinetti M E, Baker D I, MacAvay G, Claus E B, Garrett P, Gottschalk M, Kock M L,
Trainor K, Horwitz R I. A multifactorial intervention to reduce the risk of falling
among elderly people living in the community. The New England Journal of Medicine.
1994;331:821–7.
Walton W W. An evaluation of the poison prevention packaging act. Paediatrics
1982;69:363–70.
Ytterstad B. The Harstad Injury Prevention Study: Hospital-based injury recording and
community based intervention. ISM Skriftserie nr 33. 1995. Institute of Community
Medicine, University of Tromso, Norway: Troms County, Harstad Hospital.
Ytterstad B, Wasmuth H H. The Harstad Injury Prevention Study: Evaluation of hospital-
based injury recording and community-based intervention for traffic injury prevention.
Acid Anal and Prev 1995;27:111–23.
Ytterstad B. The Harstad injury prevention study: Hospital-based injury recording used
for outcome evaluation of community-based prevention of bicyclist and pedestrian
injury. Scan J Prim Health Care 1995; 13:141–9.
Ytterstad B, Sögaard A J. Harstad injury prevention study: prevention of burns in small
children by a community-based intervention. Burns 1995;21:259–66.
15
Community Psychology and Safety: A
Psychospiritual Perspective
Mohamed Seedat
INTRODUCTION
Assalamailaikum. Salaam, the Islamic salutary greeting, literally means peace be upon
you. In Arabic, salaam also denotes harmony, tranquillity, a sense of security, perfection,
accord and connectedness with those around us, contentment and freedom from any
jarring element or experience (A1 Q’uran, Surah 19, Verse 62, Yusuf Ali translation).
This multifaceted peace and associated sense of safety that salaam refers to is founded on
an inner and outer harmony. As action-oriented creatures our external harmony stems
from integrated socio-political and economic systems and structures that uphold the
principles of human rights, dignity, equal opportunity, equity, political democracy and
honesty. Inner harmony is based on a synergistic and balanced relationship between the
body, self (nafs), mind (‘aql), heart (dil) and soul (ruh).
From this perspective safety assumes a spiritual dimension in addition to the objective
and subjective dimensions delineated by safety promotion researchers (Maurice et al.,
1998). Whereas the objective dimension is assessed through behavioural and
environmental indicators, the subjective is discerned by a population’s expressed feelings
of safety and harmony or threat and disharmony. The spiritual dimension is gleaned from
the meaning and meaning systems associated with safety by different population groups.
Various texts on safety promotion echo the idea that injuries are preventable and not
mere unavoidable random acts ordained by destiny or willed by a Supreme Being (Berger
and Mohan, 1996; Laflamme et al., 1999; Maurice et al., 1998). This idea is usually
associated with an expressed concern about how notions of predestination and God’s will
may engender fatalistic thinking, and result in a resigned social acceptance of preventable
problems. These are obviously very legitimate concerns especially if we accept the
wisdom inherent in the shifts from ‘accident prevention’ to ‘safety promotion’. However,
following Svanström’s (1999) view that ‘What people themselves define as safety and
security has proved to be more important to nations, communities and organizations than
what academics have argued about’, I wish to venture an additional explanation.
Accordingly, I will attempt a closer examination of ideas on safety with specific
reference to one South African community and that community’s own meaning system.
Thereafter, I will proceed to describe the linkages between safety and the broader context
of psychological dispossession. More specifically, I shall seek to fulfil three objectives.
Firstly, I will present an account of recent personal experiences and reflections at a
mosque in South Africa, which may serve to illustrate the spiritual dimension of safety
Community psychology and safety 211
and perhaps offer an alternative understanding of the relationship between the divine and
prevention. Secondly, drawing on the ideas of various scholars and additional anecdotal
materials, I will attempt to contextualize so-called fatalistic behaviour within the
subtleties of prolonged oppression. For this purpose I will make special reference to the
legacy of apartheid in South Africa which continues to compromise the emergence of
community and individual safety. Thirdly, and by way of delineating a way forward, I
aim to highlight four research-related challenges that face community safety researchers.
ACCIDENTS: ACTS OF GOD?
if freedom requires the risk of life, then oppression too requires the fear of
physical death
On a recent visit to a mosque that I usually frequent in my childhood town, I saw a large
poster, centrally placed on the notice board, which sought to offer an explanation for
calamities, murder and crime. After a cursory look at the contents (see Box 1), I
dismissed the poster as some religious group’s limited view on the causes of injurious
events. My academic training alerted me to the tone and causal relationship assumed in
the poster. I am sure some of you may, like me, develop a counter intuitive response to
the language on an initial reading. On further examination I noted that the poster was
featured alongside various other messages including calls for relief actions in Turkey,
Albania, Kosovo and Cape Town, and notices highlighting a range of social welfare
services. Interestingly, the notice board also featured a series of messages censoring
uncontrolled harmful practices such as smoking, alcohol consumption and violent
behaviour. These messages also called on the congregation to observe the mosque and
other places of worship as smoke-free zones. I then re-examined all the posters within the
broader meaning system prevalent among the sector of Muslims I am most familiar with
and the Imam’s earlier sermon. I noted that consistent with the views expressed in the
poster all of the messages invoked Q’uranic injunctions and practices, and sayings of
Prophet Muhammed and his companions, including classical scholars of jurisprudence, to
censure risk-taking behaviours and to alternatively encourage community safety in all its
dimensions. That the protagonist of these messages also engaged in various welfare,
health promotion and community development initiatives, suggests that for them and
their followers injuries and ill health are definitely preventable. However, this Muslim
group’s engagement in social welfare activism is punctuated by three ideas. First,
calamities, murders, crimes and other injurious incidents are only preventable in so far as
they are blessed and willed by the Supreme Creator. Second, protection, prevention and
community safety are predicated on individual and collective behaviour, which occur
within a particular prescribed socio-moral framework. Third, safety promotion as
exercised within a broader ethos of community welfare, is a psycho-social-cum-spiritual
endeavour that fosters a dynamic connection to the Supreme Being and a synergistic
relationship between divine and material.
BOX 1
AMITIES, MURDER, CRIME.

WHY? WHY? WHY?
HERES WHY !!!

Abdullah ibn Umar (radhi allahu anhu) narrates. ‘I was among ten people who
came to Nabi (sallallahu alahyi was sallam) from the Muhaajireen.
He said ‘O Muhaajireen! There are five things I ask Allah’s refuge lest you
fall prey to them:
1. SHAMELESSNESS
When shamelessness becomes prevalent and the norm amongst people, they
would become entangled in plague and in such diseases that
never existed during the time of their forefathers.
2. CHEATING IN TRADE (FRAUD)
When a nation cheats in measure (practices fraud) they would be afflicted
by drought, hardships and an oppressive ruler.
3. FAILURE TO PAY ZAKAAT (CHARITY)
When a nation stops paying zakaat they would be deprived of rain. If it were
not for the animals it would never have rained.
4. DISHONOURING ONE’S PROMISES
When people break promises, Allah sets over them enemies from another
nation, who will usurp (steal) their belongings.
5. VERDICTS IN CONFLICT WITH COMMANDS OF ALLAH
When their leaders refrain from giving verdicts according to the commands
of Allah Ta’ala, He subjects them to murder
and civil strife.
******************************************************************
• FORNICATION, ADULTERY AND INTEREST
Rasoolullah (sallallahu alayhi wa sallam) is reported to have said:
‘The people among whom the taking of interest and
indulgence in adultery appears openly have indeed invited
Allah’s wrath and anger upon themselves]. (Targheeb)
• BRIBERY
Rasoolullah (sallallahu alayhi wa sallam) also said: “Fear for others
will engulf the hearts of people among whom there is much bribery”.
• DISOBEDIENCE TO PARENTS
Further Rasoolullah (sallallahu alayhi wa sallam) also said: “Allah postpones
(and defers) the punishment for every sin for as long as He wishes.
But the calamity which follows disobedience to parents
is very soon. Before death occurs one becomes overtaken by it”. (Durr Manthoor)
• TO TURN A BLIND EYE TO VICE!
Rasoolullah (sallallahu alayhi wa sallam) warned: “Through the
iniquitous deeds of a few people Allah does not cause

widespread calamity to descend upon all until such time
that those deeds are performed in their presence and
they, in spite of being capable of stopping it, refrain therefrom.
At such a time the punishment descends upon all, high and low”.
• MAKE AMENDS
* Perform salaat with Jamaat
* Recite Sura Yaseen and AyatuI Qursite regularly
*Make Zikr and Istighfaar abundantly
* Discharge your Zakaat
* Increase your charity
*Do not indulge in extravagance
* Respect your parents
* Promote virtue and prohibit vice
* Earn halaal rizk
*Stop giving bribery
*Read Qunoot Nazilaa in Fajr salaah (with intention of seeking protection
for yourself and all the Muslims in South Africa and the whole Ummah)
Such spiritualized safety promotion behaviour is not unique to Muslims. Everyday, all
over the world, individuals and groups representing major religious systems-Islam,
Christianity, Buddhism, Hinduism and Judaism—assume various precautionary measures
to protect themselves from injury and a range of other violations. Yet they
conscientiously invoke the blessings of the Supreme Being through prayer and a variety
of related rituals. Some people place stickers containing prayers in their cars, homes,
work places and recreational facilities. Others carry various forms of prayers on their
persons. Some people engage in prayer for safety before taking a long trip. These rituals
and prayers, as integral features of a larger spiritually inspired belief system, provide
meaning and solace especially when precautionary behaviours seem so insignificant in
the presence of large-scale threats and structural obstacles generated by oppressive and
exploitative socio-economic arrangements of society (see Box 2).
prolonged oppression in the form of discrimination places children and women, in

particular, at great risk for mental health problems and disorders
BOX 2
FOR YOUR…
SAFETY AND SECURITY

In these times when mugging, robberies, murder and a string of
crimes are committed daily, people have become extremely fearful of

their safety and many are even thinking of migrating.
On his recent visit to Durban, Faqihul Ummah, Hadhrat Mufti
about some Duas which could be read in these troubled times,
Mahmoodul Hassan Saheb Gangohi (Madda Zilluhu) was asked
Among other Duas, Hadhrat Mufti Saheb has prescribed reading
the following two Duas as often as possible.
‘O Allah, enable us to love Imaan (faith) and decorate our hearts
with it and lead us to hate disbelief, immorality and sin and include us
among the righteous ones’.
‘O Allah, cause us to die as Muslims and join us with the pious
people so that we may not be among those who are put to shame or
tested.
It is safe to assume that many of the world’s people, as guided by divine scriptures or
divinely inspired text, pursue safety in a manner that is punctuated by consciousness of
the divine, unlike the academic tendency to dichotomize the secular and religious. Once
we begin focusing our attentions on the study of risk and, alternatively, safety promoting
beliefs and behaviours within specific meaning systems and normative structures we are
also encouraged to take a contextualized approach. Such an approach will require us to
examine the subtleties of psychological dispossession that denies the marginalized, in
particular, the right to safety and security and perpetuates fatalistic patterns of behaviour
for ideologically suspect reasons.
PSYCHOLOGICAL DISPOSSESSION
In preparing for this section of the paper, I considered and debated the various materials
that I could use to convey my thoughts on psychological dispossession. I specifically
wanted material that could humanize abstract ideas about oppression. So 1 decided to
search for the manifestations of subtle psychological dispossession in present day South
Africa. Within hours of my search, a series of very disturbing personal encounters
prompted me to reflect on the social and human phenomenon of oppressive subservience
that remains common place in South Africa and perhaps in many other parts of the
former colonized and enslaved world. While on my usual morning shopping chores at a
mall east of Johannesburg, I noticed an elderly woman moving from car to car in the
parking lot entreating motorists to purchase grass brooms from her. Before she
approached me directly I paid very little attention to her physical demeanour and speech.
Within seconds of the elderly lady addressing me I felt embarrassed, uneasy and
ashamed. The language of baaskap (subservience) punctuated the lady’s demeanour and
sales manner. She addressed me in the language of apartheid servitude. In silence my
family and I drove out of the mall. Lost in our own thoughts we drove along for at least
fifteen minutes when once again our thoughts were jolted by a group of street children at
a very busy intersection. Like the elderly lady they too addressed everyone with
deference and subservience, hoping that among the generous and guilt-ridden motorists
they approached, someone would pass on food or money. Realizing that the same culture
of subservience ran across the age spectrum, I wondered whether I, like so many other
South Africans, had become anaesthetized and blind to the ubiquitous dehumanization
produced by apartheid. I could hear the poetic echoes of an internationally renown South
African poet, Don Omarrudin Mattera, written thirty years ago, entitled ‘For a Cent’.
Mattera (1983) writes:
Each morning
Corner of prichard and joubert,
Leaning on a greasy crutch
Near a pavement dustbin
An old man begs
Not expecting much.
His spectacles are cracked and dirty

He does not see my black hand
Drop a warm cent into his scurvy palm
But instinctively he mutters:
Thank you my Baas!
Strange
That for a cent
A man can call his brother. Baas!
A contrasting and somewhat more affirming phenomenon is contained in the story of

Thabo, a man from Transkei whom I met a year ago. In explaining how he came to seek
out our Centre for Peace Action, in the south of Johannesburg, he tells the story of police
harassment, personal injuries and a series of other harsh experiences at the gold mines
west of Johannesburg. In describing the consequences of the adversity he faced before
arriving at our Centre he talks about the experience of psycho-spiritual death. In his own
words: ‘I was dead inside and just a body’. However, a deep connection to family, a
profound sense of morality and a love for responsibility enabled Thabo to transcend his
self-reported transient homicidal, suicidal and criminal thoughts. He was able to muster
up sufficient inner psychological and outer social resources to re-link with his heart and
spiritual essence and connect with a community-based safety promotion service.
courtesy, affection and concern is reserved for the oppressor and his/her
representative institutions while the other is dehumanized and inferiorized
through a plethora of ideas and a dubious discourse that contain no empirical
truth
Thabo’s resourcefulness, ability to bounce back and seek help with dignity as a contrast
to the many occurrences of subservience, timidity and psychological dispossession still
evident in South Africa and perhaps in many other parts of the former colonized world,
warrants a nuanced explanation. One reality has become extremely clear in seeking this
explanation; even with the repealing of apartheid laws and the introduction of a new
constitution, in some ways these acts have served to merely tear down what we now
realize was only the scaffolding of apartheid leaving the building behind. Despite the
demise of institutionalized political apartheid its mutative psychosocial oppressive
presence continues to haunt many people in their dreams and waking life.
Scholars of community psychology, focusing on the dynamics and subtleties of
psychological oppression, provide some insight into the collective and individual impact
of oppression and their implications for the development of safety as a human right. In
explicating the master-slave dialectic Hussein Bulhan (1985) asserts, ‘If freedom requires
the risk of life, then oppression too requires the fear of physical death’. Apartheid as a
crude institutionalization of racism and economic exploitation could therefore not be
implemented and maintained without the threat and exercise of violence. The fear of
physical death takes root when armed resistance is defeated, the mind’s reasoning is
subverted through the perverted logic that justifies and rationalizes oppression, and the
heart’s links to goodness is denied. At best courtesy, affection and concern is reserved for
the oppressor and his/ her representative institutions while the other is dehumanized and
inferiorized through a plethora of ideas and a dubious discourse that contain no empirical
truth. This fear of physical death ‘crystallizes’ as a conflict between individual physical
survival and the more profound wish for happiness and contentment that may be achieved
through a balance of the various constituent elements of human nature, expressed in the
greeting of salaam. Fear pits the baser self with all of its rage, selfish, instinctual and
aggressive forces against the human self as located in reason, spirituality and the drive
towards inner good.
Fear of physical death diminishes the capacity for struggle and search for safe
behaviour as individual survival and the material qualities of life assume a higher
premium than the preservation of collective history, community ideals and cohesion, and
the search for psycho-spiritual development and safety. The fear of physical death,
together with the severe restrictions placed on oppressed people’s time, space, thoughts,
actions and nature, may well manifest as a repertoire of fatalistic, subservient and ego-
dystonic behaviours.
Another scholar of oppression, James C.Scott (1990), drawing on examples from
literature, history and the politics of culture from around the world, discerns both public
and hidden transcripts in the master-slave relationship. Public transcripts, including acts
of deference, silence, ritualistic subservience and fatalistic behaviours, are merely a mask
designed to produce conformity in line with what the dominant group may want things to
appear in public.
The public transcript is part and parcel of the dialectic of disguise and surveillance that
punctuates relations between the weak and the strong; it is integral to the art of resistance.
The hidden transcripts unfold outside the direct surveillance of the dominant group and
includes speeches, behaviours, practices, folk tales, rumours, gossip, songs, jokes and
gestures that ‘confirm, contradict, or inflect what appears in the public transcript’ (Scott,
1990). The hidden transcripts in Scott’s view provide a critique of the oppressor behind
the cloak of secrecy.
Following Bulhan (1995) and Scott (1990), it may be held that the dominated have a
vested interest in maintaining a public aura that reinforces hegemonic appearances. Such
adaptation and physical survival tactics may well include a personal toll that may
manifest as a higher risk for psychopathology, somatic afflictions, sudden eruptions of
rage, fatalism and various forms of risky behaviours. Social coercion that produces an
excessive focus on physical survival undermines the search for internal harmony,
alienates the oppressed from their true nature and the spiritualized quest for safety. It
would be presumptuous and arrogant to assume that fear and psycho-social and spiritual
death characterize the elderly lady and the street children, whom I referred to earlier.
However, it is safe to assert that their public displays of public subservience—whether
conscious or unconscious—must produce internal psychic dissonance that may manifest
as passive aggression, repressed rage and anger and maladaptive coping mechanisms.
Evidence from the United States and South Africa show that prolonged oppression in the
form of discrimination places children and women, in particular, at great risk for mental
health problems and disorders (Pillay and Lockhat, 1999; Bulhan, 1985).
The impact of oppression does not escape the historical and current oppressor. For
instance, in order for apartheid to have been institutionalized in the way it was, it required
the production of depravity, torturous behaviour, the deracination of compassion, and
individuals who could serve as agents of fear and terror. These agents of terror and fear
underwent a process of dehumanization themselves, who through mechanisms of consent
creation, came to assume and justify a racial superiority in exchange for economic, social,
and political privilege. Ironically these agents of fear historically and currently continue
to be motivated by fear. The impact of ‘swart gevaar’, which refers to the idea of black
peril, should not be underestimated. Whereas colonial and apartheid oppression produced
notions of bestiality, uncontrolled impulses and intellectual immaturity to inferiorize the
colonized person, it also conjured up fears of sexual occupation and geographical
expulsion among the dominant sector who then turned inward to create a lager mentality.
The lager mentality served to uphold fears and irrational thoughts about blacks and it also
subverted the humanity of the oppressor; values and morality were perverted to the extent
that compassion, care, mercy and kindness tended to be selfishly reserved for the flesh of
your flesh, your fellow oppressor. Today, the black peril has assumed a different form
and content, sometimes subtle and sometimes obvious. The interrelated prejudiced
notions that blacks are incapable of governance, corrupt and criminal by nature, given to
living in squalid and overcrowded conditions even when they move into the cities and
formerly white suburbs perpetuates fear and propels a physical exodus wherever and
whenever blacks are perceived as a majority. The desire to return to a militaristic order
and maintain relics of the apartheid era perhaps continues to manifest itself at the
interpersonal, economic and social levels. The nonchalant use of derogatory terms to
refer to African gardeners and domestic helpers, and the inadvertent tendency to reinforce
the language of subservience when interacting with the most marginalized and
dispossessed does little to eradicate the master-slave relationship. At the socio-economic
level a cursory survey of business enterprises and places of leisure and entertainment,
including public places for dining, will reveal old patterns of socio-political domination
and the emergence of a classist society.
the fear of physical death, together with the severe restrictions placed on
oppressed people’s time, space, thoughts, actions and nature, may well manifest as
a repertoire of fatalistic, subservient and ego-dystonic behaviours
Following Bulhan (1985) and Scott (1990), it maybe argued that for healing and a sense
of safety to occur among the oppressed and oppressor the slumbering slave and haughty
master within must be respectively ejected. The ejection process, a psychological one, is
fraught with many struggles and conflicts. It therefore requires individuals and collectives
to re-connect with their humanity through a concerted process of moral and psycho-
ethical reconstruction. The dismantling of oppressive political systems is only the first
step towards psychological healing and safety promotion, especially within fledgling
nation-states struggling to engender democratic forms of governance and social relations.
The establishment of safety within a nation-state requires economic equity, social
egalitarianism, and moral and spiritual integrity. Most importantly, the historically
oppressed who came to assume positions of authority and acquire new found wealth must
guard against reproducing the offences and ideology of the oppressor. Otherwise safety
promotion will remain an ideal.
FUTURE DIRECTIONS: IMPLICATIONS AND RESEARCH

CHALLENGES
In summary, I have employed two key ideas to frame the substantive arguments in my
presentation. First, the definition of safety needs be extended to include a spiritual
dimension, where relevant, so as to give due consideration to the spiritually minded for
whom safety promotion is possible in so far as the Supreme Creator may ultimately will
it. Second, efforts programmed to promote safety as a human right must be
contextualized and critically considered within oppressive ideology’s mutative presence
in this era of increasing globalization. Such a contextualization also enables an alternative
analysis that views fatalism as symptomatic of either a profound sense of psychological
dispossession or part of a public façade maintained by dispossessed groups in the interest
of limited psychic survival. Safety efforts, especially in the South and marginalized
sectors of the North, must today grapple with the globalization of materialism, greed,
money politics; political expediency, and the commercialization of values and identities.
Just as economic globalization restricts the nation-states ability to develop equitable
economic structures and policies, the impact of psychological dispossession undermines
the promotion of safety. Therefore, our research efforts need to proceed beyond our
present evidence lead search for risk and determinants. So what are the intellectual
challenges facing the safety promotion movement? I endeavour to describe four
interrelated research challenges.
The first intellectual challenge centres on the need to re-examine current dominant
notions of human nature within the social sciences and the associated helping
professions. The historical secularization of science and knowledge production has
tended to create a situation of delimited appropriation within the basic and applied
sciences. Delimited appropriation refers to the intellectual tendency to limit specific ideas
and discourse to what is considered their relevant and appropriate place. Hence for
instance, ideas about spirituality are to be confined to religious institutions and politics to
parliament following the mythical logic that science and the applied professions are
neutral, value free and non-prescriptive. The consequent secularized and fragmented
definitions of human nature ignore the transcendental and spiritual nature of the
individual. The psycho-spiritual conception of human nature detailed by many Islamic
scholars, and alluded to earlier in this paper, offers one an alternative to the current
secular notions of human essence (Haeri, 1998; Laleh, 1996; Mohamed, 1996; Shafii,
1988). For example, following a close study of the Qu’ran, classical and contemporary
Islamic scholars have discerned that the human personality structure comprises of the
physical body, the self (nafs), the mind (‘aql), the heart (dil) and the soul (ruh) . Whereas
the body serves as the source of physical nourishment and mechanical energy, the self
(nafs) includes the depraved or commanding animal self, which is equated to the Id, and
the human/commanding self. Thus the self can be fearful, impulsive, spiteful, and selfish
or in contrast it can be generous, courageous, peaceful and helpful. The mind (‘aql) is the
site of practical and abstract intellect, rationality and judgement. Some scholars conclude
that the mind (‘aql) is the ‘cornerstone of civilization, cultural reliability and
tradition’ (Haeri, 1998; Shafii, 1988). The heart (dil) as the organ of cognition, is the
home of emotions and intuitive knowledge. The heart (dil) as the synergistic force
harnesses the baser self and steers the energies of the body and mind (‘aql) towards the
inclination for goodness, a quality that humans are biologically programmed for. The
spirit or Divine spark energizes and provides life to the self (nafs) and the physical body.
It is the reflection point for the self (nafs).
When the heart (‘aql) is free of defects such as hatred, suspicion and lust, it will radiate
with the source of enlightenment vis-à-vis the soul (uh) and the inclination for goodness.
Many such alternative ways of looking at healing and understanding the workings of
humans remain unexamined within mainstream thinking, theoretical formulations and
prevention procedures.
the dismantling of oppressive political systems is only the first step towards
psychological healing and safety promotion
The second intellectual challenge relates to forging operational methodologies that

support the contextualization of safety promotion research and applied initiatives within
universally acceptable human rights programmes. Due to the intimate relationship
between safety and human rights we are required to question the kind of society and
associated community safety initiatives we wish to advance. Safety promotion research
within a movement that strives for the creation of a tolerant, egalitarian society free of
social oppression, economic exploitation, and foreign and local political domination
presents theoretical and research imperatives that require innovation and the willingness
to transcend the usual boundaries of the empirical social and medical sciences. When
operating from a human rights framework the injury epidemiologists will be challenged
to forge methods for the operationalizing and incorporation of oppressive systems and the
ensuing master-slave dialectic into considerations of risks and determinants. This
challenge requires precise methods for how and where we begin operationalizing
exploitative ideologies and systems as risk inducing phenomena. If and once we accept
that safety embraces a spiritual dimension, in addition to the objective and subjective
dimensions, we may be able to discern a third challenge as well. This third challenge
relates to developing replicable strategies for forging linkages with existing collective
modes, rituals and spaces of healing and safety that offer solace and contentment
alongside all of the current well known individual, group and family oriented modalities
of safety promotion. Millions of citizens across various nation-states regularly congregate
to participate in various politically and/or spiritually constructed nodes of healing, safety,
faith and psycho-spiritual cleansing and development. For instance, the weekly Christian
Mass, Jewish Sabbath, Muslim Ju’mah (Friday) Prayer and Buddhist Meditation are
occasions for profound reflection, psychological safety, healing, hope, compassion and
spiritual bonding to others, nature and the Eternal Reality. Similarly, the many public
holidays such as the South African hero’s day, youth day, women’s day and heritage day;
are commemorative occasions for reflecting on the atrocities and struggles of the past and
honouring the fallen heroes and heroines.
These are intensely emotionally and spiritually laden days intended to engender a
recommitment to the human rights struggles of the present and future in the search for
inner and outer harmony and security. Apart from their cathartic benefits, these religious
and spiritualized public holidays contain elements of psycho-ethical and moral
reconstruction so vital and essential for the community safety advocate to understand and
link with. How do we as public health advocates make the connection to such
spiritualized and therapeutic spaces in our endeavours to promote community safety
within a human rights framework?
The fourth and final challenge deals with the question of developing safety promotion
into a psycho-ethical practice. Approaches and practices that integrate socio-spiritual
values based on justice, equity, wisdom, temperance, a balanced regard for self and
others are the greatest source of safety, peace and resilience. Such liberatory approaches
and practices may enable the curtailment of the baser instinct that continuously demands,
‘I want what I want when I want it’ irrespective of the social, psychological, emotional,
physical and spiritual costs. This challenge requires the public health practitioner to
reflect on her/his own values and beliefs and engage in self-dialogue as part of the safety
promotion solution. It also means wrestling with oppressive systems in their mutated
form in our own collective minds as people and our hearts as prevention activists. In
concrete terms it requires us to debate and develop an ethical code of conduct that
informs our safety promotion work within a human rights framework.
If we are to translate these challenges into well-defined basic and applied research
projects we may require substantial intellectual, political and financial resources. Most
significantly, we require open-minded modes of enquiry, which are underlined by the
quest for innovative explanatory paradigms; the quest for innovative paradigms may be
greatly strengthened if safety promotion researchers consider employing multiple ways of
knowing and investigating safety phenomena. Because logicoempiricist traditions in
western ethnosciences have negated humanistic, spiritual, artistic and intuitive ways of
knowing, safety promotion researchers do not need to be hostage to the logic that
primarily stresses measurement and control for the creation of predictive explicatory
models of human behaviour. The search for liberatory discourse requires openness,
freedom of enquiry, assessment of all information obtained by the individual, and the
incorporation of multiple methods of research into intellectual endeavours. Religion,
spiritual expression, morality, ethics, literature, oral history and intuitive ways of
knowing can all make substantial contributions towards the forging of emancipatory
paradigms and safety promotion methodology (Asante, 1990; Gordon, 1988; Stanfield,
1985).
These challenges imply that the safety promotion movement and its advocates embrace
multiple roles so as to serve as basic and applied researchers, programme developers and
human rights activists at the macro and micro levels of society.
In closing, I hope that this paper will ultimately serve as a modest addition to the
extensive body of knowledge on the conceptual and operational aspects of safety and
safety promotion (Maurice et al., 1998). Whatever the criticisms levelled at this paper,
the limitations are my own and the products of fallible human reasoning. But in the words
of one of my favourite philosophers, Ali Shariati: ‘Knowledge means struggling,
thinking, making observations and adjusting one’s views to reach something. It is not that
we should not say nothing until we reach ‘the knowledge of certainty’ and the ‘truth of
certainty’ for it is impossible to reach it all at once…We should continue saying and
thinking as we reach towards it’ (Sharati, 1974).
REFERENCES
Al Quran. English Translation. Yusuf Ali. International Islamic Publishers.

Asante, M.K., 1990, Kemet, Afrocentricity and Knowledge. (New Jersey: African World
Press, Inc.).
Berger, L.R. and Mohan, D., 1996, Injury Control. A Global View. (New Delhi: Oxford
University Press).
Bulhan, H.A., 1985, Frantz Fanon and the Psychology of Oppression. (New York:
Plenum Press).
Gordon, E.W., 1988, Coping with communicentric bias in knowledge production through
educational research. Paper presented at the American Educational Research
Association Annual Meeting. New Orleans, Louisiana.
Haeri, F., 1998, Ripples of Light. (London: Zahra Publications).
Laflamme, L., Svanström, L. and Schelp, L., 1999, Safety Promotion Research.
(Stockholm: Karolinska Institutet).
Laleh, B., 1996, Sufi Women of America. Angels in the Making. (Chicago: Institute of
Psychoethics and Guidance).
Mattera, D.C., 1983, Azanian Love Song. (Johannesburg: Skotaville Publishers).
Maurice, P. et al., 1998, Safety and Safety Promotion: Conceptual and operational
Aspects. (Quebec: WHO Collaborating Centre for Safety Promotion and Injury
Prevention).
Mohamed, Y., 1996, Fitra. The Islamic Conception of human Nature. (London: Ta-Ha
Publishers).
Pillay, A. and Lockhat, R., in press, Models of service delivery for children. In
Community Psychology in the South. Theory, Methods and Practice. Edited by:
M.Seedat, Duncan, N. and S.Lazarus, International Thompson Publishers.
Scott, J.C., 1990, Domination and the Arts of Resistance. Hidden Transcripts. (New
Haven: Yale University Press).
Shafii, M., 1988, Freedom from the Self. Sufism, Meditation and Psychotherapy. (New
York: Human Sciences Press Inc.).
Shariati, A., 1974, Art. Awaiting the Saviour. (Houston: Free Islamic Literature Inc.).
Stanfield, J.H., 1985, The ethnocentric bias of social science knowledge production.
Review of Research in Education, 12, pp. 320–331.
Svanström, L., 1999, Foreword. In Safety Promotion Research. Edited by: Lucie
Laflamme, Lief Svanström and Lothar Schelp, pp. 6–7, (Stockholm: Karolinska
Institutet).
16
Labour and Safety: Science, Technology and
Society
A.K.(Dunu) Roy
PRINCIPLES OF SAFETY
On the night of 2 August 1999, two express passenger trains rammed into each other near
Gaisal in Assam, India. Both trains were packed with sleeping passengers and over 300
were reported to have died in the crash. It took a week to clear the debris and conduct an
actual body count. Soon after the accident the Inspector of Safety of the Indian Railways
undertook a detailed investigation, while almost three weeks later the Government
announced the setting up of an Enquiry Commission headed by a retired judge of the
Supreme Court. However, long before any of these bodies could submit their reports and
immediately after the accident, several ‘authorities’ announced that the crash was due to
the ‘negligence’ of the dead driver of one of the trains. Thus, one man amongst several
hundreds was singled out not only for his own death but that of many others.
The ideology of the ‘careless worker’, still being resurrected almost a century after the
beginnings of the struggle to combat the idea of the accident-prone worker. In fact, the
findings of the Safety Inspector, made public several weeks later, clearly indicated that
the stretch of track on which the accident had taken place was known to be dangerous
with several previous reports of malfunctioning signals and faulty safety procedures. The
railway management was severely indicted for not having taken adequate safeguards
earlier and for having ignored repeated warnings. This was not the first instance of such
reports having been made public. But the myth of the careless worker persists in the
public mind. A similar accident near London, about three months later, elicited similar
public responses.
Roots of the myth go far back in time. Towards the end of the nineteenth century,
industry associations in USA had developed the compensation-safety apparatus to deal
with the problem of industrial accidents. This apparatus emphasized compensation over
prevention and safety over health. Thus, in 1890, the steel industry was prosecuting
workers for disobeying safety orders rather than looking at causes of equipment failure.
In 1910, US Steel devised the Voluntary Accident Relief Plan for paying fixed amounts
for job related injuries causing death or disability, against a commitment by workers not
to sue for damages. Such an approach obviously focused on mechanical, electrical, and
fire hazards. However, the injury caused by chemicals in the workplace could not be
ignored for long. The first list of standards of Maximum Allowable Concentrations
(MACs) for hazardous airborne contaminants was made available only in 1946, after
World War II, even though these MACs were not legally enforceable. It took three more
decades into the seventies for a new activism in health and safety to take root in US
industry, which squarely blamed worker injuries and diseases on corporations’
unwillingness to spend and on the drive to speed up production (Berman, 1978).
In the eighties progressive labour unions tried to move beyond these simple arguments.
For instance, a tripartite committee on occupational safety, health and welfare established
in 1984 by the South Australian government, enunciated seven principles of health and
safety (Matthews, 1985):
• The toll of injury and disease can be reduced by adopting preventive measures,
which have to be balanced against clinical measures.
• A preventive strategy needs to focus on underlying unhealthy work systems,
and not on making workers ‘aware’.
• A basic level of safety has to be legally imposed on all industries to place all
of them in the same competitive position.
• Standards of health and safety cannot be determined a priori, but only as a
social process of evaluation involving workers, employers and government.
• Legal rights and powers have to be conferred upon workers’ representatives
to enable them to participate in this social process.
• Fora have to be provided for workers and employers to resolve their
conflicts over health and safety measures and standards.
• Preventive measures need to be complemented by provisions for the care of
victims of occupational injuries and disease.
as society has changed, new production technologies have emerged giving rise to
new hazards and it is possible that the established principles of health and safety
at work may not be adequate
Accordingly, the South Australian government signed an accord with the unions
recognizing the rights and powers of workers’ representatives. Some of these included
the right to participate in inspections, receive information on safety and health, prevent
continuation of unsafe work, initiate prosecutions, and function on paid time.
In this context, it would be appropriate to recall that the standards for hazardous
chemicals are particularly important since they directly impact on the nature and extent of
occupational diseases. Toxicological data on chemicals is crucial for setting of such
standards. There are a variety of techniques for obtaining such toxicological data which
include Short-term LD50, Short-term Irritancy, Sub-acute, Chronic, Short term
Mutagenic, Reproductive, and Behavioural tests. Of these tests, the LD50 is irrational
because it specifies the dose at which 50 per cent of the test animals die; while the Short-
term Irritancy or Draize test is exceptionally cruel. The Chronic and Reproductive tests
are superior indicators of toxicity over the long term but they require large investments of
money and time to complete. The Behavioural test is excellent for understanding effects
on the nervous system; while the Short-term Mutagenic or Ames test is rapid, cheap, and
valuable (Mathews, 1985). World over, industry and regulatory systems depend heavily
on the LD50 as a marker of toxicity—as embodied in the Threshold Limit Values set by
the American Conference of Governmental Industrial Hygienists, an unofficial body of
Labour and safety 225
‘experts’. Hence, the participation of workers in evaluating and setting standards is
crucial to the development of health and safety programmes.
In the nineties environmental activists in USA added another dimension to the control
of workplace hazards. They began with an emphasis on pollution prevention as opposed
to pollution control and evolved a Toxics Bill of Rights. This contained the Right to
Know, the Right to Cleanup, the Right to Compensation, the Right to Law Enforcement,
the Right to Participate, Inspect, and Negotiate, the Right to Prevention, and the Right to
Freedom from Toxics (Cohen, 1990). However, they also recognized that workers had
led the way for citizens in the fight for a clean environment. Thus, technological solutions
to pollution included the evolution of substitutes, the production of less waste-intensive
products, waste reduction, and recycling linked to job creation and the reduction of
medical costs. Community groups even advanced a set of principles for occupational
safety and health for chemical hazards (Cohen, 1990):
• Maximum levels above which worker exposure is prohibited.
• Exposure levels that trigger actions such as medical surveillance of workers.
• Comprehensive labelling.
• Protective equipment and control procedures.
• Access by employees to company records and to chemical hazard information.
• Employee training on safe handling of chemicals.
Thus evolution of principles regarding the safety of workers has marked a long trajectory.
From a simple principle of payment of minimal compensation for damage to life and
limb, workers’ struggles have forced employers and government to accept that it is work
systems which are responsible for the creation of hazards. Thus a set of principles have
emerged which not only speak of compensation and treatment but also of the need to
reorganize work so as to prevent injury and disease. These principles have been further
advanced by the activity of environmentalists who have linked hazards at the workplace
to pollution outside and have built a bridge between workers and citizens to influence the
manner in which production and technological decisions are taken. However, as society
has changed, new production technologies have emerged giving rise to new hazards and
it is possible that the established principles of health and safety at work may not be
adequate. These rapid changes in contemporary society are discussed in the next section.
CHANGING SOCIETIES
Changes in society are the basis on which the foundations of labour safety are
established. As Berman (1978), commenting on the situation in USA, notes, ‘In the first
two decades of this century, monopoly corporations such as US Steel responded to the
movement around occupational safety and health by setting up a business-controlled
“compensation-safety apparatus”, a stalling operation which, by appearing to be doing
something, withheld the issue of working conditions from the public agenda until the late
1960s…In fact, the workers’ compensation system stabilized compensation costs to
employers at 1 percent of payroll by almost totally ignoring the problems of long-term
disability, occupational disease, and worker rehabilitation. By setting up a closed
compensation bureaucracy, companies avoided the costs and embarrassments of jury
trials.’ In other words, investments in safety and compensation were determined by costs,
and labour agitation on these issues also meant increased costs to the company.
It was after the massive (and disastrous) strike by steelworkers in 1902 that the
steelmakers started reorganizing the industry. It should be noted that it was a time when
steel was in massive demand and provided the basis for industrialization and transport.
Physical labour was replaced by huge electrical cranes and intricate intra-plant railway
systems. Gigantic new furnaces and heaters were built, while formerly separate
operations were integrated into single complexes. Production was carried out under
military-style chains of command and skilled workers were demoted to semi-skilled
status. Union activists were fired and socialists blacklisted or driven underground. The
dirtiest and most unpleasant jobs were filled by blacks and immigrants. The pace of work
was speeded up beyond belief and the effect on working conditions was disastrous
(Herman, 1978). The entire effort was geared to produce as much steel as possible as
cheaply as possible.
there is a direct relationship between the mechanisms and policies of international

financial agencies and the worsening of working conditions in areas most
compliant to models of development based on market criteria
Other industrial sectors followed the lead given by the steel industry. One of the most
profitable was the chemicals industry and the two world wars witnessed a huge expansion
in the number of chemicals in the workplace and the environment. Petroleum,
petrochemicals, and automobiles began to replace steel and railroad as the premier areas
for investment. Computing machines and systems transformed the nature of work in
offices. The profits from the incredible productivity increases were gradually cornered by
fewer and fewer large corporations. Independent farmers and small businesses were
marginalized by the alliance of big business with big banks. Such was the magnitude of
the industrial explosion that it was estimated in 1988 that between 20 to 400 billion
pounds of toxic substances were being emitted to air, land, and water. The Federal
government was forced to set up 1,219 Superfund sites, costing more than $100 billion
over some decades, to deal with the waste. These disposal sites did not include those of
the States and the military, and the budgets did not include the industrial and social costs
(Cohen, 1990). Toxic-related jobs were declining but the profits from and production of
toxics were rising. Toxic production had become a global enterprise for US Corporations
(PHI, 1993).
Trends in the American economy were similar to those that emerged in Europe and,
later, in the rest of the world. For instance, in 1985, Intel dismissed 6000 employees,
abandoned the memory business to the Japanese, and wagered its future on the 386
microprocessor (SWOP, 1994). A few years later Intel abandoned everything for the
Pentium even though senior managers knew that within six years or so the Pentium
would be outdated. The rapidly changing nature of the industry, therefore, made hiring
and firing the norm. The industry preferred to hire small, female workers of colour who
were seen to be more passive, obedient, and easier to manage. Hence, unionization was
strictly controlled in the microelectronics industry and the labour force was purposely
divided by race, nationality, and language. Such a work organization not only prompted
the physical movement of the industry to the less developed nations, where wages were
lower and health issues of less concern, but also submerged the health and safety issues,
particularly those related to exposure to glycol ether used as solvents.
As the Permanent People’s Tribunal observed in its 1988 verdict delivered at Berlin,
there is a direct relationship between the mechanisms and policies of international
financial agencies and the worsening of working conditions in areas most compliant to
models of development based on market criteria (PPT, 1992). The Tribunal was receiving
evidence on industrial and environmental hazards and the violation of human rights at its
session at Bhopal in 1992, where it heard several cases from different parts of the world.
Among these were the case of spraying Agent Orange in Vietnam to destroy 14 per cent
of the South’s woody vegetative cover, and the large-scale mining by Benguet
Corporation in the Phillippines. Also presented at the session were studies of the effect of
methyl mercury discharges in Japan, the mining by Asian Rare Earths in Malaysia, the
occupational health of textile workers in South Korea and Sri Lanka, the occurrence of
silicosis in workers in Thailand and India, and the incidence of occupational deaths in
factories in Taiwan and the People’s Republic of China.
Even in India, there were parallel patterns. A review of industrial development up to
the late eighties (Qadeer, 1989) revealed that there had been significant diversification,
leading to an increase in the industrial work force, particularly in the unorganized sector.
The new industries had a faster growth rate and were also more hazardous, as shown by
accident records. Thus, the risk to workers had increased because of new and unfamiliar
technologies at the workplace. The majority of the labour force came from the
impoverished migrating peasantry, and those with a better bargaining power, that is, from
the middle and upper castes, dominated the jobs. Women and children had a lower status
at work, while gender, caste, and regional identities were used to divide the workers.
In the nineties there was an impressive growth in the Indian plastics industry, higher
than anywhere else in the world (MOEF, 1997). The use of plastics was spreading in
virtually all high-growth sectors of the economy—in infrastructure, agriculture, building
and construction, telecommunications, consumer goods, packaging, health and medicare.
From 1.88 million tonnes in 1995–96, the demand was expected to cross 4 million tonnes
by 2001–02, even though India’s materials recycling rate of 60 per cent is the highest in
the world. The PVC industry, in particular, was set to vastly expand production as
plastics were replacing traditional materials and creating new markets. For this purpose,
large-scale chlorine-based units were being built on greenfield sites in Gujarat. This was
part of the rapid, unplanned growth of the chemical manufacturing sector in India. The
industry planned to deal with the increased waste crisis by promoting recycling, but this
would only postpone eventual disposal in landfills and incinerators (Greenpeace, 1996).
What is of particular interest in India is the attempt to amend various laws to make the
economy more ‘compatible’ with the global one (IFT, 1999). The proposed changes in
the Factories Act, Trade Union Act, Contract Workers (Abolition and Regulation) Act,
and Industrial Disputes Act ‘indicate the confluence of interests of the Indian capitalist
class and the Indian State’. The convergence of interests is further illustrated by the
report of the Task Force of Indian Trade and Industry, the Ninth Plan document, and the
mandate of the Second National Commission on Labour: all arguing for making the
labour market more ‘flexible’. This is further supported by the recommendations of the
Task Force on Administrative and Legal Simplifications—a task force exclusively
constituted from the ranks of the big Indian industrial houses—as well as those of the
Task Force on Knowledge-Based Industries. Roy, gives examples from New Zealand,
Japan, South Korea, and India, to argue that the changes in labour laws taking place in
Asia are meant to minimize the legal protection of employment and conditions of work as
well as to reduce the legal right and power of unions to bargain for wages and conditions
of employment and work IFT, 1999).
All these changes in society and technology clearly indicate that workers are beginning
to face new dangers which need not necessarily fit into the traditional classification of
occupational hazards as defined earlier. We shall now briefly consider what these new
hazards are.
NEW HAZARDS
workers often do not know what it is that they are handling and what are its toxic
effects. Many effects are discovered only after long periods of exposure and the
actual incidence of disease in workers or animals
One of the major hazards is the growth of chemicals in the workplace. According to
various estimates, there maybe as many as 100,000 different branded chemicals being
used by workers and over 3,000 new chemicals are introduced every year. Only a fraction
of these have been adequately identified and studied for their health effects. For instance,
Threshold Level Values have been published for only 500 chemicals, while some
provisions have been codified for their use and storage, packaging, labelling, and
transport. There is no legal control for the introduction of new chemicals. Nor is there a
requirement for disclosure of trade names (Mathews, 1985). Hence, workers often do not
know what it is that they are handling and what are its toxic effects. Many effects are
discovered only after long periods of exposure and the actual incidence of disease in
workers or animals. Thus, cancer in coke oven workers was detected on the basis of a
study of mortality data over almost 100 years, and the first US regulations were issued
only in 1977. Similarly, the carcinogenic effects of Vinyl Chloride Monomer were
noticed quite accidentally in 1970 in an experiment with rats, but it was only when deaths
occurred in a Goodrich factory that the exposure limits were reduced dramatically from
500 to 1ppm. In the case of poly vinyl chloride, the product itself is a poison, containing a
wide variety of cancer causing and hormone disrupting chemicals as additives
(Greenpeace, 1996).
Chemicals are also being widely used in agri-business. Farmers and farm workers are
exposed to more and higher levels of pesticides than any other segment of the population,
both at work and at home. But there is little information about the incidence or the extent
of exposure. Birth defects are suspected but there is insufficient chronic toxicity testing
(Goldburg, 1990). The use of chemicals in the Vietnam war to defoliate the land is
known to have released large quantities of dioxin—one of the most toxic compounds
known (PPT, 1992). There are now reports that Monsanto is marketing the same
chemicals for domestic and commercial use as weed killers. The presence of herbicides in
ground water is now widely documented. These include atrazine, triazines, metribuzin,
metalochlor, and trifluralin—all of which are now the subjects of herbicide tolerance
research. In other words, the latest trends in biotechnology are going to increase the use
of chemical toxins to kill weeds and pests, which is going to increase the exposure levels
of the workers manufacturing and applying these chemicals.
Another growing area of chemical use is the manufacture of microelectronic
components, now perceived as the basis for the third industrial revolution. Studies of
women in the telecom industry have documented a miscarriage rate double that of
women outside. These women were exposed to glycol ether widely used as solvents in
the industry. The actual health effects in a small group of 7–8 women in GTE Lenkurt
covered an amazing range of diseases: carpal tunnel syndrome, reflex sympathetic
distropy, headaches, fatigue, memory loss, attention switching and poor attention spans,
slowed reflexes, encephalopathy, positive MRIs (of the brain), hypothyroidism, adrenal
gland failure, colour pulps, lupus, cancer, menstrual problems, cervical precancerous
tissue, reactive airway disease, multiple chemical sensitivities, high liver readings, sinus
surgery, irritability, depression, anxiety, and higher rates of infection. Solvent induced
encephalopathy, in fact, is known to induce changes in personality and intellect. This has
to be coupled with the existence of a dynamic industry with short term workers and new
combinations of chemicals. Since the effects are cumulative over time, the damage may
be done long before the worker becomes aware of the danger (SWOP, 1994).
The telecom industry has also given rise to radio frequency and microwave radiations
in the 100Hz to 300GHz range. The health effects of these radiations are fiercely
contested. Independent scientists say that they have all-pervading biological effects,
particularly on fundamental cellular processes. Studies on animals have demonstrated
genetic effects, cancer, reproductive effects, cataracts, nervous system effects, blood
forming system effects, and immune system effects (Mathews, 1985). However, the
industry claims that the effects are local and mediated by heating. Instead of following
the precautionary principle and funding research to investigate the actual impacts on
worker safety and health—both in the office and the factory—the industry is frenetically
pushing a high growth strategy. Coupled with the advances in information technology is
the growth of Repetitive Strain Injuries, as workers cope with boring, repetitive tasks in
cramped spaces. The overuse of certain muscles and tendons leads to strain. There is a
physiological cost to what has been described as a ‘permanent state of arousal’ (Mathews,
1985). This reflects in the greater risk of accidents due to fatigue and long-term organic
disease and body changes. It is part of the new technological development which
increases mental and physical pressure on workers.
What is of greater concern as a new hazard is how these developments affect the
availability of work itself. As the Public Health Institute and the Labour Institute
demonstrate (PHI, 1993) in USA, toxic-related jobs may be high paying but are in
decline. The loss of a job not only means financial disaster to the worker but also
increases the levels of death and violent crimes. Even a full time job does not yield
security because wages could be low enough or the tenure of employment short enough
to bring the worker below the poverty level. As industry reorganizes to move capital into
more profitable areas, it is not only toxic-related jobs that are at stake. New
documentation in India indicates that the incidence of closure is increasing (CWM,
1998). What is common to these closures is the deliberate mismanagement by employers,
the collusion by financial institutions and regulatory institutions, and the rejection of
worker-proposed plans for revival. In a way it is reminiscent of the trends in the US at the
beginning of the century when competition was controlled through an alliance of big
business with banks and the federal government, forestalling unionization through
welfare and mechanization.
what is common to these closures is the deliberate mismanagement by employers,

the collusion by financial institutions and regulatory institutions, and the rejection
of worker-proposed plans for revival
For instance, the Bharat Gold Mines Limited at Kolar, with a peak employment of 36,000
workers, was branded sick in 1992 and put up before the Board of Industrial Financing
and Restructuring (BIFR) for possible revival or closure. An independent detailed study
of the unit indicated that the mines were still viable but had been deliberately allowed to
deteriorate through ill advised and untenable diversification and the use of a faulty
process of extraction. There had been inadequate asset utilization and no investment in
research and development since 1962. However, various proposals for revival were
rejected by the BIFR and the eventual aim of the management appears to be to sell the
mines to a foreign company once the labour force has been reduced to 2,500 workers.
The proposed joint venture with a foreign firm envisages introduction of a technology
which is already existing in the mines, and extraction of gold from the old dumps.
However, the Memorandum of Understanding also provides for prospecting leases.
Which means that there may still be ore reserves which are extractable from viable
depths. Studies conducted by the Geological Survey in this regard have not been made
public and this further strengthens the suspicion of collusion.
Such processes of disinvestment are being advocated all over the public sector, but
they are also implicit in the private sector. Thus, Kamani Engineering Corporation in
Mumbai became sick due to financial mismanagement in the mid-seventies. The Public
Financial Institutions with a stake in the company, took over control in 1975, through the
intervention of the Kamani Employees Union, and there was a dramatic turnaround in the
fortunes of the company. But in 1992 the management was subverted by another private
company, the R.P.Goenka Group, which raised its stakes through a questionable merger
and then began to divest the Kamani company of its revenues and capital assets. The
Public Financial Institutions have abdicated their responsibilities and the survival of
20,000 workers and a Rs.1 billion-company is at stake. Similar cases have been
documented by the Centre for Workers’ Management for Oriental Power Cables at Kota,
Kirpal Ispat at Gorakhpur, Deepak Insulated Cable in Mysore, Ganga Vanaspati in
Durgawati, and Saha Keil of Hosur. When conjoined with the number of factories closed
down on environmental grounds by court orders, it is clear that hundreds of thousands of
workers are facing job losses. In many ways, loss of work is the biggest work hazard.
It is, therefore, clear that new principles of work safety have to be evolved in order to
meet the new and emerging hazards which face workers in contemporary society. What
could these principles be?
EMERGING PRINCIPLES
Two decades ago it had become fairly clear that it was not enough to demand the right to
available information. Equally, if not more, important was the implementation of an
honest national data-gathering procedure which would avoid wasting limited health and
safety resources on trivia (Berman, 1978). However, such a procedure has still not
become a regulatory requirement in any country. In India, the data on occupational injury
and disease remains hopelessly inadequate and there is little or no funding available to
sponsor research in these areas. Incipient public efforts to obtain data remain confined to
lobbying for a Right to Information legislation (Mander, 1999).
Disclosure of all available information still remains an extremely valuable principle,
particularly in the context of the huge number of untested chemical substances present in
the workplace. But the demand for disclosure is no longer confined to the health effects
of the substance. New principles are emerging for a cradle-to-grave approach, in which
information is provided about where and when the chemicals are going. In addition,
preparation for worst-case scenarios of accidents and releases is becoming a mandatory
requirement (SWOP, 1994). Such an approach has been adopted in what are known as the
Silicon Principles for the microelectronics industry:
• Establish a comprehensive toxics use reduction programme

• Develop health and safety education programmes and health monitoring
• Work with local communities to establish ‘Good Neighbour Agreements’.
• Implement Worker Improvement Programmes and Economic Impact
Statements.
• Support National Research and Development policy directed by civilian (as
opposed to military) needs
• Establish corporate policies requiring equal standards for sub-contractors
and suppliers.
• Establish corporate standards that are enforced equally, domestically and
internationally.
• Establish a life-cycle approach to all manufacturing, from Research and
Development to Disposal.
• Work closely with local communities and workers to ensure full oversight
and participation.
These principles are supported by three other requirements relating to the protection of
work and security:
• Open, binding commitment goals for employment.

• Affirmative action in hiring at all levels.
• Benefits packages to include medical benefits and child-care services.
Greenpeace International has proposed three similar principles for toxics
reduction (Greenpeace, 1996):
• Search for substitutes and alternatives.
• Increase funding for pollution monitoring and control.
• Re-evaluate all materials and processes to reduce use and release of
persistent, toxic pollutants.
Both these approaches support victim’s assertions as documented at the hearings of the
Permanent People’s Tribunal (PPT, 1992):
• I do not want to be a victim; steps must be taken to guard against victimization.

• If I am a victim I want all available help from industry, government, and the
community.
• I do not want to be re-victimized by governments, companies, courts, and
the medical and legal professions.
Based on these assertions, the Tribunal set out four major findings as follows:
• The clear violation of universally recognized human rights of victims—the
right to life, the right to health, the right to livelihood and the right to a healthy,
pollution free environment.
• The inadequacy of the law and legal procedures and existing regulatory
mechanisms to prevent such violations, to provide redress to victims and to
punish those responsible for violations.
• The irresponsibility of those engaged in various industrial and other
hazardous activities, which involve release of toxic substances or making of
products injurious to human health and the environment, in failing to exercise a
duty of care to those afflicted by such activities.
• The positive role of victims’ organizations and human rights activists in
struggling to create countervailing power to enable victims to obtain redress and
to press for the accountability of those responsible for causing death and
damage.
the biggest threat to a worker’s life is the possibility of having no work at all, or
having work that creates products that are dangerous to the entire community
The Tribunal proceeded to further lay out a strategy for providing adequate protection
against these human rights violations, inadequacy of the law, and irresponsible behaviour:
• Formulating legally enforceable rights to life, health, livelihood, and healthy

environment.
• Strengthening legal and regulatory mechanisms accordingly.
• Asserting the right to information of workers and the community.
• Strengthening the public capacity to monitor.
At a seminar on the Sick Industrial Companies Bill held in New Delhi, organized by the
Centre for Workers’ Management (CWM, 1998), several recommendations were made by
the participants regarding how sickness should be defined and what could be done to
protect jobs. These may be summarized as follows:
• Define industrial sickness using the net worth concept
• Potential sickness should be flagged where the net worth drops to 50 per
cent of equity
• A tripartite mechanism has to be set up for monitoring incipient and
potential sickness
• The reference to BIFR has to be mandatory with the responsibility lying
with the management but the right of referral has to be extended to all interested
parties
• The responsibility for providing full information to BIFR lies with the
company
• The revival scheme should be assessed on the basis of techno-economic
criteria, not on the financial interests of secured creditors
• The BIFR should ensure that all concerned interests are heard and protected,
expert assistance is engaged, and the financial institutions cooperate
• There should be a deterrent for defaulting on commitments
• There should be continuous and regular monitoring of revival procedures
• The BIFR should have the status of a civil court for enforcement
• There should be complete participation of workers in rehabilitation or sale in
all BIFR proceedings
• The workers have a right to all information
• When exercising their right to take over a sick unit, special consideration
should be given to workers’ equity.
When all these recommendations are put together what emerges are principles of safety
which translate into a veritable Charter of Rights for workers which may be enumerated
as follows:
• Right to work and to participate in all decisions regarding the security and
safety of work.
• Right to determine and regulate socially safe products, materials, equipment,
and processes in all work.
• Right to organize without fear of victimisation.
• Right to collect all information, particularly with respect to the nature of
long-term hazards, and the accuracy of reporting and evaluating systems.
• Right to competent and safe co-workers, supervisors, managers, researchers,
and experts.
• Right to health, both within and without the workplace.
• Right to prosecute irresponsible authorities, managers, and polluters.
Such a Charter of Rights is clearly a major advance on the hitherto restricted

understanding of health and safety at work, primarily because it takes the concept of
safety and health out of the persona of the worker alone and locates it firmly within the
realm of production in society. Thus, it recognizes that the biggest threat to a worker’s life
is the possibility of having no work at all, or having work that creates products that are
dangerous to the entire community. This is a universal vision of health and safety, and
such a vision challenges those basic social forces which determine how a society may
grow. In other words, it no longer focuses on the harm to the individual worker but on the
potential destruction of the whole world.
REFERENCES
Berman, Daniel M., 1978, Death on the Job: Occupational Health and Safety Struggles
in the United States. (New York and London: Monthly Review Press).
Cohen, Gary. and O’Connor, John (eds.), 1990,. Fighting Toxics: A Manual for
Protecting your Family, Community, and Workplace. Washington DC and Covelo,
California: Island Press).
CWM, 1998, Papers of the Seminar on Progressive and Plural Corporate Legislation,
Sick Industrial Companies Bill 1997, and Companies Bill, 1997. (New Delhi: Centre
for Workers’ Management).
Goldburg, Rebecca, Rissler, Jane, Shand, Hope, and Hassebrook, Chuck, 1990,
Biotechnology’s Bitter Harvest: Herbicide-Tolerant Crops and the Threat to
Sustainable Agriculture. A Report of the Biotechnology Working Group
Greenpeace, 1996, The Stranger. (Amsterdam: Greenpeace International).
Greenpeace, 1997, Heavy Burden: A Case Study on Lead Waste Imports into India.
(Amsterdam: Greenpeace International).
IFT, 1999. Changes in labour laws: A strategy of metropolitan capital, work at night …in
the day: intensification of women’s labour. Labour File, Vol.5, No.5, pp 1–11.
Mander, Harsh, 1999, The Movement for the Right to Information in India. (Pune:
National Centre for Advocacy Studies).
Mathews, John, 1985, Health and Safety at Work: Australian Trade Union Safety
Representatives Handbook. (Sydney and London: Pluto Press).
MOEF, 1997, National Plastics Waste Management Task Force—Report. (New Delhi:
Government of India, Ministry of Environment at Forests).
PHI, 1993, Jobs and the Environment. (New York: The Public Health Institute and The
Labor Institute).
PPT, 1992, Findings and Judgements of the Third Session on Industrial and
Environmental Hazards and Human Rights: 19–24 October, Bhopal-Bombay [India].
(New York: Permanent Peoples’ Tribunal).
Qadeer, Imrana, and Roy, Dunu, 1989, Work, wealth, and health: sociology of workers’
health in India. Social Scientist, Vol.17, No.5–6, pp 45–92.
SWOP, 1994, Intel Inside New Mexico. (Albuquerque, New Mexico: South West
Organizing Project).
17
Achievements in Consumer Product Safety and
the Challenges of Globalization
Wim Rogmans
INTRODUCTION
This paper reviews the achievements in consumer product safety measures since the early
seventies and looks into the perspectives of global markets. It will also address the
opportunities and threats of liberalization of markets, in view of ensuring the provision of
a sustainable development of minimum safety requirements an extending part of the
world’s population.
This paper presents a short review of the development of product safety policies in the
West since the seventies and a summary of basic regulatory tools for ensuring effective
product safety policy, a bird’s-eye view on the main domains of regulatory activities
related to consumer product safety. Lessons will be drawn on what works in product
safety regulation and what does not. Finally, developments in regulatory policies in the
nineties, initiated in view of the growing internationalization of trade and commerce, are
discussed. These developments are clearly linked with the globalization of markets and
may contribute to a wider sharing of the attainments in the West in the field of consumer
product safety.
In this paper consumer product safety is to be understood as being related to non-food
products. Compared to food regulations the history of the regulation of non-food products
is shorter. It is only in this century that non-food products are being regulated, although
not unambiguously, under a separate body of law, enforcement and international
exchange. Irrespective of the critical events that we still witness in the food sector
nowadays, the regulation and control of food has matured much more and has become
more consistent than it has for non-food. Within the category of non-foods, medicinal
drugs and motor-vehicles as fields of interest have been excluded, as there are well-
developed, separate structures for regulation and control in these areas (Drug Control
Agencies and Transport Safety Agencies, respectively).
The concept of consumer products has a wider scope in this paper than that of some
product safety agencies, as it includes the entire domestic environment, i.e. private
dwellings and neighbourhood environment, such as play yards and schools. By including
these locations, the concept of ‘consumer product safety’ encompasses the domain also
referred to as ‘home and leisure area’. However, it does not include work-and traffic-
related safety issues. It is evident that these boundaries cannot be drawn too rigidly, as
there are apparent overlaps, such as injuries related to bicycling which can be classified
as a leisure activity but is often transport-related also.
The focus here is on consumer product safety measures, voluntary or mandatory
measures that aim at improving the inherent and physical characteristics of a product or
of environmental features, in order to lower the risk of an injury event or the severity of
injury outcome. These measures typically focus on those that Haddon (1980) claimed to
be more effective, such as separating vulnerable risk groups from harmful agents
(fencing, guarding, increasing child resistance to packaging) and reducing the impact of
energy transfer during the injury event (providing risk groups with shock absorbing
materials, such as helmets and shin guards, or improving the shock absorption capacity of
environmental features such as floor coverings and play ground surfaces). Measures that
aim at changing risk group’s behaviour towards safer handling dangerous situations, are
out of the scope of this paper, as these are covered by information campaigns, educational
programmes and community interventions. However, in this respect again a sharp
distinction is sometimes difficult to make, as a number of interventions include
environmental adaptations as well. (Recent reviews of such programmes refer to Towner
et al., 1993 and Svanström, 1999).
A final limitation of this review is caused by the fact that most consumer product
safety initiatives have been developed and implemented in high income countries, mostly
OECD member states (North America, western Europe, Japan and Australia in
particular). It is evident that safety has its price and even in high income countries the
willingness to pay that price is not unlimited.
authorities of the EU member states are bound to monitor the compliance of

products with the general requirements and to adopt wide-ranging powers of
regulatory control
For the time being we have to learn our lessons mainly from experiences in the high
income countries, not ignoring the fact that increasing globalization of markets will
facilitate opportunities for low income countries as well.
PRODUCT SAFETY POLICY: DEVELOPMENT OF REGULATORY

FRAMEWORK
Protection of health and safety of consumers by means of product safety regulation was
an important governmental concern long before the development of product safety as a
separate policy sector was. However, regulatory and monitoring activities in this field,
which in many countries date to the turn of the century, concentrated on certain specific
product areas—food and pharmaceutical products. On a piecemeal basis other specific
product groups, such as electrical and gas appliances, were added to the body of statutory
requirements.
Consumer concerns
Changing consumption patterns and the availability of large quantities of technically
complex consumer products in the post-war period of economic recovery, prompted a
Achievements in consumer product safety 237
more systematic and consistent approach of product safety issues. In past decades, most
industrialized countries incorporated four categories of basic actions in their product
safety policies, i.e. (OECD, 1983):
• preparatory action: surveillance of consumer products on the market in order

to identify product hazards
• regulatory action: development of product related safety standards and
regulations
• monitoring action: inspection and testing of consumer products to ascertain
compliance with existing safety standards
• corrective action: immediate intervention in case of the detection of
unexpected hazards and defective products.
These developments were inspired by the growing recognition of consumer concerns in

the sixties, which led to the creation of private and semi-public consumer organizations.
In USA, vocal leaders attracted the attention of the press and Congress, on the issue of
automobile safety. In western Europe family organizations started a consumer movement
in the early fifties, followed by cooperative movements and trade unions in the sixties.
Finally, consumer organizations established themselves as private bodies, often supported
by public grants, and developed an important service business. Examples are Consumers
Association (UK), Consumentenbond (NL) and Stiftung Warentest (Germany).
The growing economic and political importance of the consumer movement in the
sixties prompted governments to incorporate consumer viewpoint in policy development
and to protect consumer interest through adequate regulation and enforcement. It was
also a response to statistical evidence that safety concerns of public authorities should not
be limited to traditional domains, but also encompasses the area of product-related
injuries: evidence which is still valid today (Table 1).
Table 1
Comparative statistics (USA-EU) on the three traditional safety domains (work, traffic
and consumer product related)
Eu UbA
GBP $
Work related:
deaths 6,000 6,500
injuries 10 million 1 3,2 million
estimated costs 96 billion GBP 1 45 billion $
Traffic related:
deaths 45,000 42,000
injuries 3 million 3,4 million
estimated costs 195 billion GBP 150,5 billion $

Consumer product related:
deaths 80,000 22,0002
injuries 40 million 29,5 million2
estimated costs 70 billion1 400 billion $3
1 medical costs only 2 product related injuries under CPSC’s jurisdiction only J direct and
indirect costs such as suffering and lost earnings
The response of governments generally took the form of the creation of a separate
consumer policy unit, often within the trade department, the creation of advisory bodies
such as a consumer affairs councils and the establishment of specialized agencies for
consumer product safety.
General Safety Acts

For Consumer Product Safety Policy, the year 1972 may be considered as an important
milestone as this was the year in which the US Consumer Product Safety Commission
was created, and in Europe (Paris), the heads of states of the European Communities
agreed on the basic rights of consumers, granting the highest priority to the right of safety
and establishing an independent service for consumer safety within the European
Commission.
Concurrent with these institutional frameworks, new regulatory frameworks were
introduced by means of general ‘product safety acts’. In principle these acts apply to a
wide range of products and in addition provide powers to establish safety requirements
on a product by product basis. The available penalties under these laws vary in severity
and include fines, legal provisions for product recall and bans.
General laws of this type were introduced firstly in Canada (Hazardous Product Act,
1969) and USA (Consumer Product Safety Act, 1972), followed by the UK (Consumer
Safety Act in 1978, revised and integrated into the Consumer Protection Act in 1987) and
the Consumer Product Safety Act in Japan in 1984 (Table 2). In other countries, such as
Germany and the Netherlands, the general product safety requirements were included in
framework regulation that had originally been designed for a broader policy domain, such
as the safety of work equipment (Germany) or the safety of foods and other commodities
(Netherlands). In the European region, the General Product Safety Directive (GPSD)
promulgated by the European Community in 1992, is intended to harmonize the
European member states’ current general safety regulations. It imposes on producers to
market safe products exclusively and to provide customers with the relevant information,
in order to enable them to assess the risks inherent in a product in normal or reasonably
foreseeable use. They must also monitor in an adequate way the safety of products put on
the market and take appropriate action (including withdrawal of products from the
market) to avoid safety problems.
the important pre-condition remains that all parties concerned can contribute to
the rule/ standard-making process
Authorities of the EU member states are bound to monitor the compliance of products
with the general requirements and to adopt wide-ranging powers of regulatory control. In
case of conflicting approaches among the Member States, the European
Table 2
Summary of relevant consumer product safety legislation in EU (and UK, France,

Germany, specified at national levels) and in USA, Japan and Australia
REGION/COUNTRY EUROPEAN USA

UNION UK FRANCE GERMANY
Relevant Legislation General Consumer La Loi sur la Gerate Consum
Product Safety Protection protection et la Sicherheisgesatz Product
Directive Act (1987) formation du (1968) Safety A
(1992) consommateur (1972)
Special (1978) +
Directives La Loi relative State
Low Voltage a la securite des Flamma
(1973) consommateurs Fabrics
Aerosols (1983) (1953)
(1975) Fed
Pressure Hazard
Vessels Substan
(1987) Act (19
Food Poison
Imitations Prevent
(1987) Packagi
Toys (1988) Act (19
Machinery Refrige
(1989) Safety A
Pers. (1950)
Protective
Equipment
(1981)
Gas
Appliances
(1990)
Main Responsible DG Consumer Dept. Trade DG de la Minisrty of US

Authority Policy and and Industry Conairrance, de Labour and Consum
Consumer (Consumer la Social Affairs Product
Health Safety Unit) Consommation Federal Board Safety

Protection Local et de for Commi
Authority laRepression Occupational
Coordinating des Fraudes Safety
Body for (DGCCRF) (Bundesanstalt
Trading Commission de Arbeitsschutz)
Standards la Securite des
(LACOTS) Consommateurs
(advisory
body/clearing
house function)
Enforcement Structure Responsibility 467 Local 22 Regional 71 Local Trade Enforce
of 1 5 Authorities Departments an Supervisory units in
Member 100 Local Offices states
States Authorities
Remarks no duty of Volunatry

notification Scheme for
Safety
Certificates
Geprufte
Sicherheit
Commission consults a Committee on Product Safety Emergencies, upon the conclusion

of which the Council of Ministers is entitled to take a final decision. How well thought
these measures and mechanisms for interventions may appear, the crucial issue in the EU
today remains the lack of a proper implementation of the GPSD in national law, in
national enforcement practice and in efficient decision-making processes at EU-level. In
general there is still a North-South-division in Western Europe as regards the quality of
product safety control and the Commission lacks competent steering power in solving
safety disputes among member states efficiently (Micklitz, 1990).
Although the GPSD provides, in theory at least) national authorities with a powerful
tool to intervene in case of unreasonable failures in product/service delivery to
consumers, it also needs further upgrading (CEG, 1998; Rogmans, 1999) by:
majority of these regulations focus on construction safety and fire-resistance

only…Since most of the critical features, such as stairways and domestic
appliances, are implemented in the stage of final construction and habitation,
enforcement and control are most inadequate.
• establishing a duty for producers to inform (notify) authorities at once in case

of a safety problem
•requiring member states to adopt powers to recall or ban products in case of
an emergency
• improving the transparency and rapidity of information exchange and
subsequent decision making (including an upgrade of risk assessment
methodologies)
• using stricter guidelines for the application, design, durability and
limitations of warnings and safety instructions provided with the products.
These recommendations are also valid for general safety acts in other regions such as
Japan and Australia. The US act may serve as a model in this respect.
Specific regulations
Since the turn of this century specific product regulations were introduced on a piecemeal
basis. Consequently, in addition to the general safety act (CPSA of 1982) the US-CPSC
administers the more specific rules related to the Federal Hazardous Substances Act, the
Flammable Fabrics Act, the Poison Prevention Act and the Refrigerator Safety Act (Table
2).
In the EU-region the piecemeal-approach has resulted in a wide range of (high risk)
product category-specific Directives, a number of which are relevant for consumer safety,
e.g. the directives on electrical appliances (Low Voltage Directive), toys and power
equipment (‘machinery’). The majority of these directives have been (re-)designed in
accordance with the socalled ‘new approach’ with respect to harmonization of European
Community legislation. The former approach of harmonization consisted of a long and
cumbersome process of full and detailed integration of all parts and parcels of individual
member state’s legislation into harmonized community legislation. The core elements of
the ‘new approach’ are (Brack, 1999):
• European product law is limited to so-called fundamental legal requirements

concerning the safety of (groups of) products with regard to the health of
product users;
• these abstract, rather vague, fundamental prescriptions are worked out in
further technical detail by appointed European normalization organizations
which deliver standards;
• all products marketed in any of the EU member states must conform with
the fundamental legal safety requirements, which ought to be indicated by the
CE-mark (‘Conformité Européenne’) on each commodity, which conformity
with technical standardized specifications is optional, not mandatory
• the mandatory legal part of this product law system of rules is linked to the
optional technical standardization part of the system by the concept of
presumption of conformity.
A product is presumed to conform to fundamental legal requirements if it is manufactured

according to the standardized specifications issued by the officially appointed
standardization bodies, i.e. the Comité Européen de Normalization (CEN) and the Comité
Européen de Normalisation Electrotechnique (CENELEC). The key concept of the
presumption of conformity demonstrates clearly that the proper functioning of this ‘new
approach’ depends greatly upon the work that is being done within the normalization
organizations. They are responsible for the conscientious interpretation and analysis of
abstract legal safety requirements in order to transform them into workable specifications.
The process of delegating the definition of technical details to private standardization
bodies is being regarded by some observers as privatization of the rule making powers
(Brack, 1999). Others underline the opportunities this process may offer for speeding up
the process of standardization and for increasing the flexibility of technical specifications
in response to market development. The important pre-condition remains that all parties
concerned can contribute to the rule/standard-making process, a reason why the European
Commission supports the input from consumer, experts in this process through the
coordinating body, ANEC. However, even in case of a product category as vital as ‘toys’,
the result of consumer expert input in the standardization process appears to be very
limited (Petre, 1995).
Liability regulations
Concurrent with the evolution of a more consistent set of ‘preventive’ regulations, the
‘strict liability’—principle made its entrance into courtrooms in the late sixties. Liability
schemes had been available in most countries for many decades and were intended to
provide consumers with a just compensation for damage and injuries incurred and to
deter producers of goods and services from delivering sub-standard quality. Proof of
failure in product safety had to be provided by the plaintiff and the plaintiff had to
establish that the breach of the legal duty by the producer was due to negligence or fault
(‘fault based’—liability or liability ‘under negligence’). However, under ‘strict liability’
the plaintiff does not need to show fault or negligence on the part of the defendant-
manufacturer and the court’s focus shifts from the complicated study of the knowledge
and behaviour of the manufacturer to the more simple and objective analysis of the
characteristics of the product/service delivered. Some nations allow the defendants to
defend themselves on the state-of-the-art or development risk principle, which protects
them from liability with regard to defects in their products that were unknown at the time
of production, given the then existing technology.
The impact of strict liability was significant in the USA in the seventies and led to
what is described as the ‘product liability’-crisis: an outcry of business representatives,
stating that product liability insurance was becoming either unobtainable or not
affordable at all. A more careful study of the situation revealed that although the costs of
product liability insurance did increase in the seventies, the average cost amounted to less
than one per cent of sales in high-risk industries (Abott, 1980). It also concluded that
strict liability and increased insurance costs forced many manufacturers of high-risk
products to devote more time and resources to product liability cost prevention. Also
insurers are providing high-risk companies with cost prevention advice more often and
are including stricter requirements for quality assurance, including safety, in their
insurance policies.
the wide use of blister packs has short-cut the need for further regulation in that
domain, as it is assumed that blister packs reduce effectively the possibility for
young children to get access to a potentially dangerous number of tablets
The European Liability Directive (1985), however, did not result in the same outcry as
did US legislation in the seventies, which is particularly due to the fact that:
• the European Directive being ‘less strict’ than US-legislation, provided

producers with a number of means of defence
• in Europe liability lawsuits are tried before judges (unlike in the USA before
juries). Judges tend to adhere more strictly to the stipulations in the law and to
be less tempted to award punitive damages
• contingent-fees (no cure-no pay), as claimed by attorneys in the USA, are
not allowed in Europe
• in Western Europe the social security system in general covers already a
greater part of the losses in income suffered by injured and/or disabled victims
of product failures.
Naturally, business representatives and insurers were most vigilant when the debate on
the European Liability Regulation commenced in the mid-seventies, however, a
constructive dialogue resulted in a directive acceptable to both consumer advocates and
business.
In many countries the implications of this strict liability for consumers, however, is
still the subject of controversy (OECD, 1995). Many critics argue that the current
developments in the product liability area have brought about higher prices for
consumers, less innovation and a more litigious society, where individual responsibility is
gradually being replaced by judicial policy-making. On the other hand strict liability
creates strong incentives that influence business behaviour and performance towards
greater responsibility, as well as a more diligent monitoring to prevent defective products
from reaching the market place.
The policies of regulating product safety

In reconstructing the history of product safety policies and its evolution during the past
decades, the process and the results seem to be well-organized and logical: a ‘triptych’ of
legislative measures (first of all a general product safety regulation) secondly, a separate
set of requirements for high risk categories and thirdly, compensation requirements in
case of damage or injuries), including provisions for information gathering, market
control, consumer information (e.g. labelling) and emergency measures (among which
withdrawal of products) as well.
However, in most countries product safety regulation has been a highly political issue
with strong ideological pressures during the last decades (Brannigan, 1991). The debate
included both the question of whether any regulation is needed at all (Viscusi, 1984) and
the issue of public authorities’ capacity to regulate and control product safety (Shavell,
1984).
The political conflict was sharpened by business lobbies, media pressure and consumer
activism. The critical debate is on the interpretation of two basic consumer rights: the
right to safety and the right to choose, which is one of the reasons why most regulations
were initiated in response to critical events and pressures from public and industry, and a
minor number by consistent risk assessment methodologies.
It is evident that the general safety legislation, which became effective in most
countries in the late seventies and early eighties, amended a lot of gaps that the piecemeal
approach and the resulting Vertical’ safety requirements left. This was not the result of a
shift from a ‘freedom of choice’ ideology towards a ‘right to safety’ principle, but was
clearly inspired by the advent of regional trade markets: it is the free-competition and
free-trade principle that led to abolishing discriminating technical barriers to trade, i.e.
bringing technical requirements for safety at a common level within the ‘single market’
of the EU and other trade blocks, such as NAFTA (USA, Canada and Mexico). This has
been the most powerful force for governments to line up current product safety policies,
to speed up the process of harmonising existing regulations (as the European Commission
did through its ‘new approach’) and to promote international standardization through
CEN and ISO-International Standardization Organization. It put product safety on the
national and international agenda of the European Union member states, USA, Japan and
Australia in particular and has resulted in a stronger and more consistent institutional and
regulatory infrastructure for product safety.
Single markets—single product safety regulations?

Focusing now on the framework for safety regulations within the trade blocks, the
question is how they succeeded in promoting inter-state trade, given the existing
diversities in safety requirements that still exist among them. In this respect two
examples, namely the US-model and the EC-approach, are instructive.
To many the US serves as the model of ‘one single market’. Yet the situation with
respect to product safety is much more complex. Under the US Constitution, inter-state
commerce should be free from any unreasonable burden of state interference. This may
be viewed as being similar to the contents of Article 30 of the EU Treaty concerning
national restrictions on trade between member states (Brannigan and Meeks, 1991).
However, as under article 36 of the EU Treaty, the states of the USA are also free to set
safety requirements for both internal and external products, unless the motive or effect
would be discriminatory to out-of-state producers. Concerning inter-state trade, the
federal government can nevertheless pre-empt any power of the state, but often federal
statutes set minimum requirements only, allowing any state to set higher standards (for
example, seat belt legislation). With regard to products regulated under the CPSA,
however, the ruling is that in case a product safety standard is effective under the CPSA,
no state shall have the authority to establish mandatory requirements, unless these are
identical to the requirements of ‘the Federal Standard’.
The US-market in essence is a single market, but product safety is regulated at
different levels and the respective states may respond to hazards in an entirely different
way, reason why a dynamic tension between federal and state regulations is noticeable,
almost a kind of competition on the opportunity to regulate, as the US-regulatory activity
is highly political. Given the inherent openness of the US-governmental system (Freedom
of Information Act) and effective public control on decision-making by administrative
agencies (‘due process’ doctrine), this leads to a transparent process of risk analysis and
decision making. It puts a halt effectively to unreasonable state actions and allows for a
diversity of addressing safety issues according to local needs in terms of risks, demands
of industry and consumer expectations.
product safety measures, through voluntary standards or mandatory regulations,

cannot be implemented effectively without concurrent investments in raising risk-
awareness, willingness to implement measures and in creating a sustainable
culture of safety
As regards establishing a European single market, until 1984 within the European Union
the object was to fully harmonize the different technical regulations and standards.
However, this approach proved to be unsuccessful due to the time-consuming and
cumbersome procedures. It certainly was not realized for 50 per cent by the year 1993,
the then magic year the EC had set as deadline for completing the internal market.
To speed up the process the ‘new approach’ was introduced in 1984 (see the section in
this paper on Product safety policy: development of regulatory framework). An important
feature of this approach is that member states are obliged to recognize products
manufactured in other states in conformity with harmonized European standards or
transposed national standards. In order to facilitate manufacturers to provide evidence of
conformity to a new approach directive, the European Commission developed a complex
system of conformity assessment and marketing approval. According to this system, for
low risk products, such as ‘bandages’ and ‘toys’ (sic), a manufacturer’s declaration of
conformity can be sufficient, whereas for more complex products, such as ‘chain saws’
and ‘circular saws’, assessment and certification by independent testing houses is
required. These testing houses have to be approved by competent authorities for this
purpose and notified to the Commission and other Member States and are therefore called
‘notified bodies’.
Unlike the US system, the EU kept focusing on the prerogatives of European
legislation. Failing to implement these in detailed regulation, the EC invented the
‘solution’ of the presumption that national regulations and standards are not conflicting
with European Directives and Standards, unless proven to be so by one of the member
states. The end result is that as in USA, the European Union allows for different levels of
regulation (federal vs. state), that reflect different levels of risk acceptance, standards and
requirements. On top of that we have to mention the major differences in enforcement
practices among EU member states (Micklitz, 1990; Sutherland, 1994).
EFFECTIVENESS OF PRODUCT SAFETY MEASURES
As stated earlier, product safety measures may deal with a vast array of environmental
features and product characteristics that are relevant for ensuring safety at home and in
leisure time. A large number of these measures have been initiated inspired by ‘common
sense’ and are not being evaluated at all. The outcome of these measures is at best being
monitored by authorities who consult routine statistics (and apply economic models such
as the CPSC, who calculate the savings as a result of for example its standard for walk-
behind mowers at $ 300 million per year), as well as feedback from professional groups
such as the fire brigades in the case of fire prevention measures and the toxicology
centres with regard to poison prevention measures. This leaves us with a limited number
of effectiveness studies reported in the literature only, which differ considerably in:
• problem area being addressed, e.g. the prevention of bath tub drowning,
residential fires, or dental injuries in field hockey
• scientific rigour of the research design, mostly being pre-and post-tested
only and randomized controlled trials being the exception
So far a consistent meta-analysis on product safety measures in general has
not been performed, with the exception of the short study published by
Vrijenhoek and Weperen (1997), who conclude that:
• effectiveness studies on product safety measures are scarcely being reported
• the majority of these reported studies measure the results in proxy measures,
the relation of which with injury reduction targets is mostly defined vaguely
• the research design does not allow for unequivocal conclusions on the
relation between trends and safety measures as introduced.
Irrespective of these flaws in rigour in research design, analysis and reporting, a great
number of reported studies are instructive and several have been instrumental in
implementing good practices worldwide. An example of the latter category is the
mandatory provision of child safety closures for toxic household products (caustics for
example), which after a successful introduction in the USA in 1970, has been adopted by
most of the countries in the western hemisphere. But there are other instructive examples
such as the successful introduction of smoke detectors, window guards and personal
protective equipment (helmets, shin guards, mouth guards and so on).
In the following section we will briefly review a number of studies that we ordered
with respect to subjects the primary field of application of which is either related to the
domestic area (residential setting including domestic products) or to the community
environment (schools, play grounds and recreational areas).
Methods addressing the domestic environment and related products

This section addresses measures that aim to improve the safety of private dwellings as
well as specific products and appliances used in these. The review is certainly not
exhaustive and will focus on the domestic environment, i.e. on building regulations and
measures to control fire hazards, hazards of falling from windows or stairs and measures
to control toxic risks.
Building codes
A large number of accidental injuries and deaths are related to environmental hazards at
home, among which slippery surfaces, loose rugs, objects on floor, inadequate lightning,
poorly designed stairs, flat glass in doors, substandard gas and electrical facilities.
In most countries building codes and standards intend to control a number of these
hazards by setting minimum safety requirements. With the exception of the requirements
for gas and electrical facilities (and the appliances, which are governed under separate
international product regulations), the details in the building codes vary from country to
country (Pauls, 1991), and even within countries uniformity is lacking, as in many cases
the implementation of most technical details for private homes is left to local authorities.
The majority of these codes lack clear cut minimum safety requirements for hazardous
features such as stairs, windows, balconies, flat glass and hot water supply, regardless of
the apparent effectiveness of these measures as we will see in the following sections.
Enforcement is another flaw in current building regulations. As in view of the aspect of
safety the majority of these regulations focus on construction safety and fire-resistance
only, enforcement is effected by construction engineers solely and only in the design and
construction stages. Since most of the critical features, such as stairways and domestic
appliances, are implemented in the stage of final construction and habitation,
enforcement and control are most inadequate. The fact that the potential aspect of
enforcement can be improved, is illustrated by a study in Massachusetts (Ghallangher et
al., 1985). While conducting home safety inspections in accordance with the State
Sanitary Code, this study found an average of 11.1 code violations per household.
Follow-up visits revealed full correction of these violations, thus demonstrating the
effectiveness of this approach.
before introducing new product safety measures one should consider seriously any
unforeseen side effects of regulations
Window guards and safety gates

A guard is a most effective tool to prevent children from falling from windows or stairs.
Studies have documented the importance of security windows and balconies. In New
York city the ‘Children can’t Fly’ programme, by the implementation of window-guards
on high-risk houses, brought about a 50 per cent drop of serious injuries and deaths due
to window falls (Spiegel and Lindamann, 1977).
Gates on stairways are potentially very effective in preventing young children from
falling, however, the major problem in most dwellings is how to find a way to attach
them firmly, permanently mounted and closed child-proof (Joffer, 1988). This item
should be part and parcel of standard building codes, whereas now parents are struggling
with temporary provisions of dubious quality or, when the accordion-type of gates is
used, of even greater risk—namely suffocation—than the hazard they intend to address.
Fire safety
Numerous studies have examined the efficacy of smoke-detector-give-away campaigns,
with and without installation, or of low-cost purchase opportunities (Neily et al., 1993).
Most of these campaigns addressed high-risk neighbourhoods, such as low-income and
ethnic communities and neighbourhoods with a high proportion of children or elderly
residents. These studies report favourable results although long-term compliance, for
instance regular checks on batteries and proper maintenance, is weak. The potential
efficacy of smoke alarms, which in the USA is reported to be at about 86 per cent of all
fire deaths (Federal Emergency Agency, 1980) is thereby seriously hampered.
New technologies have made smoke detection more reliable as well as affordable at
low costs. For this reason, undoubtedly, standard requirements for reliable smoke
detectors, which are installed properly with the mains connected, should be included in
building codes, as it is reported already to be in half the number states in the US. In the
European Union, Norway is the only country applying a national regulation in that
respect.
Another measure that may reduce the damage of fires is by mandating the use of
sprinkler systems in high-risk buildings, i.e. the old and wooden apartment houses and
densely populated apartment buildings. By mandating cigarettes to be fire-safe and
cigarette lighters to be child-safe, important ignition sources can also be addressed
effectively. This is also true for regulating the flammability of childrens’ clothing as well
as carpets and curtains. Most product safety authorities are struggling with introducing
such legislation as the counter pressure from manufacturers is heavy and the evidence of
the number of lives being saved as well as the amount of cost reductions achieved, are
not incontestable.
Toxic household products and medicines

The regulation of the safety of packaging containing cleaners and pesticides for domestic
purposes is one of the most evident success-stories in product safety. During a period of
fifteen years the US-Poison Prevention Packaging Act (issued in 1970) resulted in a
decline in the number of deaths by one-fifth the number reported before 1970. The PPPA
also requires childproof containers for a number of medicinal drugs, including over-the-
counter-drugs such as aspirins. Also in this respect the Act proved to be most effective
and resulted in a decline of half the number of deaths and hospitalizations in this respect
in ten years’ time (see among others Walton, 1982).
Inspired by this success almost all Western countries now have legal requirements for
the childproof packaging of toxic domestic products. For medicinal drugs the wide use of
blister packs has short-cut the need for further regulation in that domain, as it is assumed
that blister packs reduce effectively the possibility for young children to get access to a
potentially dangerous number of tablets. However, it would be advisable to provide the
tablets in opaque plastic cells and sufficiently strong strip materials, to prevent children
from opening them.
Unlike the dramatic reductions in poisonings by liquids and solids, poisoning by gases
and vapours remains a huge problem. Carbon monoxide is the major cause of such
poisonings, mainly due to gas operated water heaters installed in poorly ventilated homes
(EC, 1998). Also unventilated space heaters are responsible for a significant number of
deaths each year (CPSC, 1984). The advent of modern and central heating systems has
reduced considerably the number of CO-deaths, as well as of deaths due to fires.
However, in rural and low-income areas and in dwellings designed for temporary
habitation (recreational homes) the CO-hazard remains an important threat. Regulation
and regular control of heating systems should be enforced more strictly, while in addition
the campaigns for installing CO-detectors should be promoted, similarly to the successful
promotion of smoke detectors. Finally, there is a need to inform the new urban
generation, that is unaware of the CO-risks, how to cope with the potentially lethal
equipment in their temporary residences.
standardization process is not as flexible and rapid as expected and, even

more so, still dominated by business interest and financing and therefore
leads too often to rather biased end products
Measures addressing the community environment

This section looks somewhat beyond the domestic setting and addresses safety issues that
are relevant in the wider community, such as safety at day care centres and schools, at
playgrounds, of surface water in the neighbourhood and in swimming pools, on sport
fields and while cycling.
Day care centres and schools

In most countries safety and health of day care centres is under the responsibility of
health authorities, whereas that of elementary and secondary schools is under the
educational authorities. Most of the safety requirements are included in a broader set of
instructions that are part of the licensing contract between authorities and the
centres/schools. In addition, in most countries the occupational safety and health
requirements that employers are obliged to provide for their work force, apply to the
educational sector as well (Laflamme and Menckel, 1997). However, this often leads to
ambiguous situations, as the exact safety requirements are not made explicit to all parties
at all times, nor is the enforcement responsibility clearly defined.
As a result of this in the day care centres and school environments lots of initiatives are
left to the management itself and the local authorities, without proper exchange and
synergy at a national level. An exception may be those countries that have mandatory
insurance schemes for pupils and teachers alike, like Germany. These insurance schemes
include important facilities for safety check programmes, teacher training and school
education, as well as technical requirements for school buildings and equipment in use
(BAGUV, 1996).
Not surprisingly the number of carefully evaluated school and day care safety
programmes is very minor. This is due to the fragmentation of regulatory and enforcing
powers, the culture and the professional attitudes in this particular setting, which
reinforce self-control and auto-determination in managing educational centres. A new
development is this respect is the certified inspection scheme based on agreed national
standards of good practice (Weperen, 1998).
Playgrounds
Playground-related injuries constitute an important part of childhood trauma. Most of
them are due to falls in play areas (collision with other children, stumbling) or falls from
monkey bars or other equipment, reason why impact absorption of playground surfacing
is the first and most effective measure one can take. There is ample evidence (Mayer,
1996; Sibert, 1999, Mott et al., 1997) that under specific conditions and maintenance
requirements sand, wood chips and artificial surfaces (rubber tiles) can significantly
reduce the severity. In most countries guidelines on playground surfacing are made
available through safety agencies and standardization bodies, however, only in a few
countries these guidelines are made mandatory through regulation and enforcement
schemes. This is also true in respect of requirements and standards set for playground
equipment. Although a large number of guidelines for design, installation and
maintenance are available nowadays, the majority of them are being applicable on a
voluntary basis. Only in a few countries (like the Netherlands and the UK) regulations are
effective and therefore also inspection schemes to enforce full compliance.
Contrary to the effectiveness of improving shock absorption of surface materials, the
results of stricter requirements for playground equipment are less well proven. This does
not imply, of course, that we can continue to expose young and vulnerable children to
substandard and potentially hazardous equipment. For this reason many local authorities
and an increasing number of national authorities are willing to implement the available
guidelines and standards such as those of CEN (1997 and 1998) on a voluntary basis.
Surface water
Drowning rates are highest among pre-scholars, while low income groups and some
ethnic minorities are disproportionately affected as well. This may be due to greater
exposure, insufficient swimming skills or lower risk-awareness. The majority of
drowning among this age group occur in the neighbourhood, i.e. residential swimming
pools, garden ponds and in nearby lakes, canals and ponds.
In Australia a few states apply compulsory requirements for private swimming pools to
be provided with child-proof barriers (Dept. Local Gov., 1991; Pitt, 1991). Owing to its
effectiveness the implementation has been also considered in other regions of the world,
such as the USA. Studies by the US-CPSC indicate that a 70 per cent-reduction of private
swimming pool drowning among young children may be achievable by fencing all
private pools (Baxter, 1987).
These measures will not make adult supervision redundant, neither re-engineering of
other hazardous areas in the neighbourhood, such as the water side of ponds, canals and
lakes. Effective measures for lowering the risk for children of getting from the waterfront
into surface waters are available (Jaartsveld, 1994) but are not yet included in a uniform
set of standards for local authorities and urban designers.
Sports
Product safety measures in sports focus mainly on the design of effective protective
equipment, such as helmets and braces and on improving the safety of sports
accommodations and equipment. Examples of the latter are the provision of slip-resistant
floors, improving the construction strength of equipment and optimizing impact
absorption of landing pads. Another example in this category is the development of so-
called ‘breakaway’-bases used in baseball and soft ball in order to reduce the damage of
injuries due to collisions with ‘sports furniture’ (Janda et al., 1993).
Specific consumer products that have proven to help reducing the impact of injuries in
sports are: horse riding helmets, mouth guards, eye and face protectors, shin guards, knee
braces, high top shoes and taping. There is ample, biomechanical, evidence that such
products have a productive effect and lower the risk of a serious injury (Munro et al.,
1995). However, the quality of products offered to amateurs sportsmen should be
improved remarkably, in order to have the protective effect reflected in a significant drop
of serious injuries also. Both mandatory and voluntary requirements, i.e. standards, will
facilitate that process in due course.
too often product safety measures are developed and designed upon ‘crisis
management’—in response to an event or series of events that necessitate
authorities to intervene
Bicycling
There is ample evidence from the US and Australia suggesting that both incidence and
severity of head injuries are lower in cyclists wearing helmets, when compared to those
who are not (Rivara, 1998). However, the target groups continue to oppose strongly to
the wearing of helmets: inconvenience, costs, and the ‘nerd’-effect are major barriers
still. Yet, the state of Victoria, by means of continuous promotion and finally compulsory
legislation succeeded to increase the helmets wearing percentages from 5 per cent in the
seventies up to 83 per cent in 1992 (Graitcher, 1995). Other measures that can reduce
injury risk among bicyclists concern the stability of the construction of bicycles, cyclists’
conspicuousness and road design, i.e. separate lanes for cyclists and for motorized
vehicles. The latter type of measure in particular has proven to be most effective, if
properly applied (a line mark in the road does not provide sufficient separation).
Conclusions: What makes it work and what does not

We can draw a number of conclusions from the evaluation of product safety measures
developed so far. Firstly, apart from the fact that product safety measures do not always
require the full active participation of the target group in getting things safer, the many
examples quoted above show ample evidence that it is a fallacy to believe that product
safety measures would be ‘passive’ measures. Proper installation, use and maintenance of
smoke detectors, stair gates, child safety caps, cycle helmets, protective taping, pool
fences and so on, require the active and continuous participation of the target group.
Therefore product safety measures, through voluntary standards or mandatory
regulations, cannot be implemented effectively without concurrent investments in raising
risk-awareness, willingness to implement measures and in creating a sustainable culture
of safety.
Secondly, some success stories may easily be transferred from one region to others,
e.g. child resistant packaging for toxic products, but others do not as easily, e.g. wearing
bicycle helmets. Environmental (road conditions), natural (landscape), social (transport
needs), economic (access to less risky modes of transport) and cultural (cycling for fun or
purely for transportation) factors determine for example the actual choice of transport
modes and the safety precautions to be taken. A blind transfer of the successful helmet
story from Australia to the Netherlands will be a waste of energy and resources. This is
even more true for the transfer of good practice in high-income countries towards low-
income countries (Berger and Mohan, 1996).
Thirdly, before introducing new product safety measures one should consider seriously
any unforeseen side effects of regulations (Farquhar, 1998). The presumed effect of ABS
on speeding behaviour of adults is a popular example, but in the consumer product safety
field there are actual examples of negative side-effects of regulations, such as parents
putting turpentine in open containers for convenience of use. The side-effect of regulating
barriers for children to prevent them from getting access to hazardous environments, is
that is may be difficult for others to obtain access easily as well. Also, it is quite harmful
for elderly if they cannot open blister packs or disclose gates easily, which need to be
opened in case of fire.
Fourthly, too often product safety measures are developed and designed upon ‘crisis
management’, i.e. in response to an event or series of events that necessitate authorities to
intervene. Of course, there are a lot of good examples where actions are based upon
thorough examination of the problem, investigation of alternative solutions, support for
change and so on. In Western countries in particular a tradition has been established in
this respect, but even that region knows a multitude of safety problems that are not
adequately dealt with and that are left to local parties to be solved, such as the safety at
day care centres, schools, in sports, recreation (in particular in preventing drowning) and
not least in home design and maintenance. Whereas there are clear incentives for national
authorities to regulate movables (see section on Product Safety Policy: developments of
regulatory framework), regulating important safety features in the neighbourhood
environment is apparently regarded as being less urgent.
Fifthly, no law is effective without proper enforcement. In most regions of the world
confidence in that is seriously jeopardized as a result of the outbreaks of serious events in
the food sector and also due to problems with non-food, such as products intended for
children, toys and pushchairs in particular. It happens that one of the firstly regulated
categories of products, i.e. toys, causes problems continuously, which even result in
conflicts between national authorities on how to interpret the standards and/or regulations
as internationally agreed upon (OECD, 1996). These conflicts occur, irrespective of the
internationally well-organized toy industry and the great efforts of European and US
authorities to reach consensus on essential requirements. Apparently the toy-market is
much too volatile and dynamic for authorities to control pro-actively, as they lack prime
information on product development, technological advances of materials, production
logistics and distribution.
So, what makes product safety measures work? A systematic approach that
transgresses ‘local thinking’ by:
• identifying the right issue to address, taking into account the number and
severity of injuries, the economic costs and personal sufferings, the community
concern about the issue and the support basis at policy and administrative level;
• identifying feasible and low cost intervention measures, based upon
thorough knowledge of community’s risk perception and willingness to adopt
such measures also on the long term. Such measures should be made as
consistent as possible with those measures that have been implemented yet
effectively, including a full assessment of side-effects and compensatory
behaviour as well;
• reassuring a consistent support level in the community and at political and
administrative level by concluding joint agreements for implementing the
envisaged measures;
• implementing measures while monitoring continuously the process and its
results, which includes permanent feedback to the parties involved in order to
sustain their commitment.
despite the influence of the consumer movement in putting consumer protection

and product safety on the political agenda, most of the product safety measures in
the past decades were clearly inspired by trade policy
GLOBAL TRADE AND PRODUCT SAFETY
Despite the influence of the consumer movement in putting consumer protection and
product safety on the political agenda, most of the product safety measures in the past
decades were clearly inspired by trade policy, in particular the wish either to protect local
markets from products of unknown qualities or, as in the case of the USA and the EU, to
open internal markets with a (semi-federal structure. The globalization of markets has
provided an even stronger impetus to national governments to harmonize many of the
relevant legislations and the detailed requirements laid down so far either in specific law
or in national standards. The question, however, is whether this will lead to a ‘race to the
bott ‘“downward harmonization’), or to a sharing of the highest possible level as attained
in the most advanced member states (‘upward harmonization’).
Globalization: a mixed blessing?

For better or worse, globalization has become the economic buzz word of the nineties.
National economics will undoubtedly become more integrated as cross-border flow of
trade increases. The positive point of view in this regard is that globalization will boost
productivity and living standards everywhere, as it allows for large-scale economics,
better division of labour between countries and more transparent competition. Critics,
however, worry among others about the erosion of the ability of governments to protect
acquired levels of health and well being of citizens.
However, regardless of the discussion on the ‘new’ global economy, globalization is
not an unprecedented phenomenon: in the second half of the nineteenth century, owing to
remarkable declines in the costs of transport (railways and steamships) and reduction of
trade barriers, the flow of products across borders was huge. This ended abruptly with the
World War I, after which the world moved into a period of strong protectionism. After
World War II, the big economic powers agreed to reduce trade barriers, which was a vital
prerequisite for recovering from economic depression. This resulted in the General
Agreement on Tariffs and Trade (GATT), which was replaced by the World Trade
Agreement (WTA) in 1995. Yet, the integration of markets is still far from complete:
although the ratio of trade to total output, i.e. merchandise trade as a percentage of GDP,
has increased considerably in most countries since 1950, it exceeds the figures of 1913
only slightly.
Even more remarkable is the fact that a large part of international trade is not global at
all but is limited to trade with neighbouring countries: e.g. 60 per cent of German export
takes place within the EU region and 80 per cent of Canadian export ends up in the USA.
So, today, instead of global trade we are witnessing a huge expansion of intra-regional
trade-activity. Naturally, the establishment of regional trade associations, such as the EU,
NAFTA and APEC, has been most instrumental in this respect. These three trade-blocks
also account for 90 per cent of the total international trade. Undoubtedly, due to the
revolutionary development of the information and communication technology, which has
only just started its impact on the consumer goods markets, an international trade, which
will be genuinely global, will expand significantly during the years to come.
From single regional markets to one global market: mutual (dis-)

agreements?
Since the agreement on opening trade markets worldwide and the implementation of the
WTA in 1995 in particular, a new impetus is given to the authorities in trade blocks to
look for common principles and frameworks in addressing product safety issues. But
there are other pragmatic reasons for reassessing the performance of the traditional
regulatory frameworks, as they did not meet sufficiently the needs for business to:
• develop new technology and materials rapidly, without being burdened by

detailed technical, often obsolete, regulations;
• shorten the period of time to market new products and services and being
free from cumbersome and costly procedures for conformity testing to be
performed in each new market; and
• adapt flexibly to different markets and cultures and be able to anticipate on
differences in consumer expectations, local traditions and cultures, without
jeopardising essential safety requirements.
The solutions found for these problems were:
• boosting the process of international standardization, which allows for a more

rapid response to changes in technology than the rule-making process did and
also reduces significantly the workload of regulators. This resulted in a
commitment from industrial states to work preferably towards international
(ISO) standards exclusively;
• setting up a programme for mutual recognition of testing procedures and
testing houses, in order to avoid multiple testing and to simplify procedures of
accreditation, such as the EC-scheme for mutual agreements and the bilateral
agreements between EU and countries like USA, Canada, Japan and Australia;
• promoting cooperation and exchange among business, governments and
consumer representatives, as through the Transatlantic Dialogue on Consumer
Affairs.
The EC-New Approach (promoting the importance of standards) and the EC-Global
Approach (furthering mutual recognition on testing and certification within the region)
can be seen as the precursors of the new policy of international governments with respect
to globalization of markets and new means for controlling product safety.
Stakeholders’ view
we have to question the effectiveness of establishing reasonable levels of safety on

the one hand and the creation of open competition, not hampered by further
restrictions, on the other hand
A recent study performed by OECD (1996) provides better insight in the point of view of
business on national and international product safety and safety regulations. The study
focused on four product categories (toys, home appliances, power equipment and
personal protective equipment), the related business and regulations. The study reveals
that first of all manufacturers regard conformity of their products with standards as
essential as it helps to protect the company’s reputation, it reduces safety risks for
consumers and reduces the risk of legal liability. To manufacturers and their trade
associations, product safety standards and conformity requirements are becoming more
important as global competition intensifies. They emphasize, however, the differences
among the great variety of product safety standards and the burdens of conformity
assessment procedures. What might appear to an outsider as insignificant variations in
product standards, create, in view of manufacturers, costly production difficulties.
It is revealing that manufacturers identify major differences in particular in a product
category with a history of study and negotiating as long as is the case with toys.
According to the study report, the divergencies between national and regional standards
on toys are ‘a source of great dissatisfaction’, which could be remarkably improved by
the harmonisation of all major standards (i.e. ISO, EU-standards and US-standards).
Another suggestion made is the use of ‘hazard-based’ approach standards design, using
injury evidence as basis for requirements. The publication of general guidelines (an
example of an ISO-Guide is to be found in Table 3) is also an important tool for getting
product safety being a core concern in all stages of production, starting with product
development and design.
Current conformity assessment procedures seem to be of even greater concern to
manufacturers and trade associations. The majority of manufacturers who responded to
the survey of OECD (1996), stated that current testing and certification requirements add
significantly to their costs: they feel that a product should only be tested once, upon
which the results should be accepted worldwide.
Table 3
ISO/IEC Guide 50: Safety aspects—Guidelines for child safety (summary of revision
dated January, 1999).
In 1987 the first edition of ISO/IEC Guide 50, Child safety and standards, was published. Since
then, standards-makers have had the possibility to consult this valuable guide in order to address
child safety in an adequate and consistent way. A few years ago, the ISO Consumer Policy
Committee (COPOLCO) decided that the Guide needed to be revised. This would create an
opportunity to incorporate recent views on child safety, and to bring it more in line with the
general Guide (ISO/IEC Guide 51:1999, Safety aspects—Guidelines for their inclusion in
Standards. An ad hoc working group with experts from all over the world prepared a draft that
was formally published early in 1999.
Hazards as a source of injury—a new focus for Guide 50
The new edition of ISO/IEC Guide 50 places more emphasis on hazards, as the sources of
injury/harm present in products and the environment. The first edition focused more on products,
accidents and injuries, and these categories were not always clearly distinguished, e.g. scalds and
other injuries appeared in a lst of ‘accidents’.
The Guide starts with a general approach to child safety including the principles for a systematic
way to address hazards. This latter issue is closely linked to ISO/ IEC Guide 51. Next, specific
developmental characteristics of children are described that place them at a particular risk of
injury: body size, motor development, cognitive development, and psychosocial factors.
Types of hazard
The Guide’s principal clause is devoted to hazards to which children may be exposed during their
use of or interaction with a product, along with specific suggestions for addressing those hazards.
Mechanical hazards are the most widespread; several sub-types are distinguished and they are
discussed in detail.
Thermal hazards can be the origin of burns (hot surfaces) and scalds (hot liquids).
Chemical hazards in the field of child safety are often linked to typical child behaviour, like
putting various objects in the mouth.
Electrical hazards are associated with all electrical equipment. Radiation is not a very common
hazard in the household, but products like microwaves and smoke alarms require special
attention. In addition, solar radiation is a natural hazard for which various products claim
protection.
Biological hazards can be associated with many types of product where growth e.g. bacteria and
moulds, is feasible. Inadequate protective function may not seem a hazard as such, but it is a
useful means to link performance requirements of e.g. bicycle helmets and stair gates to safety
levels. If the consumer expects a protective function, its absence constitutes a hazard.
Similarly, safety information is an important means for producers to communicate with
consumers. The pre-sale information should enable a well-informed choice between products;
information accompanying the product should give guidance to use and maintenance, and warn of
residual hazards that cannot be sufficiently reduced by design.
An Annex is included as a checklist for product hazards, examples of injuries, age groups at
special risk, and approaches to solutions.
Who is Guide 50 intended for?
Formally, the Guide is intended for those involved in the preparation and revision of standards.
However, the information it contains can also be directly useful to designers, architects,
manufacturers, service providers, communicators, safety professionals, and the like.
one of the challenges for safety regulations and business in forthcoming decades is
the technological revolution, which will certainly have its bearing on quality,
functional properties and safety aspects of consumer products, but will also open
new opportunities for developing new products and components for existing
products in a rapid pace
FUTURE DIRECTIONS
Given the dynamics in product safety regulation in the seventies and eighties and the
introduction of more flexible arrangements for reaching agreement on essential
requirements, we have to question the effectiveness of establishing reasonable levels of
safety on the one hand and the creation of open competition, not hampered by further
restrictions, on the other hand. In high-income countries product safety policies stand at a
crossroads. Recent events in the domain of food safety in particular, but also in non-food
product safety (e.g. phthalate in toys and children’s products, cigarette-lighters and
flammability of furniture), seem to undermine the confidence in national and
international authorities to ensure proper levels of product safety consistency.
Pitfalls
In implementing the more flexible approach towards product safety a number of serious
weaknesses became evident:
• the standardization process is not as flexible and rapid as expected and, even
more so, still dominated by business interest and financing and therefore leads
too often to rather biased end products (ANEC, 1994);
• conformity assessment is trusted to notified bodies that do not necessarily
possess the broad disciplinary background needed for products that are in use in
different settings and conditions. A recent study revealed that a number of
notified bodies in the EU miss essential competence as required for the directive
they were acknowledged for (Bryden, 1996);
• inter-administrative cooperation in enforcement and control of product
safety remains a daunting task, taking into account the great variety in
enforcement practices in the different countries and continents, insufficiency in
information exchange (regardless of obligatory structures such as the Rapid
Exchange system in the EU), and the lack of agreement on methodologies for
risk assessment and enforcement decision-making (Rogmans, 1996).
The process of deregulation and market flexibility apparently has led to a greater
ambivalence in specific requirements in vigour and in responsibilities the main
stakeholders have to carry out in view of product safety control. It is evident, however,
that in regulating product safety during the next decades, authorities should not make a U-
turn and should not restart the more technical and detailed process that may lead to
mandatory standards nailed down in legislative rulings.
Need for integrated approach

One of the challenges for safety regulations and business in forthcoming decades is the
technological revolution, which will certainly have its bearing on quality, functional
properties and safety aspects of consumer products, but will also open new opportunities
for developing new products and components for existing products in a rapid pace. In
future markets, service and product industries need to respond quickly to these
developments, not being needlessly hindered by restrictive standards and procedures. In
this respect, the trend towards more standardization instead of detailed regulation and
common general procedures for conformity testing has to be welcomed. However, the
process includes many more stakeholders, including business and consumer
representatives, than one was used to in the past decades of top-down regulation. It
therefore needs much more orchestration on behalf of government authorities in order to
lead this process of reaching consensus on safety requirements and its conformity testing
towards an effective result.
Effective decision making requires the involvement of the major stakeholders and an
open and transparent debate regarding the relevant safety requirements based on evidence
and best practices. The present challenge for the trading blocks is to establish adequate
infrastructures for orchestrating such decision-making in the respective regions as well as
at international level. This goes far beyond the efforts invested today in regulatory and
enforcement exchange and requires among others the establishing of regional agencies
responsible for consumer product safety analogous to the CPSC in USA. It includes also
(Rogmans, 1997):
• the establishment of clearing houses in the various regions, linking all relevant
data and providing easy access to vital information on consumer product-related
risks and their prevention;
• the advancement of analysis and research that supports a high quality of
standards, regulation and emergency measures as well as properly targeted
information and educational programs;
•strengthened and coordinated enforcement and control of safety provisions
and regulations; and
• increased efforts in safety promotion and coordinated programs for safety
education and information.
Free trade and freeing business from excessive constraints is essential for the economic
growth and development of all nations. Only with visionary leadership can we ensure that
free trade goes hand-in-hand with fair trade and that the global market place will
maintain one of the most fundamental values: the protection of citizens against accidental
injuries and deaths.
REFERENCES
Abott, H., 1980, Safe enough to sell?—Design and product liability, (Design Council,
London).
ANEC, 1994, Report on standardization work in the interest of consumers. (European
Association for Coordination of Consumer Representation in Standardisation,
Brussels).
BAGUV, 1996, Sicherheit im Unterricht—ein Handbuch für Lehrkrafte, Bundesverband
Unfallversicherung, München.
Baxter, L., 1987, Study on child drownings, US-CPSC, Washington DC.
Berger, L.R. and Mohan, D., 1996, Injury Control—a global view, (Oxford University
Press, Delhi).
Brack, A., 1999, The CE-mark and the new European approach to product safety law,
J.Consumer Product Safety, 6, pp. 45–58.
Brannigan, V. and Meeks, C.B., 1991, European unification, single markets and product
safety: the American experience , J.Consumer Policy, 14, pp. 63–86.
Bryden, A., 1996, Consumer product safety: a growing role for third party conformity
assessment operators. In New Challenges, W.Rogmans.
CEG, 1998, Review of the 1992 General Product Safety Directive, (Consumer in Europe
Group, London).
CEN, 1997–1998, European standards for playground equipment, their instalment and
maintenance, Part 1–5, (Comité Européen de Normalisation, Brussels).
CPSC, 1984, Status report on gas heating systems, (Consumer Product Safety
Commission, Washington, DC).
CPSC, 1997, Strategic Plan 1996–2006, (Consumer Product Safety Commission,
Washington, DC).
Dept. Local Government, 1991, Swimming pool fencing review group. (Child Accident
Prevention Foundation Australia).
European Commission (EC), 1998, CO-intoxication: a silent killer in wintertime,
Brussels.
Farquhar, B.J., 1998, What makes regulation work, Injury Prevention, 4, pp. 253–262.
Federal Emergency Movement Agency, 1980, An evaluation of residential smoke
detector performance, Washington DC.
Gallangher, S. et al., 1985, A home injury prevention programme for children, Paediatric
Clinic North America, 32, pp. 95–112.
Graitcher, P.L. et al., 1995, A review of educational and legislative strategies to promote
bicycle helmets, Injury Prevention 1, pp. 122–129.
Haddon, W., 1980, The basic strategies for reducing damage from hazards, Hazard
Jaartsveld, R., 1994, Safety of water surfaces (in Dutch only), Educatieve uitgever,
Montfoort.
Janda, D.H. et al., 1993, A prospective study comparity standard and breakaway bases in
college and professional baseball, Clinical J. Sport Medicine, 3, pp. 78–81.
Joffer, M and Ludwig, S., 1988, Stairway injuries in children, Paediatrics, 82, pp. 457–
461.
Laflamme, L. and Menckel, E., 1997, School injuries in a public health perspective,
Mayer, J. et al., 1996, Parameters correlating to injury severity score in playground-
related fall accidents, International J. Consumer Safety, 3, pp. 147–153.
Micklitz, H.J. (editor), 1990, Post-market control of consumer goods, (Nomos-Verlag,
Baden-Baden).
Mott, A. et al., 1997, Safety of surfaces and equipment for children in playgrounds,
Lancet, 349, pp. 1874–1876.
Munro, J. et al., 1995, Can we prevent accidental injury in adolescence? A systematic
review of the evidence, Injury Prevention, 1, pp. 249–255.
Neily, M. et al., 1993, Residential smoke detector performance in the USA, (US-CPSC,
Washington, DC).
OECD, 1983, Consumer policy during past ten years; main developments and prospects
(Organization for Economic Cooperation and Development, Paris).
OECD, 1995, Product liability rules in OECD-countries (Organization for Economic
Cooperation and Development, Paris).
OECD, 1996, Consumer product safety standards and conformity assessment: their effect
on international trade, (Organization for Economic Cooperation and Development,
Paris).
Pauls, J.L., 1991, Safety standards, requirements and litigation in relation to building use
and safety, Safety Sc. 14, pp. 125–154.
Petre, L., 1995, Consumer input in standardisation—a case study of toy safety,
International J. Consumer Safety, 2, pp. 209–223.
Pitt, W.R., 1991, Childhood drowning and near drowning in Brisbane: the contribution of
domestic pools, Medical J. Australia, 154, 661–665
Rivara, E.P. et al., 1998, Prevention of bicycle-related injuries: helmets, legislation and
education, Annual Review Public Health, 19, pp. 298–318.
Rogmans, W., 1996, New challenges for consumer safety in the global market, (ECOSA,
Amsterdam).
Rogmans, W., 1997, Europe signposts a safer world: the way ahead for consumer safety
in Europe, International Journal for Consumer Safety, 4, pp. 215–221.
Rogmans, W., 1999, Injury prevention and safety promotion—European Challenges,
ECOSA 2nd Convention, Edinburgh.
Shavell, S., 1984, Liability for harm versus safety regulation, J. Legal Studies, 9, p. 357.
Sibert, J., 1999, Playground injury prevention, paper at 2nd European Convention on
Injury Prevention, Edinburgh.
Spiegel, C.N. and Lindamann, F.C., 1977, Children can’t fly: a program to prevent
childhood mortality and morbidity from window falls, American J. Public Health, 67,
pp. 1143–1147.
Sutherland, R., 1994, Administrative cooperation between member states in
Implementing EC-directives, (European Commission report, Brussels).
Svanström, L., 1999, Evidence-based injury prevention and safety promotion—a review
of concepts and studies, Karolinska Institute, Stockholm.
Towner et al., 1993, Reducing childhood accidents: the effectiveness of health promotion
interventions, (Health Education Authority, London).
Van Aken, D., 1999, Childhood injury control—a new edition of ISO/IEC Guide 50,
ISO-bulletin, Geneva/
Viscusi, W.K., 1984, Regulating consumer product safety, (American Enterprise Institute
for Public policy Research, Washington DC).
Vrijenhoek, M. and Weperen, W. van, 1997, Effectiveness of product safety measures,
Proceedings Product Safety Research Conference, ECOSA, Amsterdam.
Walton, W., 1982,. An evaluation of the Poison Prevention Packaging Act, Paediatrics,
69, pp. 363–370.
Weperen, Willem van, 1998, Inspection and certification scheme for day care settings
draft produced by Keurmerkinstituut b.v., Amsterdam.
18
Technologies, Therapies, Emotions and
Empericism in Pre-hospital Care
Mathew Varghese
INTRODUCTION
Trauma patients form a heterogeneous group who have suffered tissue damage from
exposure to some form of energy. The most common denominator in any trauma patient
is bleeding. This could be external or internal bleeding. Neuronal damage and direct
damage to chest, abdominal viscera all significantly modify the outcome of trauma,
independently and along with physiological changes associated with bleeding. Over 50
per cent of trauma patient’s mortality in the first hour results from traumatic brain injury.
Traditionally accepted markers of physiological status of the patient are used for
monitoring the patient’s condition in most situations. These include pulse, blood pressure,
respiratory rate and temperature and in a head injury patient the level of consciousness of
the patient. However, monitoring parameters in the field situation and taking steps to
correct them are difficult because changes in physiological parameters in the body are
closely linked to multiple organ systems. No system can be taken in isolation.
Pre-hospital care developed as a speciality to minimize the damage to the patient
before definitive care is made available. A lot of controversy exists regarding what
should be the ideal pre-hospital management for the injured patient. No system is ideal
for all areas. Policy formulations for pre-hospital management of the injured will
obviously be determined by the availability of trained personnel and technological and
economic resources at hand. But what is the optimum level of care and who should
provide it? In other words, what pre-hospital interventions should we make when we
have all the resources to ensure that the largest number of trauma patients survive with
minimum morbidity.
BEYOND THE ABC OF RESUSCITATION?
Conventional first-aid to an injured includes resuscitation, dressing of wounds and

splintage. In principle resuscitation at the accident site and at the hospital site is not very
much different. The classical description of the ABC of resuscitation took care of the
most vital aspects of survival for a patient and also prioritized the sequence of
intervention, that is the maintenance of a clear Airway, establishment of Breathing, and
establishment and assessment of Circulation. To achieve these a large number of
technologies and therapies have developed over the years.
Technologies, therapies, emotions and empericism 263
Different centres for pre-hospital care follow different protocols. These include: Basic
Life Support (BLS), Pre-hosPital Trauma Life Support (PHTLS) and Advanced Trauma
Life Support (ATLS). Several recent reports are available on the relevance of individual
components of these protocols and experimental studies to validate them (Sampalis,
1992; 1994; Martin, 1992; Bickel, 1994; Teach, 1995; Adams, 1996; Eisen, 1998;
Bissell, 1998). Pre-hospital resuscitation theoretically results in improved end organ
perfusion and oxygen delivery in the pre-operative phase, in the field and during patient
transfer reducing subsequent morbidity and mortality (Owens, 1995). A review of the
ABC of resuscitation steps should clarify some of the controversies surrounding pre-
hospital care.
Maintenance of a clear airway

Extension of neck and elevation of chin may be the only-step required for maintenance of
a clear airway. At times this may not be sufficient. Foreign bodies, blood and vomitus in
the mouth may need to be cleared. The oesophageal obturator airway is designed to
provide ventilation by occluding the oesophagus and by default forcing oxygen through
the oropharynx into the trachea and lungs (Don Michael, 1968). This is not routinely
recommended now because of several complications associated with it.. According to the
ATLS protocol (ATLS 1993), if the patient is not breathing after clearing of airway then
insertion of an endotracheal tube may be required. However, endotracheal intubation is a
procedure, which needs training and skill. This will need a medical or paramedical person
to be present at the accident site. Some countries have restrictions on the use of
endotracheal tubes and only licensed persons are allowed to insert them. A lay person
cannot do endotracheal intubation. At times endotracheal intubation can be done only
with narcotic analgesics (Gerich, 1998), midazolam (Dickinson, 1999) for ‘clenched
teeth’ may prevent attempts at intubation. A small percentage of these endotracheal tubes
may actually be in the oesophagus (Pointer, 1988; Gerich, 1998; Dickinson, 1999).
However, Gerich (1998) reports that only 1 out of the 8 patients with surgical airways
survived. Combitube is another alternative for airway management in emergency
situations where personnel trained in tracheal intubation are not available.
a lot of controversy exists regarding zwhat should be the ideal pre-hospital

management for the injured patient
According to some of the protocols, repeated failed attempts and inability to maintain
sufficient oxygen saturation (90 per cent), during repeated intubation attempts represent
indications for surgical air way by doing a cricothyrotomy (Gerich, 1998).
Irreversible brain damage occurs between 5–10 minutes of cessation of breathing. To
provide a trained person at the accident scene within this period is very difficult. Help
from bystanders can be useful during this time. A trained bystander could call for help,
clear the airway or extend the neck, lift the chin to ensure breathing.
If after clearing of the airway the patient does not breathe he will need mouth-to-
mouth-expired-air-ventilation or if available a bag-valve-mask-ventilation. Mouth-to-
mouth-expired-air-ventilation is a highly specialized psychomotor skill, which has to be
learnt, rehearsed regularly before it can be performed successfully. In a Norwegian study
(Bjorshol, 1996), among health and rescue workers outside hospital it was found that
only 1 per cent were able to perform satisfactory cardiopulmonary resuscitation (CPR) of
cardiac arrest patients according to accepted guidelines. Of these only 17 per cent
ventilated and compressed efficiently with a rhythm of 2 breaths for 15 external cardiac
compressions or 1 breath for 5 cardiac compressions, and 50 per cent believed that they
were efficient in life saving first aid.
There is a wide variation in the effectiveness of mouth-to-mouth resuscitation provided
for cardiopulmonary arrest. The best results are obtained only when the person providing
CPR is periodically rehearsing the procedure.
In a patient with a high probability of cervical spine injury manipulating the neck for
endotracheal intubation could be dangerous. More sophisticated techniques of
nasotracheal or fibreoptic assisted intubation technique are recommended for these
patients. New technologies going well beyond the simple ABC of resuscitation!
Establishment/assessment of circulation
Most trauma patients bleed. Bleeding could either manifest externally or could be occult,
internally. Internal bleeding is particularly difficult to assess if it is occurring in the
abdominal or pelvic cavity. Bleeding into the chest is also difficult to assess but the
dramatic presentation of an associated lung or heart injury may give sufficient warning of
the damage underneath. If the patient’s assessment is done soon after the injury and pulse
and blood pressure parameters are taken as gold standards, then there could be a serious
error of judgement as haemorrhage into closed cavities may take time to manifest as
alterations in these measured values. This is because it takes sometime before
physiological signs of compensation or decompensation become apparent.
The human circulatory system functions as a closed loop. Any break in the vessels
converts this closed loop into an open system. As more and more blood leaks out, the
heart will have less and less blood available to pump. Initially the body compensates by
increasing the heart rate and then mobilizing fluids away from the non-vital parts of the
body. With decreasing blood in the system the blood pressure drops pulse becomes rapid
and feeble till it cannot be felt. Medically such a patient is described to be in a state of
haemorrhagic shock. There are numerous definitions of haemorraghic shock involving
blood pressure, heart rate and urine output. Traditionally, when blood pressure, heart rate
and urine output return to normal it was presumed that resuscitation was complete.
However, a majority of patients may not be in uncompensated shock and some patients of
compensated shock may continue to have clinical signs of uncompensated shock (Porter,
1998).
Pre-hospital medical care and care in the hospital is aimed at preventing the
development of traumatic haemorrhagic shock and the consequences of shock in an
injured patient. This can be achieved by
• control of bleeding
• replacing the blood lost with blood
• replacing the blood volume lost by fluid replacement.
Control of bleeding
If the patient is bleeding externally, direct pressure with a gauze or a clean cloth is an
effective method of controlling blood loss. If the wound is on the limbs then elevation
and direct pressure together are very effective.
Tourniquets, which were once popular as effective method of controlling blood loss,
are now not recommended at all. In fact, tourniquets are dangerous and some patients
have even lost their limbs because of improper use of tourniquets. Direct pressure and
elevation are the only safe methods recommended.
Internal bleeding cannot usually be controlled by non-surgical methods. Assessment of
blood loss internally into the chest or abdomen is very difficult in the pre-hospital phase.
Hence in patients with suspected chest or abdominal injuries the aim should be to transfer
the patient as quickly as possible to a hospital for definitive treatment.
Blood transfusion
among health and rescue workers outside hospital it was found that only 1 per
cent were able to perform satisfactory cardio-pulmonary resuscitation (CPR) of
cardiac arrest patients according to accepted guidelines
Replacing blood for blood is ideal intervention theoretically. This is neither possible nor
desirable in all injured patients receiving pre-hospital care. This is not possible in all
patients or desirable in injury patients receiving pre hospital care. O(−ve) is a rare blood
group and its availability is poor even in hospitals. With the possibility of HIV and
hepatitis infection there is an added deterrent to the use of blood widely in trauma
patients. There are some ambulances in the world which have blood banks with O(− ve)
blood. If direct cross matching is to be done this takes about 30 minutes. There does not
seem to be any additional advantage in having blood available within ambulances.
Blood group identification is recommended by some transport authorities to be
included in the driving license. Though it is good to know your blood group for donating
others, advance knowledge of blood group of the injured patients does not in any way
hasten the transfusion of blood. This is because blood matching has to be done by the
care provider in any case.
Intravenous fluids
Intravenous fluids are widely used as volume replacements for blood loss. This is done
both in the pre-hospital and in the hospital phase. Placement of intravenous access lines is
considered an important part of pre-hospital care. The ATLS protocol specifically
recommends the insertion of two large bore (16 G or more) catheters for intravenous fluid
transfusion. In adult injured patients 1–2 litres of Ringer’s lactates are recommended.
Krausz (1992) found that intravenous access placement failed in 27 per cent of cases
and an average of 10–12 minutes were lost in placement of intravenous cannula.
Placement of an intravenous cannula is particularly difficult in a shocked patient as all the
veins collapse in shock. In children this is difficult even when they are not in shock as the
normal caliber of their veins is small.
A number of studies recommend starting of intravenous line on the plea that it can
maintain the fluid volume in circulation when it is compromised as a result of injury.
Lewis (1986) has shown that intravenous infusion is of benefit only if the bleeding rate is
more than 25–100ml per minute, the pre-hospital time is more than 30 minutes and the
intravenous infusion rate is equal to the bleeding rate.
Newer technologies being made available aim at increasing the volume of fluid infused
in shorter times to keep pace with the rate of loss of blood. These include very large bore
catheters and use of rapid infusion devices. In children intraosseus infusion has been
recommended to overcome the difficulties of intravenous cannula placement. This has
been associated with complications of osteomyelitis and compartment syndromes.
Hypertonic saline has been used on the rationale that only smaller volumes need to be
transfused, and that they help mobilize body fluids faster. All these techniques are aimed
at increasing the circulating fluid volume in the presence of a leaking closed loop of the
system.
Recent animal experiments have shown that in swine models aggressive saline infusion
led to greater bleeding and failed to improve survival (Kowalenoko, 1992; Owens, 1995).
In rat tail experiments using hypertonic saline Krausz (1992) reported greater bleeding
and rapid haemodynamic worsening and early death when infusion were given without
control of bleeding. Results of experiments in mice show that haemodilution is an
unfavourable condition for the natural defense mechanism against hemorrhage
(Okumura, 1995).
A critical review of earlier experiments showing that intravenous fluids were useful
clearly reveals a design flaw in the experiments. In all these experiments a certain
percentage of blood volume of the experimental animal was withdrawn and then the
effect of fluid replacement studied. In the newer experimental design blood was
withdrawn and a cannula was left in the place (or by some other method) to simulate
continuous bleeding. This change in technique produced dramatically different results.
In normal human physiology continued bleeding leads to a reduction in blood pressure.
This reduction in blood pressure causes the initiation of a reflex narrowing of blood
vessels carrying blood to the periphery. The lowering of blood pressure and narrowing of
blood vessels reduce the pressure head of blood at the injured site. This allows clot
formation and sealing off all injured vessels.
By infusing intravenous fluids without controlling bleeding the normal physiological
response of the body is prevented. This delays clot formation and sealing of injured blood
vessels. The blood loss therefore, can be much more than when no infusions are given.
The reason for not seeing a marked increase in mortality and haemorrhage in the patients
infused with intravenous fluids is possibly because of the fact that the rate and volume of
fluid infused in most pre-hospital injured patients is very low.
The volume as well as the rate of fluid infusion seems to influence the outcome of
haemorrhage in experimental studies. Rapid infusions of large volumes increased blood
loss but seemed to improve survival. Where as moderate fluid volume was not associated
with significant blood loss regardless of rate. The greatest blood loss was seen in animals
with slower rates of infusion (Soucy, 1999). The correct rate of infusion and volume of
infusion in any given patient is still a matter of conjecture though a number of algorithms
are available.
In an evaluation of the use of intravenous fluids in penetrating truncal injury patients
(Martin, 1992) found no significant difference in post-operative complications and rate of
survival to hospital discharge. The study concluded that further studies are necessary to
determine if it is advantageous to delay fluid resuscitation until surgical intervention. In
pre-hospital times of 30–40 minutes, mortality following trauma is not influenced by the
pre-hospital administration of intravenous fluids but is related to the severity of
underlying injuries (Kaweski, 1990). Bickell (1994) found improved survival and
decreased perioperative complications in patients with penetrating torso injuries with
delayed resuscitation. Teach (1995) found pre-hospital fluid therapy was inconsequential
to outcome in 47 of 50 patients, possibly beneficial in 2 out of 50 and possibly
detrimental in one of 50 patients.
Though widely used all over the world, increasing evidence suggests that infusion of
intravenous fluids in the pre-hospital setting, may not be useful but could be harmful also.
Pneumatic antishock garments (PASG) or military antishock trousers

(MAST)
These were introduced in 1970 for use in trauma patients. PASG corrects hypotension of
increasing peripheral resistance and by autotransfusion of blood from the compressed site
to the central area of circulation. Serious complications have been associated with its use.
They involve development of neurovascular injuries, development of compartment
syndromes and pressure necrosis. In a healthy volunteers study, PASG was found to
reduce the vital capacity and tidal volume of volunteers significantly. MAST markedly
decreased the abdominal contribution of breathing (−57 per cent ± 22per cent) (Riou,
1991). Survival of injured patients is not improved despite increase in blood pressure. In
a prospective randomized study no significant differences were found in hospital stay and
mortaility following use of PASG (Chang, 1995). The use of PASG or MAST is not
recommended any more in the pre-hospital setting.
tourniquets, which were once popular as effective method of controlling blood loss,
are now not recommended at all
In summary ,the best treatment for bleeding in a traumatized patient is surgical control of
bleeding and not haemodynamic manipulations with fluids, blood or PASG.
Pre-hospital cardiac arrest

If bleeding continues and the volume of blood reduces markedly, the heart will stop
functioning as a result of hypoxia and decreased volume of blood to pump. The heart
could also arrest as a result of direct injury to the heart and lungs or as a result of head
injury causing damage to vital centres.
In trauma 60 per cent of deaths occurs at the site of injury and 40 per cent occur in the
hospital (Kries Jr., 1986). In Turnkey’s classification 50 per cent of deaths in trauma
occur immediately within 1st hour, 30 per cent occur early (1 hr-1 week) while 20 per
cent deaths occur late (after 1 week) (Trunkey, 1983). The immediate and late trauma
death rates are not affected by changes in pre-hospital care. Some of the early deaths can
possibly be prevented by interventions in the pre-hospital period. The ‘golden hour’ as
described by Cowley (1977) and the ‘platinum half hour’ as described by Pepe (1990)
underlines the importance of early interventions in trauma patients. However, if a trauma
patient develops a cardiac arrest before he reaches the hospital the chances of survival are
poor even after cardiopulmonary resuscitation. In a study of 138 cases who had CPR in
the pre-hospital period there were no survivors (Rosemurgy, 1993). In 23 patients who
required CPR in the pre-hospital phase again there were no survivors (Tsai, 1987).
Gallagher (1955) found that only 46 per cent of bystanders were able to do effective CPR
in outside hospital cardiac arrest.
Increasing number of studies are reporting the futility of attempting CPR in the pre-
hospital cardiac arrest patients in trauma.
Triage
The classification of patients according to medical needs and matching of these patients
to available care resources is called triage. This helps to save time and avoids shunting of
patients from one centre to another. For optimal allocation of resources in the treatment
of trauma it would be useful to be able to decide as early as possible which patient will
benefit most from transport to a dedicated trauma centre (Bond, 1997). This may be
difficult in patients who have neuronal injury or truncal injury which is in the early
evolving phase. Different criteria and indices have been evolved for transportation of
patients to a designated trauma centre. The pre-hospital index takes into consideration,
systolic blood pressure, pulse/min, respiration and consciousness. An index of 0 to 3 is
minor trauma, 4 to 20 is major trauma. This is reportedly better when the mechanism of
injury criteria is also taken into consideration (extrication time is more than 20 minutes,
ejection, occupant death, steering wheel deformity, fall of more than 50 feet and injuries
of automobile vs pedestrians). The ideal triage criteria in any given situation is difficult.
Ambulances and transportation of patients

Ambulances have become synonymous with transportation of the injured and the sick.
Transportation of the injured have been done in all kinds of vehicles. Faster and more
comfortable modes of transportation have been designed with a view to take the patient
as rapidly as possible to a definitive care facility.
The time interval between injury and arrival at a hospital has an influence on the
outcome of trauma. A comparison of transportation time of patients in different wars of
this century is summarized in Table 1.
Table 1
Summary of mortality rates and transportation times in different wars of this century
WAR TRANSPORTATION TIME (in hours) MORTALITY

World War I 12–16 8%
World War II 6–12 4.5%
Korean War 2–4 2.5%
Vietnam War 1.5–2 2%
The marked decrease in mortality in the Vietnam war cannot be attributed to a reduction
in pre-hospital times alone. Hospital care of injured has changed completely between
World War I and the Vietnam war. Reduction in transportation time is only one of the
parameters that had changed between World War I and the Vietnam War.
When a person is injured it is assumed that he or she will be transported by an
ambulance to the hospital. In reality, in a large number of countries this is not so. This is
particularly true in a number of rural areas of the world and urban areas of several low
income countries. Even in high income countries some of the patients are transported by
non-EMS vehicles (Demetriades, 1996). Interestingly patients with severe trauma
transported by private means in this setting were found to have better survival than those
transported via EMS system. Persons without access to telephone also often use private
transport to transfer trauma patients to a trauma centre. Of the 4 per cent patients
transported in private vehicles 50 per cent did not have access to telephone. Among the
others, fear of delay and under estimation of the severity of trauma were the other causes
(Hammond, 1993). In Philadelphia 61 per cent of Police Chiefs indicated that police
officers would occasionally ‘scoop and run’ with a critically ill child rather than wait for
the emergency medical services to arrive (Sinclair, 1991).
though it is important for the injured patient to reach a definitive care facility at
the earliest in urban situations with short transportation times excessive speeding
cannot improve transportation times
In a study done in Delhi it was found that ambulances transported only 4 per cent of
patients. Of the injured 51 per cent were transported to the hospital by taxies. About 53
per cent of these patients were transported within 30 minutes of the injury (Maheshwari,
1989). Despite the fact that they were not transported in ambulances this is comparable to
some of the best transportation times in the world with good communication facilities and
excellent ambulance services. In a comparative study of trauma mortality patterns, Mock
(1998) reported no patients were transported in ambulances to a teaching hospital in
Ghana while over 90 per cent were transported by ambulances in Mexico and Seattle. In
Ghana 58 per cent were transported by ambulances and 22 per cent by private cars. The
differences in mortality in this study could not truly be attributed to increased time of
transportation or differences in treatment.
Equipment in ambulances
The ambulance itself may be a simple vehicle with a stretcher or it could be fitted with
the most sophisticated equipment for monitoring and providing advanced cardiac life
support with defibrillators. In trauma patients the probability of revival after pre-hospital
cardiac arrest is practically nil, unlike in cardiac disease patients. Usually the injury has
caused so much of haemorrhage that the oxygen carrying capacity of blood would be
significantly deranged and the myocardium is unlikely to respond to defibrillation.
Numerous other equipment like suction machines and immobilization devices for limb
or spinal trauma also form part of ambulance equipment. However, there is no data to
suggest that use of these equipment alter the outcome of trauma.
Speed of ambulances
Though it is important for the injured patient to reach a definitive care facility at the
earliest in urban situations with short transportation times excessive speeding cannot
improve transportation times. This speeding may in fact contribute to risk of injury to
patients, other motorists and pedestrian on the road. The incidence of fatal ambulance
crashes during emergency use is reportedly higher than during non-emergency use. These
are particularly higher for lights and siren travel (Saunders, 1994; Pirrallo, 1994). There
are anecdotal reports of injuries to patients and people outside ambulances involved in
crashes.
A recent study has shown that ambulances with flashing lights and sirens do not
significantly reduce patient transportation time. The study used ambulances with lights
and sirens and a control ambulance without any of these, it revealed the mean time saved
to be 43.5 seconds in 50 trips (Hunt, 1995). Use of sirens also significantly disturbs the
patients being carried in it. The noise of sirens and traffic also disturb recording of blood
pressures of patients in moving ambulances (Prasad, 1994).
Air ambulances
Air ambulances have been promoted with a view to reduce transportation times and
hence reduce mortality, air ambulances are costly, and their health benefits are small
(Snooks, 1996). The study found that there was no improvement in response times and
the time on scene was longer for helicopter-attended patients. Logistic regression analysis
in helicopter transported trauma patients have shown that transportation by helicopter
does not affect the estimated odds of survival (Brathwaite, 1998).
Another study showed that a large majority of trauma patients transported by both
helicopter and ground ambulance has low injury severity measures. Outcomes were not
uniformly better among patients transported by helicopter. Only a very small subset of
patients transported by helicopter appear to have any chance of improved survival based
on their helicopter transport (Cunningham, 1997). Doubtful benefits have also been
reported by Koury (1998) and increased mortality, 18 per cent compared to 13 per cent
for ground transported patients for helicopter transportation of victims in urban area
Schiller (1988)..
Helicopter services may have a role in remote inaccessible areas in the sea, desert or
mountains. However, routine use of air ambulances in the urban setting is not cost
effective.
Ambulance personnel
The number and training of ambulance personnel varies from place to place. Some have
only drivers trained in emergency care while others have emergency care paramedics. In
some parts of the world there are physician-manned ambulances. Trained medics and
paramedics are posted in the emergency medical service ambulance to ensure that the
trauma patients receive optimal care from the site of injury. Physician-manned on scene
care was found to cause a significant increase in scene time and total pre-hospital time.
These delays are associated with an increase in the risk for death in-patients with severe
injuries (Sampalis, 1994). Physicians on the scene tend to try to provide more care in the
field than well trained paramedics, therefore, the time to definitive care of the
haemorrhage may be delayed (McSwain, 1995).
With the information available it seems that in an urban setting all that is required is a
comfortable vehicle with sufficient space to carry the injured safely to a hospital.
Role of medication
Analgesics for trauma patients and cardiac drugs for non-trauma patients are the most
commonly used medications. Fentanyl was used in 75.4 per cent of patients with
fractures during transportation to the hospital (DeVellis, 1998). Drugs were administered
in 8.5 per cent of urban emergency patients and 7 per cent of rural emergency patients
either at site or during transportation (Moss, 1993). So far, there is no reported evidence
that pre-hospital medications are either beneficial or cannot be delayed until the arrival at
the emergency room.
helicopter services may have a role in remote inaccessible areas in the sea, desert
or mountains. However, routine use of air ambulances in the urban setting is not
cost effective
Care of wounds
antiseptics and antibiotics are not necessary for care of wounds. All that is required is to
keep the wound clean. Healing is a natural process, which cannot be hastened by any
medicine and ointments can only delay healing. In case of small wounds if the wound is
dirty then the best treatment is to wash the wound with clean water. This is the only first
aid that may be required for small wounds and abrasions.
Splints for the injured

Fractured/dislocated limbs can be splinted to help reduce pain and prevent further injury
to the patient. This is an important first aid measure and must be attempted on scene to
make the patient more comfortable. All kinds of materials can be improvised to work as
splints and if nothing is available the opposite uninjured limb of the patient can function
as an effective splint. Air splints are available which encircle the limbs and compress
tissues. These can cause serious damage if applied too tight. Softer easily available
materials like cushions, pillows or even rolled up magazines and newspaper may be
equally effective without causing further damage.
Care of the spine

Recognizing a spinal injury is not easy even for trained medical personnel. However, a
high index of suspicion can prevent paralysis and further damage in a spinal cord injured
patient. Spinal cord injury must be suspected if the patient has a head injury, is
unconscious or has altered sensorium, has paralysis of the limbs or is complaining of pain
in the neck or back. There is, however, significant variation in clinically clearing cervical
spine practice among emergency duty physicians (Cone, 1999).
If spinal cord injury is suspected then the best first aid is to treat the patient as a ‘log of
wood’. All movements of bending, extending or rotation is to be avoided. Four or five
persons can together transfer a patient as a ‘log of wood’.
A semi-rigid collar for the neck or even a simple rigid board can be used for shifting
the patient. Repeated transfer of the patient is to be avoided in all patients suspected to
have spinal cord injury.
ATLS vs BLS
In the mid-seventies, cardiac patients were found to do much better with the availability
of ALS care. It was assumed, therefore, that all patients would do better with more being
accomplished on the field (McSwain, 1995). This assumption neglected a basic premise
of patient care: the most important factor in patient survival is the time from the onset of
the emergency to the provision of definitive care. There has been a lot of controversy
about the value of ATLS for injured patients (Trunkey, 1984). ATLS involves a greater
use of technology, psychomotor skills and medication for pre-hospital care.
BLS on the other hand focuses on basic airway support, control of bleeding,
immobilization of spine and provision of supplemental oxygen when required.
In a sample of 360 severely injured patients Sampalis (1993) found that the outcome of
trauma is not affected by ATLS on the scene. Cayten (1993) also found no benefit from
the use of ATLS for trauma patients with pre-hospital times less than 35 minutes. This
was also reported by Adams (1996) and Sampalis (1994). Jurisdictions throughout the US
and some other parts of the world have invested substantial time and resources into
creating and sustaining a pre-hospital advanced life support (ALS) system without
knowing whether the efficacy of ALS-level care had been validated scientifically. The
strongest support for ALS level care was in the area of responses to victims of cardiac
arrest (Bissell, 1998). Sampalis (1992) reported provision of ALS on scene was
associated with a higher incidence of mortality where as definitive care in level 1 or 2
compatible hospital was associated with a lower mortality.
‘Scoop-and-run’ versus ‘stay-and-stabilize’

There are proponents for and against each of these approaches, ‘scoop-and-run’ involves
extrication of the patient, maintenance of a clear airway, protection of spine and control
of haemorrhage whenever possible. ‘Stay-and-stabilize’ on the other hand involves
placement of intravenous lines, infusion of intravenous fluids, application of PASG and
endotracheal intubation whenever required.
A review of literature and the physiological processes involved suggests that in urban
areas with transportation times of less than one hour and no delay in extrication scoop-
and-run seems to be the best policy.
Backup at the hospital

All hospitals do not have the same level of expertise for managing trauma patients.
Unnecessary shifting from one hospital to another hospital can be avoided if proper
triaging is done in the beginning. The quality of a trauma system can be assessed by the
rate of preventable deaths. One question that can help is if this patient had sustained the
accident in front of the hospital in a normal working day, might death have been
prevented? The main failures in a review of trauma deaths were found to be errors and
delays during the first phases of in-hospital assessment and care. An improvement in pre-
hospital care will be almost useless if the quality of definitive in-hospital management is
not addressed (Stochetti, 1994).
a review of literature and the physiological processes involved suggests that in

urban areas with transportation times of less than one hour and no delay in
extrication scoop-and-run seems to be the best policy
FUTURE
We are in a situation where something as basic as starting of an intravenous fluid in a

traumatized patient is being labeled as controversial. Factual meta-analysis needs to be
done to separate what really works from what is perhaps useful. The future may find that
even some of our very basic parameters of measurement of end points of resuscitation
may have changed completely.
One of the dilemmas of pre-hospital care has been ‘are we doing too little for a damage
which seems too much?’ Our emotional response seems correctly to be to do whatever
possible to save as many lives as possible. There is a need, however, to avoid deification
of technology and to homogenize responses in a problem which is essentially
heterogeneous. To make scientific conclusions we must have well-controlled prospective
randomized studies. There exists a strong general feeling that randomizing pre-hospital
care is unethical (Gold, 1987). However, there are natural control populations in place in
the world where a total contrast of no pre-hospital care exists along with places where
high-tech pre-hospital care is practised. Advantage could be taken of such situations,
normalize them for different injuries to have a controlled study. Until such carefully
designed studies are carried out we will continue to grope for answers and components of
pre-hospital care will remains controversial.
REFERENCES
Adams, J., Aldag, G. and Wolford, R., 1996, Does the level of pre-hospital care influence
the outcome of patients with altered levels of consciousness?. Pre-hospital Disaster
Medicine, 11(2), pp. 101–104.
American College of Surgeons, 1993, Advanced Trauma Life Support Program for
Physicians. (Illinois: USA).
Bickell, W.H., Wall, Jr. M.J., Pepe, P.E., Martin, R.R., Ginger, V.F., Allen, M.K. and
Mattox, K.L., 1994, Immediate versus delayed fluid resuscitation for hypotensive
patients with penetrating torso injuries. New England Journal of Medicine, 331, pp.
1105–1109.
Bissell, R.A., Eslinger, D.G. and Zimmerman, L., 1998, The efficacy of advanced life
support: a review of the literature. Pre-hospital Disaster Medicine, 13(1), pp. 77–87.
Bjorshol, C.A., 1996, Cardiopulmonary Resuscitation Skills: A Survey Among Health
and Rescue Personnel Outside Hospital. Tidsskrift for Den Norske Laegeforening, 116
(4), pp. 508–511.
Bond, R.J., Kortbeck, J.B. and Preshaw, R.M., 1997, Field trauma triage: combining
mechanism of injury with the pre-hospital index for an improved trauma triage tool.
Journal of Trauma, 43(2), pp. 283–287.
Brathwaite, C.E., Rosko, M., McDowell, R., Gallagher, J., Proenca, J. and Spott, M.A.,
1998, A critical analysis of on-scene helicopter transport on survival in a statewide
trauma system. Journal of Trauma, 45(1), pp. 140–144.
Cayten, C.G., Murphy, J.G. and Stahl, W.M., 1993, Basic life support versus advanced
life support for injured patients with an injury severity score of 10 or more. Journal of
Trauma, 35(3), pp. 460–467.
Chang, F.C. et al., 1995, PASG: Dies it help in the management of traumatic shock?
Journal of Trauma, 39(3), 453–456.
Cone, D.C., Wydro, G.C. and Mininger, C.M., 1999, Current practice in clinical spinal
clearance: implication for EMS. Pre-hospital Emergency Care, 3(1), pp. 42–46.
Cunningham, P., Rutledge, R., Baker, C.C. and Clancy, T.V., 1997, A comparison of the
association of helicopter and ground ambulance transport with the outcome of injury in
trauma patients transported from the scene. Journal of Trauma, 43(6), pp. 940–946.
Demetriades, D., Chan, L., Cornwell, E., Beizberg, H., Berne, T.V., Asensio,J., Chan, D.,
Eckstein, M. and Alo, K., 1996, Paramedic vs private transportation of trauma patients.
Effect on outcome. Arch. Surgery, 131(2), pp. 133–138.
DeVellis, P., Thomas, S.H. and Wedel, S.K., 1998, Pre-hospital and emergency
department analgesia for air-transported patients with fractures. Pre-hospital
Emergency Care, 2(4), pp. 293–296.
Dickinson, E.T., Cohen, J.E. and Mechem, C.C., 1999, The effectiveness of midazolam
as a single pharmacologic agent to facilitate endotracheal intubation by paramedics . Pre-
hospital Emergency Care, 3(3), pp. 191–193.
Don Michael, T.A., Lambert, E.G. and Mehram, A., 1968, Mouth to lung airway to
cardiac resuscitation. The Lancet, 2, pp.1329.
Asensio, J.A. and Weigelt, J.A., 1991, Contemporary Problems in Trauma Surgery,
Surgical Clinics of North America, 71 -II.
Eisen, J.S. and Dubinsky, I., 1998, Advanced life support vs life support field care: an
outcome study. Academy Emergency Medicine, 5(6), pp. 592–598.
Gallagher, E.J., Lambardi, G. and Gennis, P., 1995, Effectiveness of bystander CPR and
Survival Following out of Hospital Cardiac Arrest. JAMA, 274(24), pp. 1922–1925.
Gerich, T.G., Schmidt, U., Hubrich, V., Lobenhoffer, P. and Tscherne, H., 1998, Pre-
hospital airway management in the acutely injured patient: The role of surgical
cricothyrotomy revisited. The Journal of Trauma: Injury, Infection and Critical Care,
45(2), pp. 312–314.
Gold, C.R., 1987, Pre-hospital advanced life support vs “scoop and run” in trauma
management. Annals Emergency Medicine, 16(7), pp. 797–801.
Hammond, J., Gomez, G.A., Fine, E., Eckes, J. and Castro, M., 1993, The non-use of 9–
1–1. Private transport of trauma patients to a trauma centre. Pre-hospital Disaster
Medicine, 8(1), pp. 35–38.
Hunt, R.C., Brown, L.H., Cabinum, E.S., Whitley, T.W., Prasad, N.H., Owens, C.F., Jr.
and Mayo, C.E., Jr., 1995, Is ambulance transport time with lights and siren faster than
that without? Annals Emergency Medicine, 25(4), pp. 507–511.
US Department of Health & Human Services, 1991, Injury Control, Position Papers from
The Third National Injury Control Control Conference.
Kaweski, S.M., Sise, M.J, and Virgilio, R.W., 1990, The effect of pre-hospital fluids on
survival in trauma patients. Journal of Trauma, 30(10), pp. 1215–1219.
Koury, S.I., Moorer, L., Stone, C.K., Stapczynski, J.S. and Thomas, S.H., 1998, Air vs
ground transport and outcome in trauma patients requiring urgent operative
interventions. Pre-hospital Emergency Care, 2(4), pp. 289–292.
Kowalenoko, T., Stern, S., Dronen, S. and Wang, Xu, 1992, Improved outcome with
hypotensive resuscitation of uncontrolled hemorrhagic shock in a swine model.
Krausz, M.M., Bar-Ziv, M., Rabinovici, R. and Gross, D., 1992, Scoop and Run or
stabilize hemorrhagic shock with normal saline or small volume hypertonic saline?
Kreis, D.J., Jr.,1986, Journal of Trauma, 26(7).
Larsen, C.F., 1992, Pre-hospital Treatment of Injured and Critically lll Patients, presented
at 12th Congress of the International Association for Accident and Traffic Medicine,
Helsinki.
Lewis, F.R., 1986, Pre-hospital intravenous fluid therapy—physiological computer
modelling. Journal of Trauma, 26, pp. 804–811.
Maheshwari, J. and Mohan, D., 1989, Road Traffic Injuries in Delhi: A hospital based
study. Journal of Traffic Medicine, 17(3–4), pp. 23–27.
Martin, R.R., Bickell, W.H., Pepe, P.E., Burch, J.M. and Mattox, K.L., 1992, Prospective
evaluation of preoperative fluid resuscitation in hypotensive patients with penetrating
truncal injury: A preliminary report. Journal of Trauma, 33(3), pp. 354–362.
McSwain, N.E., 1995, Usefulness of physicians functioning as emergency medical
technicians (Editorial). Journal of Trauma, 39(6), pp.1027–1028.
Mock, C.N., Jurkovich, G.J., Amon-Kotei, D.N., Arreola-Risa, C. and Maier, R.V., 1998,
Trauma mortality patterns in three nations at different economic levels: Implications for
global trauma system development. Journal of Trauma, 44(5), pp. 804–814.
Moss, R.L., Kolaric, D. and Watts, A., 1993, Therapeutic agents utilized in urban/ rural
pre-hospital care. Pre-hospital Disaster Medicine, 8(2) pp. 161–164.
Okumura, M. et al., 1995, Hypovolemic hemorrhagic shock (An experimental study),
Rev. Hosp. Clin. Fac. Med. S. Paulo, 50(3), pp. 136–139.
Owens, T.M., Watson, W.C., Prough, D.S., Uchida ,T. and Kramer, G.C., 1995, Limiting
initial resuscitation of uncontrolled hemorrhage reduces internal bleeding and
subsequent volume requirements. Journal of Trauma, 39(2), pp. 200–209.
Pirrallo, R.G. and Swor, R.A., 1994, Characteristics of fatal ambulance crashes during
emergency and non-emergency operation. Pre-hospital Disaster Medicine, 9(2), pp.
125–132.
Pointer, J.E., 1988, Clinical characteristics of paramedics’ performance of endotracheal
intubation. Journal Emergency Medicine, 6, pp. 505–509.
Porter, J.M. and Ivatury, R.R., 1998, In search of the optimal end points of resuscitation
in trauma patients: A review. Journal of Trauma, 44(5) pp. 908–914.
Prasad, N.G., Brown, L.H., Ausband, S.C., Cooper-Spruill, O., Carroll, R.G. and
Whitley, T.W., 1994, Prehospital blood pressures: Inaccuracies caused by ambulance
noise? American Journal Emergency Medicine, 12(6), pp. 61 7–620.
Riou, B., Pansard, J.L., Lazard, T., Grenier, P. and Viars, P., 1991, Ventilatory effects of
medical antishock trousers in health volunteers. Journal of Trauma, 31(11), pp. 1495–
1502.
Rosemurgy, A.R. et al., 1993, Pre-hospital traumatic cardiac arrest: The cost of futility.
Sampalis, J.S. et al., 1993, Impact of on-site care, pre-hospital time, and level of in-
hospital care on survival in severely injured patients. Journal of Trauma, 34(2), pp.
252–261.
Sampalis, J.S., Lovoie, A.., Salas, M, Nikolis, A. and Williams, J.I., 1994, Determinants
of on-scene time in injured patients treated by physicians at the site. Pre-hospital
Disaster Medicine, 9(3), pp. 178–188.
Saunders, C.E. and Heye, C.J., 1994, Ambulance collisions in an urban environment.
Pre-hospital Disaster Medicine, 9(2), pp. 118–124.
Schiller, W.R., Knox, R., Zinnecker, H., Jeevanandam, M., Sayre, M., Burke, J., Young,
D.H., 1988, Effect of helicopter transport of trauma victims on survival in an urban
trauma centre. Journal of Trauma, 28(8), pp. 1127–11 34.
Sinclair, L.M. and Baker, M.D., 1991, Police involvement in pediatric pre-hospital care.
Paediatrics, 87(5), pp. 636–641.
Snooks, H.A., Nicholl, J.P., Brazier, J.E. and Lees-Mlanga, S., 1996, The costs and
benefits of helicopter emergency ambulance services in England and Wales. Journal of
Public Health Med., 18(1), pp. 67–77.
Soucy, D.M., Rude, M., Hsia, W.C., Hagedorn, F.N., Miner, H. and Shires, G.T., 1999,
The effects of varying fluid volume and rate of resuscitation during uncontrolled
hemorrhage. Journal of Trauma, 46(2), pp. 209–215.
Stochetti, N., Pagliarini, G., Gennari, M., Baldi, G., Banchini, E., Campari, M., Bachhi,
M. and Zuccoli, P., 1994, Trauma care in Italy; evidence of in-hospital preventable
deaths. Journal of Trauma, 36(3), pp. 401–405.
Swain, A., Dove, J. and Baker, H., 1990, ABC of Major Trauma, Trauma of the Spine
and Spinal Cord-I, British Medical Journal, 301, pp. 34–38.
Teach, S.J., Antosia, R.E., Lund, D.P. and Fleisher, G.R., 1995, Pre-hospital fluid therapy
in pediatric trauma patients. Pediatric Emergency Care, 11(1), pp. 5–8.
Trunkey, D.D., 1984, Is ALS necessary for pre-hospital trauma care. Journal of Trauma,
24, p. 86.
Trunkey, D.D., 1983, Trauma. Scientific American, 249, p. 28.
Tsai, A. and Kallsen, G., 1987, Epidemiology of paediatric pre-hospital care. Annacls of
Emergency Medicine, 16(3), p. 284.
Wilson, A. and DriscoII, P., 1990, ABC of Major Trauma, Transport of Injured Patients,
British Medical Journal, 301, pp. 658–662.
19
The Burden of Violence: An International
Public Health Perspective
Etienne Krug
INTRODUCTION
In recent times, we have all been exposed daily to the terrible images of human misery
caused by deadly conflicts in Kosovo, East Timor, Sierra Leone or the Democratic
Republic of Congo. The mass graves, mass rapes, and massive exodus are the most
visible part of the iceberg of violence. More discrete, but widespread, is the daily
suffering of children who are abused by their care givers, women victims of violent
partners, elderly persons maltreated by their children or care givers, girls subjected to
genital mutilation, or students who cannot attend school without being at risk of being
threatened, beaten or shot. During the last few years, Public health professionals are
increasingly taking a stand against accepting violence as an inevitable part of our modern
world and are taking actions to reduce its incidence.
Violence has been defined in many ways. The definition of violence commonly used
by the World Health Organization is that violence is ‘the intentional use of physical force
or power, threatened or actual, against another person or against oneself or against a
group of people, that results in or has a high likelihood of resulting in injury, death,
psychological harm, maldevelopment or deprivation’ (WHO, 1996).
While there is no universally accepted typology for violence, the groupings most
commonly used are: interpersonal, self-inflicted and organized violence (WHO, 1997)
Another way of categorizing violence is by dividing it into political, economic and social
violence. Each of these groups can in turn be divided into more specific areas such as
child abuse, intimate partner abuse, elderly abuse, youth violence, sexual violence, etc.
(Frege et al., 1995). It is useful to subdivide violence into these different sub-types to be
able to study specific risk factors or trends. Sub-types based on the motivation of the
perpetrator are useful because they highlight the need of different approaches for
prevention. However, whatever the typology used, it is also important to keep in mind
that there are many links between each of these types of violence: for example,
victimization of assault or abuse has been associated with a higher risk of suicide (Stark
et al., 1995). Victimization of child abuse has been described as a risk factor for
subsequent violent behaviour (Maxwell and Widom, 1996). Exposure to intimate partner
violence is a risk factor for future violent behaviour; and variables used to measure the
social acceptability of violence, such as the death penalty or the involvement in wars,
have been associated with the homicide rate in a society (Gartner, 1990). Reducing one
type of violence may therefore also contribute to decrease levels of other types of
The burden of violence 279
violence.
Traditionally, violence has been addressed by the legal area. Only during the last
decades has violence been increasingly recognized as a Public health problem in the
United States and a few other high income countries (Mercy et al., 1993; Rosenberg and
Fenley, 1991). In the rest of the world, violence, and injuries in general, is still absent
from the Public health agendas.
The added value of the Public health approach is that it is interdisciplinary, science-
based, and focused on prevention (Frege et al., 1995). Public health is at the intersection
of medicine, epidemiology, sociology, criminology and several other fields. Bringing
together the strengths and approaches of each of these fields, allows Public health to be
innovative. Public health is guided by science. Interventions are based on the diligent
study of risk and protective factors. Interventions are, or should be, evaluated before
implemented on a wider scale. It is its focus on primary and secondary prevention,
however, that is the main strength of the Public health approach to violence. As a
complement to the more traditional judicial approach that is mainly based on punishment
of the perpetrators, a method for tertiary prevention, Public health proposes primary and
secondary prevention. Public health believes that by addressing its social, economic or
other causes, a considerable proportion of violence will be prevented.
THE GLOBAL BURDEN
reducing one type of violence may therefore also contribute to decrease levels of
other types of violence
An increasing number of countries are reporting information on their annual mortality to

the World Health Organization (WHO). In 1999, WHO has compiled the data that has
been provided by member states and has made estimates to cover regions for which less
detailed information exists. Some of the results of these analyses were presented in the
World Health Report 1999, others are presented in this paper
Mortality
It is estimated that in 1998, 2.3 million people died from violence: 42 per cent of these
deaths were classified as suicide, 32 per cent as homicide and 26 per cent as war-related
(WHO, 1999) Worldwide the homicide rate is 12.2 per 100,000 population. There are
considerable regional differences in the relative importance of these type of deaths: for
example, in China, there are 7 suicides for each homicide. However in Sub-Saharan
Africa, and in Latin America and the Caribbean there are 13 and 5 homicides for each
suicide, respectively (WHO, 1999).
A shocking statistic is that in the beginning of the third millennium, 4 per cent of all
persons who die in the world are intentionally killed by another person or by themselves!
Despite the large burden of infectious diseases, almost 5 per cent of all deaths in low and
middle income countries in 1998 were due to intentional injuries. In the high income
countries, 2 per cent of all deaths are due to intentional injuries. In Sub-Saharan Africa,
among males, the proportion of all deaths due to violence reaches almost 9 per cent.
China is the only ‘region’ where the proportion of deaths due to violence is greater for
females than for males. This is due to the high suicide rate among females. In China,
almost half (45.5 per cent) of all injury deaths among females are suicides. This is more
than double the proportion in the Emerging Market Economies—21.7 per cent (Murray
and Lopez, 1996)
Like for many other health problems, violence is not distributed evenly among income
groups. The homicide rate for countries in the low and middle income group is estimated
to be three times higher than the homicide rate in countries from the high income group
(13.6 compared to 4.3). The homicide rate in Africa (48.8 per 100 000) is 11 times higher
than the homicide rate in high income countries. Interestingly however, the gap
difference in suicide rates between high and middle/low income countries is much
smaller (12.5 and 17.4 per 100 000).
Disability adjusted life years (DALYs) lost

In 1990, war, self-directed, and interpersonal violence were ranked 16, 17 and 19th,
respectively, among the leading causes of the world’s disability adjusted life years
(DALYs) lost, the numbers of years of life lost from premature death combined with the
loss of health from disability (Murray and Lopez, 1996). It is estimated that in 2020, war
will rank 8th, interpersonal violence 12th and self-directed violence 14th.
War is among the 15 causes of DALYs lost for persons age 0 to 44. Homicide is the
4th leading cause of DALYs lost in the age group 15 to 44, the most economically
productive group. In the high income group, homicide and violence rank among the 15
leading causes of DALYs from age 0 to 44.
Other types of morbidity

It is important to realize that the DALYs rankings are certainly an underestimation of
what the global burden of violence really is. To calculate the DALYs attributable to
violence, war or self inflicted injuries, only fractures, wounds, amputations, crushing,
burns and poisonings were taken into account (Murray and Lopez, 1996). However,
victimization of violence also causes a large number of other health consequences.
Victimization by family or intimate partner violence can, for example, lead to depression,
suicidal behaviour, eating disorders, smoking, or alcohol or drug use (Kaplan et al., 1998;
Kaslow et al., 1998; Stark et al., 1995; Pederson et al., 1998). In addition to these health
consequences, victims of sexual violence can also suffer from unwanted pregnancy or
sexually transmitted diseases (Holmes et al., 1996). Populations exposed to political
violence or conflict often suffer malnutrition or from increases in infectious diseases
(Kakar et al., 1996; Toole, 1997).
In the United States alone, every year, almost 3 million children, are abused (National
Research Council/Chalk and King, 1998). WHO recently estimated that annually 40
million children are suffering from child abuse worldwide. Studies focusing on sexual
abuse of children have been conducted in many countries and have shown clearly that the
problem crosses boundaries and cultures. Although, because of the methodological and
definitional differences in the way the studies were conducted, countries should not be
compared to each other, estimates of sexual abuse compiled in a study of 20 countries
from 4 continents, ranged from 7 per cent to 36 per cent for girls and 3 per cent to 29 per
cent for boys (Finkelhor, 1993). The WHO Database on Violence against Women, a
global compilation of statistics from research reports and documents, leads us to the same
conclusion, showing the universality of violence against women (WHO, 1999). Results
from a sample of studies with similar epidemiological characteristics and representing
countries from all continents and income groups, showed that depending on the country,
from 5 per cent to 67 per cent of women interviewed, reported having been physically
abused by an intimate partner during their lifetime. The percentage of women who
reported having suffered from attempted or completed rape by an intimate partner ranged
from 6 per cent to 48.5 per cent.
PREVENTING VIOLENCE
4 per cent of all persons who die in the world are intentionally killed by another
person or by themselves!
Despite these staggering statistics, the data also shows that violence is preventable and
should not be considered as an inevitable part of society. For example, while some
countries have been involved in numerous wars, other countries have not been at war for
decades. Firearm death rates in Asia are almost 100 times lower than in the Americas
(Krug et al., 1998). Suicide rates in Africa are much lower than in high income countries.
We need to take the time, to learn from these important cross-cultural differences to
prevent further unnecessary premature deaths through the development of sound
interventions and policies.
Research has allowed to establish some principles for interventions in the area of youth
and domestic violence prevention. These principles were developed based on careful
research into the risk and protective factors and on evaluation of different types of
interventions. At this stage most of the research has been conducted in western societies
and little is known about the universality of these principles. What is certain however, is
that there are no easy remedies to violence. Violence is a complex issue with complex
causes and only a number of interventions implemented at the same time at different
levels will allow for notable improvements.
Another lesson is that interventions need to occur at early stages. To be efficient,
interventions need to target children as early as possible and before adolescence. Even
behaviours that occur during adolescence, like school violence, or even adulthood, such
as child or intimate partner abuse, need to be prevented by intervening during childhood.
Interventions can be conducted at different levels: the individual, the family, peer
groups or the larger community (Dahlberg, 1998). Examples of interventions that target
the individual are programmes that focus on anger management or self-control. These
interventions aim at changing an individual’s attitudes and beliefs. Family level
interventions focus on improving the way parents supervise and discipline their children.
Training programmes for parents or home visitation programmes are used to prevent
child abuse. Programmes that focus on changing group interactions are often peer
mediation programmes. In these programmes, youth are assisting other youth by
mediating conflicts or serving as positive role models. At the community level,
interventions focusing on improving neighbourhoods or school settings can be
undertaken. These interventions can target the physical environment (for example,
increase lighting), improve school security (for example, use of metal detectors) or also
focus on behavioural changes such as anti-violence awareness campaigns.
In addition to these primary prevention efforts, secondary prevention efforts are needed
to reduce the long term impact of violence on victims and their families. Adequate
emergency referral systems and pre-hospital and hospital care are needed for the
treatment of injuries. In many parts of the world, this is still a problem. For example, it is
estimated that 2/3 of landmine victims die because of lack of appropriate emergency
response system. Psychological support also needs to be given to victims of the different
types of violence. For example, support groups are often very helpful for victims of
intimate partner violence or sexual abuse.
A WORLD REPORT ON VIOLENCE AND HEALTH
During recent years, some important developments, have taken place in the international
arena. These developments will contribute to including violence in the national Public
health agendas. In view of what it described as a dramatic increase in the incidence of
intentional injuries, the Forty-Ninth World Health Assembly adopted resolution WHA
49.25 in 1996, declaring violence a leading worldwide Public health problem (WHO,
1996). The Resolution urged member states to assess and develop science-based solutions
to the problem. In 1997, a Plan of Action to Reduce Violence was prepared, based on the
four steps of the Public health approach, moving from problem to solution. The first step
is to determine the magnitude, scope, and characteristics of the problem. The second step
is to study the factors that increase the risk of disease, injury, or disability, and determine
which factors are potentially modifiable. The third step is to assess what can be done to
prevent the problem by using the information about causes and risk factors to design,
pilot test, and evaluate interventions. The final step is to implement the most promising
interventions on a broad scale. The WHO collaborating centres were instrumental in
assisting WHO to set the stage for these two milestones.
The WHO programme for Violence and Injury Prevention is in the process of building
upon the WHA resolution and Plan of Action. In collaboration with several centres and
experts from around the world, WHO is producing the World Report on Violence and
Health. The goals of this document are to raise worldwide awareness about the Public
health aspects of violence and to highlight the contributions of Public health to
understanding and responding to the problem of violence. More specific objectives of the
document are 1) to describe the magnitude and impact of violence cross-nationally; 2) to
elucidate cross-national patterns of violence; 3) to provide a baseline for measuring
change and progress; 4) to summarize existing information on risk factors, prevention
approaches, and policy responses; 5) to provide directions for future research; and 6) to
make recommendations for future action in Public health.
WHO recently estimated that annually 40 million children are suffering from
child abuse worldwide
The primary audience for the report will be decision makers, Public health officials and
practitioners, and journalists. A number of international experts have been invited to
write chapters based on the data and on current relevant issues. The report will be
released in January 2001 at the WHO Executive Board meeting and will be widely
distributed to governmental and non-governmental agencies. The report will also be made
available on the World Wide Web in a format that will allow easy use of the data by
researchers. We hope that the report will be another milestone in developing the global
Public health strategy for violence prevention.
FUTURE DIRECTIONS:
The World Report on Violence and Health will include clear recommendations for future
directions, targeting especially those countries that are willing to complement the work
being done by their judicial system by starting to implement the public health approach to
violence prevention and control. Some of these recommendations are summarized as
follows:
• Intersectoriality
Violence is such a complex issue and involves so many different sectors that
only a multi-sectorial approach will allow us to address the issue. Intersectoral
collaboration is needed. Many of us work with psychologists, sociologists,
medical doctors and criminologists, and many of us also work in the field of
Public health. A step further should be taken to bring together Public health
officials from ministries, agencies and universities with representatives from
organizations representing justice, police, urban planning, human rights and
other fields. Specific partnerships need to be developed at local, national and
international levels: For example, the concepts and strategies developed by
Public health should be promoted through partnerships with the media. The
media can help us to change people’s attitudes and eventually behaviours, and
we can learn from the experience of other health issues, like tobacco or
reproductive health. Well-built communication strategies could help us change
perceptions about issues like domestic violence or change the image associated
with owning firearms.
Partnerships are also needed between Public health and related agencies and
fields. For example, Public health, international diplomacy and human rights
share many concerns and issues, in particular a prevention based approach. In a
recent article, the Swedish Foreign Minister, issued a call that resembled closely
the Public health approach: ‘We need a global structure for conflict prevention.
We must identify structural risk factors, shed light on root causes, and design
efficient means to find peaceful avenues to developments that may otherwise
lead to violence’ (Lundh, 1999). Public health efforts could strengthen the
activities developed in diplomacy or human rights by sharing its tools around
which its science and data-based approach are structured. Sound
epidemiological principles would strengthen the basis and the monitoring and
evaluation of the implementation of international treaties. Public health could
gain in strength by building upon the human rights mechanisms and using its
international tools.
• National and local strategies
National and local leaders should call upon relevant organizations to jointly
develop plans, strategies and interventions to address violence. Although this
approach has been taken on a few occasions by a handful of countries, it is still a
rare phenomenon. Workers in different disciplines should join forces to develop
joint national and local strategies that would include the sharing of data, for
example between justice and health, the definition of terms and the joint
implementation and evaluation of interventions. Well-defined goals, objectives,
strategies and time lines would be instrumental in reducing violence.
• Address the root causes of violence
As stated earlier, the causes of violence are multiple and complex. In fact,
some of the causes, such as the gap between rich and poor, seem sometimes far
removed from the area of Public health or even, often seem to be out of reach
because of the magnitude of the problem. However, it is important that Public
health contributes to research and other efforts in trying to address root causes
such as poverty and the social acceptability of violence.
• Global information sharing
Although we spend a considerable amount of time writing and
communicating about our work, we still spend too much time re-inventing the
wheel. This is especially true at the international level. We should use every
opportunity to share our experiences across borders, especially with those who
do not always have access to scientific journals or international conferences. The
World Health Organization is already playing an important role in this respect
and will hopefully be able to continue to increase its activities in the
development of guidelines and recommendations for best practices.
• Joining forces
As stated earlier, separating violence into different sub-types, such as child
abuse, intimate partner and youth violence, allows for in depth studying of the
specific risk factors or prevention programmes. However, it often also leads to a
situation where experts in the different sub-types of violence have become so
specialized that they do not communicate with or read the work of experts in any
of the other sub-types. This over-fragmentation leads to isolation and some of
the common factors or links between the different types of violence are
overlooked. Worse, in some cases where joint efforts for awareness raising,
resource mobilization or priority setting would have been more powerful, these
efforts are weakened by the absence of unified vision and strategies. Artificial
barriers should be eliminated as much as possible to allow for a joint, stronger,
integrated approach.
• Distribution of resources
We should strive for a distribution of resources reflecting the burden of
injuries. At present, the donor community does not always allocate resources
toward the problems or places with the major burden. For example, although
victims of landmines and their families deserve all the support we can give, is it
cost-effective to concentrate such a large proportion of resources on a group of
victims that represent 0.3 per cent of the total number of injury deaths in the
world? Although child abuse and intimate partner violence are widespread and
cause terrible damage, a very large proportion of victims and perpetrators of
violence are adolescent and young males, who also need our attention.
firearm death rates in Asia are almost 100 times lower than in the
Americas
• Evidence base
Building our knowledge and interventions on sound epidemiological data is
the strength of Public health. All countries should start by studying the
magnitude and characteristics of the problem in its own reality. Interventions
should be based on culturally relevant data and evaluations should be
systematically incorporated in programmes.
• Be patient.
Smallpox was not eradicated in a few years and the fight against TB, AIDS
and many other health problems has already taken decades. With all of its
complexity, violence will certainly not be eliminated in the near future. The
involvement of the Public health community in violence prevention and control
is still at its very early stage. Plutarch is quoted to have said that ‘Perseverance
is more prevailing than violence; and many things that cannot be overcome
when they are taken together, yield themselves up when taken little by little’
REFERENCES
Dahlberg, L.L., 1998, Youth violence in the United States. Major trends, risk factors, and
prevention approaches. American Journal of Preventive Medicine. 14(4), pp.259–72.
Finkelhor, D., 1993, The international epidemiology of child sexual abuse. Child Abuse
and Neglect, 18(5), pp. 409–417.
Foege, W.H., Rosenberg, M.L., and Mercy, J.A., 1995, Public health and violence
prevention, Current Issues in Public Health, 1, pp. 2–9.
Gartner, R., 1990, The victims of homicide: a temporal and crossnational comparison,
American Sociological Review, 55, pp. 92–106
Holmes, M.M. et al., 1996, Rape-related pregnancy: estimates and descriptive
characteristics from a national sample of women. American Journal of Obstetrics and
Gynecology, 175(2), pp.320–325.
Kakar, F., Bassani, F., Romer, C.J., and Gunn, S.W.A., 1996, The consequence of
landmines on public health. Prehospital Disaster Medicine, 11, pp. 41–45.
Kaplan, S.J. et al., 1998, Adolescent physical abuse: risk for adolescent psychiatric
disorders. American Journal of Psychiatry, 155(7), pp. 954–959.
Kaslow, N.K. et al., 1998, Factors that mediate and moderate the link between partner
abuse and suicidal behavior in African American women. Journal of Consulting and
Clinical Psychology, 66(3), pp. 533–540.
Krug, E.G., Powell, K.E. and Dahlberg, L.L., 1998, Firearm-Related Deaths in the United
States and 35 Other High-and Upper-Middle-lncome Countries, International Journal
of Epidemiology, 27(2), pp. 214–221.
Lindh, A., 1999, Create a World-wide Culture for Conflict Prevention, International
Herald Tribune, 18 September, p. 6.
Maxfield, M.G. and Widom, C.S., 1996, The cycle of violence. Revisited 6 years later,
Archives of Pediatric and Adolescent Medicine, Apr; 150(4), pp. 390–395.
Mercy, J.A., Rosenberg, M.L., Powell, K.E., Broome, C.V., and Roper, W.L., 1993,
Public health policy for preventing violence, Health Affairs, 12, pp. 7–29.
Murray, C.J. and Lopez, A.D. (eds), 1996, The Global Burden of Disease. (Cambridge,
Mass.: Harvard University Press).
National Research Council, 1998, Violence in Families: Assessing Prevention and
Treatment Programs. R.Chalk and PA.King, editors, (Washington, DC: National
Academy Press).
Pederson, W. et al., 1998, Alcohol and sexual victimization: A longitudinal study of
Norwegian girls. Addiction, 91(4), pp. 565–581.
Rosenberg, M.L. and Fenley, M.A. editors, 1991, Violence in America: A Public Health
Approach. (New York: Oxford University Press).
Stark, E. et al., 1995, Killing the beast within: woman battering and female suicidality.
International Journal of Health Services, 25(1), pp. 43–64.
Toole, M.J., 1997, Complex emergencies: refugee and other populations, In The Public
Health Consequences of Disasters , Edited by Noji E., pp. 419–442. (New York:
Oxford University Press).
World Health Assembly, 1996, Prevention of Violence: Public Health Priority. (Geneva.
World Health Organization), 1996 (WHA 49, 25).
World Health Assembly. 1997, Violence: A Public Health Priority, (Geneva: World
Health Organization), 1997(EB).
World Health Organization, 1996, Violence: A Public Health Priority. Working
document EHA/SPI/POA. 2 December.
World Health Organization, 1999, The World Health Report 1999. (Geneva: World
Health Organization).
World Health Organization, 1999, The World Health Report 1999 database, (Geneva:
World Health Organization).
World Health Organization, 1999, http://wwwstage.who.int/violence_injury_prevention/
New Folder/pages/who prevalence of physical viole.htm. October 1.
20
A New Vision for Suicide Prevention: The
Public Health Approach
Mark L.Rosenberg and James A.Mercy
INTRODUCTION
The public health approach differs from traditional approaches to suicide prevention.
Suicide has traditionally been seen as a mental health problem. In this view, suicide was
seen as caused by mental illness, and prevention consisted of identifying and treating
persons with the types of mental illness that led to suicide. These illnesses usually include
depression, manic depressive illness, schizophrenia, and serious character and anxiety
disorders. The traditional approach has also focused on treating patients with mental
illness one at a time as they come for mental health care in clinical settings, with
treatment provided by mental health clinicians. The public health approach is based on
three fundamental principles: it is focused on prevention, based on science, and inclusive
in scope. Suicide is an important global health problem and takes a large toll in almost
every country. The examples in this paper are drawn from the US, but we believe that the
public health approach to suicide prevention advocated here will be useful around the
world.
BACKGROUND: SUICIDE IS A WIDESPREAD PROBLEM IN THE US

AND GLOBALLY
The starting point for public health involvement in suicide is, of course, its impact on
physical and emotional health. In 1996, suicide was the ninth leading cause of death in
the US (Kachur et al., 1992). Each year suicide claims more than 30,000 lives and about
80 per cent of those who die are males. Surveillance of adolescent suicidal behaviour has
shown us several alarming trends (Rosenberg et al., 1987). From 1952 through 1992
suicide rates among adolescents and young adults nearly tripled and from 1980–1992, the
rate of suicide among persons aged 15–19 years has increased by 28.3 per cent (81 per
cent of this increase was due to the increase in firearm-related suicides) and among
persons aged 10–14 years the suicide rates increased 120 per cent (CDC, 1994; 1995;
1997). From 1980–1992, the rates among African-American males aged 15–19 years
increased 165.3 per cent (CDC, 1998). In addition, a 1993 nationwide survey of high
school students found that in the 12 months preceding the survey, one-fourth had
seriously considered suicide and one in twelve had attempted suicide.
Though suicide rates are rising among adolescents, the rates continue to be highest
among persons aged 65 years and older (CDC, 1996). From 1980 to 1992, overall suicide
rates increased for persons in only two age groups: 5–19 years and those above 65 years.
The ten-year period 1980–1990, was the first decade since the forties that the suicide rate
for older US residents rose instead of declined. In 1992, persons aged over 65 years
accounted for 13 per cent of the population but over one-fifth of all suicides. Because
older persons are the fastest growing age group in the United States, the number of
suicides in this age group will probably continue to increase.
Information on the precise global health burden of suicide is not available but estimates
of the relative and absolute contribution of suicidal behaviour to the global health burden
have recently become available through a comprehensive assessment of mortality and
disability due to disease and injuries in 1990 and projected to 2020 (Murray and Lopez,
1996). The burden of suicide is quantified by measures of two general types of health
consequences: (1) premature mortality as measured by numbers, rates, and years of life
lost due to suicide, and (2) disability (sensory, cognitive functioning, pain, affective state,
etc.) as indicated by a new measure called Disability adjusted life years Lost (DALYs).
The DALY is a new method of measuring disease burden; it is based on a
quantification of disability that is derived from the incidence, duration, and severity of
the morbidity and complications associated with specific conditions (Murray, 1996). The
method was developed by the World Health Organization and the World Bank to
overcome the limitations of using mortality as the sole measure of health impact. While
the DALY measure is an advance in assessing the burden of disease, it has limited
application. The information needed to calculate the DALY is often incomplete,
particularly in many developing countries, and suitable indicators to measure such factors
as the psychological consequences of suicidal behaviour have not yet been developed or
made generally available. Nevertheless, it is useful as a crude indicator of the health
impact of nonfatal suicidal behaviour across different regions of the world and relative to
other health problems (For details on how DALYs are calculated see Tables 1, 2 and 3 at
the end of the paper; Murray, 1996).
for many years people feared that asking people about suicidal thoughts might
actually influence them to commit suicide
THE PUBLIC HEALTH APPROACH IS FOCUSED ON PREVENTION
The public health focus on prevention means that the goal is to prevent suicides in the
future, a goal that can be measured by reductions in the suicide rate. The focus on the
future means that it is important to think about persons who may not yet be at high risk
for suicide. It is important to think about persons who are still young, or about older
persons who still have their important social supports but might be at high risk later when
they lose these supports. The focus on prevention also means that we must be activists.
Many people are quite fatalistic about suicide. They believe that suicides have always
occurred throughout recorded history and they will continue to occur and there is nothing
we can do about this problem, no way we can prevent it. It is important to change this
view of suicide. Public health takes an activist stance and says that this is a cause and
effect world: if we can understand the causes, we can affect the outcomes. If we
A new vision for suicide prevention 289
understand the causes of suicide, we can prevent suicides in the future. One pre-eminent
public health practitioners in the world, William Foege, said that ‘the most important
problem facing public health today is not the problem of HIV/AIDS, not the problem of
substance abuse, and not violence. It is the problem of fatalism’.
It is also quite important to translate research findings and scientific understanding into
programmes that are actually delivered and information that is easily understood and used
by the people who need to use it (O’Carroll, 1990; O’Carroll et al., 1991; O’Carroll and
Silverman, 1994; Potter et al., 1995; CDC, 1998). The focus on prevention does not in
any way diminish the importance of providing care for victims and survivors.
It is usually helpful to describe the suicide prevention approaches and programmes
actually in place in a country, state or community. In 1992, CDC published a document
entitled Youth Suicide Prevention Programs: A Resource Guide, as part of its effort to
evaluate specific interventions (CDC, 1992). This publication outlined the eight major
suicide prevention strategies being used by the most successful youth suicide prevention
programmes in the US. The strategies include gatekeeper training, suicide education,
screening programmes, peer support programmes, crisis centres and hotlines, and
‘intervention after a suicide’ programmes. The guide serves as an information resource
for those interested in learning about the types of prevention activities in the field, and,
most importantly, indicates two major preventive strategies that seemed promising. The
two principal strategies are to reduce the prevalence of risk factors and to identify and
intervene with high-risk persons. Interventions should be developed earlier on the
pathway toward suicide. For example, if the use of alcohol increases the risk for suicide,
then young children at risk for alcohol abuse should be given the skills to resist alcohol
when they get older. If geographic mobility and lack of social connections increases the
risk for suicide, then employers should work to promote geographic stability or provide
counselling resources and parental education when families are forced to move for their
work.
Another way to intervene earlier along the pathway and to reduce risk factors would be
to involve children earlier in preventive programmes. Children and youth may be more
educable and behaviours and skills learned earlier tend to endure. Children can also be
taught skills to equip them to deal with many different types of violence. New methods of
delivering proven effective interventions also need to be developed. Involvement in
suicide prevention should extend beyond the traditionally-involved mental health
professionals. Schools could administer screening programmes to identify children and
youth at high risk for suicide. And communities could become involved in decreasing
access to lethal means of suicide. Agricultural agents, for example, might become
involved in eliminating the use of pesticides that are frequently swallowed by suicidal
young women in rural areas (CDC, 1990).
Another implication of intervening earlier is that we need to target increasingly larger
groups as we intervene earlier because we have less information in hand to help us
identify those at the highest risk. A useful approach involves dividing target groups for
suicide prevention activities into three levels of risk. The lowest level of risk group
targeted is the general population and programmes aimed at this broad range of persons
are called ‘universal’. The next highest level of risk might be those in the population who
have some indication that they are at higher-than-average risk, but not so high that an
attempt is more likely than not. Programmes aimed at this group would be called ‘select’.
Finally, programmes for those at the very highest level of risk, would be called
‘indicated’.
THE PUBLIC HEALTH APPROACH IS BASED ON SCIENCE
nowhere is there a greater need to develop an evidence-based approach to suicide

prevention than in evaluating and demonstrating ways to prevent suicidal
behaviour especially at the community and population levels
The public health approach uses science as the basis for action. A simple model for how
public health applies science to suicide prevention is based on four steps, each step
characterized by a question. First, we ask ‘What is the problem?’ This leads to try to
answer the questions a good journalist would ask: Who are the people involved? What
methods do they use? When and where does it happen? We use epidemiological data to
answer these questions, looking at tens, hundreds, or thousands of cases. In public health,
the collection of such data to describe and track a problem is called surveillance. For
many years people feared that asking people about suicidal thoughts might actually
influence them to commit suicide. This has not been shown to be a real risk. We need to
do a better job of describing and tracking the problem of suicidal behaviour (Birkhead et
al., 1993). Work is now going on to set up monitoring systems that will tell us how often
suicidal behaviour occurs. To do this we need clear definitions of the various types of
suicidal behaviours and we need to test these for validity and reliability (O’Carroll,
1989). We need definitions that will apply to both children and adults, and we need to
differentiate between suicide mortality, i.e. completed suicides, and suicide morbidity
(Rosenberg et al., 1988; CDC, 1993). For example, we can ask whether a person ever felt
so bad that they thought about hurting themselves seriously, whether they ever actually
made a plan to hurt themselves, whether they ever did hurt themselves, and whether they
ever hurt themselves so seriously that they required medical attention. Our monitoring
systems can also monitor risk factors for suicidal behaviours and tell us who is at greatest
risk, and whether the problem is improving or worsening over time.
The second step of the public health approach asks: ‘What are the causes?’ What are
the risk factors that increase the likelihood a given person will commit suicide?
(Hirschfeld and Davidson, 1988) What kind of risk factors might be inherent the make-up
of the individual, such as a susceptibility to depression or alcoholism? What kind of risk
factors might there be in the individual’s social environment, such as a family history of
suicide, domestic violence or social isolation of young brides (Smith et al., 1986; Smith
et al., 1988; Moscicki, 1995; Kellerman et al., 1992; Wintemute et al., 1999; Potter et al.,
1998; Gould et al., 1980). What kind of risk factors might there be in the physical
environment? These risk factors would include firearms, pesticides, or high bridges. We
need to increase our knowledge of the causes of suicidal behaviour. For example,
research on risk factors has shown that substance abuse, mental illnesses, family
influences, such as a history of violence or family disruption, media influences, and
access to firearms are potential reasons for the increase in adolescent suicides (DHHS,
1989; Davidson and Gould, 1989; Davidson et al., 1988; Brent et al., 1989; Brent et al.,
1989; Kellerman et al., 1993; Cummings et al., 1997; Wintemute et al., 1999; Mercy et
al., 1993). Frequent geographic moves, alcohol use and exposure to an acquaintance or
relative’s suicides may also play a part in a young person’s suicide and research is
currently being conducted to answer these questions. Impulsive and aggressive
behaviours may also play a more important role in suicide than had been previously
recognized (Tom Simon, personal communication, Oct. 1999). Characteristics for suicide
among older persons also differ from those among younger persons. The risk factors for
suicide among older persons include alcohol abuse, depression, greater use of highly
lethal methods, and social isolation. In addition, older persons make fewer attempts per
completed suicide, have often visited a health care provider shortly before their death,
and have more physical illnesses and affective disorders. New research on risk factors
should describe differences between risks for mortality and morbidity, and should help to
understand risk and protective factors in diverse populations. We also need to better
characterize risk and protective factors and their interaction, and we need to look at the
similarities between risk factors for interpersonal and self-directed violence (Mercy et al.,
1993).
The third step of the public health approach is to take what has been learned about
patterns of the problem, and about the causes of the problem and then generate and
critically evaluate interventions. Perhaps nowhere is there a greater need to develop an
evidence-based approach to suicide prevention than in evaluating and demonstrating
ways to prevent suicidal behaviour especially at the community and population levels.
We need to learn the effectiveness of specific interventions in preventing suicide, to find
how to combine specific interventions into effective programmes, and to show the
effectiveness of suicide prevention programmes and policies in a wide range of different
community settings.
The last step in the public health approach to suicide prevention focuses on the
implementation of prevention programmes and asks the question ‘How do you do it?’
How do you mobilize resources at local, state, national, regional, and global levels to put
interventions in place and sustain them over time? This step requires that we understand
how to mobilize the ‘political will’; how to communicate scientific information about
suicide prevention; and how to integrate a wide array of suicide prevention and support
services into health, education, justice, social service, military, and other sectors.
Scientific evidence is needed and will continue to be needed because suicide is an
ever-changing story: it changes as we understand more about it, and there are real
changes in the phenomenon of suicide as people, the social environment, and the physical
environment all change.
THE PUBLIC HEALTH APPROACH IS INCLUSIVE
Suicide prevention has long been a concern of psychiatrists and others who treat mental
illnesses, since mental illness is an important risk factor for suicide. However, the clinical
treatment of mental illnesses is insufficient in itself, as clearly demonstrated by rising
suicide rates among teenagers, young adults, and the elderly. The clinical approach
should be supplemented by a broader, Public health approach, for several reasons.
Suicide is not solely determined by mental illness

Mental illness is not the only relevant risk factor in the causal mechanism leading to
suicide. Research has consistently pointed to the importance of many other factors
unrelated to mental illness as important determinants of suicide, such as accessibility to
firearms, geographic mobility, parental loss, family disruption, being a friend or family
member of a suicide victim, alcohol and drug use, and social isolation. If suicide
prevention efforts focus solely on mental illness and ignore the contribution of other
factors that contribute to suicide, many lives will be lost that might otherwise be
prevented.
mental illness is not the only relevant risk factor in the causal mechanism leading
to suicide…If suicide prevention efforts focus solely on mental illness and ignore
the contribution of other factors that contribute to suicide, many lives will be lost
that might otherwise be prevented
For those suicides for which mental illness is the key risk factor, it is inappropriate to
confine prevention efforts to the mental health sector.
Mental health practitioners can only accomplish the important clinical work they do
when patients come to see them. There are many factors, however, that determine
whether suicidal patients seek help from mental health professionals. The most striking
example of progress in this area is the training of ‘gatekeepers’ across a variety of
disciplines (e.g., education, general medicine). These gatekeepers often play a critical
role in facilitating proper care by mental health professionals, but the training of these
gatekeepers is a task which is in large part beyond the scope of mental health systems.
This is but one example in which both the broader perspective and the multidisciplinary
tradition associated with public health practice at CDC can uniquely contribute to suicide
prevention.
An effective approach to suicide prevention requires the collaboration of individuals in
Public health, mental health, medicine, education, and social services in both the public
and private sectors.
One of the most important developments in the field of suicidology in the last 30 years
has been the recognition that suicide prevention cannot be accomplished solely through
the efforts of one societal sector. This realization was clearly recognized by the
Secretary’s Task Force on Youth Suicide and highlighted among the major
recommendations of this body (DHHS, 1989). Collaborative, multisectoral work, and a
clear focus on suicide prevention put public health in the best position to coordinate and
elicit the valuable contributions to suicide prevention that can be made by a wide variety
of agencies, organizations, and professionals.
Although mental illness is an important risk factor for suicide across all age groups,
mental illness plays its least important role in the aetiology of suicide among youth aged
15 to 24 years, the group in which suicide rates have been increasing most rapidly.
Research has shown that mental illness, particularly depression, may be less frequently
associated with suicide among young people than among older adults. In fact, only a
small proportion of teen suicides occur among teenagers with manic depressive or
schizophrenic psychosis because these conditions are relatively rare. Consequently,
prevention strategies which focus on mental illness are insufficient for the prevention of
youth suicide.
The principle of inclusivity also suggests that we need to have a broad view of violence
and recognize the linkage of suicide to other types of violence. Efforts directed
simultaneously toward multiple types of violence may help to reduce suicide rates more
quickly than efforts focused on suicide alone. For example, victims of child abuse face a
higher risk of suicide than do non-victims. Thus, if we can prevent child abuse, we can
reduce the risk of suicide by giving many more people lives free of abuse. Similarly,
since women who were sexually assaulted face a higher risk of suicide, by reducing the
incidence of sexual assault we can simultaneously decrease the risk of suicide for many
women.
An inclusive approach to suicide prevention also suggests that we should be effective
in communicating scientific information about suicide prevention. Governmental
agencies should work with state and local partners to conduct educational, training, and
public awareness activities to disseminate scientific information about suicide prevention
to the public, policy makers, health departments, community-based organizations and
other entities. Abroad and inclusive Public health approach should address the reporting
of suicide. For example, such guidelines were developed from a workshop composed of
researchers, clinicians, Public health officials, educators and news media professionals.
An inclusive approach will foster coalitions that integrate a wide array of suicide
prevention and support services. Integrative leadership is the operative principle. Such
private/public partnerships, coalitions, and networks can bring together national
organizations, federal agencies, foundations, businesses, state and local health
departments, community-based organizations and others to prevent suicide.
THE NEW VISION IS POSITIVE AND OPTIMISTIC
The most important value of the public health approach is the underlying notion that we
can do better. This is a cause and effect world and if we can understand the causes we
canchange the effects. We must remember that the ultimate goal is prevention (United
States Public Health Services, 1999). If we focus on prevention, use the answers that
science can give us, and are inclusive in our approach we could set ourselves the goal of
cutting suicide rates in half-and we can do that. Wherever we are.
CALCULATING DALYs
Table 1
States or territories by demographic region
Demographically developed regions

Established market economies (EME)
Andorra, Australia, Austria, Belgium, Bermuda, Canada, Channel Islands, Denmark, Faeroe
Islands, Finland, France, Germany, Gibraltar, Greece, Greenland, Holy See, Iceland, Ireland,
Isle of Man, Italy, Japan, Liechtenstein, Luxembourg, Monaco, Netherlands, New Zealand,
Norway, Portugal, San Marino, Spain, St. Pierre and Miquelon, Sweden, Switzerland, United
Kingdom, United States.
Formerly socialist economies of Europe (FSE)
Albania, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia,
Hungary, Latvia, Lithuania, The Former Yugoslav Republic of Macedonia,, Moldova, Poland,
Romania, Russian Federation, Slovakia, Solvenia, Ukraine, Yugoslavia
Demographically developing regions
India
China
Other Asia and Islands (OAl)
American Samoa, Bangladesh, Bhutan, Brunei Darussalam, Cambodia, Cook Islands, Federated
States of Micronesia, Fiji, French Polynesia, Guam, Hong Kong, Indonesia, Johnston Island,
Kiribati, Democratic People’s Republic of Korea, Republic of Korea, Lao People’s Democratic
Republic, Macao, Malaysia, Maldives, Marshall Islands, Mauritius, Midway Island, Mongolia,
Myanmar, Nauru, Nepal, New Caledonia, Niue, Northern Mariana Islands, Palau, Papua New
Guinea, Philippines, Pitcairn Island, Reunion, Seychelles, Singapore, Solomon Islands, Sri
Lanka, Taiwan, Thailand, Tokelau Island, Tonga, Tuvalu, Vanuatu, Viet Nam, Wake Island,
Wallis and Futuna Islands, Western Samoa
Sub-Saharan Africa (SSA)
Angola, Ascension, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, Central
African Republic, Chad, Comoros, Congo, Cote d’Ivoire, Djibouti, Equatorial Guinea, Eritrea,
Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau,, Kenya, Lesotho, Liberia,
Madagascar, Malawi, Mali, Mauritania, Mayotte, Mozambique, Namibia, Niger, Nigeria,
Rwanda, Sao Tome and Principe, Senegal, Sierra Leone, Somalia, South Africa, St. Helena,
Sudan, Swaziland, Tanzania, Togo, Tristan da Cunha, Uganda, Zaire, Zambia, Zimbabwe
Latin America and the Caribbean (LAC)
Anguilla, Antigua and Barbuda, Argentina, Aruba, Bahamas, Barbados, Belize, Bolivia, Brazil,
British Virgin Islands, Cayman Islands, Chile, Colombia, Costa Rica, Cuba, Dominica,
Dominican Republic, Ecuador, El Salvador, French Guiana, Grenada, Guadeloupe, Guatemala,
Guyana, Haiti, Honduras, Jamaica, Martinique, Mexico, Montserrat, Netherlands Antilles,
Nicaragua, Panama, Paraguay, Peru, Puerto Rico, St. Kitts and Nevis, St. Lucia, St. Vincent and
the Grenadines, Suriname, Trinidad and Tobago, Turks, and Caicos Islands, Uruguay, U.S.
Virgin Islands, Venezuela
Middle Eastern crescent (MEC)
Afghanistan, Algeria, Armenia, Azerbaijan, Bahrain, Cyprus, Egypt, Former Spanish Sahara,
Georgia, Islamic Republic of Iran, Iraq, Israel, Jordan, Kazakhstan, Kuwait, Kyrgyzstan,
Lebanon, Libyan Arab Jamahiriya, Malta, Morocco, Oman, Pakistan, Qatar, Saudi Arabia,
Syrian Arab Republic, Tajikistan, Tunisia, Turkey, Turkmenistan, United Arab Emirates,
Uzbekistan, West Bank and Gaza Strip, Yemen
Data from: Murray and Lopez, 1996, Annex Table 1. State or territories included in the Global
Burden of Disease Study, by demographic region.
Table 2
Numbers* and rates** of suicide-related deaths and DALYs by region, worId-1990
Suicide Mortality Suicide DALYs (1,000s)

n Crude Age-Adjusted
US
Total 31 12.4 10.7
Male 25 20.3 17.7
Female 6 4.8 4.1
EME(-US) EME (including US)
Total 81 14.8 11.5 2,155
Male 56 21.0 16.7 1,573
Female 25 8.9 6.6 582
FSE
Total 81 23.4 20.3 1,613
Male 62 37.6 34.3 1,294
Female 19 10.5 7.9 319
India
Total 99 11.7 12.4 2,803
Male 54 12.3 13.5 1,406
Female 45 11.0 11.4 1,397
China
Total 343 30.3 30.4 8,078
Male 159 27.2 28.5 3,232
Female 184 33.6 32.5 4,846

OAl
Total 67 9.8 10.3 1,918
Male 40 11.7 13.4 1,047
Female 27 7.9 7.4 871
SSA
Total 16 3.1 3.4 472
Male 13 5.2 5.7 379
Female 3 1.2 1.2 93
LAC
Total 22 5.0 5.1 593
Male 16 7.2 7.8 383
Female 7 3.1 2.7 209
MEC
Total 46 9.2 10.5 1,335
Male 32 12.5 14.7 906
Female 14 5.7 6.4 429
World
Total 786 14.9 15.5 18,967
Male 456 17.2 18.6 10,220
Female 330 12.6 12.7 8,747
Note: legal intervention-related deaths were included in the overall violence category but not in
the homicide category. Therefore, the sum of the subcategories of violence-related deaths do not
add up to the total.
* Numbers In thousands
** Rates per 100,000 0,000 populaction
Table 3
Percentage of all deaths that are suicide related, 1 990
Deaths from all causes Percentage of deaths due to suicide

Total Suicide
(n) (%)
US 2148000* 1.4
EME(-US) 4973000 1.6
FSE 3791000 2.1
India 9371000 1.1
China 8885000 3.9
OAI 5534000 1.2
SSA 8202000 0.2
LAC 3009000 0.7
MEC 4553000 1.0
World 50467000 1.6
Data from: National Center for Health Statistics. Table 6. Deaths and death rates for the W
leading causes of death in specified race-sex groups: United States, 1990. Monthly Vital Statistics
Report 1993;41(7 Suppl):20.
Percentage of war-related deaths less than 0.05.
REFERENCES
Alcohol, Drug Abuse, and Mental Health Administration, 1989, Report of the Secretary’s
Task Force on Youth Suicide. DHHS Pub. No. (ADM) 89–1624, Vol. 4. (Washington,
D.C.: Supt. Of Docs., U.S.Government Printing Office).
Birkhead, G.S., Calvin, V.G., Meehan, P.J., O’Carroll, P.W. and Mercy, J.A., 1993, The
Emergency Department in Surveillance of Attempted Suicide: Findings and
Methodologic Considerations. Public Health Reports 1993, 108, pp. 323–331.
Brent, D.A., Perper, J.A., Goldstein, C.E., et al., 1988, Risk factors for adolescent
suicide: a comparison of adolescent suicide victims with suicidal inpatients. Arch Gen
Psychiatry, 45, pp. 581–588.
Brent, D.A., Perper, J.A., Goldstein, C.E. et al., 1991, The presence and accessibility of
firearms in the homes of adolescent suicides. JAMA, 266, pp. 2989–2995.
CDC, 1991, Attempted Suicide Among High School Students—United States, 1990.
MMWR, 40, pp. 633–635, September 20.
Centers for Disease Control, 1992, Youth Suicide Prevention Programs: A Resource
Guide. (Atlanta, GA: Centers for Disease Control).
CDC, 1994, Firearm-Related Years of Potential Life Lost Before Age 65 Years—United
States, 1980–1991. MMWR, 43, pp. 609–611, August 26.
CDC, 1995, Suicide Among Children, Adolescents, and Young Adults—United States,
1980–1992. MMWR, 44, pp. 289–291, April 21.
CDC, 1995, Fatal and Nonfatal Suicide Attempts Among Adolescents—Oregon, 1988–
1993. MMWR , 44, pp. 312–315, 321–323, April 28.
CDC, 1996, Suicide Among Older Persons—United States, 1980–1992. MMWR, 45, pp.
3–6, January 12.
CDC, 1997, Rates of Homicide, Suicide, and Firearm-Related Death Among Children 26
Industrialized Countries. MMWR, 46, pp. 101–105, February 7.
CDC, 1998, Suicide Among Black Youths—United States, 1980–1995. MMWR 47, pp.
193–196, March 20.
CDC, 1998, Suicide Prevention Evaluation in a Western Athabaskan American Indian
Tribe—New Mexico, 1988–1997. MMWR, 47, pp. 257–261, April 10.
Cummings, P., Koepsell, T.D., Grossman, D.C., Savarino, J. and Thompson, R.S., 1997,
The association between the purchase of a handgun and homicide or suicide. AJPH,
87, pp. 974–978.
Davidson, I.E., Rosenberg, M.L., Mercy, J.A., Franklin, J. and Simmons, J.T., 1989, An
Epidemiologic Study of Risk Factors in Two Teenage Suicide Clusters. JAMA, 262,
pp. 2687–2692.
Davidson, L. and Could, M.S., 1989, Contagion as a Risk Factor for Youth Suicide. In:
Alcohol, Drug Abuse, and Mental Health Administration: Report of the Secretary’s
Task Force on Youth Suicide. Vol. 2: Risk Factors for Youth Suicide, pp. 88–109.
DHHS publication no. (ADM) 89–1622, (Washington, DC: U.S. Government Printing
Office).
Gould, M.S., Wallenstein, S., Kleinman, M.H., O’Carroll, P.W., Mercy, J.A., 1990,
Suicide clusters: An examination of age-specific effects. American Journal of Public
Health, 80, pp. 211–212.
Hirschfeld, R.M.A. and Davidson, L., 1988, Risk factors for suicide. In American
Psychiatric Press Review of Psychiatry edited by Frances, A.J, and Hales, R.E., Vol 7.
(Washington, DC: American Psychiatric Press), pp. 307–333.
Ikeda, R.M., Gorwitz, R., James, S.P., Powell, K.E. and Mercy, J.A.,1997, Trends in fatal
firearm-related injuries, United States, 1962–1993. American Journal of Preventive
Medicine, 13(5), pp. 396–400.
Kachur, S.P., Potter, L.B., James, S.P. and Powell, K.E., 1995, Suicide in the United
States, 1980–1992.Violence Surveillance Summary Series, No. 1. (Atlanta: Centers for
Disease Control and Prevention, National Center for Injury Prevention and Control).
Kellermann, A.L., Rivara, F.P., Rushforth, N.B. et al., 1993, Gun ownership as a risk
factor for homicide in the home. NEJM, 329, pp. 1084–1091.
Kellermann, A.L., Rivara, F.P., Somes, G., Reay, D.J., Francisco, J., Benton, J.G.,
Prodzinski, J., Fligner, C. and Hackmann, B.B., 1992, Suicide in the home in relation
to gun ownership. NEJM, 327(7), pp. 467–472.
Mercy, J.A., Rosenberg, M.L., Powell, K.E., Broome, C.V. and Roper, W.L., 1993,
Public health policy for preventing violence. Health Affairs Winter, 12(4), pp. 7–29.
Moscicki, E.K., Muehrer, P., Potter, L.B., 1995, Introduction to Supplemental Issue:
Research Issues in Suicide and Sexual Orientation. Suicide and Life-Threatening
Behavior, Supplement:25, pp. 1–3.
Murray, Christopher. 1996, Rethinking DALYs, In Murray, Christopher and Alan Lopez
(eds.) The Global Burden of Disease: A Comprehensive Assessment of Mortality and
Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020,
pp. 1–98, (Geneva: World Health Organization).
Murray, Christopher and Alan Lopez (eds). 1996, The Global Burden of Disease: A
Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and
Risk factors in 1990 and Projected to 2020. (Geneva: World Health Organization).
O’Carroll, P.W., 1989, A consideration of the validity and reliability of suicide mortality
data . Suicide and Life Threatening Behavior, 19, pp. 1–16.
O’Carroll, P.W., 1990, Community strategies for suicide prevention and intervention. In
Suicide Over the Life Cycle. Edited by Blumenthal, S.J. and Kupfer, D.J. (Washington,
DC: American Psychiatric Press), pp. 499–514.
O’Carroll, P.W., Rosenberg, M.L. and Mercy, J.A., 1991, Suicide. In Violence in
America: A Public Health Approach edited by Rosenberg, M.L. and Fenley, M.A.,
(New York: Oxford University Press), pp. 184–196.
O’Carroll, P.W. and Silverman, M.M., 1994, Community suicide prevention: The
effectiveness of bridge barriers [case consultation]. Suicide and Life-Threatening
Behavior, Spring, 24, pp. 89–99.
Potter, L.B., Powell, K.E. and Kachur, S.P., 1995, Suicide prevention from a public
health perspective. Suicide and Life-Threatening Behavior, Spring, 25, pp. 82–91.
Potter, L.B., Kresnow, M., Powell, K.E., O’Carroll, P.W., Lee, R.K., Frankowski, R.F.,
Swann, A.C., Bayer, T.L., Bautista, M.H. and Briscoe, M.G., 1998, Identification of
nearly fatal suicide attempts: Self-inflicted injury severity form. Suicide and Life-
Threatening Behavior, 28(2), pp. 174–186.
Reza, A., Krug, E. and Mercy, J.A., 1999, A Global Concern: Mortality Associated with
Violence throughout the World. Submitted for publication.
Rosenberg, M.L., Smith, J.C., Davidson, L.E., Conn, J.M., 1987, The emergence of youth
suicide: An epidemiologic analysis and public health perspective. Annual Review of
Rosenberg, M.L., Davidson, L.E., Smith, J.C., Berman, A.L., Buzbee, H., Gantner, G.,
Gay, G.A., Moore-Lewis, B., Mills, D.H., Murray, D. and O’Carroll, P.W., 1988,
Operational criteria for the determination of suicide. Journal of Forensic Sciences 33,
pp. 1445–1456.
Smith, J.C, Mercy, J.A., Conn, J.M., Marital status and the risk of suicide. American
Journal of Public Health, 78, pp. 78–80.
Smith, J.C., Mercy, J.A. and Rosenberg, M.L., 1986, Suicides and homicides among
hispanics in the Southwest. Public Health Reports, 101, pp. 265–270.
United States Public Health Service, 1999, The Surgeon General’s Call to Action to
Prevent Suicide. Washington, DC:.
Wintemute, G.J., Parham, C.A., Beaumont, J.J., Wright, M. and Drake, C., 1999, The
mortality experience of recent purchasers of handguns, NEJM, November.
21
Barbarism and Solitude: Reflections on
Globalization and Violence
Emmanuel Rozental-Klinger
‘Total war and cold war have brainwashed us into accepting barbarity. Even worse: they
have made barbarity seem unimportant, compared to more important matters like making
money’ (Hobsbawm, 1997). In his lecture titled ‘Barbarism: A Users Guide’, Eric
Hobsbawm makes this statement that clearly and directly outlines the relationship
between globalization and violence. In fact, it is my view that the very process by which
we have become brainwashed to accept and to perpetrate barbarity is what we currently
know as globalization.
The process of accumulation of capital begins by turning a certain amount of money
into means of production and labour. This operation takes place in the market, within the
orbit of circulation. The second stage of the movement, the production process, ends as
soon as the means of production are turned into merchandise whose value exceeds the
value of its parts, containing the initial capital invested plus a surplus value. In turn, this
merchandise is sent out again into the orbit of circulation. It is necessarily sold, realizing
its value in money, in order to turn this money into new capital, and so the cycle repeats
itself endlessly (Marx, 1958).
‘I was invited to witness the trade. My guest was a gang member and only 15
years old. He trusted me and wanted to get out of trouble. He knew he was
going nowhere and that he would likely end up murdered at a young age. He
wanted to break the cycle. A police agent in uniform brought the weapon to the
dark alley. He was paid the price that had been agreed upon, and left warning
against any leaks in secrecy. I saw the gun. That was just the beginning of a
cycle. The same weapon had belonged to the leader of another gang and taken
away from him by the officer. A few days later, the same policeman, assisted by
two others, chased and arrested his client and “found” him illegally carrying a
gun. They beat him and let him go. He had three hours to get a set amount of
money or else they would lay charges. He had no choice, plus, he needed the
gun. He paid the price and recovered the weapon. The cycle continued. The next
time he was caught, after the new beating, the gun was never returned to him. A
new client bought it for a better price. Each time, crimes were committed with
the weapon. Each time, the price of the recycled weapon went up’ (or an
excellent ethnographic study of urban violence see Vanegas, 1998)
Key resources such as money, labour and means of production, and the market, constitute
the essence of the process of capital accumulation. Specific roles and relationships
Barbarism and solitude 301
between human beings are established for this process to occur: those between the
owners of capital and the labour force.
The labour force is constituted by people who sell their time and effort to the owners of
the means of production. Their labour generates the surplus value, the source of profit for
the owners of capital. In exchange, labourers are remunerated at set market prices.
Cheaper, more efficient, skilled labour generates more surplus value and greater capital
accumulation. Within this process, human beings, particularly those who find themselves
having to sell their work as labour, are reduced to another resource of the production
process aimed at the accumulation of profit for the owners of capital.
The market is a necessary stage in the process of capital accumulation, a stage where
trade of goods, services and people (labour), takes place.
Technology is another key resource in this equation. Technologies have affected and
indeed transformed the productive processes extending beyond them to almost every
sphere of human activity (Ellul, 1964).
‘As I see it, technology has built the house in which we live. The house is
continually being extended and remodelled. More and more of human life takes
place within its walls, so that today there is hardly any human activity that does
not occur within this house. All are affected by the design of the house, by the
division of its space, by the location of its doors and walls. Compared to people
in earlier times, we rarely have a chance to live outside this house. And the
house is still changing; it is still being built as well as being
demolished’ (Franklin, 1999).
given these facts and processes, globalization is itself a form of violence,

generalized and exercised against the majority of the world’s population
Technology is the outcome of productive processes, hence its development and use is
determined by capital accumulation, while simultaneously transforming the relations of
production between labour, means of production, capital and the market. Control over the
production and use of technology is determinant of power over markets, production and
accumulation.
Technology is not the sum of the artifacts, of the wheels and gears, of the rails
and electronic transmitters. Technology is a system. It entails far more than its
individual components. Technology involves organization, procedures,
symbols, new words, equations, and, most of all, a mindset’ (Franklin, 1999).
Societies are involved in complex conflicting processes whereby specific social actors
represent their interests in their struggle for power over the production and distribution of
essential resources. Within the modern capitalist State, the main actors involved in
conflictive processes over key resources have traditionally been organized labour and
capital. The State has provided the scenario and the ‘rules of the game’ for such processes
to take place. The outcome is an expression of the balance of power between the
majorities (labour and other social actors) and capital (minorities). In general, the
interests of economic development tend to take precedence over other priorities,
particularly when labour and other social interests are weak or fragmented. Economic
development and the generation of wealth serve the accumulation of profit for the
interests of capital. Yet, it is argued that this accumulation is simultaneously the means
towards social development. Consequently, markets expand as a result of increased
production and accumulation, which constitutes measurable growth.
Each individual State is immersed in a global market. The actors defining ‘rules of the
game’ are increasingly multinational and transnational. Modern development is based on
the premise of unlimited growth and the universalization of the market. Individuals
within societies and societies within world economies have entered the realm of
expanding, merging and stronger global capital accumulation interests, with greater and
growing power to produce and use key resources. Weakened nation-states and
increasingly fragmented and disorganized labour, and other social actors, strive to
survive. Such is, in essence, the process of globalization. A process that has given rise to
the fourth world (Castells, 1998).
It was part of the daily newscast, a story of yet another factory being closed and of its
workers being laid-off. A woman had her face against the wired fence. As the reporter
approached her; her tears became visible and her moans of despair louder. ‘What am I
going to do now? What am I going to do? I have four children at home and nothing to
feed them with’.
The most visible and expected outcome of globalization is inequality with gross
concentration of power and wealth in a few hands and massive exclusion and deprivation
among growing majorities. A market for profit is necessarily amoral and immoral.
Amoral, because it is not intended for human well being, but for profit accumulation.
Immoral, because it promotes excessive accumulation by a privileged minority while
others (the immense majority) are deprived from their most basic needs and rights.
According to the Human Development Report 1999 (UNDP):
• The global labour market is increasingly integrated for the highly skilled
corporate executives, scientists, entertainers and the many others who form the
global professional elite with high mobility and wages.
• Inequality has been rising in many countries since the early eighties.
• Inequality between countries has also increased. The income gap between
the 20 per cent of the world’s people living in the richest countries and the 20
per cent in the poorest was 74:1 in 1997, up from 60:1 in 1990 and 30:1 in 1960.
By the late nineties the 20 per cent of the world’s people living in the highest
income countries had:
* 86 per cent of world GDP while the bottom 20 per cent had just 1 per
cent
* 82 per cent of world export markets while the bottom 20 per cent
had just 1 per cent
* 68 per cent of foreign direct investment while the bottom 20 per
cent just 1 per cent
* 74 per cent of world telephone lines, today’s basic means of
communication while the bottom 20 per cent just had 1.5 per cent.
• OECD countries with 19 per cent of the global population have 71 per cent
of global trade in goods and services, 58 per cent of foreign direct investment
and 91 per cent of all internet users.
• The world’s 200 richest people more than doubled their net worth between
1994 and 1998, to more than one trillion dollars. The assets of the top 3
billionaires are more than the combined GNP of all least developed countries and
their 600 million people.
• The recent wave of mergers and acquisitions is concentrating industrial
power in mega corporations, at the risk of eroding competition. By 1998 the top
10 companies in pesticides controlled 85 per cent of a 31 billion dollar global
market/ and the top 10 telecommunications, 86 per cent of a 262 billion dollar
market.
• In 1993 just 10 countries accounted for 84 per cent of global research and
development expenditures and controlled 95 per cent of the US patents of the
past two decades. Moreover, more than 80 per cent of patterns granted in
developing countries belong to residents of industrial countries.
New technologies and modes of production are increasingly capital intensive. Workers are
either absorbed in lower-paying jobs or become marginalized. In order to gain
comparative advantage workers and smaller firms enter a ‘race downwards’ whereby they
decrease wages in order to remain competitive. Under these circumstances, child labour,
prostitution and other forms of exploitation become commonplace and are understandably
defended by their victims as a source of income, when other sources are denied to them.
State structural reforms at the service of global capital are leading to an increasingly
serious problem of unemployment, turning to unemployability in an environment where
there is declining responsiveness to those with the least power and the most need. This
phenomenon, added to chronic employment deficits (particularly in the most dependent
economies) has led to the rise of the ‘underclass’. Beyond the ‘reserve labour force’, these
people have become superfluous and unnecessary—they have been expelled from
productive activities. They have been made invisible.
Large urban ghettos occupy most of the Third World cities where the majority of the
population survives under precarious conditions. In fact, entire countries are ‘more
marginal today than they were at independence from colonialism’ (Apter, 1997).
social exclusion, within contexts characterized by inequity and poverty is fertile

ground for all kinds of violence
Given these facts and processes, globalization is itself a form of violence, generalized and
exercised against the majority of the world’s population. Within the context of
extraordinary technological development and given our global unprecedented ability to
generate wealth, poverty is the outcome of inequity and as such it constitutes violence
against the poor (Hobsbawm, 1994).
In spite of the growing evidence of a crisis in terms of increasing inequity, exclusion
and human suffering generated by capital, the most prevalent attitudes towards it are:
• An almost religious faith in the goodness of capital It is the perspective from

within, professed mostly by those directly benefiting from the system and its
expansive perspectives. Data and evidence describing exceptions are presented
as proof to the likelihood of a better future. Widespread failures are discarded as
exceptions. The invitation is to trust the system, for it will eventually solve its
problems.
• A pragmatic position recognizes the risks and damages caused but, given
that there are no realistic alternatives, a damage containment process must be in-
place and all practical efforts should be made to choose the best of all evils.
• Practical indifference to calamities and sufferings of others is taken to be a
legitimate mode of personal conduct. This is probably the dominant moral
culture for most of the duration in most of the societies known as liberal (Geras,
1999).
These structural determinants of violence are manifested in the context of particular

societies and nations where they occur within specific historical processes, among actors
and their relationships. Social exclusion, within contexts characterized by inequity and
poverty is fertile ground for all kinds of violence. When deprivation and opulence
coexist, violence is perceived as functional to the satisfaction of needs that appear
otherwise inaccessible.
‘A Sunday morning gathering was taking place in a health center within one of
the most violent neighborhoods in the city of Cali. A group of young people had
come to talk about their most pressing problems and needs. The issues were
poverty, lack of access to education and health care, abuses by the authorities,
drug abuse and drug trade, racism and many other issues that combined with
each other to become almost inseparable. A young man talked about the need
for a profound transformation that would not come about until and unless local
people suffering and understanding these problems would gather political power
to act on behalf of these communities. He spoke about the need for new
leadership and the role youth (such as himself) should play in this regard.
Someone spoke about “sicarios” (hired assassins). When the young man was
asked if he would consider murdering somebody he did not know for, say, a
million pesos (US $ 500) his reply was: “I would give it serious consideration.
We could do a lot with that money at home”.’
Beyond certain threshold of inequity, the distribution of income and wealth seems
arbitrary. Frustrations exacerbate when differences are perceived as unfair. In the eyes of
the ‘have-nots’, the ‘haves’ are unfairly blessed by undeserved privileges or luck. The
relative positions of individuals in society do not seem to correspond to any objective
criteria. Social positions and benefits are not the outcome of efforts or of the objective
evaluation of merits but of favorable or unfavorable circumstances. The entire social
structure appears illegitimate. What is the meaning and value of on-going individual
effort and work if they will have minimal or no impact on generating better social and
living conditions? Under such circumstances, the past, along with established class
structures, pre-determines the present and denies access to a better future. When the past
is the major determinant of the future, violent means appear as legitimate options
(Fitoussi and Rosanvallon, 1997).
These structural determinants are compounded by institutional and interpersonal
determinants.
Institutions-formal or informal, governmental and non-governmental—constitute ‘the
rules of the game’ and the provision of services within societies. The processes of market
liberalization and globalization are having a direct deleterious impact on institutions,
particularly on those established for the provision of essential services (health, education)
for those in most need in the less developed world (UNDP, 1999).
Under critical circumstances, institutions lose legitimacy and they either tend to
reproduce the violence that benefits dominant interests, or are weakened, distorted or
their function is replaced by illegal organizations that assume their role. The relationship
between institutional determinants and violence has been studied (Apter, 1997; for the
case of Colombia see Deas et al., 1999). In Colombia, where homicide constitutes the
first cause of death, more than 90 per cent impunity for this crime encourages its
recurrence and explains the appearance of extra-institutional justice mechanisms
(Romero, 1999).
The northern front of ELN, the National Liberation Army, the second most powerful
leftist guerrilla group in Colombia, sends a public communiqué dated 20 November 1999
with regards to the case of Orlando Rodriguez, a member of Barranquilla’s city council
and whom, according to the source, has had control of the budget since the eighties.
Quoting government and official sources, the front denounces continuous theft of public
resources and requests more public information from credible sources in order to carry
out a criminal process in view of the fact that his wrongdoings have been treated with
absolute impunity.
The ethics of interpersonal relationships are altered under circumstances of poverty.
Quite often the preservation of personal integrity requires that one threatens the integrity
of others, and vice versa. Under circumstances of growing exclusion, encouraged rugged
individualism, and aggressive competition for survival, the use of force is preferred to
solidarity, and aggression overrides conciliation. Conflicts tend to be resolved by the
elimination (physical and otherwise) of the counterpart. Differences become unacceptable
and social intolerance grows out of proportion. The weakest become the most
vulnerable—elderly, women, children, immigrants, minorities, youth (UNESCO, 1998;
Castells, 1998).
Under the influence of globalization, political, criminal and interpersonal forms of
violence are, not surprisingly, on the rise (Guerrero Baron, 1999). Although easy to
distinguish in theory, the boundaries between these violent expressions are often
indistinguishable.
when the past is the major determinant of the future, violent means appear as
legitimate options
Political violence is an organized attempt to change the power relationships between

social actors. Its expressed intent is to alter the control over key social resources.
According to Apter, it is carried out ‘for a designated and reordering purpose: to
overthrow a tyrannical regime, to redefine and realize justice and equity, to achieve
independence or territorial autonomy, to impose one’s religious or territorial
beliefs’ (Apter, 1997). ‘It is a rational phenomenon, no different generically speaking
than normal politics except in terms of its means’ (Apter, 1997). Within the context of
growing inequality and exclusion promoted by the interests of capital accumulation, ‘(w)
ithout suggesting a one-to-one correspondence between negative conditions and political
violence, what is happening today and developmentally, increases the possibilities for
violent discourse’ (Apter, 1997). Class, ethnicity, religion, race, language and doctrines
are revalidated as ways of defining redeeming projects, discourses and their communities.
Fundamentalisms are also on the rise. Political violence in our world can be easily
explained (although not always easily justified) as a necessary reaction to a global
process that leaves few other viable alternatives for real and effective democratic political
participation. The threat and destruction of social movements encourages a wide
spectrum of political organizations that resort to violent means (Hobsbawm, 1996).
Criminal violence includes a wide range of expressions from the sporadic disorganized
criminal activity to the globalized organized criminal economy. (For an excellent analysis
of the subject see ‘The Perverse Connection: The Global Criminal Economy’ in Castells,
1998) The latter can be understood as illegal global capital. Its purpose is the
accumulation of profit through the extraction of surplus value from a wide variety of
productive illegal processes. It is a source of labour for increasing numbers of people
throughout the world. The concentration of power and wealth in the hands of the owners
of this illegal capital follows the economic pattern of the legal accumulation processes.
The type of illegal commercial activities in which organized crime is involved generates
risk. Risk calls for the use of violent means. The combination of risk and violence exerts
economic, cultural and political impacts. These impacts are felt at every level of the legal
and illegal market, as well as on political processes. As a result, the number of those
affected, involved and responsible for the criminal economy is growing. According to the
Human Development Report 1999 (UNDP, 1999):
‘Deregulated capital markets, advances in information and communications

technology and cheaper transport make flows easier, faster and less restricted
not just for medical knowledge but for heroin-not just for books and seeds but
for dirty money and weapons.
Illicit trade—in drugs, women (sic), weapons and laundered money—is
contributing to the violence and crime that threaten neighborhoods around the
world…The Internet is an easy vehicle for trafficking in drugs, arms and women
(sic) through nearly untraceable networks. In 1995 the illegal drug trade was
estimated at 8 per cent of world trade, more than the trade in motor vehicles or
in iron and steel. Money laundering—which the International Monetary Fund
(IMF) estimates at equivalent to 2–5 per cent of global GDP—hides the traces
of crime in split seconds, with the click of a mouse. At the root of this is the
growing influence of organized crime, estimated to gross $1,5 trillion a year,
rivaling multinational corporations as an economic power. Global crime groups
have the power to criminalize politics, business and the police, developing
efficient networks, extending their reach deep and wide’.
The prevalent assumption is that criminal violence refers to the activities of mafias, drug
trade, prostitution, black markets and the like. It is important to recognize as criminal and
violent, a broad spectrum of activities that increasingly involve white-collar criminals,
governments and government officials, prestigious international figures and even heads of
State. Crime and violence are not only what the media and the governments denounce as
such. A few examples will illustrate this point. The Iran Contra affair, where the US
government at the highest level was involved in cocaine and arms trade in order to
overturn foreign governments by violent means and against its own national constitution.
Augusto Pinochet, one of several former dictators and heads of State involved in crimes
against humanity. The massive crises generated by unregulated capital speculators whose
direct result in barbarity and human suffering has not been adequately recognized.
And there are other types of international crime. The global economy is promoted by
capital in order to free the markets to open and fair competition. Evidence demonstrates
(as stated earlier) the growth of monopolies, the concentration of wealth and power in
fewer hands and the growing restrictions imposed on the ‘freedom’ of the market by
merging capital mega interests. While governments are being pressured to lift their
barriers to trade, privatize and enter open competition under the threats of economic and
political strangulation, giant multinational conglomerates close the door to open-trade and
impose monopolies.
As Anatole Kaletsky reported (November, 1999), Microsoft business genius Bill gates,
has maintained ‘highly profitable levels in software prices by establishing and exploiting
monopoly power’. A recent anti-trust decision demonstrated how ‘there can be no further
doubt that Microsoft deliberately used its absolute control over key operating software to
deter computer manufacturers from marketing cheap, simplified computers and to stifle
quality improvements in innovations’.
In the meantime, the more commonly recognized global criminal economy grows,
mostly by taking advantage of the opportunities created by the expansive global legal
economy. There is a growing consumer illegal market for drugs and arms. Any measures
taken against these harming processes have had little impact. The markets keep growing
and expanding. At most, transient control is obtained. It seems increasingly plausible that
the growth of the global legal capital necessarily promotes the growth of its counterpart
(and sometimes associate) illegal side (Castells, 1998).
Political violence in our world can be easily explained (although not always easily
justified) as a necessary reaction to a global process that leaves few other viable
alternatives for real and effective democratic political participation
Some obvious contradictions with devastating outcomes in terms of loss of human life
and extreme suffering need to be approached directly. One of the most obvious examples
of such contradictions is the emphasis placed on repression of production by the US war
against drugs (mostly involving military aid and arms transfers and trade).
Simultaneously, Colombia is a market for American produced armament. No war against
arms production and trade is launched within the United States. The producers of
weapons are protected (while the US is not exempt from internal arms-related criminal
offences, such as shootings from school yards to the stock market).
The expanding global economies have created a world of individualized consumers, in
need of a ‘quick fix’ that can be bought and which will provide immediate and
predictable results (Williams, 1983, Hobsbawm, 1994). It is not surprising that addictive
behaviours are on the rise. Yet, few efforts are made to unveil the determinants of
widespread consumption of illegal drugs (Fitoussi and Rosanvallon, 1997). The approach
is a repressive-moralistic one, which can only promote the growth of this market. It
seems likely that the combination of repressive measures taken against organized crime
with predominant social marketing strategies that invite and manipulate consumerism and
addictive behaviours constitute the best promotion for illegal drug trade.
‘Their strategy is to base their management and production functions in low-risk areas,
where they have relative control of the institutional environment, while targeting as
preferential markets those areas with the most affluent demand, so that higher prices can
be charged’. This is clearly the case for the weapons and drug cartels (Castells, 1998).
For the masses of people increasingly involved in criminal activities, the distinction
between legal and legitimate becomes a matter of survival. When obeying the law under
circumstances of increasing exclusion leads to further deprivation and hopelessness, the
illegal activities are justified as legitimate and find increasing support in masses of urban
poor (particularly young) throughout the world. If they do not find violence and crime,
violence finds them (Castells, 1998; Hobsbawm, 1997; Apter, 1997; Penaranda-Guerrero,
1999).
As Castells (1997) sunimarizes it: ‘In many contexts, daring successful criminals have
become role models for a young generation that does not see an easy way out of poverty,
and certainly no chance of enjoying consumption and live adventure. From Russia to
Colombia, observers emphasize the fascination of local youth for the Mafiosi. In a world
of exclusion, and in the midst of a crisis of political legitimacy, the boundary between
protest, patterns of immediate gratification, adventure and crime becomes increasingly
blurred.
In German Castro Caycedo’s partial interview of Pablo Escobar, former leader of the
Medellin drug cartel, explains how he became involved in criminal activities when he
was young and poor working for less than minimum wages in downtown Medellin
carrying heavy loads. At the end of the day, he was just as poor, hungry and tired.
Nobody noticed him, especially not the girls. At the time he used to watch those of his
age involved in the emerald and marihuana trades. They dressed well, looked clean, had
fancy cars and were respected and recognized by everyone. Is that not what everyone
wants, especially when young and full of energy?
Unfortunately and in view of the impact of globalization, criminal or otherwise, the
real question is: why wouldn’t youth (and everyone for that matter), join organized
crime?
In conclusion, given the complex but all too obvious relationships between violence
and globalization, and the risk both these processes pose to the survival of our species
and of life on our planet, it is imperative to overcome the pragmatic indifference that
allows us to co-exist and perpetrate human suffering. What is the alternative? This is not
a question for experts, but a challenge for humanity. The fact that we do not have the
necessary answers to it cannot continue to justify our complicity with an order that
imposes death and pain to most human beings in the world. The only acceptable reaction
to this challenge is to find the means and the ways to overcome this age of wealth and
darkness.
Finally, markets and capital have been extraordinary means for the unprecedented
generation of wealth. But we need to find ends different to those of limitless
accumulation and greed, and transform markets into regulated means for the enjoyment
and protection of life. The hints we need are probably found within realities such as the
one affecting Colombia, plagued by violence and by contrast, stubborn and creative joy.
Learning from them might guide us in the direction of the strength and true meaning of
the human spirit, whereby there will be, out of our own will, an end to the seemingly
eternal solitude. As Gabriel Garcia Marquez said it when he accepted the Nobel Prize in
1982: ‘I dare to think that it is this outsized reality, and not just its literary expression,
that has deserved the attention of the Swedish Academy of Letters.
A reality not of paper, but one that lives within us and determines each instant of our
countless daily deaths, and that nourishes a source of insatiable creativity, full of sorrow
and beauty, of which this roving and nostalgic Colombian is but one cipher more, singled
out by fortune. Poets and beggars, musicians and prophets, warriors and scoundrels, all
creatures of that unbridled reality, we have had to ask but little of imagination, for our
crucial problem has been a lack of conventional means to render our lives believable.
This, my friends, is the crux of our solitude’.
REFERENCES
Apter, David, 1997, Political violence in analytical perspective. In The Legitimization of

Violence edited by David Apter. (London: Macmillan). pp 23–25.
Castells, Manuel, 1998, The rise of the fourth world: Informational capitalism, poverty,
and social exclusion. In End of Millenium. (Oxford: Blackwell)
ElluI, Jacques, 1964, The Technological Society. (Toronto: Vintage Books)
Fitoussi, Jean-Paul and Rosanvallon, Pierre, 1997, La Nueva Era de las Desigualidades.
(Buenos Aires: Manantial).
Franklin, Ursula, 1999, The Real World of Technology. (Toronto: Anansi).
Geras, Norman, 1999, The Contract of Mutual Indifference. (London: Verso).
Guerrero Baron, Javier. La sobre-politizacion del narcotrafico en Colombia en los anos
ochenta y sus interferencias en los procesos de paz. En Penaranda, Ricardo, Guerrero,
Javier. De Las Armas a la Politica, pp 219–296 (Bogota: Tercer Mundo).
Hobsbawm, Eric, 1994, Age of Extremes.(London: Abacus)
Hobsbawm, Eric, 1997, Barbarism: A user’s guide. In On History Eric Hobsbawm.
(London: Orion House).
Kaletsky, Anatole, 1999 When corporate interests dwarfed the consumer The Times.
London. Tuesday, 9 November p. 35
Marx, Karl, 1958, El Capital. (Mexico: Fondo de Cultura Economica), Vol. 1.
Vanegas, Gildardo, 1998, Call Tras el Rostro Oculto de las Violencias. (Cali: Editorial
Univalle).
Rubio, Mauricio, 1999, La justicia en una sociedad violenta. In Reconocer La Guerra
para Construir la Paz compiled by Deas, Malcolm and Llorente, Maria Victoria.
(Bogota: Uniandes, Cerec, Norma). pp 201–235
Romero, Mauricio, 1999, Elites regionales, identidades y paramilitares en el sinu. In
Penaranda, Ricardo, Guerrero, Javier. De Las Armas a la Politica. (Bogota: Tercer
Mundo).
UNDP, 1999, World Development Report 1999. (New York: Oxford University Press).
UNESCO, 1998, World Culture Report 1998. (Paris: UNESCO).
Williams, Raymond, 1983, Towards 2000. (London: Hogarth Press).
Index
Alzheimer 154, 157, 158

Anti-sipping act 184
anxiety disorders 287
back injuries 23
behavioural adaptation 127, 128, 131, 133, 136, 139, 144, 148, 149
behavioural change 98, 102, 126, 128, 134, 136, 138, 194, 282
biases 44, 58
burden of injury 20, 27, 62 40, 65, 70
capture-recapture statistical methods 43

casualty insurance 28, 29
Causation 16, 179,
children 179, 180, 181, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192 186
and safety 186, 187
Committee on Trauma Research 127, 147
community and cluster trials 44
competition 25, 231, 245, 254, 256, 258, 303, 305, 307
complex systems 7
comprehensive costs 54, 61, 63
conflict 5, 8, 75, 82, 85, 88, 218, 219
cost data 51, 63
cost savings 51, 52, 61, 64, 65
criminal violence 65
crisis centres 289
DALYs 54, 57, 92, 280, 288, 295

Declaration of the Rights of the Child 181
Dementia 154, 157, 158
dementia 154, 157
denominator 41, 44, 63, 263
depression 92, 230, 280, 287, 290, 293
Disability adjusted life years 56, 92, 280, 287
economic pressure 25
epidemiology 37, 38, 39, 42, 48
ethics 1, 190, 222, 305
European Commission 105, 239, 242, 244, 245, 260, 261
EuroQol scale 55
exposure 38, 41, 44, 46
Index 312
exposure control 98, 103
exposure risk 38, 41
fire insurance 28, 31

Firearm 281
firearm 52, 63, 66, 169, 199, 283, 287, 290, 291
first party injury insurance 31
forgiving systems 2
Functional Capacity Index 56, 65
gatekeeper 292
gatekeeper training 289
Haddon 103
head injuries 251
Health Utility Index 55
heterogeneity 6, 81
heterogeneous traffic 75, 81, 82, 84, 85
hierarchical societies 6
high bridges 290
homicide 17, 23, 162, 278, 279, 305
hotlines 289
incidence-based costs 52
informal sector 77, 79, 80, 81, 89
injuries 194, 195, 196, 197, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 210
injury costing worldwide 62
injury data 37, 38, 39, 40, 48
Injury Impairment Index 55
institutional determinants 304
insurance 28, 29, 31, 34, 139, 151, 191, 243, 250
insurance pricing 29, 31
intentional injuries 23, 279, 281
interpersonal relationships 305
iterative algorithms 45
jury verdict method 58
land-use patterns 72, 77, 80, 88

legal drinking age 23
Liability insurance 31
liability insurance 31, 243
life insurance 28
Lloyds 28, 29, 32
loss reduction 15, 17, 28, 29, 31, 34
low income countries 93, 103
low social status 6
Index 313
manic depressive 287, 293

marine insurance 28, 29
Medical costs 52
meta-analysis 40, 47
middle class 5
missing data 39, 44, 47
mixed traffic 74, 75, 85, 89, 91
modal choice 153
narrative data 42
natural experiments 44, 46
no-fault 31
non-motorized modes 76, 80, 88
non-motorized transport 7, 78, 79, 81
numerator problem 40
observational studies 43, 44
pacifiers 26
pedestrian fatalities 105
peer pressure 25
peer support programmes 288
pesticides 23, 229, 249, 289, 290, 303
political violence 280, 305
poor communities 7
prevalence-based costs 51
private insurers 28, 34
private sector 28, 102, 231, 291
productivity pressure 25
protection of children 183, 186, 187
protective behaviour 161, 163, 165, 166, 167
Psychology 128, 131, 134, 146
public 5, 6, 9, 10, 28, 32, 34
Public health 1, 3, 4, 5, 28, 37, 40, 42, 48, 102, 160, 191, 205, 278, 282, 283, 284, 285, 287, 288,
289, 290, 291, 292, 298
public policy 12, 132, 191, 261
public transport system 79
punishment 9, 181, 184, 185, 214, 279
quality of life 17, 53, 54, 55, 56, 58, 59, 60, 61, 64, 66, 69, 150
quality of well-being 55
(QALYs) 54
randomized controlled trials 44, 46, 201, 247

resource costs 52, 59, 60, 61
risk compensation 126, 127, 130, 133 134, 135, 136, 137, 138, 139, 141, 142, 143, 144, 146, 148
Index 314
risk factors 278, 282, 283, 284, 288, 289, 290, 297
risk taking 167
safety belts 30, 31, 165, 196

SamPeltzman 129, 144
Save the Children Fund 181
schizophrenia 287
scientists 9, 10
screening programmes 289
seat belt use laws 131, 135, 136, 143
socially distributed knowledge 10
societal arrangements 1, 5, 12
Societal complexity 5
societal obligations 1, 2
societal responsibility 1
society 1, 2, 5, 6, 7, 8, 9, 10, 11, 34, 51, 60, 128, 130, 133, 145, 161, 174, 179, 180, 188, 194, 215,
219, 221, 222, 224, 225, 229, 231, 234, 244, 278, 281, 304
strategy 14, 15, 16, 48, 99, 103, 127, 132, 186, 190, 196, 203, 225, 230, 233, 283, 308
structural determinants 304
suicide 23, 66, 142, 160, 161, 164 165, 167, 278, 279, 287, 288, 289, 290, 291, 292, 295
surveillance 42, 43
survival analysis 40, 45, 46
systematic Analysis 17
technology 3, 7, 9, 56, 72, 88, 204

traditional research 10
traffic safety 92, 98, 99, 102, 103
trans-disciplinary 10, 11
under reporting 96, 101, 102

unintentional injuries 160, 161, 66, 177
urban poverty 76, 78, 79
value of reducing risk 58

value of statistical life 56, 58, 60, 64
Violence 278, 280, 281, 282, 283
violence 23, 62, 64, 65, 163, 167, 173, 218, 278, 279, 280, 281, 282, 283, 284, 285, 289, 290, 293,
300, 303, 304, 305, 308
vulnerable road users 72, 74, 75, 88, 89, 90, 98, 102, 105
welfare payments 60
work loss costs 51, 59, 61
Youth Suicide Prevention Programs:

A Resource Guide 289

Document PDF

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Document PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Document PDF

Uploaded by

Copyright:

Available Formats

Injury Prevention and Control

London and New York

Tailor & Francis is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2005.

ISBN 0-203-30533-7 Master e-book ISBN

ISBN 0-203-34380-8 (Adobe e-Reader Format)

1. Injury Control and Safety Promotion: Ethics, Science and Practice 1

The editors gratefully acknowledge the support of:

INJURY CONTROL, SOCIETY AND ETHICS

therefore, we have a societal and moral responsibility to design our products,

• Safety is a fundamental human right.

INJURY CONTROL, PUBLIC HEALTH AND TECHNOLOGY

Figure 1 Decrease in disease specific mortality rates in USA 1900–1980 (Data

safety is a pre-requisite to the maintenance and improvement of the wellbeing and

Societal complexity and injury control

infections, contagious diseases, and other health problems due to

Consequences of increasingly complex systems in LICs

A good example of this is the role of expressways in inter-city travel. When an

SCIENTISTS, SOCIETY AND INJURY CONTROL

Features of traditional and socially distributed knowledge generation (Adapted from

TRADITIONAL KNOWLEDGE SOCIALLY DISTRIBUTED

Injury control scientists work on generation of socially distributed knowledge. Such

THE WAY FORWARD

• A conscious effort to examine the theoretical directions in disciplines which

TEN STRATEGIES FOR REDUCING THESE LOSSES

there is no logical reason why the rank order…of loss-reduction counter-measures

SEPARATION OF LOSS REDUCTION AND CAUSATION

FOR EACH STRATEGY AN ANALOGOUS OPPOSITE

A SYSTEMATIC ANALYSIS OF OPTIONS

WHERE HAVE WE BEEN?

WHERE ARE WE NOW?

• Developing surveillance systems with uniform coding of the external cause

ALL INJURIES YEAR

RussianFederation 347,600 16 1994

WHAT ARE WE DOING RIGHT?

WHAT ARE WE DOING WRONG?

WHAT ELSE ARE WE DOING THAT IS WRONG?

people to take risks

government activities and funding of injury prevention programmes/ research,

INSURANCE RISK AND LOSS REDUCTION

Insurance allows organizations and/or individuals to assume otherwise unacceptable

INDIRECT LOSS REDUCTION BY INSURERS

insurance companies are financial institutions subject to government

• A five-year programme to promote safety belt use in North Carolina

UNIQUE ASPECTS OF INJURY DATA

CURRENT STATE OF THE LITERATURE OF INJURY ANALYSIS

State-of-the-art methodologies applied to the study of injuries

Quantification of the burden of injury

• to provide descriptive epidemiology of injuries and possibly provide insight

Identification of the causes of injury

an exciting new methodology for a research design is the case-crossover for

Roberts (1995) addresses several methodological issues in case-control studies, from

• Time series analyses

Evaluation of the impact of interventions to prevent and/or control injuries

The process of random assignment to intervention is considered as the most rigorous

Challenge for the injury community

systematic literature reviews and meta-analysis are appropriate tools for