Legal Access To Tobacco Products

This PDF is available from The National Academies Press at http://www.nap.edu/catalog.php?
record_id=18997
Public Health Implications of Raising the Minimum Age of Legal

Access to Tobacco Products
ISBN
Richard J. Bonnie, Kathleen Stratton, and Leslie Y. Kwan, Editors;
978-0-309-31624-8 Committee on the Public Health Implications of Raising the Minimum Age
for Purchasing Tobacco Products; Board on Population Health and Public
360 pages Health Practice; Institute of Medicine
6x9
PAPERBACK (2015)
Visit the National Academies Press online and register for...
Instant access to free PDF downloads of titles from the
NATIONAL ACADEMY OF SCIENCES
NATIONAL ACADEMY OF ENGINEERING
INSTITUTE OF MEDICINE
NATIONAL RESEARCH COUNCIL
10% off print titles
Custom notification of new releases in your field of interest
Special offers and discounts
Distribution, posting, or copying of this PDF is strictly prohibited without written permission of the National Academies Press.
Unless otherwise indicated, all materials in this PDF are copyrighted by the National Academy of Sciences.
Request reprint permission for this book
Copyright © National Academy of Sciences. All rights reserved.

Public Health Implications of Raising the Minimum Age of Legal Access to Tobacco Products
Committee on the Public Health Implications of

Raising the Minimum Age for Purchasing Tobacco Products
Board on Population Health and Public Health Practice
Richard J. Bonnie, Kathleen Stratton, and Leslie Y. Kwan, Editors

THE NATIONAL ACADEMIES PRESS 500 Fifth Street, NW Washington, DC 20001
NOTICE: The project that is the subject of this report was approved by the Govern-
ing Board of the National Research Council, whose members are drawn from the
councils of the National Academy of Sciences, the National Academy of Engineer-
ing, and the Institute of Medicine. The members of the committee responsible for
the report were chosen for their special competences and with regard for appropri-
ate balance.
This study was supported by Contract/Grant No. HHSF22301031T between the

National Academy of Sciences and the Department of Health and Human Services:
Food and Drug Administration. Any opinions, findings, conclusions, or recommen-
dations expressed in this publication are those of the authors and do not necessar-
ily reflect the views of the organizations or agencies that provided support for the
project.
International Standard Book Number 13: 978-0-309-31624-8

International Standard Book Number-10: 0-309-31624-3
Library of Congress Catalog Card Number: 2015942038
Additional copies of this report are available for sale from the National Academies
Press, 500 Fifth Street, NW, Keck 360, Washington, DC 20001; (800) 624-6242 or
(202) 334-3313; http://www.nap.edu.
For more information about the Institute of Medicine, visit the IOM home page
at: www.iom.edu.
Copyright 2015 by the National Academy of Sciences. All rights reserved.
Printed in the United States of America
The serpent has been a symbol of long life, healing, and knowledge among almost
all cultures and religions since the beginning of recorded history. The serpent ad-
opted as a logotype by the Institute of Medicine is a relief carving from ancient
Greece, now held by the Staatliche Museen in Berlin.
Suggested citation: IOM (Institute of Medicine). 2015. Public health implications

of raising the minimum age of legal access to tobacco products. Washington, DC:
The National Academies Press.

“Knowing is not enough; we must apply.

Willing is not enough; we must do.”
—Goethe
Advising the Nation. Improving Health.

The National Academy of Sciences is a private, nonprofit, self-perpetuating society

of distinguished scholars engaged in scientific and engineering research, dedicated to
the furtherance of science and technology and to their use for the general welfare.
Upon the authority of the charter granted to it by the Congress in 1863, the Acad-
emy has a mandate that requires it to advise the federal government on scientific
and technical matters. Dr. Ralph J. Cicerone is president of the National Academy
of Sciences.
The National Academy of Engineering was established in 1964, under the charter
of the National Academy of Sciences, as a parallel organization of outstanding en-
gineers. It is autonomous in its administration and in the selection of its members,
sharing with the National Academy of Sciences the responsibility for advising the
federal government. The National Academy of Engineering also sponsors engineer-
ing programs aimed at meeting national needs, encourages education and research,
and recognizes the superior achievements of engineers. Dr. C. D. Mote, Jr., is presi-
dent of the National Academy of Engineering.
The Institute of Medicine was established in 1970 by the National Academy of

Sciences to secure the services of eminent members of appropriate professions in
the examination of policy matters pertaining to the health of the public. The Insti-
tute acts under the responsibility given to the National Academy of Sciences by its
congressional charter to be an adviser to the federal government and, upon its own
initiative, to identify issues of medical care, research, and education. Dr. Victor J.
Dzau is president of the Institute of Medicine.
The National Research Council was organized by the National Academy of Sci-
ences in 1916 to associate the broad community of science and technology with the
Academy’s purposes of furthering knowledge and advising the federal government.
Functioning in accordance with general policies determined by the Academy, the
Council has become the principal operating agency of both the National Academy
of Sciences and the National Academy of Engineering in providing services to
the government, the public, and the scientific and engineering communities. The
Council is administered jointly by both Academies and the Institute of Medicine.
Dr. Ralph J. Cicerone and Dr. C. D. Mote, Jr., are chair and vice chair, respectively,
of the National Research Council.
www.national-academies.org

COMMITTEE ON THE PUBLIC HEALTH

IMPLICATIONS OF RAISING THE MINIMUM AGE
FOR PURCHASING TOBACCO PRODUCTS
RICHARD J. BONNIE (Chair), Harrison Foundation Professor of

Medicine and Law, Professor of Psychiatry and Neurobehavioral
Sciences, Director of the Institute of Law, Psychiatry, and Public
Policy, University of Virginia
ANTHONY J. ALBERG, Blatt Ness Distinguished Endowed Chair in
Oncology, Professor, Public Health Sciences, Interim Director of
Hollings Cancer Center, Medical University of South Carolina
REGINA BENJAMIN, NOLA.com/Times Picayune Endowed Chair in
Public Health Sciences, Xavier University, New Orleans
JONATHAN CAULKINS, Professor, Operations Research and Public
Health Policy, Heinz College of Public Policy and Management,
Operations Research Department, Carnegie Mellon University
BONNIE HALPERN-FELSHER, Professor, Department of Pediatrics,
Director of Research, Associate Director of Adolescent Medicine
Fellowship Program, Division of Adolescent Medicine, Stanford
University
SWANNIE JETT, Executive Director, Florida Department of Health in
Seminole County
HARLAN JUSTER, Director, Bureau of Tobacco Control, New York State
Department of Health
JONATHAN D. KLEIN, Associate Executive Director, Julius B. Richmond
Center of Excellence for Children and Secondhand Smoke, American
Academy of Pediatrics
PAULA M. LANTZ, Professor and Chair, Department of Health Policy
and Management, Milken Institute School of Public Health, George
Washington University
ROBIN MERMELSTEIN, Director of the Institute for Health Research
and Policy, Professor of Psychology, Clinical Professor of Community
Health Sciences, School of Public Health, University of Illinois,
Chicago
RAFAEL MEZA, Assistant Professor, Department of Epidemiology,
University of Michigan
PATRICK O’MALLEY, Research Professor, Institute for Social Research,
University of Michigan
KIMBERLY THOMPSON, Professor of Preventive Medicine and Global
Health, University of Central Florida College of Medicine, President,
Kid Risk, Inc.

Consultants
THEODORE R. HOLFORD, Susan Dwight Bliss Professor of Public
Health (Biostatistics) and Professor of Statistics, Yale School of
Medicine, Yale University
DAVID T. LEVY, Professor, Lombardi Comprehensive Cancer Center,
Georgetown University Medical Center
MARIA RODITIS, Postdoctoral Research Fellow, Adolescent Medicine,
Division of Adolescent Medicine, Department of Pediatrics, Stanford
University
IOM Staff
KATHLEEN STRATTON, Study Director
LESLIE Y. KWAN, Research Associate
BETTINA RITTER, Research Assistant
ANNA MARTIN, Senior Program Assistant
DORIS ROMERO, Financial Associate
ROSE MARIE MARTINEZ, Senior Board Director, Board on Population
Health and Public Health Practice
vi

Reviewers
T
his report has been reviewed in draft form by individuals chosen for
their diverse perspectives and technical expertise, in accordance with
procedures approved by the National Research Council’s Report
Review Committee. The purpose of this independent review is to provide
candid and critical comments that will assist the institution in making its
published report as sound as possible and to ensure that the report meets
institutional standards for objectivity, evidence, and responsiveness to the
study charge. The review comments and draft manuscript remain confiden-
tial to protect the integrity of the deliberative process. We wish to thank the
following individuals for their review of this report:
ANNETTE M. BACHAND, Colorado State University

SANJAY BASU, Stanford Prevention Research Center
CHRISTINE DELNEVO, Rutgers School of Public Health
EDWARD EHLINGER, Minnesota Department of Health
MICHAEL P. ERIKSEN, Georgia State University
THOMAS J. GLYNN, Stanford University and American Cancer
Society
STEVEN A. SCHROEDER, University of California, San Francisco
JOSHUA M. SHARFSTEIN, Maryland Department of Health and
Mental Hygiene
LAURENCE STEINBERG, Temple University
JENNIFER IRVIN VIDRINE, University of Texas MD Anderson
Cancer Center
vii

viii REVIEWERS
KENNETH W. WACHTER, University of California, Berkeley

ALEXANDER C. WAGENAAR, University of Florida
Although the reviewers listed above have provided many constructive

comments and suggestions, they were not asked to endorse the conclusions
or recommendations, nor did they see the final draft of the report before
its release. The review of this report was overseen by SUSAN J. CURRY,
University of Iowa, and RONALD S. BROOKMEYER, University of Cali-
fornia, Los Angeles. Appointed by the National Research Council and
the Institute of Medicine, they were responsible for making certain that
an independent examination of this report was carried out in accordance
with institutional procedures and that all review comments were carefully
considered. Responsibility for the final content of this report rests entirely
with the authoring committee and the institution.

Preface
T
he Surgeon General’s clarion call in 1964 for “appropriate remedial
action” to address the hazards of smoking is often credited with hav-
ing launched the nation’s public health campaign against cigarettes.
Effective federal action was impeded for more than three decades by a sym-
bolic congressional action in 1965 mandating weak package warnings and
then by the regressive decision by Congress in 1969 to preempt the states
from regulating tobacco advertising “based on smoking and health.” The
1969 legislation also banned tobacco advertising on television and thereby
erased the country’s first major tobacco control initiative—the hugely sig-
nificant ruling by the Federal Communications Commission that broadcast-
ers who aired tobacco advertisements were required by the agency’s fairness
doctrine to make time available for antismoking messages.
Attention then shifted to the states, largely driven by a grassroots
movement for public smoking restrictions. The campaign was given major
boosts by an important Surgeon General report emphasizing the addictive
properties of nicotine (1988) and an Environmental Protection Agency re-
port on the environmental hazards of tobacco smoke (1992). Another key
building block of contemporary tobacco control was the initiative aiming to
reduce youth smoking spearheaded by Congressman Mike Synar in 1992.
The Synar Amendment requires states to enact and enforce youth access
restrictions or else forfeit 40 percent of their block grants for substance
abuse prevention and treatment. Within 2 years, the Synar Amendment was
followed by two major reports by the Surgeon General and by the Institute
of Medicine (IOM) on preventing the onset of nicotine addiction in adoles-
cents and by a rhetorically and politically important initiative by Food and
ix

x PREFACE
Drug Administration (FDA) Commissioner David Kessler characterizing

nicotine addiction as a “pediatric disease.” Despite some dissension within
the ranks of tobacco control advocacy, preventing youth initiation took its
place as one of the core strategic components of tobacco control.
The campaign against secondhand tobacco smoke and the new focus
on child protection and the prevention of addiction played pivotal roles in
the gradual evolution of public support for aggressive tobacco control in
the 1990s. The cause of tobacco control was also fundamentally acceler-
ated by the emerging evidence that cigarettes have been engineered to be
addictive and by the public distaste for industry advertising campaigns
that seemed so obviously targeted at children and adolescents. In 1995, as
the policy context for tobacco control rapidly evolved, FDA announced
its innovative initiative to declare jurisdiction over cigarettes as “nicotine
delivery devices” and its intention to develop a new rule aiming to reduce
youth smoking. FDA’s proposed rule included limitations on advertising
and promotion as well as federal restrictions on youth access. Although the
age of access in FDA’s regulation was 18, the agency considered setting the
minimum age at 21. Whatever the reasoning within FDA may have been,
the consensus within the IOM committee that authored the 1994 report on
youth smoking was that setting the age at 21 was too large a leap for reform
in a political and social context in which existing youth access restrictions
were largely unenforced and cigarettes were easily available to children old
enough to put coins in a vending machine.
FDA’s Tobacco Rule was proposed in 1995, promulgated in 1996,
and invalidated by the Supreme Court in 2000. However, momentum for
aggressive tobacco control continued to build throughout this period. The
state attorney generals’ lawsuits against the tobacco companies to recover
Medicaid costs attributable to smoking—and the accompanying disclosures
of industry documents—led to the Master Settlement Agreement in 1998
and to aborted negotiations regarding federal tobacco regulation. Mean-
while, social norms toward smoking have been transformed, prevalence has
gradually declined, more reports on tobacco have been issued by the IOM
and by Surgeons General, and the Family Smoking Prevention and Tobacco
Control Act was enacted in 2009. Tobacco advocates have begun to focus
on the “end game” for cigarette smoking.
It is in this context that Congress directed FDA in the Tobacco Control
Act to commission a report on the public health implications of raising the
minimum age of legal access to tobacco products. Many states and locali-
ties are considering proposals to raise the age, and some have already done
so. In light of the extraordinary momentum achieved by tobacco control
advocacy over the past three decades, talking about raising the age of youth
access may seem anticlimactic. However, cigarette smoking is a stubborn
and costly public health problem, and the tobacco industry is resourceful

PREFACE xi
and creative. Adult prevalence remains about 18 percent, and smoking-

related deaths approach 480,000 per year.
Although initiation rates have been dropping in recent years, history
shows that they can reverse course just as easily. And investments in to-
bacco control tend to erode whenever the economy weakens. The develop-
ment and marketing of new products is a wild card in the epidemiology of
tobacco use. E-cigarettes and modified-risk tobacco products may eventu-
ally reduce the prevalence of cigarette smoking, but it is also possible that
these products could serve as starter products for people who would not
otherwise have begun smoking cigarettes and could also reduce incentives
for cessation by addicted smokers who otherwise would have quit. Bringing
these products within FDA’s regulatory jurisdiction is imperative.
Vigilance is always advisable in tobacco control. It is prudent for
federal policy makers and state and local authorities to strengthen all poli-
cies aimed at reducing the initiation of smoking, including the design and
enforcement of youth access restrictions. The minimum age of legal access
to tobacco products was set at 18 by the states more than two decades
ago in response to federal incentives and is now required by federal law.
However, states and localities remain free to raise the age. By assessing the
public health implications of raising the minimum age, this report aims
to provide the scientific guidance the states and localities need. In return,
I urge states and localities that decide to raise the age to make sure that
the necessary data are collected to evaluate the new policy in achieving its
ultimate goal—the reduction and eventual elimination of tobacco use by
children and youth.
Richard J. Bonnie, Chair

Committee on the Public Health
Implications of Raising the Minimum
Age for Purchasing Tobacco Products


Acknowledgments
T
he committee would like to express its sincere gratitude to the many
people who contributed time and expertise in the development of
this report. The work would not have been possible without the
support of our sponsor, the Center for Tobacco Products of the Food and
Drug Administration.
The committee would also like to acknowledge several consultants who
contributed to this study. First and foremost, the committee extends its
immense gratitude to Theodore R. Holford (Yale University) and David T.
Levy (Georgetown University Medical Center), whose development, knowl-
edge, and application of the Yale Lung Cancer/Cancer Intervention and
Surveillance Modeling Network and SimSmoke models were integral to the
deliberations of the committee and contributed significantly to the quality
of the report. We thank them for their patience, expertise, and many hours
of hard work. The committee is also grateful to Maria Roditis (Stanford
University), who provided consultation and editorial support on draft ma-
terials on adolescent and young adult development, and Robert Pool for
his assistance in editing the report.
Many individuals volunteered significant time and effort to address and
educate the committee during our information-gathering meetings (see Ap-
pendix E for the names of these speakers). We are grateful to each of them
for sharing their expertise and responding to our questions. The committee
would like to add special thanks to Neal Benowitz (University of California,
San Francisco) for his additional consultation and technical review of mate-
rial on developmental neurobiology and neurological response to nicotine.
The committee also expresses its deep appreciation to the staff of the
xiii

xiv ACKNOWLEDGMENTS
Institute of Medicine and the National Academies for their invaluable as-
sistance on this study. We thank the National Academies Research Center
staff for their diligent help with research and references. We are grateful for
the leadership of Rose Marie Martinez, senior board director of the Board
on Population Health and Public Health Practice, and for the energetic and
resourceful contributions of research associate Leslie Kwan, research assis-
tant Bettina Ritter, and senior program assistant Anna Martin. Finally, we
extend special thanks to study director Kathleen Stratton, whose wisdom
and guidance throughout the study process were indispensable.

Contents
SUMMARY 1
Statement of Task, 1
Interpreting the Statement of Task, 2
Adolescent and Young Adult Developmental Trajectories and
Patterns of Tobacco Use, 3
Current Practices Regarding Youth Access Restrictions, 3
Effects of Raising the MLA on Tobacco Use, 4
Adolescents Less Than 18 Years of Age, 5
Young Adults 18 to 20 Years of Age, 7
Health Effects of Raising the MLA, 8
Considerations for Policy Makers, 10
References, 12
1 INTRODUCTION 15
Tobacco Use in Adolescents and Young Adults, 16
High-Risk Populations, 17
Brief History of Tobacco Control, 17
Statement of Task, 19
Interpreting the Statement of Task, 22
Use of Models in This Report, 23
Outline of the Report, 25
References, 25
xv

xvi CONTENTS
2 PATTERNS OF TOBACCO USE BY ADOLESCENTS AND

YOUNG ADULTS 31
Prevalence of Cigarette Smoking, 31
Socioeconomic Status, 34
Geographic Variation, 36
Metropolitan Status, 40
Other Individual Risk Factors for Tobacco Use, 41
Mental Illness, 41
Sexual Orientation, 41
Initiation, 41
A Note on the Definition of Initiation, 42
Smoking Intensity, 46
Emerging Patterns, 48
Other Tobacco Products, 48
Patterns of Use and Progression of Nicotine Dependence, 52
Age of Initiation and Smoking Intensity, 55
Age of Initiation and Continued Smoking, 55
Tobacco Cessation Among Adolescents and Young Adults, 56
References, 58
3 THE DEVELOPMENTAL AND ENVIRONMENTAL CONTEXT

OF ADOLESCENT AND YOUNG ADULT TOBACCO USE 63
Cognitive, Psychosocial, and Biological Development in
Adolescents and Young Adults, 64
Cognitive Development, 64
Psychosocial Development, 66
Biological Development of Adolescents and Young Adults, 72
Tobacco-Related Decision Making by Adolescents and Young
Adults, 79
Tobacco Industry Targeting Adolescents and Young Adults, 80
Implications, 82
References, 83
4 THE EFFECTS OF TOBACCO USE ON HEALTH 91

Time Horizon for the Health Effects of Cigarette Smoking, 92
Spectrum of Health Effects, 92
Morbidity, 96
Immediate Health Effects, 96
Intermediate-Term Effects on Morbidity, 102
Long-Term Morbidity, 106
Maternal/Fetal and Infancy Health Effects, 108
Age of Initiation and Health Outcomes, 111
Other Tobacco Products and Sources of Exposure, 113

CONTENTS xvii
Impact of Cigarette Smoking on Mortality, 121

Cancer, 122
Cardiovascular Disease, 122
Diabetes, 122
COPD, 122
Increased Susceptibility to Infectious Lung Diseases, 123
Impact of Exposure to Secondhand Smoke on Mortality, 123
References, 123
5 RESTRICTIONS ON YOUTH ACCESS TO TOBACCO

PRODUCTS 129
Youth Tobacco Access Laws in the United States, 129
Federal Youth Tobacco Access Laws, 129
State and Local Youth Access Laws, 132
Enforcement of Youth Access Laws, 133
Enforcing Restrictions Against Licensed Retailers, 134
Enforcing Restrictions Against Internet Sellers, 138
Enforcing Restrictions Against Non-Licensed Sellers and
Social Distributors, 139
Summary, 140
Sources of Cigarettes for Underage Individuals, 140
References, 151
6 EVIDENCE ON THE EFFECTS OF YOUTH ACCESS

RESTRICTIONS 155
The Impact of Enacting or Raising the Minimum Legal Age to
Purchase Tobacco Products, 156
Lessons from Alcohol, 158
Summary, 161
A Logic Model for Predicting the Effects of an MLA, 161
Declarative Effects and Deterrent Effects of Legal Restrictions, 162
Reducing Availability by Increasing Search-Time Costs, 163
Penalties for Users, 165
Measures of Availability, 165
The Tobacco Control Context, 167
Effects of Retailer Interventions on Access to and Use of
Tobacco, 167
Effect of Retail Enforcement and Other Interventions on
Retailer Compliance, 167
Relationship Between Retail Interventions and Underage
Tobacco Use, 170
Relationship Between Retail Interventions and Perceived
Availability, 173
Summary, 175

xviii CONTENTS
Underage Access Restrictions in the Context of Other

Tobacco Control Policies, 176
Multiple Statewide Retailer Interventions and Underage
Tobacco Consumption, 177
Comprehensive Tobacco Control Policies and Underage
Tobacco Consumption, 178
Summary, 179
Tobacco Purchase, Use, and Possession Laws, 180
Summary, 182
References, 183
7 THE EFFECT ON TOBACCO USE OF RAISING THE

MINIMUM AGE OF LEGAL ACCESS TO TOBACCO
PRODUCTS 193
Methods, 193
Rationale for Expected Impact of Raising the Minimum Age of
Legal Access on Initiation of Tobacco Use, 195
Adolescents Less Than 15 Years of Age, 197
Adolescents 15 to 17 Years of Age, 198
Rebound, 200
Intensity, 200
Summary of Committee Estimates and Conclusions of the
Likely Effects of Raising the MLA on Tobacco Use
Initiation, 201
Estimated Initiation Effect Sizes, 204
Modeling, 205
Effects of Raising the MLA on Smoking Initiation, 208
Smoking Prevalence, 209
References, 216
8 HEALTH BENEFITS OF RAISING THE MINIMUM AGE OF

LEGAL ACCESS TO TOBACCO PRODUCTS 219
Premature Deaths Prevented, 219
Lung Cancer Deaths, 227
Maternal and Child Health Outcomes, 229
Time to Accrue Benefits, 232
Other Health Effects, 232
Immediate Health Effects, 233
Intermediate Health Effects, 234
Long-Term Health Effects, 235

CONTENTS xix
Implications of Raising the Minimum Age of Legal Access to

Tobacco Products on Health, 237
References, 240
9 OTHER CONSIDERATIONS FOR POLICY MAKERS 241

National or State Enactment of MLA, 242
Effects of Other Tobacco Control Policies, 243
Scope and Enforcement of MLA Restrictions, 245
Enforcement Against Retailers, 245
Enforcement Against Social Sources, 248
Black Market Supply to Adolescents and Young Adults, 249
Enforcement of PUP Restrictions, 249
Adolescent Development and the MLA for Tobacco, 251
Possible Public Health Effects of New Tobacco Products, 254
Possible Effects of Raising the Tobacco MLA on Use of
Alcohol and Other Drugs, 256
Concluding Remarks, 258
References, 259
APPENDIXES
A State and Local Laws on the Minimum Age of Legal Access to
Tobacco Products 265
B State Laws—Tobacco Transfers to Minors 287
C State Laws—Tobacco Purchase–Use–Possession by Minors 315
D Supplemental Information About the Models 327
E Open Meeting Agendas 369
F Committee Biographical Sketches 373


Summary
S
moking rates in the United States have declined substantially since the
release of Smoking and Health: Report of the Advisory Committee to
the Surgeon General of the Public Health Service in 1964, when the
prevalence of current cigarette smoking was around 42 percent. Recent es-
timates reveal that since 1964, tobacco control in the United States has led
to 8 million fewer premature deaths and has extended the mean life span
at age 40 by about 2 years (Holford et al., 2014). However, tobacco use
continues to have major public health implications; while the prevalence
of current cigarette smoking among U.S. adults has declined to around 18
percent (Schiller et al., 2014), more than 42 million American adults still
smoke (HHS, 2014).
STATEMENT OF TASK
The Family Smoking Prevention and Tobacco Control Act of 2009
(hereafter referred to as the Tobacco Control Act) amended the Federal
Food, Drug, and Cosmetic Act, granting the Food and Drug Administration
(FDA) broad authorities over tobacco products. The Tobacco Control Act
directed FDA to, among other things, issue regulations to restrict cigarette
and smokeless tobacco retail sales to youth and to restrict tobacco product
advertising and marketing to youth. The act, however, prohibits FDA from
taking several specific steps, including establishing a minimum age of sale

2 MINIMUM AGE OF LEGAL ACCESS TO TOBACCO PRODUCTS
of tobacco products to persons over 18 years of age.1 On the other hand,

the Tobacco Control Act directed FDA to convene a panel of experts to
conduct a study on “the public health implications of raising the minimum
age to purchase tobacco products” and to submit a report to Congress on
the issue.
In August 2013 FDA contracted with the Institute of Medicine (IOM)
to convene a committee to:
1. Examine existing literature on tobacco use initiation, and

2. Use modeling and other methods, as appropriate, to predict the
likely public health outcomes of raising the minimum age for pur-
chase of tobacco products to 21 years and 25 years.
The resulting IOM Committee on the Public Health Implications of Raising

the Minimum Age for Purchasing Tobacco Products, assembled to address
these issues, was composed of experts in public health law, the epidemiol-
ogy of tobacco use and tobacco risks, adolescent and young adult develop-
ment, risk behaviors and perceptions, public health policy and practice, and
public policy modeling.
Interpreting the Statement of Task

During a discussion at the first public meeting of the committee, a rep-
resentative of the Center for Tobacco Products of FDA urged the committee
to include in its analysis the impact of raising the minimum age of legal
access to tobacco products (MLA) to 19 years of age. The public health
impacts examined in this report include tobacco initiation, prevalence,
morbidity, and mortality. The committee uses the term “tobacco product”
to mean any product covered by FDA regulatory authority, although most
of the literature and the modeling focus on cigarettes. The committee did
not consider the economic impact of raising the MLA, nor did it compare
the effects of raising the MLA with other youth-oriented tobacco control
policies.
The Tobacco Control Act refers to both minimum age for purchase2
and minimum age for sale.3 The committee focused on the implications
of raising the MLA in the context of the body of youth access laws and
enforcement policies currently in place across the country. These laws and
policies vary considerably, not only in the scope of conduct that is prohib-
1
Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31 § 906.
111th Cong. (June 22, 2009).
2 Id. § 104.
3 Id. § 906.

SUMMARY 3
ited but also in the prescribed penalties for violations. What they all have in
common, however, is a focus on curtailing retail access to tobacco products
by underage persons, with little, if any, emphasis on punishing the under-
age users of tobacco products. The committee’s charge requests conclusions
regarding the public health implications of raising the MLA without any
recommendations regarding whether the MLA should be raised.
ADOLESCENT AND YOUNG ADULT DEVELOPMENTAL

TRAJECTORIES AND PATTERNS OF TOBACCO USE
Brain development continues until about age 25. While the develop-
ment of some cognitive abilities is achieved by age 16, the parts of the brain
most responsible for decision making, impulse control, sensation seeking,
future perspective taking, and peer susceptibility and conformity continue
to develop and change through young adulthood. Adolescent brains are
uniquely vulnerable to the effects of nicotine and nicotine addiction. Ado-
lescent and young adult developmental trajectories may be altered by social
and environmental contextual influences, including normative developmen-
tal transitions into and out of school or work or changes in living arrange-
ments or relationships.
According to the most recent results from an annual survey of adoles-
cents in grades 8, 10, and 12, American teens are smoking less than ever
before (Johnston et al., 2014b). Cigarette smoking in this age group peaked
in 1996–1997 before beginning a fairly steady and substantial decline that
continued through the mid-2000s. This decline in adolescent smoking has
continued since then, but at a slower rate (HHS, 2014). Data from 2012
show that 34.1 percent of Americans between 21 and 25 were current
cigarette users, making that the age group with the highest prevalence of
cigarette smoking (SAMHSA, 2013). While almost 90 percent of people
who have ever smoked daily first tried a cigarette before 19 years of age, the
fact that nearly all others who ever smoked daily tried their first cigarette
before the age of 26 should not be overlooked (see Table 2-8 in Chapter 2).
Additionally, only 54 percent of daily smokers are smoking daily before age
18, but 85 percent are doing so by age 21 and 94 percent before age 25.
These data strongly suggest that if someone is not a regular tobacco user
by 25 years of age, it is highly unlikely they will become one.
CURRENT PRACTICES REGARDING

YOUTH ACCESS RESTRICTIONS
Although most states currently set the minimum age of legal access to
tobacco at 18, four states set it at 19, and New York City and several other
localities around the country have raised the MLA to 21. All 50 states and

the District of Columbia prohibit commercial transfers to underage persons,

while 48 states and the District of Columbia also prohibit noncommercial
transfers (e.g., giving, exchanging, bartering, furnishing, or otherwise dis-
tributing tobacco). Based on random, unannounced compliance inspections
of tobacco retailers, the national average rate of tobacco sales to underage
individuals (i.e., noncompliance) in 2013 was 9.6 percent.
Active enforcement of tobacco minimum age restrictions, including
meaningful penalties for violations, increases retailer compliance and de-
creases the availability of retail tobacco to underage persons. However, it is
difficult to know precisely how much increasing retailer compliance reduces
the availability of retail tobacco to underage persons or how much the de-
creased retail availability of tobacco affects underage tobacco use because
of the continued availability of tobacco from noncommercial sources. Un-
derage users rely primarily on “social sources” (friends and relatives) to get
tobacco, and there is little evidence that underage individuals are obtaining
tobacco from the illegal commercial market. Bans on the noncommercial
distribution of tobacco by friends, proxy purchasers, and other social
sources are not well-enforced.
EFFECTS OF RAISING THE MLA ON TOBACCO USE

Through an iterative and consensus-driven process, the committee con-
sidered how these age-related effects would translate into potential changes
in the rates of initiation across different age segments through adolescence
and young adulthood for each of the three policy options (raising the MLA
to 19, 21, or 25 years of age). The committee assigned ordered, categorical
labels to its estimates as small, medium, or large. The committee attached
numeric ranges to each of the magnitude estimate descriptors for use in
the modeling. The committee used increments of 5 percent, ranging from 5
to 30 percent, to quantify the range of possible changes in initiation rates
for use in the models. The committee has more confidence in its estimates
pertaining to raising the MLA to 19 or 21 than in its estimates pertaining to
raising the MLA to 25 because of the greater level of extrapolation needed
for estimating change and also other factors that appear with increased age.
Conclusion 7-1: Increasing the minimum age of legal access to tobacco

products will likely prevent or delay initiation of tobacco use by ado-
lescents and young adults.
The definition of “initiation” used in this report, including in the

modeling, is having smoked 100 cigarettes. This definition is based on data
obtained from the National Health Interview Survey. Smoking at least 100
cigarettes in one’s lifetime goes beyond occasional trying or “experimenta-

SUMMARY 5
tion.” To achieve the benchmark of 100 cigarettes, one must have access
to cigarettes over a period of time and have developed symptoms of depen-
dence and stronger motives for use beyond perceived peer or social group
pressure (Dierker and Mermelstein, 2010).
A critical component in the development of dependence and continued
tobacco use is the reinforcing effects of nicotine. Adolescent brains have a
heightened sensitivity to the rewarding effects of nicotine, and this sensitiv-
ity diminishes with age (Adriani et al., 2006; Jamner et al., 2003). Thus,
the probability that a user escalates to dependence after the first few trials
is likely to decrease the further one moves away from adolescence.
Changes in the initiation of tobacco use would not necessarily be
linear with increases in the MLA or be equal for all segments of under-
age individuals. Changing the MLA has an indirect effect of helping to
change norms about the acceptability of tobacco use, but this effect may
take time to build. In addition, the norms about acceptability of tobacco
use are also likely to vary by age, with greater perceived unacceptability
for those the farther away from the MLA. If the MLA increases to 21, the
social unacceptability of smoking will be greater for a 16-year-old than for
a 20-year-old.
Given the assumption that changes in the MLA could have differential
effects on adolescents at different ages, the committee considered possible
changes in initiation rates for three age divisions: (1) adolescents under age
15; (2) adolescents between the ages of 15 and 17; and (3) individuals at
age 18 for estimates with an MLA of 19, or individuals at ages 18 to 20 or
21 to 24 for an MLA of 21 or 25, respectively. These age groupings reflect
not just differences in years from the MLA but also several important de-
velopmental transitions that play a role in tobacco use.
Conclusion 7-2: Although changes in the minimum age of legal access

to tobacco products will directly pertain to individuals who are age 18
or older, the largest proportionate reduction in the initiation of tobacco
use will likely occur among adolescents 15 to 17 years old.
Conclusion 7-3: The impact on initiation of tobacco use of raising

the minimum age of legal access to tobacco products (MLA) to 21
will likely be substantially higher than raising it to 19, but the added
effect of raising the MLA beyond age 21 to age 25 will likely be con-
siderably smaller.
Adolescents Less Than 18 Years of Age

Many adolescents under age 15 are not yet in high school or of driving
age. Adolescents under age 15 are less likely to have coworkers or members

of their peer networks who are over the MLA (with the likelihood decreas-
ing as the MLA increases). Thus, social network sources and mobility are
most restricted for adolescents under age 15. For adolescents under 15 years
of age, raising the MLA from 18 to 19 may have only a modest impact on
reducing social sources, given the small difference in age. Increasing the
MLA to 21, however, would provide a greater distancing of social sources.
Although 19-year-olds may still be in high schools and thus potentially in-
fluence those under 15, it is far less likely that 21-year-olds are in the same
social networks. On the other hand, increasing the MLA from 21 to 25
will not be likely to achieve many additional notable reductions in social
sources for those under 15 beyond what is achieved with an MLA of 21.
Although social sources play a central role in establishing adolescent
tobacco use patterns, other factors that contribute to early adolescent
tobacco use (for those who initiate before age 15) may limit the reduc-
tions that would be achieved with increases in the MLA. Adolescents
who reach a level of 100 cigarettes before 15 may be those who are most
susceptible to the reinforcing effects of nicotine, who have higher levels of
psychological or substance use comorbidities, who have a combination of
problem behaviors (of which tobacco use is one manifestation), and who
have social networks within which tobacco and other substances are more
readily available, regardless of age. Thus, the committee also expects that
there may be limits to how much changes in the MLA will affect this sub-
set of adolescents. Considering the balance of these factors, the committee
estimates that for adolescents under age 15 reductions in initiation will be
small for an MLA of 19 and medium for an MLA of 21 and an MLA of 25.
The committee expects that the greatest gains in reducing tobacco use
will be achieved for adolescents between the ages of 15 and 17. Negative
consequences for tobacco use, through parental or school controls, are
still relevant, and changes in the MLA are likely to increase these negative
consequences as social norms adjust. Adolescents in this age group are still
most likely to get tobacco through social sources (committee analysis of
Arrazola et al., 2014; Johnston et al., 2014a). Between the ages of 15 and
17 adolescent mobility increases with driving privileges. Social networks
and potential social sources of tobacco start to increase as some adolescents
take on formal, part-time jobs with coworkers who may be over the MLA.
Changing the MLA to 19 may not change social sources substantially for
these adolescents, but the committee expects that raising the MLA to 21
will substantially impact initiation. Raising the MLA to 25 may provide
only a modest additional reduction in initiation over that achieved with
an MLA of 21, given that changes to social network sources may not be
substantially different.
Balancing these factors, the committee estimates that the reduction in
initiation in this age group will likely exceed that seen in adolescents less than

SUMMARY 7
15 years of age for all policy options. Furthermore, the committee estimates
that the higher the MLA, the greater the effect on initiation rates will be.
Young Adults 18 to 20 Years of Age

By age 18, many adolescents graduate from high school and have
numerous life transitions, including entering higher education, exposure
to more adults in the workforce, leaving home, and significant changes
in social networks. Patterns of initiation to date also show a tailing off of
initiation by age 18 (committee analysis of Johnston et al., 2014a). Given
that the social networks of 18-year-olds overlap more with 19-year-olds,
the committee expects a small reduction in initiation for 18-year-olds for
an MLA of 19. The committee expects similar effects on initiation rates
for 19- and 20-year-olds as for 18-year-olds with an MLA of 21 or 25.
This expectation of increased effect is due primarily to the increased social
distancing expected when the MLA is raised to 21 or 25, but it also takes
into account the benefit of the additional maturing of executive functions
among young adults, the decreased sensitivity to the rewarding properties
of nicotine, the additional social norms proscribing tobacco use, and tobac-
co’s decreased social value and the decreased motives for use as individuals
enter the workforce or parenthood.

Changes in initiation for young adults in the 21–24 age group were
considered only for the case of raising the MLA to 25. Even under the cur-
rent MLA of 18, the probability of initiation at these ages is substantially
lower than for adolescents and younger adults. However, current patterns
of tobacco marketing suggest that young adults are increasingly targeted in
tobacco promotions (Ling and Glantz, 2002), and tobacco promotions are
frequently linked with bar settings and alcohol consumption, which may
also keep this age group susceptible to initiation (Ling and Glantz, 2002).
In addition, the committee considered that there may be more lax enforce-
ment for an MLA of 25. Considering the balance of factors, the committee
expects that some reduction in initiation will still occur with an MLA of
25 but that this reduction will be small.
Conclusion 7-4: Based on the modeling, raising the minimum age of

legal access to tobacco products, particularly to age 21 or 25, will likely
lead to substantial reductions in smoking prevalence.
Two tobacco simulation models commissioned by the committee,

SimSmoke and the Cancer Intervention and Surveillance Modeling Net-

work (CISNET) smoking population model, suggest significant reductions

in smoking prevalence from 2015 to 2100 in the United States, even under
a status quo scenario with regard to the MLA; these declines reflect ongoing
benefits from prior tobacco control policies. The models predict that rais-
ing the MLA would lead to considerable additional reductions in smoking
prevalence based on the committee’s conclusions about the likely reductions
in smoking initiation described above. Specifically, both models estimate
that raising the MLA will lead to approximately a 3 percent decrease in
smoking prevalence for an MLA of 19, a 12 percent decrease for an MLA
of 21, and a 16 percent decrease for an MLA of 25 above and beyond the
decrease predicted in the status quo scenario.
HEALTH EFFECTS OF RAISING THE MLA

Given the likelihood that raising the MLA would decrease the rates of
initiation of tobacco use by adolescents and young adults, it follows that
tobacco-related disease and death would also decrease, generally in propor-
tion to the decrease in tobacco use.

legal access to tobacco products will likely lead to substantial reduc-
tions in smoking-related mortality.
Conclusion 8-2: Based on a review of the literature, raising the mini-

mum age of legal access to tobacco products (MLA) will likely imme-
diately improve the health of adolescents and young adults by reducing
the number of those with smoking-caused diminished health status. As
the initial birth cohorts affected by the policy change age into adult-
hood, the benefits of the reductions of the intermediate and long-term
adverse health effects will also begin to manifest. Raising the MLA will
also likely reduce the prevalence of other tobacco products and expo-
sure to secondhand smoke, further reducing tobacco-caused adverse
health effects, both immediately and over time.
Adolescents and adults most commonly use tobacco in the form of

cigarettes, and the adverse health effects of cigarettes are best documented
among all the various forms of tobacco use. Cigarette smoking is causally
associated with a broad spectrum of adverse health effects that begin soon
after the onset of regular smoking and significantly diminish the health
status of the smoker compared to nonsmokers. Cigarette smoking causes
many adverse health effects with an intermediate latency, such as subclinical
atherosclerosis, impaired lung development and function, diabetes, peri-
odontitis, exacerbation of asthma, subclinical organ injury, and adverse sur-

SUMMARY 9
gical outcomes. Cigarette smoking is also causally associated with a broad

spectrum of long-latency adverse health effects, such as chronic obstructive
pulmonary disease, coronary heart disease, and numerous cancers, that
cause suffering, impaired quality of life, and premature death. Results from
both models suggest that reductions in smoking-related mortality following
an increase in the MLA will be large but will not be observed for at least 30
years after the increased MLA takes effect. For example, if the MLA were
raised now to age 21 nationwide, modeling suggests that for the cohort
of people born between 2000 and 2019 there would be approximately 10
percent fewer lifetime premature deaths, lung cancer deaths, and years of
life lost (YLL) from cigarette smoking. Given the status quo projections,
this translates to approximately 249,000 fewer premature deaths, 45,000
fewer deaths from lung cancer, and 4.2 million fewer YLL.4
Smoking combustible tobacco products other than cigarettes, such as
pipes and cigars, is causally associated with a broad spectrum of adverse
health effects. The impact of raising the MLA on morbidity and mortal-
ity from these products would depend on the risk profile of each product
and the degree to which that product is used in the population over time.
Raising the MLA can also be expected to lessen exposure to secondhand
smoke from cigarettes and other combustible tobacco products. Second-
hand smoke exposure is causally associated with a number of adverse
health effects.
Conclusion 8-3: Based on a review of the literature and on the model-

ing, an increase in the minimum age of legal access to tobacco products
will likely improve maternal, fetal, and infant outcomes by reducing the
likelihood of maternal and paternal smoking.
Maternal smoking during pregnancy and secondhand smoke exposure

during infancy are causally associated with many adverse health outcomes.
Such exposures not only leave exposed infants prone to various short- and
long-term health risks but can also result in death. The SimSmoke model
projected the effects of raising the MLA on the incidence of select m
aternal–
child outcomes. Relative to the status quo, if the MLA were raised now
to age 21 nationwide, modeling projects that by 2100 there would be an
estimated 286,000 fewer pre-term births, 438,000 fewer cases of low birth
4 All absolute differences, including the numbers of premature deaths, lung cancer deaths,
and YLL, are relative to underlying status quo projections. These status quo projections esti-
mate decreases in smoking prevalence and thus smoking-attributable morbidity and mortality.
As such, the committee encourages the reader to focus on the percentage reduction rather than
on the absolute numerical estimates.

weight, and roughly 4,000 fewer sudden infant death syndrome (SIDS)
cases among mothers age 15 to 49.5
CONSIDERATIONS FOR POLICY MAKERS

The Tobacco Control Act sets a “floor” of 18 on the MLA, while al-
lowing states and localities to raise the age. Unless Congress acts to raise the
age on a national basis or delegates authority to FDA to do so, one might
expect a patchwork of different MLAs in different states and localities, as
existed for alcohol for many decades, rather than a uniform MLA across
all of the 51 jurisdictions. The simulations described in Chapters 7 and 8
model a situation in which increases in the MLA would be adopted and
implemented on a nationwide basis. In the absence of a national MLA, the
public health impact of raising the MLA for tobacco would be dependent,
first and foremost, upon the degree to which local and state governments
take up this policy. To the extent that states choose not to raise the MLA,
the effects estimated in Chapters 7 and 8 are not likely to be realized.
The strength and efficacy of existing state and local tobacco control
programs vary significantly, reflecting differences in the number and in-
tensity of tobacco control activities and the resources allocated to support
them. The modeling essentially aggregates each state’s tobacco control ac-
tivities, whether they are strong or weak. To the extent that policy makers
in individual states want to derive state-based estimates from the findings
of a national modeling exercise, they will have to take into account whether
the existing levels of tobacco control activity in their states are comparable
to the “average” state. If they are much weaker or stronger, extrapolation
from the modeling used in this report may not be suitable.
The committee expects social sources, especially proxy purchases, to
remain the primary sources of tobacco for underage persons, and it has
been realistic about the high level of continuing availability to underage
adolescents and young adults who are in the workforce or in college en-
vironments. Our estimates in this respect are predicated on relatively con-
servative assumptions. Although access to social sources could be reduced
significantly if the laws prohibiting transfers to underage persons were
aggressively enforced, the committee does not expect such a radical change
in enforcement policy in the foreseeable future, especially under a higher
MLA, because of likely public resistance. However, if a state or locality
ramped up the threat of detection and punishment against social sources,
5 All absolute differences, including the numbers of cases of pre-term births, low birth
weight, and SIDS, are relative to underlying status quo projections. These status quo projec-
tions predict that there will be decreases in smoking prevalence, and thus smoking-attributable
morbidity and mortality.

SUMMARY 11
the impact on adolescent and young adult consumption could be greater

than the committee has projected.
Concerns about adolescent vulnerability to addiction and immaturity
of judgment support an underage access restriction, but they do not resolve
the policy question about the specific age at which the line should be drawn.
The argument against raising the MLA above 18 is predicated on the as-
sumption that adolescents older than 17 are mature enough to make their
own decisions about what is in their best interests. However, evidence sug-
gests that capacities related to mature judgment, especially in emotionally
charged situations or in situations in which peer influence plays a role, are
still developing into the early 20s. Many young people in their late teens
and early 20s may also still be at elevated risk, developmentally speaking,
to becoming addicted to nicotine. A balance needs to be struck between the
personal interest of young adults in making their own choices and society’s
legitimate concerns about protecting the public health and discouraging
young people from making decisions they may later regret (IOM, 2007;
IOM and NRC, 2004). Although some line is required, 18 is not the only
developmentally plausible place to draw it. Every state sets the legal age for
certain activities higher or lower for different policy purposes, and state leg-
islators will likely continue to draw the line in different places in different
policy contexts (Bonnie and Scott, 2013; Hamilton, 2010; Steinberg, 2012).
The committee assumes that the MLA will be increased for all tobacco
products, including electronic nicotine delivery systems (ENDS), and that
the intensity of enforcement will be the same for all products. The com-
mittee sees no reason to believe that the effects of the legal norm and its
enforcement on retailer compliance, retail availability, or access to social
sources would differ materially for ENDS as compared with other tobacco
products. Given the evidence that adolescents who currently initiate to-
bacco use with ENDS rather than with conventional tobacco products are
younger (Wills et al., 2014), the main effect of raising the MLA for ENDS
will likely be to reduce the number of adolescents and young adults who
initiate tobacco use with ENDS. However, recent trends suggest that ENDS
initiation is already increasing and is likely to increase even if the MLA is
raised. Increased initiation of ENDS use may reduce initiation of cigarette
use because some adolescents and young adults who otherwise would have
initiated cigarette users will become ENDS users instead. It may also delay
initiation of cigarette use for others, including some proportion who would
not have otherwise used traditional cigarettes. Presumably FDA and state
policy makers will take these possibilities into account in setting the MLA
and will carefully monitor the promotion and use of ENDS, especially by
adolescents and young adults.
Although the full benefits of preventing initiation of tobacco use will
take decades to accrue, some direct health benefits, including those from

reduced secondhand smoke exposure, will be immediate. Perhaps the great-

est uncertainty in the committee’s assessment is the currently unpredictable
effects of the marketing and use of ENDS and other novel tobacco prod-
ucts. However, in the absence of transformative changes in the tobacco
market, social norms and attitudes, or the epidemiology of tobacco use,
the committee is reasonably confident that raising the MLA will reduce
tobacco initiation, particularly among adolescents 15 to 17 years of age,
will improve health across the life span, and will save lives.
REFERENCES
Adriani, W., V. Deroche-Gamonet, M. Le Moal, G. Laviola, and P. V. Piazza. 2006. Preexpo-
sure during or following adolescence differently affects nicotine-rewarding properties in
adult rats. Psychopharmacology 184(3–4):382–390.
Arrazola, R. A., N. M. Kuiper, and S. R. Dube. 2014. Patterns of current use of tobacco
products among U.S. high school students for 2000–2012—Findings from the National
Youth Tobacco Survey. Journal of Adolescent Health 54(1):54–60.
Bonnie, R. J., and E. S. Scott. 2013. The teenage brain: Adolescent brain research and the law.
Current Directions in Psychological Science 22(2):158–161.
Dierker, L., and R. Mermelstein. 2010. Early emerging nicotine-dependence symptoms: A
signal of propensity for chronic smoking behavior in adolescents. Journal of Pediatrics
156(5):818–822.
Hamilton, V. E. 2010. Immature citizens and the state. Brigham Young University Law Review
1055(4):1055–1148.
HHS (Department of Health and Human Services). 2014. The health consequences of
smoking—50 years of progress: A report of the Surgeon General. Atlanta, GA: U.S.
Department of Health and Human Services, Centers for Disease Control and Preven-
tion, National Center for Chronic Disease Prevention and Health Promotion, Office on
Smoking and Health.
Holford, T. R., R. Meza, K. E. Warner, C. Meernik, J. Jeon, S. H. Moolgavkar, and D. T.
Levy. 2014. Tobacco control and the reduction in smoking-related premature deaths in
the United States, 1964–2012. JAMA 311(2):164–171.
IOM (Institute of Medicine). 2007. Ending the tobacco problem: A blueprint for the nation.
Edited by R. J. Bonnie. Washington, DC: The National Academies Press.
IOM and NRC (Institute of Medicine and National Research Council). 2004. Reducing under-
age drinking: A collective responsibility. Edited by R. J. Bonnie. Washington, DC: The
National Academies Press.
Jamner, L. D., C. K. Whalen, S. E. Loughlin, R. Mermelstein, J. Audrain-McGovern, S.
Krishnan-Sarin, J. K. Worden, and F. M. Leslie. 2003. Tobacco use across the forma-
tive years: A road map to developmental vulnerabilities. Nicotine & Tobacco Research
5(Suppl 1):S71–S87.
Johnston, L. D., P. M. O’Malley, J. G. Bachman, J. E. Schulenberg, and R. A. Miech. 2014a.
Monitoring the Future: National survey results on drug use, 1975–2013: Volume 1, Sec-
ondary school students. Ann Arbor: Institute for Social Research, University of Michigan.
Johnston, L. D., P. M. O’Malley, R. A. Miech, J. G. Bachman, and J. E. Schulenberg. 2014b.
Monitoring the Future: National survey results on drug use, 1975–2013: Overview, key
findings on adolescent drug use. Ann Arbor: Institute for Social Research, University of
Michigan.

SUMMARY 13
Ling, P. M., and S. A. Glantz. 2002. Why and how the tobacco industry sells cigarettes to
young adults: Evidence from industry documents. American Journal of Public Health
92(6):908–916.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2013. Results from
the 2012 National Survey on Drug Use and Health: Summary of national findings.
Rockville, MD: Substance Abuse and Mental Health Services Administration; Center for
Behavioral Statistics and Quality; Department of Population Surveys.
Schiller, J. S., B. W. Ward, and G. Freeman. 2014. Early release of selected estimates based on
data from the 2013 National Health Interview Survey. U.S. Department of Health and
Human Services, Centers for Disease Control and Prevention and National Center for
Health Statistics. http://www.cdc.gov/nchs/data/nhis/earlyrelease/earlyrelease201406.pdf
(accessed February 22, 2015).
Steinberg, L. 2012. Should the science of adolescent brain development inform public policy?
Issues in Science and Technology 28(3):67–78.
Surgeon General’s Advisory Committee on Smoking and Health. 1964. Smoking and health:
Report of the Advisory Committee to the Surgeon General of the Public Health Service.
Washington, DC: U.S. Government Printing Office.
Wills, T. A., R. Knight, R. J. Williams, I. Pagano, and J. D. Sargent. 2014. Risk factors for
exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pedi-
atrics 135(1):e43–e51.


Introduction
T
he study of the relationship between tobacco use and health prob-
lems has a long history. The classic papers by Doll and colleagues
began to appear in 1950, with the first prospective study linking
cigarette smoking and lung cancer published in 1954 (Doll and Hill, 1954),
following up on many cross-sectional studies. A number of other impor-
tant studies added to the growing evidence base about the health risks of
smoking (e.g., Cornfield et al., 1959; Dorn, 1959; Hammond and Horn,
1958; Wynder and Graham, 1950). A seminal report, Smoking and Health:
Report of the Advisory Committee to the Surgeon General of the Public
Health Service,1 was published in 1964, and since that time Surgeons
General have released 32 other reports on a variety of topics related to
tobacco use (HHS, 2014).
Smoking rates in the United States have declined substantially since
1965 when the prevalence of current cigarette smoking was approximately
42 percent (HHS, 2014). Furthermore, it has recently been estimated that
tobacco control policies in the United States since 1965 have led to 8 mil-
lion fewer premature deaths and have extended the mean life span by 19
to 20 years per death postponed, corresponding to an increment of about
2 years in life expectancy at age 40 (Holford et al., 2014). However, to-
bacco use continues to have major public health implications: While the
prevalence of current cigarette smoking among U.S. adults declined from
24.7 percent in 1997 to 17.8 percent in 2013 (NCHS, 2014), more than 42
1 This report is often referred to as the first Surgeon General’s report on smoking; however,
the authors were actually a nongovernmental advisory committee to the Surgeon General.
15

million American adults still smoke, leading to about 480,000 premature

deaths each year (HHS, 2014).
TOBACCO USE IN ADOLESCENTS AND YOUNG ADULTS

According to the most recent results from an annual survey of ado-
lescents in grades 8, 10, and 12, American teens are smoking less than
ever before (Johnston et al., 2014). Smoking in this age group peaked in
1996–1997 before beginning a fairly steady and substantial decline that
continued through the mid-2000s (HHS, 2014). In 2013 the number of
adolescents who reported having smoked in the previous 30 days had de-
creased from peak levels seen in the mid-1990s by 79 percent in grade 8,
70 percent in grade 10, and 56 percent in grade 12 (Johnston et al., 2014).
Other surveys show similar trends (Kann et al., 2014; SAMHSA, 2013).
While tremendous strides have been made, each day more than 3,000 ado-
lescents try their first cigarette, and, if current trends continue, 5.6 million
adolescents alive today in the United States are likely to die prematurely of
smoking-related illness (HHS, 2014).
Tobacco use by young adults (those between 18 and 24 years of age)
also poses serious concerns. While nearly 90 percent of people who have
ever smoked daily first tried a cigarette before 19 years of age, the fact that
another 9.4 percent tried their first cigarette before the age of 26 should not
be overlooked (see Table 2-8 in Chapter 2). Additionally, only 54 percent
of daily smokers are smoking daily before age 18, but 85 percent are doing
so by age 21, and 94 percent before age 25 (see Table 2-8 in Chapter 2).
These data strongly suggest that if someone is not a regular tobacco user
by 25 years of age, it is highly unlikely they will become one.
Data from 2012 show that current cigarette use among adults was
highest among persons ages 21 to 25 years (34.1 percent) (SAMHSA,
2013). Certain emerging patterns of tobacco use among young adults are
also of concern, including an increase in the number of young adults who
smoke lightly (fewer than five cigarettes per day) or intermittently (non-
daily) (Fagan and Rigotti, 2009; Pierce et al., 2009) but do not consider
themselves “smokers” (Leas et al., 2014). There has also been a very recent
increase in the use of other tobacco products, such as electronic cigarettes
and hookahs, among college students (HHS, 2012; Johnston et al., 2014).
Research suggests that brain and psychosocial development continues
past the age of 18 years (IOM and NRC, 2014), the age of legal tobacco
purchase in the United States. The self-regulatory system matures gradu-
ally, beginning in pre-adolescence and continuing through young adulthood
(Steinberg, 2012). High-risk behaviors, including tobacco use, are generally
more common in adolescents and young adults than in older adults. Addi-
tionally, the tobacco industry, prohibited from marketing to those younger

INTRODUCTION 17
than 18 years of age, has for decades targeted marketing and promotional
activities to young adults (Sepe et al., 2002). The convergence of the neu-
robiological factors and the tobacco use epidemiology reinforces the impor-
tance of preventing young adults, in addition to children and adolescents,
from becoming tobacco users.
HIGH-RISK POPULATIONS
Neither the prevalence of cigarette smoking nor the use of other to-
bacco products is evenly distributed in the population; rather, both are
more heavily concentrated in certain population subgroups than in others.
Over time in the United States, cigarette smoking has become more and
more concentrated in lower socioeconomic groups defined by few years of
schooling and lower income (Fagan et al., 2007). Smoking prevalence also
varies across racial and ethnic groups, with the highest prevalence among
American Indians and Alaskan natives and the lowest among Asian Ameri-
cans (Fagan et al., 2007).
Sexual orientation is also strongly associated with the prevalence of
current smoking. Smoking prevalence is much higher among sexual minori-
ties than in the population as a whole (Lee et al., 2009; Ryan et al., 2001).
The prevalence of smoking among persons with a history of mental illness
is approximately double the prevalence in the general population (Lasser
et al., 2000). This increased likelihood of smoking in those with a history
of mental illness is not limited to one or a few psychiatric diagnoses but
rather is a cross-cutting association that applies to psychiatric diagnoses
across the board (Lasser et al., 2000). Historically, the prevalence of smok-
ing has been higher among active duty military personnel (Bray et al., 2006)
and veterans of the military (Brown, 2010) than in the general population.
There is evidence that this disparity is diminishing in the veteran population
(Hamlett-Berry et al., 2013).
BRIEF HISTORY OF TOBACCO CONTROL

The release of the 1964 report on smoking and health spurred our cur-
rent tobacco control activities, and efforts increased dramatically beginning
in the 1990s. In the early 1990s tobacco control advocates and policy mak-
ers focused on preventing children from initiating tobacco use. Congress
included an important policy lever, known as the Synar Amendment to the
Alcohol, Drug Abuse, and Mental Health Administration Reorganization
Act,2 aimed at decreasing youth access to tobacco. The Synar program re-
2ADAMHA Reorganization Act of 1992, Public Law 102-321, 102nd Cong. (July 10,
1992).

quires states to have laws in place prohibiting the sale and distribution of
tobacco products to persons under the age of 18 and to enforce those laws
effectively (SAMHSA, 2014). Failure to meet these requirements may result
in a state losing 40 percent of its substance abuse prevention and treatment
block grant. The Synar program is described in detail in Chapter 5.
In 1994 a committee convened by the Institute of Medicine (IOM)
released the report Growing Up Tobacco Free (IOM, 1994). The report
called for a comprehensive youth-oriented tobacco control strategy. The
strategy included Congress establishing a regulatory program for tobacco
products within an appropriate agency of the Public Health Service. In 1995
the commissioner of the Food and Drug Administration (FDA), Dr. David
Kessler, famously declared smoking a “pediatric disease” because “nicotine
addiction begins when most tobacco users are teen-agers” (Hilts, 1995). In
1996 FDA issued a final rule prohibiting the sale of cigarettes and smokeless
tobacco to any person under age 18 and imposing restrictions on the mar-
keting, labeling, and advertising of tobacco products (HHS, 1996). While
this 1996 rule was invalidated in 2000 by a Supreme Court decision ruling
that FDA did not have the authority to regulate tobacco products,3 it was
specifically incorporated in the Family Smoking Prevention and Tobacco
Control Act of 20094 (hereafter referred to as the Tobacco Control Act).
The Master Settlement Agreement of 1998 (MSA) resulted from settle-
ments between the attorneys general of 46 states and the 4 largest tobacco
manufacturers (NAAG, 1998). The MSA required the companies to make
annual payments to the states as compensation for some of the medical
costs of caring for people with smoking-related diseases; to curtail or
end certain tobacco marketing practices; and to dissolve tobacco industry
organizations. The MSA also called for the establishment of a national
foundation, which led to the creation of the American Legacy Foundation,
a nonprofit tobacco control research and education organization known for
its early and aggressive media campaigns about the dangers of tobacco use.
The child-focused strategy, although not universally embraced (Craig
and Boris, 2007; Glantz, 1996), galvanized attention and resources, and
significant successes followed. For example, the proportion of students in
grades 9 through 12 who had used tobacco products in the past 30 days
(including cigarettes, smokeless tobacco products, and cigars) decreased
46.1 percent between 1997 and 2011, from 43.4 percent to 23.4 percent
(CDC, 2012b). This remarkable progress sprung from a number of well-
established policy levers: increased state and federal excise taxes, compre-
3 FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 120 S. Ct. 1291, 146 L. Ed.
2d 121 (2000).
4 Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31, 111th
Cong. (June 22, 2009).

INTRODUCTION 19
hensive state tobacco control programs, smoke-free policies that help to

denormalize smoking behavior and to decrease secondhand smoke expo-
sure, national and local media campaigns to alert children and adolescents
to the dangers of tobacco use and to de-glamorize the behavior, promo-
tion of cessation strategies, school-based programs, and surveillance and
evaluation.
Today, most tobacco control programs are administered at the state and
local levels. States fund their tobacco control programs through a variety of
revenue streams, including state general funds, federal government funding,
tobacco industry settlement payments, cigarette excise taxes, and funding
from nonprofit organizations. The Office on Smoking and Health at the
Centers for Disease Control and Prevention (CDC) compiles and publishes
an evidence-based guide to help states plan and establish effective tobacco
control programs (CDC, 2014). CDC recommends that state programs be
funded at $10.53 per person in the state population. While most states
spend significantly less than that (CDC, 2012a), funding for state tobacco
control programs has nonetheless been shown to be associated with de-
creases in adolescent and young adult smoking (Farrelly et al., 2013, 2014).
STATEMENT OF TASK
The Tobacco Control Act amended the Federal Food, Drug, and Cos-
metic Act to grant FDA broad authority over tobacco products adminis-
tered by a newly created Center for Tobacco Products (CTP) funded with
user fees paid by the tobacco industry. The Tobacco Control Act directed
FDA to, among other things, issue regulations to restrict cigarette and
smokeless tobacco retail sales to youth and restrict tobacco product adver-
tising and marketing to youth. (See Box 1-1 for a summary of the major
components of the Tobacco Control Act.) On the other hand, the act
specifically prohibits FDA from taking certain actions, including reducing
nicotine levels in tobacco products to zero, requiring a prescription to pur-
chase tobacco products, banning the face-to-face sale of tobacco products
in any one specific category retail environment, banning specific classes
of tobacco products, and establishing a minimum age of sale of tobacco
products higher than 18 years of age.5 The Tobacco Control Act did, how-
ever, direct FDA to convene a panel of experts to conduct a study on “the
public health implications of raising the minimum age to purchase tobacco
products” and to submit a report to Congress on the issue.
5Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31 § 906,
111th Cong. (June 22, 2009).

BOX 1-1
Key Components of the Family Smoking Prevention and
Tobacco Control Act of 2009
What the Tobacco Control Act does
Restricts cigarettes and smokeless tobacco retail sales to youth by directing

FDA to issue regulations which, among other things:
• Require proof of age to purchase these tobacco products—the federal
minimum age to purchase is 18—Sec. 102
• Require face-to-face sales, with certain exemptions for vending machines
and self-service displays in adult-only facilities—Sec. 102
• Ban the sale of packages of fewer than 20 cigarettes—Sec. 102
Restricts tobacco product advertising and marketing to youth by directing

FDA to issue regulations which, among other things:
• Limit color and design of packaging and advertisements, including audio-
visual advertisements—Sec. 102 (However, implementation of this provi-
sion is uncertain due to pending litigation. See Discount Tobacco City &
Lottery v. USA, formerly Commonwealth Brands v. FDA.)
• Ban tobacco product sponsorship of sporting or entertainment events
under the brand name of cigarettes or smokeless tobacco—Sec.102
• Ban free samples of cigarettes and brand-name non-tobacco promotional
items—Sec. 102
Note: Among its many provisions, the Tobacco Control Act required FDA to
reissue its 1996 final regulations aimed at restricting the sale and distribution of
cigarette and smokeless tobacco products—Sec. 102
The Tobacco Control Act specifically
Requires bigger, more prominent warning labels for cigarettes and smoke-
less tobacco products:
However, the implementation date of more prominent warning labels for
cigarettes is uncertain, due to ongoing proceedings in the case of R. J. Reynolds
Tobacco Co. v. U.S. Food and Drug Administration, No. 11-1482 (D.D.C.), on ap-
peal, No. 11-5332 (D.C.Cir.).
Gives FDA authority over, among other things:

• Registration and inspection of tobacco companies—Sec. 905 of the
FDCA
• Standards for tobacco products—Sec. 907 of the FDCA
• “Premarket Review” of new tobacco products—Sec. 910 and 905 of the
FDCA
• “Modified risk” products—Sec. 911 of the FDCA
• Enforcement action plan for advertising and promotion restrictions—Sec.
105

INTRODUCTION 21
The Tobacco Control Act also requires
• Tobacco industry must disclose research on the health, toxicological,

behavioral, or physiologic effects of tobacco use—Sec. 904 of the FDCA
• Tobacco industry must disclose information on ingredients and constitu-
ents in tobacco products, and must notify FDA of any changes—Sec. 904
of the FDCA
How FDA oversees the implementation of the Tobacco Control Act
Among other things, FDA:

• Established the Center for Tobacco Products to implement the Tobacco
Control Act—Sec. 901 of the FDCA
• Established the Tobacco Products Scientific Advisory Committee to pro-
vide advice, information, and recommendations to FDA—Sec. 917 of the
FDCA
• Assesses user fees on tobacco product manufacturers and importers
based on their market share. The fees are used to fund FDA activities
related to the regulation of tobacco products—Sec. 919 of the FDCA
• Reports to Congress on how best to encourage companies to develop in-
novative products that help people stop smoking—Sec. 918 of the FDCA
• Issues regulations and conducts inspections to investigate illicit trade in
tobacco products—Sec. 920 of the FDCA
• Convenes a panel of experts to study the public health implications of
raising the minimum age to purchase tobacco products—Sec. 104
Limits on FDA’s authority:

FDA cannot:
• Ban certain specified classes of tobacco products—Sec. 907 of the FDCA
• Require the reduction of nicotine yields to zero—Sec. 907 of the FDCA
• Require prescriptions to purchase tobacco products—Sec. 906 of the
FDCA
• Ban face-to-face tobacco sales in any particular category of retail outlet—
Sec. 906 of the FDCA
The Tobacco Control Act preserves the authority of state, local, and tribal
governments to regulate tobacco products in certain specific respects. It also
prohibits, with certain exceptions, state and local requirements that are different
from, or in addition to, requirements under the provisions of the FDCA relating to
specified areas.
SOURCE: FDA, 2014.

In August 2013 FDA contracted with the IOM to convene a commit-

tee to:
1. Examine existing literature on tobacco use initiation, and

2. Use modeling and other methods, as appropriate, to predict the
likely public health outcomes of raising the minimum age for pur-
chase of tobacco products to 21 years and 25 years.
The resulting Committee on the Public Health Implications of Raising

the Minimum Age for Purchasing Tobacco Products comprises experts in
public health law, epidemiology of tobacco use and tobacco risks, adoles-
cent and young adult development, risk behaviors and perceptions, public
health policy and practice, and public policy modeling. (See Appendix F
for the biographical sketches of committee members.) The committee met
five times, including holding a public workshop. (See Appendix E for the
agendas of public meetings.)
Interpreting the Statement of Task

At its first meeting, a representative of CTP discussed the charge with
the committee. During that discussion, CTP urged the committee to in-
clude in its analysis the impact of raising the minimum age to 19, 21, and
25 years of age and the committee has done so. CTP also encouraged the
committee to conceive broadly the definition of “public health impact.” As
described in future chapters, the committee assessed the effects of possible
policy changes on tobacco initiation, prevalence, morbidity, and mortal-
ity. However, because the charge is limited to public health implications,
the committee did not analyze the overall economic impact of raising the
minimum age.
Because the Tobacco Control Act refers to both minimum age for pur-
chase6 and minimum age for sale,7 there is some ambiguity regarding the
scope of the legal restriction the committee has been instructed to assess.
The committee interpreted its charge to focus on the minimum age of legal
access to tobacco products (MLA) in the context of the body of youth ac-
cess laws and enforcement policies currently in place across the country.
As will be discussed at length in this report, these laws and policies vary
considerably, not only in the scope of the conduct that is prohibited but also
in the prescribed penalties for violations. What all of the laws and policies
have in common, however, is a focus on curtailing retail access to tobacco
products by underage persons, with little, if any, emphasis on punishing
6 Id. § 104.
7 Id. § 906.

INTRODUCTION 23
the underage users of tobacco products themselves. As requested by CTP,

the committee has made no recommendations on whether the MLA should
be raised. The report is limited to findings and conclusions bearing on the
public health implications of raising the MLA as well as a review of relevant
policy considerations.
As the reader will see, there exists an abundance of relevant data on
adolescent tobacco use, risks of tobacco use, effects of youth access restric-
tions and their enforcement, and adolescent and young adult brain and
psychosocial development. However, there are many important unknowns,
including a rapidly changing landscape of tobacco products. The recent
increase in the use of electronic nicotine delivery systems and hookahs by
adolescents and young adults could have a substantial effect on the use of
cigarettes and other combustible tobacco products, but it is too early to
make informed predictions about these effects.
Additionally, there is no direct empirical evidence on the effects on
adolescent and young adult tobacco use of raising the MLA above 18 years
of age. Four states have an MLA of 19 years, but the effect of setting the
age at 19 has not been studied. Several small jurisdictions in Massachusetts
have raised the MLA above 18 years, but, again, the effect of doing so has
not been evaluated. New York City raised the MLA to 21 years as of May
2014, but insufficient time has passed to study its effect. In the absence of
pertinent studies of the effect of raising the MLA for tobacco, the commit-
tee drew on the relevant bodies of literature on adolescent and young adult
development, the epidemiology of tobacco use, enforcement of youth access
restrictions, studies of the effect of raising the minimum legal drinking age
for alcohol, and the effects of other tobacco control policies to estimate the
likely effects of raising the MLA for tobacco on initiation of tobacco use
and the health consequences of that level of tobacco use.
Use of Models in This Report

The charge to the committee specifically includes the use of modeling.
Simulation modeling is the primary tool used to assess the potential out-
comes, benefits, and costs of public health and policy interventions (Feuer
et al., 2004; Habbema et al., 2006; NRC, 1994; Thompson and Graham,
1996). Models complement traditional statistical and epidemiological ap-
proaches, and they translate and synthesize available evidence into an
integrated framework to assist with decision making. Notable examples of
the application of simulation models in non-tobacco public health policy
include pandemic preparedness (Halloran et al., 2008), the design of op-
timal vaccination strategies (Elbasha et al., 2009; Kim and Goldie, 2008;
Kim et al., 2009; Thompson, 2013; Thompson et al., 2015; Van de Velde
et al., 2012), cocaine use simulations (Caulkins et al., 2007; Rydell and

Everingham, 1994), and the assessment of effective cancer screening strate-

gies (de Koning et al., 2014; Knudsen et al., 2007; Mandelblatt et al., 2009;
Zauber et al., 2008). Some guidelines exist to support the development of
policy models in some domains, and generally they suggest that compar-
ing and contrasting the predictions from different models can enhance the
validity of the conclusions and allow for the exploration of a wider range
of assumptions and of potential policy and health outcomes (Caro et al.,
2012; Eddy et al., 2012; Habbema et al., 2006; Mandelblatt et al., 2009;
Weinstein et al., 2003; WHO, 2008; Zauber et al., 2008).
To date, tobacco control simulation models have focused primarily
on cigarette smoking and have provided estimates of the impact of cur-
rent policies (program evaluation), forecasts of their future effects (status
quo projections), and assessments of the possible effects of new policies
(Ahmad, 2005a,b; HHS, 2014; Holford et al., 2014; Levy et al., 2005,
2010, 2012; Mendez and Warner, 2000; Mendez et al., 2013; NCI, 2007;
Warner and Mendez, 2012). Reports from the U.S. government have high-
lighted the important insights of these models (HHS, 2014; NCI, 2007).
For this report, the committee commissioned the use of two established
cigarette smoking macro-simulation models to complement its conclusions
about the effects of a change in the MLA on tobacco initiation by provid-
ing quantitative estimates of how the likely effects on initiation would
affect future smoking prevalence and select measures of smoking-related
morbidity and mortality. The models are the Cancer Intervention and Sur-
veillance Modeling Network (CISNET)8 smoking population model and
the SimSmoke model. Both models simulate annual age-specific smoking
prevalence and smoking-attributable mortality. In addition, CISNET mod-
els the variation in smoking patterns by birth cohort and can account for
the effects of smoking intensity. SimSmoke models the effects of important
tobacco control policies and supports the simulation of maternal and child
health outcomes. While increasing the MLA is currently the purview of
states and localities, the models project the effects of a policy change on
the United States as a whole and cannot take into consideration important
differences across the country that could influence the magnitude of the
effect of raising the MLA in states or localities.
8 CISNET is a consortium of National Cancer Institute–sponsored investigators who use sta-
tistical modeling to improve the understanding of cancer control interventions in prevention,

screening, and treatment and also their effects on population trends in incidence and mortality.
As noted, for simplicity, the committee uses CISNET to refer both to the consortium as well
as to the CISNET smoking population model used in this report.

INTRODUCTION 25
OUTLINE OF THE REPORT

The next five chapters provide foundational material on tobacco use
patterns (Chapter 2), brain and psychosocial development in adolescents
and young adults (Chapter 3), health effects of tobacco use (Chapter 4),
the current legal landscape regarding minimum age laws and the enforce-
ment of youth access restrictions (Chapter 5), and the effectiveness of youth
access restrictions (Chapter 6). The committee’s conclusions regarding the
likely impact of raising the MLA on initiation and prevalence of tobacco
use are set forth in Chapter 7 and the conclusions on the likely impact of
raising the MLA on morbidity and mortality are found in Chapter 8. The
report concludes with a discussion of the considerations for policy makers.
The details of the models used can be found in Appendix D, along with
comprehensive results.
REFERENCES
Ahmad, S. 2005a. Closing the youth access gap: The projected health benefits and cost sav-
ings of a national policy to raise the legal smoking age to 21 in the United States. Health
Policy 75(1):74–84.
———. 2005b. The cost-effectiveness of raising the legal smoking age in California. Medical
Decision Making 25(3):330–340.
Bray, R. M., K. L. Rae Olmsted, J. Williams, R. P. Sanchez, and M. Hartzell. 2006. Progress
toward Healthy People 2000 objectives among U.S. military personnel. Preventive Medi-
cine 42(5):390–396.
Brown, D. W. 2010. Smoking prevalence among U.S. veterans. Journal of General Internal
Medicine 25(2):147–149.
Caro, J. J., A. H. Briggs, U. Siebert, K. M. Kuntz, and ISPOR–SMDM Modeling Good
Research Practices Task Force. 2012. Modeling good research practices—Overview: A
report of the ISPOR–SMDM Modeling Good Research Practices Task Force-1. Medical
Decision Making 32(5):667–677.
Caulkins, J. P., P. Dietze, and A. Ritter. 2007. Dynamic compartmental model of trends in
Australian drug use. Health Care Management Science 10(2):151–162.
CDC (Centers for Disease Control and Prevention). 2012a. State tobacco revenues compared
with tobacco control appropriations—United States 1998–2010. Morbidity and Mortal-
ity Weekly Report 61(20):370–374.
———. 2012b. Trends in the prevalence of tobacco use—National YRBS: 1991–2011. http://
www.cdc.gov/healthyyouth/yrbs/pdf/us_tobacco_trend_yrbs.pdf (accessed January 20,
2015).
———. 2014. Best practices for comprehensive tobacco control programs—2014. Atlanta,
GA: U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention, National Center for Chronic Disease Prevention and Health Promotion, Of-
fice on Smoking and Health.
Cornfield, J., W. Haenszel, E. C. Hammond, A. M. Lilienfeld, M. B. Shimkin, and E. L.
Wynder. 1959. Smoking and lung cancer: Recent evidence and a discussion of some
questions. Journal of the National Cancer Institute 22(1):173–203.
Craig, M. J., and N. W. Boris. 2007. Youth tobacco access restrictions: Time to shift resources
to other interventions? Health Promotion Practice 8(1):22–27.

de Koning, H. J., R. Meza, S. K. Plevritis, K. ten Haaf, V. N. Munshi, J. Jeon, S. A. Erdogan, C.

Y. Kong, S. S. Han, J. van Rosmalen, S. E. Choi, P. F. Pinsky, A. Berrington de Gonzalez,
C. D. Berg, W. C. Black, M. C. Tammemagi, W. D. Hazelton, E. J. Feuer, and P. M.
McMahon. 2014. Benefits and harms of computed tomography lung cancer screening
strategies: A comparative modeling study for the U.S. Preventive Services Task Force.
Annals of Internal Medicine 160(5):311–320.
Doll, R., and A. B. Hill. 1954. The mortality of doctors in relation to their smoking habits.
BMJ 1(4877):1451–1455.
Dorn, H. F. 1959. Tobacco consumption and mortality from cancer and other diseases. Public
Health Reports 74(7):581–593.
Eddy, D. M., W. Hollingworth, J. J. Caro, J. Tsevat, K. M. McDonald, J. B. Wong, and ISPOR-
SMDM Modeling Good Research Practices Task Force. 2012. Model transparency and
validation: A report of the ISPOR–SMDM Modeling Good Research Practices Task
Force-7. Value in Health 15(6):843–850.
Elbasha, E. H., E. J. Dasbach, R. P. Insinga, R. M. Haupt, and E. Barr. 2009. Age-based pro-
grams for vaccination against HPV. Value in Health 12(5):697–707.
Fagan, P., and N. A. Rigotti. 2009. Light and intermittent smoking: The road less traveled.
Nicotine & Tobacco Research 11(2):107–110.
Fagan, P., E. T. Moolchan, D. Lawrence, A. Fernander, and P. K. Ponder. 2007. Identifying
health disparities across the tobacco continuum. Addiction 102(Suppl 2):5–29.
Farrelly, M. C., B. R. Loomis, B. Han, J. Gfroerer, N. Kuiper, G. L. Couzens, S. Dube, and
R. S. Caraballo. 2013. A comprehensive examination of the influence of state tobacco
control programs and policies on youth smoking. American Journal of Public Health
103(3):549–555.
Farrelly, M. C., B. R. Loomis, N. Kuiper, B. Han, J. Gfroerer, R. S. Caraballo, T. F. Pechacek,
and G. L. Couzens. 2014. Are tobacco control policies effective in reducing young adult
smoking? Journal of Adolescent Health 54(4):481–486.
FDA (Food and Drug Administration). 2014. Overview of the Family Smoking Prevention
and Tobacco Control Act. http://www.fda.gov/downloads/TobaccoProducts/Guidance
ComplianceRegulatoryInformation/UCM336940.pdf (accessed November 20, 2014).
Feuer, E. J., R. Etzioni, K. A. Cronin, and A. Mariotto. 2004. The use of modeling to under-
stand the impact of screening on U.S. mortality: Examples from mammography and PSA
testing. Statistical Methods in Medical Research 13(6):421–442.
Glantz, S. A. 1996. Preventing tobacco use—The youth access trap. American Journal of
Public Health 86(2):156–158.
Habbema, J. D., C. B. Schechter, K. A. Cronin, L. D. Clarke, and E. J. Feuer. 2006. Model-
ing cancer natural history, epidemiology, and control: Reflections on the CISNET breast
group experience. Journal of the National Cancer Institute Monographs 36:122–126.
Halloran, M. E., N. M. Ferguson, S. Eubank, I. M. Longini, Jr., D. A. Cummings, B. Lewis,
S. Xu, C. Fraser, A. Vullikanti, T. C. Germann, D. Wagener, R. Beckman, K. Kadau, C.
Barrett, C. A. Macken, D. S. Burke, and P. Cooley. 2008. Modeling targeted layered
containment of an influenza pandemic in the United States. Proceedings of the National
Academy of Sciences of the United States of America 105(12):4639–4644.
Hamlett-Berry, K., D. E. Christofferson, and R. A. Martinello. 2013. Smoking and tobacco use
within the Department of Veterans Affairs. American Journal of Public Health 103(7):e3.
Hammond, E. C., and D. Horn. 1958. Smoking and death rates: Report on forty-four months
of follow-up of 187,783 men, 2: Death rates by cause. JAMA 166(11):1294–1308.
HHS (Department of Health and Human Services). 1996. 21 CFR parts 801, 803, 804, 807,
820, and 897, Regulations restricting the sale and distribution of cigarettes and smoke-
less tobacco to protect children and adolescents; final rule. Federal Register 61(168):
44396–44618.

INTRODUCTION 27
———. 2012. Preventing tobacco use among youth and young adults: A report of the Sur-
geon General. Atlanta GA: U.S. Department of Health and Human Services, Centers for
Disease Control and Prevention, National Center for Chronic Disease Prevention and
Health Promotion, Office on Smoking and Health.
———. 2014. The health consequences of smoking—50 years of progress: A report of the
Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers
for Disease Control and Prevention, National Center for Chronic Disease Prevention and
Hilts, P. J. 1995. F.D.A. Head calls smoking a pediatric disease. The New York Times (March
9, 1995). http://www.nytimes.com/1995/03/09/us/fda-head-calls-smoking-a-pediatric-
disease.html (accessed February 4, 2015).
Holford, T. R., R. Meza, K. E. Warner, C. Meernik, J. Jeon, S. H. Moolgavkar, and D. T.
Levy. 2014. Tobacco control and the reduction in smoking-related premature deaths in
the United States, 1964–2012. JAMA 311(2):164–171.
IOM (Institute of Medicine). 1994. Growing up tobacco free: Preventing nicotine addiction in
children and youths. Edited by B. S. Lynch and R. J. Bonnie. Washington, DC: National
Academy Press.
IOM and NRC (Institute of Medicine and National Research Council). 2014. Investing in the
health and well-being of young adults. Edited by R. J. Bonnie, C. Stroud, and H. Breiner.
Washington, DC: The National Academies Press.
Johnston, L. D., P. M. O’Malley, R. A. Miech, J. G. Bachman, and J. E. Schulenberg. 2014.
Monitoring the future: National survey results on drug use, 1975–2013: Overview, key
Michigan.
Kann, L., S. Kinchen, S. L. Shanklin, K. H. Flint, J. Hawkins, W. A. Harris, R. Lowry, E.
O’Malley Olsen, T. McManus, D. Chyen, L. Whittle, E. Taylor, Z. Demissie, N. Brener,
J. Thornton, J. Moore, and S. Zaza, Centers for Disease Control and Prevention. 2014.
Youth risk behavior surveillance—United States, 2013. Morbidity and Mortality Weekly
Report Surveillance Summaries 63(4):1–168.
Kim, J. J., and S. J. Goldie. 2008. Health and economic implications of HPV vaccination in
the United States. New England Journal of Medicine 359(8):821–832.
Kim, J. J., J. Ortendahl, and S. J. Goldie. 2009. Cost-effectiveness of human papillomavirus
vaccination and cervical cancer screening in women older than 30 years in the United
States. Annals of Internal Medicine 151(8):538–545.
Knudsen, A. B., P. M. McMahon, and G. S. Gazelle. 2007. Use of modeling to evalu-
ate the cost-effectiveness of cancer screening programs. Journal of Clinical Oncology
25(2):203–208.
Lasser, K., J. W. Boyd, S. Woolhandler, D. U. Himmelstein, D. McCormick, and D. H.
Bor. 2000. Smoking and mental illness: A population-based prevalence study. JAMA
284(20):2606–2610.
Leas, E. C., R. W. Zablocki, S. D. Edland, and W. K. Al-Delaimy. 2014. Smokers who report
smoking but do not consider themselves smokers: A phenomenon in need of further at-
tention. Tobacco Control. Epub February 5.
Lee, J. G., G. K. Griffin, and C. L. Melvin. 2009. Tobacco use among sexual minorities in the
USA, 1987 to May 2007: A systematic review. Tobacco Control 18(4):275–282.
Levy, D. T., N. Nikolayev, and E. A. Mumford. 2005. Recent trends in smoking and the role
of public policies: Results from the SimSmoke tobacco control policy simulation model.
Addiction 100(10):1526–1537.
Levy, D. T., A. L. Graham, P. L. Mabry, D. B. Abrams, and C. T. Orleans. 2010. Modeling
the impact of smoking cessation treatment policies on quit rates. American Journal of
Preventive Medicine 38(3S):S364–S372.

Levy, D., L. M. de Almeida, and A. Szklo. 2012. The Brazil SimSmoke policy simulation
model: The effect of strong tobacco control policies on smoking prevalence and smoking-
attributable deaths in a middle income nation. PLoS Medicine 9(11):1–12.
Mandelblatt, J. S., K. A. Cronin, S. Bailey, D. A. Berry, H. J. de Koning, G. Draisma, H.
Huang, S. J. Lee, M. Munsell, S. K. Plevritis, P. Ravdin, C. B. Schechter, B. Sigal, M. A.
Stoto, N. K. Stout, N. T. van Ravesteyn, J. Venier, M. Zelen, E. J. Feuer, and Breast
Cancer Working Group of the Cancer Intervention Surveillance Modeling Network.
2009. Effects of mammography screening under different screening schedules: Model
estimates of potential benefits and harms. Annals of Internal Medicine 151(10):738–747.
Mendez, D., and K. E. Warner. 2000. Smoking prevalence in 2010: Why the Healthy People
goal is unattainable. American Journal of Public Health 90(3):401–403.
Mendez, D., O. Alshanqeety, and K. E. Warner. 2013. The potential impact of smoking control
policies on future global smoking trends. Tobacco Control 22(1):46–51.
NAAG (National Association of Attorneys General). 1998. Master settlement agreement.
http://publichealthlawcenter.org/sites/default/files/resources/master-settlement-agreement.
pdf (accessed September 5, 2014).
NCHS (National Center for Health Statistics). 2014. Early release of selected estimates based
on data from the 2013 National Health Interview Survey. http://www.cdc.gov/nchs/data/
nhis/earlyrelease/earlyrelease201403.pdf (accessed February 6, 2015).
NCI (National Cancer Institute). 2007. Greater than the sum: Systems thinking in tobacco
control. Tobacco Control Monograph No. 18. Edited by A. Best, P. I. Clark, S. J.
Leischow, and W. M. K. Torchim. Bethesda, MD: U.S. Department of Health and Human
Services, National Institutes of Health, National Cancer Institute.
NRC (National Research Council). 1994. Science and judgment in risk assessment. Washing-
ton, DC: National Academy Press.
Pierce, J. P., M. M. White, and K. Messer. 2009. Changing age-specific patterns of cigarette
consumption in the United States, 1992–2002: Association with smoke-free homes and
state-level tobacco control activity. Nicotine & Tobacco Research 11(2):171–177.
Ryan, H., P. M. Wortley, A. Easton, L. Pederson, and G. Greenwood. 2001. Smoking among
lesbians, gays, and bisexuals: A review of the literature. American Journal of Preventive
Medicine 21(2):142–149.
Rydell, C. P., and S. S. Everingham. 1994. Controlling cocaine: Supply versus demand pro-
grams. Santa Monica, CA: RAND Corporation.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2013. Results from
the 2012 National Survey on Drug Use and Health: Summary of national findings. Rock-
ville, MD: Substance Abuse and Mental Health Services Administration.
———. 2014. About the Synar amendment and program. http://beta.samhsa.gov/synar/about
(accessed September 8, 2014).
Sepe, E., P. M. Ling, and S. A. Glantz. 2002. Smooth moves: Bar and nightclub tobacco pro-
motions that target young adults. American Journal of Public Health 92(3):414–419.
Thompson, K. M. 2013. Modeling poliovirus risks and the legacy of polio eradication. Risk
Analysis 33(4):505–515.
Thompson, K. M., and J. D. Graham. 1996. Going beyond the single number: Using proba-
bilistic risk assessment to improve risk management. Human and Ecological Risk As-
sessment 2(4):1008–1034.
Thompson, K. M., R. J. Duintjer Tebbens, M. A. Pallansch, S. G. F. Wassilak, and S. L. Cochi.
2015. Polio eradicators use integrated analytical models to make better decisions. Inter-
faces 45(1):5–25.

INTRODUCTION 29
Van de Velde, N., M. C. Boily, M. Drolet, E. L. Franco, M. H. Mayrand, E. V. Kliewer, F.

Coutlee, J. F. Laprise, T. Malagon, and M. Brisson. 2012. Population-level impact of the
bivalent, quadrivalent, and nonavalent human papillomavirus vaccines: A model-based
analysis. Journal of the National Cancer Institute 104(22):1712–1723.
Warner, K. E., and D. Mendez. 2012. Accuracy and importance of projections from a dynamic
simulation model of smoking prevalence in the United States. American Journal of Public
Health 102(11):2045–2048.
Weinstein, M. C., B. O’Brien, J. Hornberger, J. Jackson, M. Johannesson, C. McCabe, B. R.
Luce, and ISPOR Task Force on Good Research Practices—Modeling Studies. 2003.
Principles of good practice for decision analytic modeling in health-care evaluation:
Report of the ISPOR Task Force on Good Research Practices—Modeling studies. Value
in Health 6(1):9–17.
WHO (World Health Organization). 2008. WHO guide for standardization of economic
evaluations of immunization programmes. Geneva, Switzerland: Initiative for Vaccine
Research of the Department of Immunization, Vaccines, and Biologicals.
Wynder, E. L., and E. A. Graham. 1950. Tobacco smoking as a possible etiologic factor in
bronchiogenic carcinoma: A study of 684 proved cases. JAMA 143(4):329–336.
Zauber, A. G., I. Lansdorp-Vogelaar, A. B. Knudsen, J. Wilschut, M. van Ballegooijen, and
K. M. Kuntz. 2008. Evaluating test strategies for colorectal cancer screening: A deci-
sion analysis for the U.S. Preventive Services Task Force. Annals of Internal Medicine
149(9):659–669.


Patterns of Tobacco Use by

Adolescents and Young Adults
S
everal national surveys provide data for estimation of smoking behav-
ior among adolescents and young adults in the United States. These
data sources include Monitoring the Future Study (MTF), National
Health Interview Survey (NHIS), National Longitudinal Study of Ado-
lescent Health (Add Health), National Survey on Drug Use and Health
(NSDUH), National Youth Tobacco Survey (NYTS), and Youth Risk Be-
havior Surveillance System (YRBSS). This chapter summarizes the rates of
adolescent and young adult tobacco use as reported in these sources as well
as in the 2012 Surgeon General’s report Preventing Tobacco Use Among
Youth and Young Adults (HHS, 2012). When discussing the rates of to-
bacco use, “tobacco use” is defined to include use of cigarettes, smokeless
tobacco, cigars, and electronic nicotine delivery systems (ENDS), or “e-
cigarettes.” Data on rates of tobacco use among different groups, including
at-risk populations, are also presented. A comprehensive synthesis of these
data is described in the 2012 Surgeon General’s report. This chapter then
continues with evidence about the effect of age of initiation on patterns of
nicotine dependence and cessation.
PREVALENCE OF CIGARETTE SMOKING

Cigarette smoking is the most common way that adolescents and young
adults use tobacco, and data on prevalence of cigarette smoking are the
most comprehensive and systematic and have the longest history of collec-
tion among all data on tobacco use. Additionally, combusted tobacco such
as traditional cigarettes is responsible for the vast majority of tobacco-
31

related death and disease in the United States (HHS, 2014). Thus, much of
this review will focus on adolescent and young adult cigarette use; however,
data on other tobacco products will be provided where available.
Table 2-1 provides current cigarette smoking rates from the 2013 YRBSS
by gender, race/ethnicity, and grade. Data from the 2013 YRBSS show that
slightly fewer than one in five high school seniors (19 percent) were current
cigarette smokers, defined as having smoked within the 30 days immediately
before the survey (Kann et al., 2014). Monitoring the Future reports similar
data, with 16 percent being current smokers (Johnston et al., 2014b). These
prevalence data indicate that there has been a continued decline in smoking
among high school students in recent years, although the decline has been
occurring at a slower rate than in the early 2000s (HHS, 2012). Both YRBSS
and MTF show a substantial increase in cigarette use with increasing grade
level (although YRBSS shows a decline from the 11th to the 12th, which is
likely due to the fact that a number of students drop out between the 11th
and 12th grades). For comparable grades (10th and 12th), the estimates
for YRBSS are slightly and consistently higher than for MTF, probably due
to differences in how questions are asked. The different estimates from the
surveys could result from a variety of factors, and each of the surveys has
relative strengths and weaknesses (SAMHSA, 2012b). YRBSS and MTF are
school-based samples, so these surveys exclude school dropouts and young
adults who have graduated from high school. NSDUH, on the other hand,
TABLE 2-1 Percentage of High School Students Who Currently Smoke

Cigarettes by Gender, Race/Ethnicity, and Grade—YRBSS, 2013
Female Male Total
Race/Ethnicity:
White Non-Hispanic 18.1 19.1 18.6

Black Non-Hispanic 6.2 10.5 8.2
Hispanic 13.1 15.0 14.0
Grade:
9 10.0 10.3 10.2
10 12.6 13.6 13.2
11 18.9 23.4 21.1
12 18.7 19.6 19.2
Total 15.0 16.4 15.7
NOTE: Current smoking defined as having smoked on at least 1 day during the 30 days
before the survey.
SOURCE: Kann et al., 2014.

PATTERNS OF TOBACCO USE BY ADOLESCENTS AND YOUNG ADULTS 33
includes dropouts and has all ages 12 and older. Current smoking prevalence
is highest among white adolescents, followed by Hispanic and black ado-
lescents. Trends are similar among young adults (HHS, 2012). While black
and Hispanic males smoke more than females, prevalence rates of current
smoking are the same for males and females among whites.
Table 2-2 shows NSDUH estimates of monthly cigarette use by age.
Note that prevalence of use continues to increase post-high school, with a
TABLE 2-2 Percentages Used Cigarettes in

Past Month by Age, NSDUH, 2012
Age Percentage
12 0.5
13 1.8
14 3.3
15 6.0
16 11.1
17 16.1
18 25.1
19 27.7
20 32.1
21 33.4
22 35.1
23 33.0
24 35.4
25 33.6
26–29 33.4
30–34 31.9
35–39 26.7
40–44 24.3
45–49 26.0
50–54 24.5
55–59 21.5
60–64 16.9
65+ 10.0
Total 22.1
SOURCE: Table 2.12B from SAMHSA, 2013a.

Percent smoking in past 30 days
35.0
30.0
25.0
20.0
15.0
10.0
5.0
*
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
18 26.7 24.4 25.0 23.2 21.6 21.6 20.4 20.1 19.2 18.7 17.1
19-20 29.8 27.0 27.9 27.5 24.6 22.6 21.8 21.2 19.6 18.5 16.8
21-22 32.6 30.5 31.3 29.2 27.3 27.8 24.5 25.2 22.8 23.3 18.9
23-24 31.9 31.0 31.5 29.3 28.1 26.7 26.5 24.1 23.0 22.0 20.4
25-26 27.3 27.0 29.6 30.7 29.1 27.5 24.5 22.6 24.3 23.4 20.7
FIGURE 2-1 Trends in 30-day cigarette smoking prevalence by age group, 18–26,
MTF, 2002 through 2012.
SOURCE: Johnston et al., 2013.
sharp increase at age 18, then leveling off around ages 21 to 22. The sharp
increase from 16.1 percent at age 17 to 25.1 percent at age 18 is presum-
ably due at least in part to the fact that the minimum legal age for purchase
of tobacco products is 18.
Although MTF is a school-based sample, the study includes a longitu-
dinal component, allowing for estimates for smoking rates for young adults
who are high school graduates. Figure 2-1 shows trends from 2002 to 2012
in prevalence of 30-day cigarette smoking by age groups, from 18 to 26.
The trends show continuing declines in cigarette use among young adult
high school graduates, with some convergence among age groups. N SDUH,
which includes dropouts, also shows declines through 2012 among those
ages 18 to 25 (SAMHSA, 2013a).
Finding 2-1: Almost one in five high school seniors is a current (in
the past 30 days) cigarette smoker, compared with one in three young
adults.
Socioeconomic Status
For some years cigarette smoking has been more concentrated among
those with lower socioeconomic status but in recent years that concen-

tration has become more pronounced. In adolescents and young adults,

socioeconomic status is typically assessed using measures of parental (often
maternal) educational attainment or of the adolescent or young adult’s edu-
cational goals. Table 2-3 provides data from 10th graders in two MTF stud-
ies to illustrate the point. In 1997, when smoking rates among adolescents
reached their recent peak, smoking rates were slightly higher among 10th
graders whose parents had less education, but by 2013 the discrepancies
had substantially widened due to a greater decline among students whose
parents had more education compared with those whose parents had less
education, with 13–14 percent of those whose parents had less education
being smokers compared to 5–6 percent of those whose parents had more
education.
The relationship between socioeconomic status and smoking also dif-
fers by racial/ethnic category (HHS, 2012). Bachman and colleagues (2011),
for example, used data from MTF and found that for white students in 8th
through 12th grades there was a clear negative linear relationship between
parental education and smoking rates. For black students, a similar nega-
tive relationship existed between smoking and parental education, but the
relationship was much smaller. For Hispanic students, the relationship was
nonlinear, with smoking rates relatively high among Hispanic students with
parents of higher education levels compared to white and black students,
and relatively low among Hispanic students with the least educated parents
compared to whites and blacks. It is possible, however, that these findings
TABLE 2-3 Percentage of 10th Graders Who Smoked Cigarettes in the

Past 30 Days, by Parental Education, MTF, 1997 and 2013
Parental Education 1997 2013
1.0–2.0 (low) 28.2 12.8
2.5–3.0 33.2 13.6
3.5–4.0 30.9 10.2
4.5–5.0 28.5 6.0
5.5–6.0 (high) 24.6 4.9
Total 29.8 9.1
NOTE: Parental education is an average of mother’s education and father’s education. Re-
sponse categories are (1) completed grade school or less, (2) some high school, (3) completed
high school, (4) some college, (5) completed college, and (6) graduate or professional school
after college.
SOURCE: Johnston et al., 2014a.

TABLE 2-4 Percentage of 10th Graders Who Smoked Cigarettes in the

Past 30 Days, by 4-Year College Plans, MTF, 1997 and 2013
4-Year College Plans? 1997 2013
No 47.2 23.3
Yes 26.8 7.4
Total 29.8 9.1
NOTE: Respondents indicated how likely they were to graduate from a 4-year college pro-
gram; those who said “definitely” or “probably will” were coded “yes.”
SOURCE: Johnston et al., 2014a.
are due to the high number of Hispanic parents with low socioeconomic
status (as defined by parental education).
As a further illustration, Table 2-4 provides smoking rates for two
groups: those who expect to complete a 4-year college program versus those
who do not. In 1997 those in the latter group were almost twice as likely
to be current smokers as those in the former group (47.2 percent versus
26.8 percent), but by 2013 the ratio was more than 3 to 1 (23.3 percent
versus 7.4 percent).
Among young adults, smoking rates similarly differ by education. Those
who do not enroll in college are more likely to have started smoking at a
younger age and to be current smokers, and they are less likely to attempt
to quit smoking than their peers who enroll in college (Green et al., 2007).
Also, among young adults not attending college, full-time employment is
associated with higher rates of tobacco use (Welte et al., 2011).
GEOGRAPHIC VARIATION
Tables 2-5 and 2-6 provide the percentage of 12- to 17-year-olds who
smoked cigarettes in the past month by region of the country and by state
of residence, in combined years 2002–2003 and 2010–2011. (Combining
two years of data is necessary because of the small numbers of cases avail-
able in many states.) Figure 2-2 provides a visual display of the considerable
variation by state for 2010–2011. Utah had the lowest rate (5.1 percent),
and West Virginia had the highest (11.8 percent) (SAMHSA, 2012c). The
10 highest states (red color in Figure 2-2, greater than 9.7 percent) were, in
descending order: West Virginia, Montana, Kentucky, Missouri, Wyoming,
Iowa, New Hampshire, Vermont, Oklahoma, and South Dakota. The nine
lowest states, plus the District of Columbia (white color in Figure 2-2,
7.11 percent or less), were, in descending order: Texas, Virginia, Nevada,

TABLE 2-5 Percentage Using Cigarettes in the Past Month, Ages 12–17,
by Region, NSDUH, 2002–2003 and 2010–2011
Region 2002–2003 2010–2011 Percentage Point Change
Total U.S. 12.57 8.07 −4.50
Northeast 12.72 8.14 −4.58

Midwest 14.63 8.84 −5.79
South 13.21 8.19 −5.02
West 9.47 7.13 −2.34
SOURCE: SAMHSA, 2012c.
Florida, New York, Hawaii, California, District of Columbia, Maryland,

and Utah. Of the four regions of the country, the Midwest had the highest
rates of smoking among 12- to 17-year-olds (8.8 percent), and the West had
the lowest (7.1 percent). Between 2002–2003 and 2010–2011, all regions
and all states showed declines.
FIGURE 2-2 Cigarette use in the past month among adolescents ages 12 to 17, by
state. Average annual percentages, NSDUH, 2010 and 2011.
SOURCE: SAMHSA, 2012a.

38
TABLE 2-6 Percentage Using Cigarettes in the Past Month, Ages 12–17, by State, NSDUH, 2002–2003 and
2010–2011
State 2002–2003 2010–2011 Change State 2002–2003 2010–2011 Change
Alabama 13.69 8.84 −4.85 Montana 16.1 11.73 −4.37

Alaska 13.25 7.84 −5.41 Nebraska 16.36 8.87 −7.49
Arizona 12.84 8.76 −4.08 Nevada 12.73 6.95 −5.78
Arkansas 16.05 9.35 −6.7 New Hampshire 14.03 10.7 −3.33
California 7.48 6.25 −1.23 New Jersey 11.83 8.35 −3.48
Colorado 13.74 8.64 −5.1 New Mexico 12.34 9.08 −3.26
Connecticut 13.45 8.02 −5.43 New York 11.81 6.83 −4.98
Delaware 14.07 8.16 −5.91 North Carolina 14.78 8.86 −5.92
District of 7.1 5.99 −1.11 North Dakota 17.53 9.64 −7.89
Columbia
Florida 12.26 6.91 −5.35 Ohio 14.52 9.62 −4.9

Georgia 12.83 9.41 −3.42 Oklahoma 14.96 10.21 −4.75

Hawaii 8.78 6.68 −2.1 Oregon 11.29 7.45 −3.84
Idaho 12.48 8.57 −3.91 Pennsylvania 14.73 9.58 −5.15
Illinois 13 7.28 −5.72 Rhode Island 13.72 8.39 −5.33
Indiana 14.39 8.06 −6.33 South Carolina 12.21 9.7 −2.51
Iowa 14.27 10.8 −3.47 South Dakota 19.79 10.18 −9.61
Kansas 13.95 9.62 −4.33 Tennessee 14.33 9.18 −5.15
Kentucky 17.62 11.66 −5.96 Texas 11.65 7.11 −4.54
Louisiana 15.01 9.14 −5.87 Utah 6.57 5.13 −1.44
Maine 12.16 8.87 −3.29 Vermont 14.84 10.48 −4.36
Maryland 11.08 5.9 −5.18 Virginia 14.17 6.99 −7.18
Massachusetts 11.69 8.16 −3.53 Washington 10.84 8.56 −2.28
Michigan 13.59 8.4 −5.19 West Virginia 17.34 11.8 −5.54
Minnesota 15.67 8.7 –6.97 Wisconsin 15.32 8.1 –7.22
Mississippi 12.83 9.37 –3.46 Wyoming 12.78 11.26 –1.52
Missouri 17.88 11.6 –6.28
SOURCE: SAMHSA, 2012c.

39
Metropolitan Status
Table 2-7 provides prevalence of 30-day smoking by age group and by
metropolitan status for adolescents and young adults. In each age group,
the nonmetropolitan segment has the highest rate of smoking and the large
metropolitan has the lowest, with the small metropolitan segment being
intermediate.
Use of other tobacco products similarly varies by metropolitan status,
with greater use in more rural communities and less use in more urban ar-
eas. Adolescents and young adults residing in rural communities are more
likely to use tobacco and, particularly, smokeless tobacco or chew because
of the cultural norms set within their communities (Melnick et al., 2001;
Peek et al., in preparation). Rural life is often associated with the rodeo
or being a cowboy (Peek et al., in preparation), with males playing sports
such as baseball (whose athletes use smokeless tobacco at high rates), and
with men being more macho and tough (Melnick et al., 2001; Peek et al.,
in preparation). These attitudes often translate into a situation in which
it is socially normative to use tobacco in order to mirror these images.
Furthermore, in these often insular, small communities where everyone is
connected and knowledgeable of each other’s action, younger adolescents
are able to obtain chew and other tobacco products from members of their
community more easily (Peek et al., in preparation).
TABLE 2-7 Percentage Smoking in the Past 30 Days by Age and

Metropolitan Status, NSDUH, 2012
12–13 14–15 16–17 18–20 21–25
Large Metropolitan 0.9 3.7 11.6 26.3 32.0
Small Metropolitan 1.1 5.4 15.2 28.9 34.5
Non-Metropolitan 1.7 8.1 18.1 34.3 42.0
Total 1.1 4.9 13.7 28.3 34.2

SOURCE: Committee analysis from SAMHSA, 2013a.

OTHER INDIVIDUAL RISK FACTORS FOR TOBACCO USE
Mental Illness
Tobacco use is also more common among those with mental illness, in
part because these individuals use nicotine as a means of “self-medication,”
mood regulation, and stress mitigation (Ziedonis et al., 2008). On the other
hand, Goodman and Capitman (2000) assessed 8,704 adolescents and
found that depressive symptoms did not predict smoking. Instead, smok-
ing predicted subsequent depressive symptoms. Similarly, greater levels of
smoking during adolescence and early adulthood have been associated with
a higher risk for agoraphobia, general anxiety disorder, and panic disorder
(Johnson et al., 2000), suggesting that while there is a strong relationship
between mental illness and smoking, the nature of this relationship is still
unclear.
Sexual Orientation
Lesbian, gay, bisexual, transgender, questioning, or queer (LGBTQ)
adolescents and young adults appear to use smoking as a means of coping
with the stigma associated with their sexual identity (Rosario et al., 1997).
Higher smoking rates among LGBTQ youth persist even after accounting
for psychosocial factors such as depression, self-esteem, and familial smok-
ing habits (Austin et al., 2004). However, it is also the case that supportive
social environments (operationalized by assessing the proportion of same
sex couples living in the counties studied, the proportion of schools with
gay–straight alliances, the proportion of schools with policies protecting
gay students, and the proportion of schools with antidiscrimination poli-
cies) have been associated with lower rate of tobacco use (Hatzenbuehler
et al., 2011).
Finding 2-2: Significant disparities in tobacco use remain among ado-

lescents and young adults nationwide. The lowest rates are found in
the western United States, in large metropolitan areas, among African
Americans, adolescents who plan to go to college, and adolescents
whose parents’ education includes graduate school or a professional
degree.
INITIATION
The Surgeon General’s 2012 report stated that one of the most im-
portant and widely cited findings from the 1964 Surgeon General’s report
on smoking and health was that cigarette smoking almost always begins

before adulthood (HHS, 2012). The 2012 report corroborated that the find-
ing still held. Table 2-8 in this report updates that information and shows
that the finding is still true. Among adults ages 30 to 34 who ever smoked
daily, 89.8 percent had first tried a cigarette before age 19, and 99.2 per-
cent before age 26. The 2012 Surgeon General’s report emphasized that a
relatively high proportion of adult smokers initiate at a relatively early age.
For example, more than one-third (36.7 percent) of adults who had ever
tried a cigarette reported trying their first cigarette by age 14. The figure in
Table 2-8 (36.2 percent) is virtually identical to this number (36.2 percent).
A Note on the Definition of Initiation

The preceding data on initiation has used the typical definition of ini-
tiation as being the point in time at which one first tries a cigarette, which
is the way that initiation is measured by most surveys of adolescents and
young adults. However, the NHIS survey used a different definition of age
of initiation, which is often used in surveys of adults, and does not treat a
person as having initiated smoking until that person has smoked at least
100 cigarettes. In the models reported in Chapter 8, NHIS data are used
as the basis for estimating the effects that changing the minimum legal age
has on initiation. This raises the question of how different these definitions
are in practice. NSDUH asks about both the age of first use and whether
the respondent has ever used 100-plus cigarettes in his or her life, so these
data can be used to compare the distributions of ages of initiation for all
NSDUH respondents versus just those who progressed to smoking 100-plus
cigarettes. To summarize the results of this comparison, while the distribu-
tions are not identical, they are quite close, suggesting that this adjustment
is not a major concern.
The age of first use was cross-tabulated with having smoked at least
100 cigarettes across the lifetime for 26- to 34-year-old respondents and
separately for all respondents 26 and older in the NSDUH surveys of 2002
through 2012, combined. The results for all respondents 26 and older have
the advantage of being based on a larger number of respondents, but the
restriction to 26- to 34-year-olds limits the analysis to younger respondents,
whose cigarette initiation patterns may differ from those of older respon-
dents from earlier generations.
Figure 2-3 shows the distributions of ages of initiation for 26- to
34-year-olds. The comparison of interest is between the thick black line
(for all respondents) and the thick red line (just for those who progressed
to 100-plus cigarettes). The black line in some sense corresponds to MTF
and other data that ask about age of first use for all who have ever smoked
any cigarettes; the red line corresponds to the NHIS data, the input for the
models.

TABLE 2-8 Cumulative Percentage of Recalled Ages at Which Respondents First Used a Cigarette and Began
Smoking Daily, by Smoking Status Among 30- to 34-Year-Olds, NSDUH, 2012
Persons Who Had Ever Tried
All Persons a Cigarette Persons Who Ever Smoked Daily
First Tried a Began Smoking First Tried a
Age Cigarette Daily First Tried a Cigarette Cigarette Began Smoking Daily
≤10 3.8 0.4 5.4 7.0 1.1
11 5.9 1.1 8.5 11.4 2.7
12 11.9 2.3 17.0 21.4 5.8
13 17.4 3.9 25.0 30.6 9.8
14 25.2 6.1 36.2 45.6 15.4
15 34.6 9.8 49.7 62.3 24.8
16 43.5 15.9 62.4 75.0 40.1
17 48.9 21.3 70.2 81.9 53.9
18 56.7 27.3 81.5 89.8 69.0
19 60.0 30.5 86.1 94.1 77.0
20 63.0 33.6 90.5 95.9 84.9
21 64.9 35.1 93.3 97.1 88.5
22 66.1 36.0 94.9 98.0 90.7

23 66.8 36.4 95.9 98.5 92.0

24 67.1 37.1 96.4 98.5 93.6
continued
43
TABLE 2-8 Continued
44
Persons Who Had Ever Tried

All Persons a Cigarette Persons Who Ever Smoked Daily
First Tried a Began Smoking First Tried a
Age Cigarette Daily First Tried a Cigarette Cigarette Began Smoking Daily
25 68.0 38.4 97.7 99.2 96.8
26 68.4 38.6 98.2 99.4 97.5
27 68.8 38.8 98.9 99.5 98.0
28 69.0 38.9 99.1 99.7 98.1
29 69.2 39.2 99.3 99.9 98.9
30 69.5 39.5 99.8 100.0 99.8
31 69.5 39.5 99.8 100.0 99.8
32 69.6 39.5 100.0 100.0 99.8
33 69.6 39.6 100.0 100.0 99.9
34 69.6 39.6 100.0 100.0 100.0
Never smoked 100 100 NA NA NA
SOURCE: Committee analysis of data from HHS et al., 2014.

FIGURE 2-3 Age distribution of cigarette initiation reported by 26- to 34-year-olds,

broken down by those who did versus those who did not progress to using 100-plus
cigarettes in their lifetimes (62 percent progressed; 38 percent did not), NSDUH,
2002 through 2012.
Those who ended up smoking more heavily have a distribution of

ages of initiation that skews slightly younger, with more initiating at ages
12–16 and fewer initiating after 17. The largest difference is at age 13; 9.7
percent of all smokers initiated at age 13, but 11.6 percent of those who
progressed did so.
Figure 2-4, which shows data for all respondents age 26 and above,
shows even smaller differences between those who did and those who did
not progress to smoking 100-plus cigarettes.

FIGURE 2-4 Age distribution of cigarette initiation reported by those 26 years old
and older, broken down by those who did versus those who did not progress to us-
ing 100-plus cigarettes in their lifetimes, NSDUH, 2002 through 2012.
SOURCE: Committee analysis from HHS et al., 2014.
Finding 2-3: Among adults who become daily smokers, nearly all re-
port first use of cigarettes before 19 years of age (90 percent), with 99
percent reporting first use before 26 years of age.
SMOKING INTENSITY
The most commonly used metric of smoking intensity is the number of
cigarettes smoked per smoking day. Table 2-9 provides the average number
of cigarettes smoked per smoking day for those who smoked cigarettes in
the past 30 days, by age, based on 2012 NSDUH data. The right-most
two columns compare the data for those who smoked less than about half
a pack per day with those who smoked half a pack or more per smoking
day. There are substantial increases between ages 12 through 15 and age
16, and between ages 17 and 18, but then relatively little increase in the
average number of cigarettes smoked per smoking day in the age range
from 18 to 20. Intensity increases substantially after that. An alternative
metric for gauging the overall exposure to cigarettes is the number of days
that an individual has smoked in the past month. This metric captures the
frequency or regularity of use.

TABLE 2-9 Average Number of Cigarettes per Smoking Day in Past 30 Days, NSDUH, 2012
Number of (½ Pack) (1 Pack) (1½ Packs) (2+ Packs) Row Less than ½ Pack or
cigarettes <1 1 2–5 6–15 16–25 26–35 35+ Total ½ Pack More
Age:
12–15 33.3 26.9 33.6 4.5 1.2 0.1 0.5 100 93.7 6.3
16 24.8 20.3 36.4 16.9 1.1 0.0 0.5 100 81.5 18.5
17 21.8 23.1 35.8 14.7 4.7 0.0 0.0 100 80.7 19.4
18 15.5 20.5 34.9 20.6 5.6 1.2 1.6 100 70.9 29.0
19 17.7 18.1 34.9 19.8 8.3 0.8 0.4 100 70.7 29.3
20 12.9 19.2 35.7 22.9 8.3 0.7 0.4 100 67.8 32.3
21 13.2 16.1 32.7 25.8 9.0 2.4 0.8 100 62.0 38.0
22–23 14.1 14.1 33.8 25.0 10.9 1.8 0.3 100 62.0 38.0
24–25 10.9 11.6 32.1 27.3 15.2 2.2 0.6 100 54.6 45.3
26–29 11.9 12.3 30.4 27.6 14.5 2.5 0.7 100 54.6 45.3
30–34 7.6 6.7 26.7 31.6 21.5 3.8 2.1 100 41.0 59.0
35–49 6.5 6.2 20.4 29.7 28.4 6.9 2.0 100 33.1 67.0
50–64 3.6 6.4 19.2 29.5 27.2 9.3 4.8 100 29.2 70.8
65+ 2.8 5.7 23.9 31.2 22.8 8.0 5.4 100 32.4 67.5
NOTE: Entries are percentages. The survey question was, “On the days you smoked cigarettes during the past 30 days, how many cigarettes did

you smoke per day, on average?”

47
Emerging Patterns
In general, the rates of cigarette smoking have been declining, although
there have been recent signs of a deceleration in that decline (SAMHSA,
2013b). In addition to this general decline, there has also been clear evi-
dence of an increasing trend toward lighter use. One indication of this is
that among those who smoked cigarettes in the past 30 days, the proportion
of those who smoke every day has been decreasing, and, conversely, the
proportion of nondaily smokers has been increasing. Table 2-10 provides
the percentages, from 1991 to 2013, of past-30-day smokers who smoked
less often than daily for four age groups, based on the Monitoring the
Future study. Among young adults ages 19 to 28, the percentage of cur-
rent smokers who were nondaily smokers rose steadily from 23 percent in
1991 to 40 percent in 2013. The rise in this population of lighter smokers
has important implications for the understanding of nicotine addiction and
dependence (Shiffman, 2009).
OTHER TOBACCO PRODUCTS

While there are several sources of reliable information on cigarettes,
there is less extensive information on other tobacco products, particularly
for trends in their use. The situation regarding tobacco products other than
traditional cigarettes is currently highly volatile, with new products being
introduced and existing products being modified. NYTS asks questions
about a range of non-cigarette tobacco products, including cigars, smoke-
less tobacco, tobacco smoked with a hookah, pipes, electronic cigarettes,
snus, kreteks, bidis, and dissolvable tobacco. Table 2-11 provides results
from the 2011 and 2012 surveys.
The various products are ordered in Table 2-11 according to their total
prevalence in 2012, from highest to lowest. Overall, just about one in four
high school students reported using at least one tobacco product during the
previous 30 days (23.3 percent in 2012). Cigarettes were the most com-
monly used, at 14.0 percent, but cigars were not far behind at 12.6 percent.
“Cigars” includes cigars, cigarillos, and little cigars. This class of products
has seen major changes in the types and number of products available and
in the marketing of the products. Little cigars may be very similar to ciga-
rettes in size and shape, and may be flavored with fruit or candy. They are
typically taxed at lower rates than cigarettes and may therefore be more
affordable. While the rates of current cigarette use have seen a significant
decrease, the rates of smokeless tobacco use, including the use of chew,
dip, or moist snuff, have remained stable or even increased. For example,
among Americans age 12 or older, 3.1 percent were current (past-month)
users of smokeless tobacco in 1998, and that figure was at 3.5 percent in
2012 (SAMHSA, 2013a).

TABLE 2-10 Percentage of Past 30-Day Smokers Who Smoked Less

Than Daily, MTF, 2013
Grade Young adult
Year 8th 10th 12th (19–28)
1991 49.7 39.4 34.6 23.0
1992 54.8 42.8 38.1 26.1
1993 50.3 42.5 36.5 25.7
1994 52.7 42.5 37.8 26.1
1995 51.3 41.6 35.5 27.4
1996 50.5 39.8 34.7 27.6
1997 53.6 39.6 32.6 31.1
1998 53.9 42.8 36.2 29.1
1999 53.7 38.1 33.2 29.0
2000 49.3 41.4 34.4 27.6
2001 54.9 42.7 35.6 29.8
2002 52.3 42.9 36.7 27.4
2003 55.9 46.7 35.2 28.5
2004 52.2 48.1 37.6 28.8
2005 57.0 49.7 41.4 31.5
2006 54.0 47.6 43.5 31.1
2007 57.7 48.6 43.1 34.0
2008 54.4 52.0 44.1 32.1
2009 58.5 51.9 44.3 35.6
2010 59.2 51.5 44.3 33.9
2011 60.7 53.4 44.9 35.5
2012 61.2 53.7 46.7 35.0
2013 60 51.6 47.9 39.5
SOURCE: Johnston et al., 2014c.
According to the 2011 YRBSS, 12.8 percent of adolescent males and

2.2 percent of adolescent females in the United States reported current use
of smokeless tobacco (Eaton et al., 2012). Overall, current use of smokeless
tobacco was higher among whites (9.3 percent) than among Hispanics (5.9
percent) or blacks (3.1 percent). In the NYTS survey, smokeless tobacco,
which includes chewing tobacco, snuff, and dip, was used by 6.4 percent

TABLE 2-11 Percentage of High School Students Using Tobacco

Products in Past 30 Days, by Gender, NYTS, 2011–2012
Male/
Total Female Male Female
Ratio in
2011 2012 2011 2012 2011 2012 2012
Any tobacco 24.3 23.3 19.0 18.1 29.4 28.3 1.56
Cigarettes 15.8 14.0 13.8 11.7 17.7 16.3 1.39
Cigars 11.6 12.6 7.4 8.4 15.7 16.7 1.99
Smokeless 7.3 6.4 1.6 1.5 12.9 11.2 7.47
tobacco
Hookahs 4.1 5.4 3.5 4.5 4.8 6.2 1.38

Pipes 4.0 4.5 2.8 3.2 5.1 5.8 1.81
Electronic 1.5 2.8 0.7 1.9 2.3 3.7 1.95
cigarettes
Snus 2.9 2.5 0.8 0.9 5.1 3.9 4.33

Kreteks 1.7 1.0 0.8 0.5 2.4 1.5 3.00
Bidis 2.0 0.9 1.0 0.5 2.9 1.3 2.60
Dissolvable 0.4 0.8 0.1 0.6 0.6 1.0 1.67
tobacco
SOURCE: Arrazola et al., 2013.
of those surveyed. Another form of smokeless tobacco is snus, which is a

relatively new product in the United States and was used by 2.5 percent.
Approximately 5 percent of the respondents reported using the hoo-
kah (waterpipe), and an equal number reported smoking pipes. Electronic
cigarettes are an increasingly visible part of the tobacco product scene, but
as of 2012 less than 3 percent of high school students reported using them.
Nevertheless, ENDS use is increasing rapidly among adolescents. In 2014,
for the first time in a U.S. national study, Monitoring the Future reported
that more high school students used e-cigarettes than traditional cigarettes
or any other tobacco product. The difference in the use of e-cigarettes ver-
sus traditional cigarettes was greater among younger students: 9 percent
of 8th grade students reported using an e-cigarette in the past 30 days, as
compared with 4 percent for traditional cigarettes; 16 percent of 10th grade
students reported using an e-cigarette, as compared with 7 percent for tra-
ditional cigarettes; and 17 percent of 12th grade students reported using an
e-cigarette, as compared with 15 percent for traditional cigarettes (Wadley
and Bronson, 2014). Kreteks, bidis, and dissolvable tobacco (another recent

addition to the group of smokeless tobacco products) all were used by 1

percent or less of respondents.
Males were more likely than females to use at least one tobacco prod-
uct (28.3 percent versus 18.1 percent in 2012), and, for any given product,
males were more likely than females to report using that product. The male/
female ratios were particularly high for smokeless tobacco and the newer
smokeless snus. The Surgeon General’s 2012 report noted that as of 2010,
about 1 in 10 high school senior males was a current smokeless tobacco
user and about 1 in 5 high school senior males was a current cigar smoker
(HHS, 2012). The 2013 YRBSS found that about one in six high school
senior males was a current smokeless tobacco user, and about one in four
high school senior males was a current cigar smoker (Kann et al., 2014).
The use of these two classes of tobacco products clearly has not declined
in recent years (HHS, 2012; Kann et al., 2014).
Table 2-12 provides some limited information on the use of tobacco
products among adolescents and adults, as reported by NSDUH. As with
NYTS, cigars are found to make up a relatively high proportion of tobacco
product use, particularly among young adults.
Some limited trend data on smoking tobacco with a hookah are avail-
able from the Monitoring the Future study. Table 2-13 shows that smoking
tobacco with a hookah is particularly popular among college students, with
26 percent reporting in 2013 having done so at least once in the previous
12 months. Even among 12th graders, the behavior is relatively common,
with 21 percent reporting having done so in 2013. However, much of this
behavior is light or experimental, with only 9 percent of 12th graders re-
porting having smoked with a hookah more than five times in the previous
12 months (Wadley and Barnes, 2013).
TABLE 2-12 Percentage Who Used Tobacco Products in Past 30 Days by

Age, NSDUH, 2012
Smokeless Any Tobacco
Age Tobacco Cigars Pipe Cigarettes Products
12–13 0.4 0.4 0.3 1.2 1.6
14–15 1.7 1.7 0.5 4.6 6.3
16–17 4.0 5.6 1.2 13.6 17.6
18–20 5.4 11.9 2.1 28.2 34.9
21–25 5.6 10.0 1.6 34.1 40.0
26–34 4.7 7.3 1.1 32.6 37.5
35+ 3.0 3.9 0.8 20.1 24.7
SOURCE: SAMHSA, 2013a.

TABLE 2-13 Prevalence of Hookah Use in Past 12 Months, MTF, 2010

Through 2013
2010 2011 2012 2013
12th grade 17.1 18.5 18.3 21.4
College students 27.9 25.7 26.1
Young adults (19–28) 20.1 19.1 20.4
SOURCES: Johnston et al., 2014b; Wadley and Barnes, 2013.
The Surgeon General’s report of 2012 data concluded that concurrent

use of multiple tobacco products (poly-tobacco use; usually using cigarettes
and another tobacco product) was prevalent among adolescents. Among
those who use tobacco, nearly one-third of high school females and more
than one-half of high school males report having used more than one to-
bacco product in the past 30 days. By 2012, more than one-third of high
school female users were poly-tobacco users. In the 2012 NYTS, of the
15.4 percent of high school females who reported tobacco use, 38 percent
of them—or 5.9 percent of all high school females—reported using more
than one tobacco product; the corresponding figures for high school males
were 55 percent of 25.3 percent, or 13.8 percent of all high school males
(Arrazola et al., 2014).
Finding 2-4: Concurrent use of multiple tobacco products is prevalent

among adolescents.
Finding 2-5: It is difficult to assess trends in non-cigarette products

because the products themselves are changing. While cigarette use has
been declining, the use of some other products has not.
PATTERNS OF USE AND PROGRESSION

OF NICOTINE DEPENDENCE
Tobacco use in adolescents and in young adults is not a unitary phe-
nomenon; instead it is best characterized by a series of events that involve
multiple behaviors and feelings (Mayhew et al., 2000) and transitions in
a sequence from initial trials with tobacco to more occasional use, to the
development of dependence and regular use, through to cessation. Tobacco
use in adolescence is highly variable in terms of both frequency of use and
intensity of use (Mermelstein et al., 2002).
Age-based prevalence data for tobacco use provide cross-sectional
views of tobacco use from which one can assume patterns of progression.
However, while such cross-sectional prevalence data can provide infor-

mation about the total number of smokers at a given age and even offer
insights into how many individuals have started or stopped smoking in a
given year, they are less directly informative about individual differences in
progression of tobacco use behavior. With the use of newer data analytic
techniques (e.g., latent variable growth mixture modeling), researchers
have identified various trajectories of smoking behavior among adolescents
and young adults (e.g., Bernat et al., 2008; Brook et al., 2008; Chassin
et al., 2008; Colder et al., 2001, 2008; Costello et al., 2008; Jackson et
al., 2008; Lessov-Schlaggar et al., 2008; Riggs et al., 2007; Stanton et al.,
2004; Tucker et al., 2006). These approaches may help to better describe
the heterogeneity of longitudinal patterns of use and to identify factors
that discriminate among the different trajectories. Among the trajectories
that have been identified are groups of adolescents who experiment but
have non-escalating trajectories and other groups that escalate rapidly.
Unfortunately, these studies have not to date provided the fine-grained
age detail during the young adult period necessary to reliably identify the
differences between those individuals who initiate and escalate starting in
young adulthood and those who initiate during the earlier adolescent years.
In addition, most of these studies have provided data on the broad popula-
tion of adolescents, most of whom fall into the nonsmoking trajectories. As
such, they provide less in-depth information on the patterns of progression
of those adolescents who try tobacco use. Furthermore, all of these studies
have focused exclusively on cigarette use, and none have considered how
the use of other tobacco products (e.g., cigars, smokeless tobacco, hookah,
etc.) may affect these trajectories. In addition, to date there has been no
systematic data collected concerning how patterns of tobacco product use
may vary by product or by combinations of products, including product
switching. The changing landscape of available tobacco products may well
affect overall patterns of use.
The often irregular pattern of tobacco use behavior presents a challenge
for clearly identifying exactly when nicotine dependence develops in the
progression of tobacco use. The level of dependence symptoms that indi-
viduals experience is believed to be the most important factor contributing
to smoking persistence and failed cessation efforts. Nicotine dependence is
characterized by physiological adaptations (e.g., tolerance, withdrawal) and
other accommodating behaviors (e.g., time spent in activities necessary to
obtain and use nicotine and to recover from its effects and the forfeiting or
reduction of important social, occupational, or recreational activities) re-
sulting from chronic smoking. Nicotine dependence predicts smoking regu-
larity and quantity across adolescence into young adulthood (Dierker and
Mermelstein, 2010; O’Loughlin et al., 2003; Selya et al., 2013). Ongoing
longitudinal studies of adolescent smoking that have examined the develop-
ment of nicotine dependence symptoms suggest that nicotine dependence

follows different developmental trajectories in different individuals (Hu et

al., 2008) and that for some adolescents, nicotine dependence symptoms
emerge very soon after the onset of smoking and at low levels of nicotine
exposure, well before the establishment of daily smoking patterns (Dierker
and Mermelstein, 2010; DiFranza et al., 2002). The McGill Study on the
Natural History of Nicotine Dependence in Teens confirmed individual
differences in the emergence of dependence and identified adolescents meet-
ing the criteria for ICD-10 nicotine dependence even among sporadic and
monthly smokers (O’Loughlin et al., 2003). Demographic (gender, ethnic-
ity) differences may also affect the development of nicotine dependence at
low levels of smoking exposure. For example, women have been shown to
have higher rates of dependence than men who engage in the same amount
of smoking (Kandel and Chen, 2000). In addition, compared to other racial
groups, whites have been found to have lower rates of lifetime nicotine
dependence (Hu et al., 2006; Kandel and Chen, 2000) and higher quit
rates (Fagan et al., 2007). It may well be that some of the differences in
the patterns of development of nicotine dependence, especially with regard
to age sensitivity, may be explained by individual differences in patterns
of brain development, genetics, or initial sensitivity to nicotine (Swan and
Lessov-Schlaggar, 2007).
There is considerable evidence that age of initiation is associated with
levels of nicotine dependence. As presented in the 2012 Surgeon General’s
report (HHS, 2012), data from the NSDUH 2007–2010 surveys show
that a younger age of initiation is strongly associated with greater nicotine
dependence in both young adulthood (18 to 25 years old) and older adult-
hood (26 years and older). Consistent dose–response gradients were pres-
ent, indicating that the younger the age of initiation, the greater the degree
of nicotine dependence. Furthermore, these associations held true regardless
whether age of initiation was measured as the age of the first puff or the age
an individual first smoked daily and were also independent of the length
of the transition from the first cigarette to daily smoking (HHS, 2012).
Longitudinal studies following participants from adolescence to young
adulthood also showed a statistically significant gradient, with younger ages
of initiation associated with greater nicotine dependence (Buchmann et al.,
2013; Hu et al., 2006). The association between earlier age of initiation
and greater nicotine dependence in early life also persists into adulthood.
Cross-sectional data in 21- to 30-year-olds (Breslau and Peterson, 1996)
and in later adulthood (Lando et al., 1999; Park et al., 2004) also show
clear gradients indicating that the earlier the age of starting cigarette smok-
ing, the greater the nicotine dependence; in both studies the strong associa-
tion between a younger age of initiation and greater nicotine dependence
was clearly evident across ages of initiation ranging from adolescence to
25 years of age and older. These findings suggest that there is no apparent
threshold beyond which this association does not apply.

Finding 2-6: Symptoms of nicotine dependence can develop even at

low levels of exposure to smoking, well before the establishment of
daily smoking.
Age of Initiation and Smoking Intensity

Smoking intensity, defined as the number of cigarettes smoked per
day, is strongly related to nicotine dependence and to all health out-
comes. Strong associations between younger ages of smoking initiation
and heavier smoking are evident even in studies that have examined this
question among adolescents who started smoking before high school and
assessed smoking intensity in high school (Escobedo et al., 1993; Everett
et al., 1999; Reidpath et al., 2014). Strong and statistically significant
associations were also observed in longitudinal studies that followed in-
dividuals from adolescence to young adulthood (Buchmann et al., 2013;
Hu et al., 2006).
U.S. national cross-sectional data indicate that an earlier age of first
puffing a cigarette or of smoking cigarettes daily were both strongly as-
sociated with a greater likelihood of being a heavier smoker both in 18- to
25-year-olds and in those 26 years and older and that this association re-
mained consistent regardless of the transition time from first trying a ciga-
rette to becoming a daily smoker (HHS, 2014). Additional cross-sectional
studies document a strong dose-dependent association between a younger
age of initiation and a greater number of cigarettes smoked per day in
young adulthood (Breslau, 1993) and in older adulthood (Chen and Millar,
1998; D’Avanzo et al., 1994; Fernandez et al., 1999; Hu et al., 2006; Lando
et al., 1999; Taioli and Wynder, 1991).
Age of Initiation and Continued Smoking

The evidence reviewed above that a younger age of initiation is as-
sociated with greater nicotine dependence and greater smoking intensity
supports the suggestion that an earlier age of initiation would be associ-
ated with an increased likelihood of remaining a smoker throughout the
life span, and the empirical data on this association supports that assump-
tion. An earlier age of starting to smoke cigarettes has been associated with
an increased likelihood of remaining a smoker (or reduced likelihood of
quitting) in several studies that span periods of life starting at various points
from pre-high school to high school (Everett et al., 1999) and progressing
to young adults (Breslau and Peterson, 1996) and older adulthood (Chen
and Millar, 1998; D’Avanzo et al., 1994; Eisner et al., 2000; Khuder et al.,
1999). The influence of the age of initiation on smoking cessation does not
appear to simply be an artifact of an early initiation of smoking being asso-

ciated with a longer duration of smoking, all else being held equal (Breslau
and Peterson, 1996).
Finding 2-7: An earlier age of initiation is associated with greater levels

of nicotine dependence.
Finding 2-8: An earlier age of initiation is associated with greater in-

tensity and persistence of smoking beyond adolescence and through
adulthood.
TOBACCO CESSATION AMONG ADOLESCENTS

AND YOUNG ADULTS
As noted above, a sizable portion of adolescent smokers, even those
who are infrequent and light smokers, show signs of nicotine addiction and
are likely to continue smoking into adulthood. The fact that adolescents
do not seem to spontaneously “mature out” of smoking (Mermelstein,
2003) does not necessarily reflect a lack of motivation to quit. Rather,
a majority of adolescent smokers want to quit, and many of them make
serious attempts to do so (Bancej et al., 2007; Marshall et al., 2006).
However, tobacco cessation among adolescents is challenging. Despite a
lower frequency and intensity of use in adolescents compared with adults,
the rates of cessation among adolescents are low, and most adolescents
experience difficulty in quitting (Mermelstein, 2003; O’Loughlin et al.,
2009). Most adolescents who want to quit attempt to do so without any
formal assistance, and of the few who have formal assistance, even fewer
use evidence-based approaches (Curry et al., 2009). Although there are
a number of good behaviorally based interventions for adolescents, and
these interventions increase the chances of adolescent smokers achieving
cessation, their reach is limited and their overall success rates are lower
than one finds with adult evidence-based programs (Curry et al., 2009).
A recent Cochrane meta-analysis of tobacco cessation interventions for
regular smokers younger than 20 reported mixed findings for interventions,
with the more complex counseling approaches showing some promise, but
few trials showing pharmacotherapy to be effective in helping adolescent
smokers quit (Stanton and Grimshaw, 2013). The review concluded that
there is not yet sufficient evidence to recommend one specific approach for
widespread implementation for adolescent smokers.
The developmental challenges of adolescence may also interfere with
an adolescent smoker’s ability to quit. These challenges include the ado-
lescent’s stage of cognitive development and ability to problem-solve and
maintain coping skills under periods of emotional arousal, particularly
arousal brought on during nicotine withdrawal, as well as other age-based

challenges that come with an adolescent’s lack of control over his or her
environment and lack of ability to modify cues that may promote smoking
(Curry et al., 2009). Thus, not all adolescents who smoke may have the cog-
nitive, environmental, and emotional resources to make cessation attempts
successful. Cessation attempts are also less successful among adolescents
who smoke more or who smoke daily (Bancej et al., 2007). In one of the
few studies to examine the discontinuation of smoking among adolescents
who are light and mostly intermittent smokers, O’Loughlin et al. (2014)
found that males and older adolescents were more likely to discontinue
smoking, and suggested that older adolescents may be more successful for
a variety of reasons, including moving into adult roles, developing increased
skills to manage a quit attempt, and having more exposure to cessation
aids.
Young adults also find cessation challenging, and the evidence is mixed
as to whether young adults are more successful than older adults, with
relatively few studies having compared cessation rates across age groups.
Messer et al. (2008) found that young adults ages 18 to 24 were more likely
to quit successfully than older adults. However, Villanti et al. (2010) found
that there is limited evidence for the efficacy of cessation interventions spe-
cifically geared to young adults. In a meta-analysis addressing the question
of whether cessation interventions that are successful for older adults work
equally well for young adults, Suls et al. (2012) found that interventions
that are efficacious for the general adult population are equally effective
for young adults. The larger problem, however, is attracting young adults
to evidence-based cessation programs (Suls et al., 2012).
In sum, adolescents, even those who are light and intermittent ad-
olescent smokers, have difficulty stopping their tobacco use, especially
once dependence symptoms have emerged, even if the symptoms have not
yet reached the level of fully developed nicotine dependence. In addition,
evidence-based cessation interventions for adolescents are not as easily or
widely available as they are for adults, and pharmacological approaches are
limited in both reach and effectiveness (Curry et al., 2009). More cessation
options are available for young adults, and success in quitting may be easier
to achieve during the young adult years.
Finding 2-9: Tobacco cessation among adolescents is difficult to achieve,

with few, if any, well-supported interventions that are available for
widespread dissemination. More effective treatment options are avail-
able for young and older adults.

REFERENCES
Arrazola, R. A., S. Dube, and B. A. King. 2013. Tobacco product use among middle and high
school students—United States, 2011 and 2012. Morbidity and Mortality Weekly Report
62(45):893–897.
Arrazola, R. A., N. M. Kuiper, and S. R. Dube. 2014. Patterns of current use of tobacco
products among U.S. high school students for 2000–2012—Findings from the National
Youth Tobacco Survey. Journal of Adolescent Health 54(1):54–60.
Austin, S. B., N. Ziyadeh, L. B. Fisher, J. A. Kahn, G. A. Colditz, and A. L. Frazier. 2004.
Sexual orientation and tobacco use in a cohort study of U.S. adolescent girls and boys.
Archives of Pediatrics and Adolescent Medicine 158(4):317–322.
Bachman, J. G., P. M. O’Malley, L. D. Johnston, J. E. Schulenberg, and J. M. Wallace, Jr. 2011.
Racial/ethnic differences in the relationship between parental education and substance
use among U.S. 8th-, 10th-, and 12th-grade students: Findings from the Monitoring the
Future project. Journal of Studies on Alcohol and Drugs 72(2):279–285.
Bancej, C., J. O’Loughlin, R. W. Platt, G. Paradis, and A. Gervais. 2007. Smoking cessation
attempts among adolescent smokers: A systematic review of prevalence studies. Tobacco
Control 16(6):e8.
Bernat, D. H., D. J. Erickson, R. Widorne, C. L. Perry, and J. L. Forster. 2008. Adolescent
smoking trajectories: Results from a population-based cohort study. Journal of Adoles-
cent Health 43(4):334–340.
Breslau, N. 1993. Daily cigarette consumption in early adulthood: Age of smoking initiation
and duration of smoking. Drug and Alcohol Dependence 33(3):287–291.
Breslau, N., and E. L. Peterson. 1996. Smoking cessation in young adults: Age at initiation
of cigarette smoking and other suspected influences. American Journal of Public Health
86(2):214–220.
Brook, D. W., J. S. Brook, C. Zhang, M. Whiteman, P. Cohen, and S. J. Finch. 2008. Develop-
mental trajectories of cigarette smoking from adolescence to the early thirties: Personality
and behavioral risk factors. Nicotine & Tobacco Research 10(8):1283–1291.
Buchmann, A. F., D. Blomeyer, C. Jennen–Steinmetz, M. H. Schmidt, G. Esser, T. Banaschewski,
and M. Laucht. 2013. Early smoking onset may promise initial pleasurable sensations
and later addiction. Addiction Biology 18(6):947–954.
Chassin, L., C. Presson, D. C. Seo, S. J. Sherman, J. Macy, R. J. Wirth, and P. Curran. 2008.
Multiple trajectories of cigarette smoking and the intergenerational transmission of
smoking: A multigenerational, longitudinal study of a Midwestern community sample.
Health Psychology 27(6):819–828.
Chen, J., and W. J. Millar. 1998. Age of smoking initiation: Implications for quitting. Health
Reports 9(4):39–46.
Colder, C. R., P. Mehta, K. Balanda, R. T. Campbell, K. P. Mayhew, W. R. Stanton, M. A.
Pentz, and B. R. Flay. 2001. Identifying trajectories of adolescent smoking: An applica-
tion of latent growth mixture modeling. Health Psychology 20(2):127–135.
Colder, C. R., B. R. Flay, E. Segawa, D. Hedeker, and TERN members. 2008. Trajectories
of smoking among freshmen college students with prior smoking history and risk for
future smoking: Data from the University Project Tobacco Etiology Research Network
(UpTERN) study. Addiction 103(9):1534–1543.
Costello, D. M., L. C. Dierker, B. L. Jones, and J. S. Rose. 2008. Trajectories of smoking from
adolescence to early adulthood and their psychosocial risk factors. Health Psychology
27(6):811–818.
Curry, S. J., R. J. Mermelstein, and A. K. Sporer. 2009. Therapy for specific problems: Youth
tobacco cessation. Annual Review of Psychology 60:229–255.

D’Avanzo, B., C. La Vecchia, and E. Negri. 1994. Age at starting smoking and number of
cigarettes smoked. Annals of Epidemiology 4(6):455–459.
156(5):818–822.
DiFranza, J. R., J. A. Savageau, N. A. Rigotti, K. Fletcher, J. K. Ockene, A. D. McNeill,
M. Coleman, and C. Wood. 2002. Development of symptoms of tobacco dependence
in youth: 30 month follow up data from the DANDY study. Tobacco Control 11(3):
228–235.
Eaton, D. K., L. Kann, S. Kinchen, S. L. Shanklin, K. H. Flint, J. Hawkins, W. A. Harris, R.
Lowry, T. McManus, D. Chyen, L. Whittle, C. Lim, and H. Wechsler. 2012. Youth risk
behavior surveillance—United States, 2011. Morbidity and Mortality Weekly Report
61(4):1–162.
Eisner, M. D., E. H. Yelin, P. P. Katz, S. C. Shiboski, J. Henke, and P. D. Blanc. 2000. Predic-
tors of cigarette smoking and smoking cessation among adults with asthma. American
Journal of Public Health 90(8):1307–1311.
Escobedo, L. G., S. E. Marcus, D. Holtzman, and G. A. Giovino. 1993. Sports participation,
age at smoking initiation, and the risk of smoking among U.S. high school students.
JAMA 269(11):1391–1395.
Everett, S. A., C. W. Warren, D. Sharp, L. Kann, C. G. Husten, and L. S. Crossett. 1999. Ini-
tiation of cigarette smoking and subsequent smoking behavior among U.S. high school
students. Preventive Medicine 29(5):327–333.
Fagan, P., E. T. Moolchan, D. Lawrence, A. Fernander, and P. K. Ponder. 2007. Identifying
health disparities across the tobacco continuum. Addiction 102(Suppl 2):5–29.
Fernandez, E., A. Schiaffino, C. La Vecchia, J. M. Borras, M. Nebot, E. Salto, R. Tresserras,
L. Rajmil, J. R. Villalbi, and A. Segura. 1999. Age at starting smoking and number of
cigarettes smoked in Catalonia, Spain. Preventive Medicine 28(4):361–366.
Goodman, E., and J. Capitman. 2000. Depressive symptoms and cigarette smoking among
teens. Pediatrics 106(4):748–755.
Green, M. P., K. L. McCausland, H. Xiao, J. C. Duke, D. M. Vallone, and C. G. Healton.
2007. A closer look at smoking among young adults: Where tobacco control should focus
its attention. American Journal of Public Health 97(8):1427–1433.
Hatzenbuehler, M. L., N. F. Wieringa, and K. M. Keyes. 2011. Community-level determinants
of tobacco use disparities in lesbian, gay, and bisexual youth: Results from a population-
based study. Archives of Pediatrics and Adolescent Medicine 165(6):527–532.
HHS (Department of Health and Human Services). 2012. Preventing tobacco use among
youth and young adults: A report of the Surgeon General. Atlanta GA: U.S. Department
of Health and Human Services, Centers for Disease Control and Prevention, National
Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and
Health.
———. 2014. The health consequences of smoking— 50 years of progress: A report of the
HHS, SAMHSA, and CBHSQ (Department of Health and Human Services, Substance Abuse
and Mental Health Services Administration, and Center for Behavioral Health Statistics
and Quality). 2014. National survey on drug use and health, 2012. ICPSR34933-v2:
Ann Arbor, MI: Inter-University Consortium for Political and Social Research [distribu-
tor], 2014-10-06. http://doi.org/10.3886/ICPSR34933.v2 (accessed February 6, 2015).

Hu, M. C., M. Davies, and D. B. Kandel. 2006. Epidemiology and correlates of daily smoking
and nicotine dependence among young adults in the United States. American Journal of
Hu, M. C., B. Muthen, C. Schaffran, P. C. Griesler, and D. B. Kandel. 2008. Developmental
trajectories of criteria of nicotine dependence in adolescence. Drug and Alcohol Depen-
dence 98(1–2):94–104.
Jackson, K. M., K. J. Sher, and J. E. Schulenberg. 2008. Conjoint developmental trajecto-
ries of young adult substance use. Alcoholism: Clinical and Experimental Research
32(5):723–737.
Johnson, J. G., P. Cohen, D. S. Pine, D. F. Klein, S. Kasen, and J. S. Brook. 2000. Association
between cigarette smoking and anxiety disorders during adolescence and early adulthood.
JAMA 284(18):2348–2351.
Johnston, L. D., P. M. O’Malley, J. G. Bachman, and J. E. Schulenberg. 2013. Monitoring the
future: National survey results on drug use 1975–2012: Volume 2, College students &
adults ages 19–50. Ann Arbor, MI: Institute for Social Research, University of Michigan.
Demographic subgroup trends among adolescents in the use of various licit and illicit
drugs 1975–2013. Ann Arbor, MI: Institute for Social Research, University of Michigan.
———. 2014b. Monitoring the future: National survey results on drug use 1975–2013. Vol-
ume 1: Secondary school students. Ann Arbor: Institute for Social Research, University
of Michigan.
———. 2014c. Monitoring the future: National survey results on drug use 1975–2013: Vol-
ume 2, college students & adults ages 19–55. Ann Arbor: Institute for Social Research,
the University of Michigan.
Kandel, D. B., and K. Chen. 2000. Extent of smoking and nicotine dependence in the United
States: 1991–1993. Nicotine & Tobacco Research 2(3):263–274.
O’Malley Olsen, T. McManus, D. Chyen, L. Whittle, E. Taylor, Z. Demissie, N. Brener,
J. Thornton, J. Moore, and S. Zaza, Centers for Disease Control and Prevention. 2014.
Youth risk behavior surveillance—United States, 2013. Morbidity and Mortality Weekly
Report Surveillance Summaries 63(4):1–168.
Khuder, S. A., H. H. Dayal, and A. B. Mutgi. 1999. Age at smoking onset and its effect on
smoking cessation. Addictive Behaviors 24(5):673–677.
Lando, H. A., D. T. Thai, D. M. Murray, L. A. Robinson, R. W. Jeffery, N. E. Sherwood, and
D. J. Hennrikus. 1999. Age of initiation, smoking patterns, and risk in a population of
working adults. Preventive Medicine 29(6 Part 1):590–598.
Lessov-Schlaggar, C. N., H. Hops, J. Brigham, K. S. Hudmon, J. A. Andrews, E. Tildesley, D.
McBride, L. M. Jack, H. S. Javitz, and G. E. Swan. 2008. Adolescent smoking trajectories
and nicotine dependence. Nicotine & Tobacco Research 10(2):341–351.
Marshall, L., M. Schooley, H. Ryan, P. Cox, A. Easton, C. Healton, K. Jackson, K. C. Davis,
and G. Homsi. 2006. Youth tobacco surveillance —United States, 2001–2002. Morbidity
and Mortality Weekly Report 55(3):1–56.
Mayhew, K. P., B. R. Flay, and J. A. Mott. 2000. Stages in the development of adolescent
smoking. Drug and Alcohol Dependence 59(Suppl 1):S61–S81.
Melnick, M. J., K. E. Miller, D. F. Sabo, M. P. Farrell, and G. M. Barnes. 2001. Tobacco use
among high school athletes and nonathletes: Results of the 1997 Youth Risk Behavior
Survey. Adolescence 36(144):727–747.
Mermelstein, R. 2003. Teen smoking cessation. Tobacco Control 12(Suppl 1):i25–i34.
Mermelstein, R., S. M. Colby, C. Patten, A. Prokhorov, R. Brown, M. Myers, W. Adelman,
K. Hudmon, and P. McDonald. 2002. Methodological issues in measuring treatment
outcome in adolescent smoking cessation studies. Nicotine & Tobacco Research 4(4):
395–403.

Messer, K., D. R. Trinidad, W. K. Al-Delaimy, and J. P. Pierce. 2008. Smoking cessation rates
in the United States: A comparison of young adult and older smokers. American Journal
of Public Health 98(2):317–322.
O’Loughlin, J., J. DiFranza, R. F. Tyndale, G. Meshefedjian, E. McMillan-Davey, P. B. S.
Clarke, J. Hanley, and G. Paradis. 2003. Nicotine-dependence symptoms are associ-
ated with smoking frequency in adolescents. American Journal of Preventive Medicine
25(3):219–225.
O’Loughlin, J., A. Gervais, E. Dugas, and G. Meshefedjian. 2009. Milestones in the process
of cessation among novice adolescent smokers. American Journal of Public Health
99(3):499–504.
O’Loughlin, J. L., M. P. Sylvestre, E. N. Dugas, and I. Karp. 2014. Predictors of the occur-
rence of smoking discontinuation in novice adolescent smokers. Cancer Epidemiology,
Biomarkers & Prevention 23(6):1090–1101.
Park, S. M., K. Y. Son, Y. J. Lee, H. C. Lee, J. H. Kang, Y. J. Lee, Y. J. Chang, and Y. H. Yun.
2004. A preliminary investigation of early smoking initiation and nicotine dependence
in Korean adults. Drug and Alcohol Dependence 74(2):197–203.
Peek, K. J., J. Kusel, S. L. Eyre, C. L. Auerswald, and B. L. Halpern-Felsher. In preparation
(unpublished). Understanding smokeless tobacco initiation and use among rural youth:
An unexplored population.
Reidpath, D. D., T. M. Davey, A. Kadirvelu, I. N. Soyiri, and P. Allotey. 2014. Does one
cigarette make an adolescent smoker, and is it influenced by age and age of smoking ini-
tiation? Evidence of association from the U.S. Youth Risk Behavior Surveillance System
(2011). Preventive Medicine 59:37–41.
Riggs, N. R., C. P. Chou, C. Li, and M. A. Pentz. 2007. Adolescent to emerging adulthood
smoking trajectories: When do smoking trajectories diverge, and do they predict early
adulthood nicotine dependence? Nicotine & Tobacco Research 9(11):1147–1154.
Rosario, M., J. Hunter, and M. Gwadz. 1997. Exploration of substance use among lesbian,
gay, and bisexual youth: Prevalence and correlates. Journal of Adolescent Research
12(4):454–476.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2012a. 2010–2011
National Survey on Drug Use and Health national maps of prevalence estimates by
state. http://www.samhsa.gov/data/NSDUH/2k11State/NSDUHsae2011/Maps/NSDUH
saeMaps2011.htm (accessed January 28, 2015).
———. 2012b. Comparing and evaluating youth substance use estimates from the National
Survey on Drug Use and Health and other surveys. Rockville, MD: Substance Abuse
and Mental Health Services Administration; Center for Behavioral Statistics and Quality;
Department of Population Surveys.
———. 2012c. National Survey on Drug Use and Health: Comparison of 2002–2003 and
2010–2011 model-based prevalence estimates (50 states and the District of Columbia).
Rockville, MD: Substance Abuse and Mental Health Services Administration; Center for
Behavioral Statistics and Quality; Department of Population Surveys.
———. 2013a. Results from the 2012 National Survey on Drug Use and Health: Detailed
tables. Rockville, MD: Substance Abuse and Mental Health Services Administration;
Center for Behavioral Statistics and Quality; Department of Population Surveys.
———. 2013b. Results from the 2012 National Survey on Drug Use and Health: Summary of
national findings. Rockville, MD: Substance Abuse and Mental Health Services Adminis-
tration; Center for Behavioral Statistics and Quality; Department of Population Surveys.
Selya, A. S., L. C. Dierker, J. S. Rose, D. Hedeker, X. Tan, R. Li, and R. J. Mermelstein. 2013.
Time-varying effects of smoking quantity and nicotine dependence on adolescent smoking
regularity. Drug and Alcohol Dependence 128(3):230–237.

Shiffman, S. 2009. Light and intermittent smokers: Background and perspective. Nicotine &
Tobacco Research 11(2):122–125.
Stanton, A., and G. Grimshaw. 2013. Tobacco cessation interventions for young people.
Cochrane Database of Systematic Reviews 23(8).
Stanton, W. R., B. R. Flay, C. R. Colder, and P. Mehta. 2004. Identifying and predicting ado-
lescent smokers’ developmental trajectories. Nicotine & Tobacco Research 6(5):843–852.
Suls, J. M., T. M. Luger, S. J. Curry, R. J. Mermelstein, A. K. Sporer, and L. C. An. 2012.
Efficacy of smoking-cessation interventions for young adults: A meta-analysis. American
Journal of Preventive Medicine 42(6):655–662.
Swan, G. E., and C. N. Lessov-Schlaggar. 2007. The effects of tobacco smoke and nicotine on
cognition and the brain. Neuropsychology Review 17(3):259–273.
Taioli, E., and E. L. Wynder. 1991. Effect of the age at which smoking begins on frequency of
smoking in adulthood. New England Journal of Medicine 325(13):968–969.
Tucker, J. S., P. L. Ellickson, M. Orlando, and D. J. Klein. 2006. Cigarette smoking from
adolescence to young adulthood: Women’s developmental trajectories and associates
outcomes. Women’s Health Issues 16(1):30–37.
Villanti, A. C., H. S. McKay, D. B. Abrams, D. R. Holtgrave, and J. V. Bowie. 2010. Smoking-
cessation interventions for U.S. young adults: A systematic review. American Journal of
Preventive Medicine 39(6):564–574.
Wadley, J., and S. Barnes. 2013. Teen smoking continues to decline in 2013. Ann Arbor, MI:
University of Michigan News Service. http://home.isr.umich.edu/releases/teen-smoking-
continues-to-decline-in-2013 (accessed February 6, 2015).
Wadley, J., and A. Bronson. 2014. E-cigarettes surpass tobacco cigarettes among teens. Ann
Arbor, MI: University of Michigan News Service. http://ns.umich.edu/new/multimedia/
videos/22575-e-cigarettes-surpass-tobacco-cigarettes-among-teens (accessed February 6,
2015).
Welte, J. W., G. M. Barnes, M. C. O. Tidwell, and J. H. Hoffman. 2011. Tobacco use, heavy
use, and dependence among adolescents and young adults in the United States. Substance
Use and Misuse 46(9):1090–1098.
Ziedonis, D., B. Hitsman, J. C. Beckham, M. Zvolensky, L. E. Adler, J. Audrain-McGovern,
N. Breslau, R. A. Brown, T. P. George, J. Williams, P. S. Calhoun, and W. T. Riley. 2008.
Tobacco use and cessation in psychiatric disorders: National Institute of Mental Health
report. Nicotine & Tobacco Research 10(12):1691–1715.

The Developmental and Environmental

Context of Adolescent and
Young Adult Tobacco Use
T
obacco use is the result of a complex and dynamic interplay of mul-
tiple converging developmental, social, and environmental factors.
Many of these factors are developmentally related, with adolescence
and young adulthood as a key period of vulnerability to tobacco use and
the progression to nicotine dependence (Jamner et al., 2003).
The development of adult decision-making skills and abilities is a
continuous process that begins in early adolescence and continues into and
through young adulthood, with no firm age periods for when specific de-
velopmental milestones occur. Furthermore, there are individual variations,
with spurts of change and disjuncture resulting from social and environ-
mental factors that influence the normative developmental process. These
social influences are particularly salient in later adolescence and young
adulthood.
Although previously considered a relatively short transition period, the
late teens through the early 20s (ages approximately 18 to 26) is now con-
sidered a distinct period of life known as young adulthood (IOM and NRC,
2014). The newfound focus on this developmental period is due in part to
prolonged education, delayed marriage, and delayed parenthood—events
that historically marked adulthood, adult roles, and adult responsibility
(Settersten and Ray, 2010)—and in part to studies showing that the brain
continues to develop until the mid-20s (Giedd, 2008; Luna et al., 2004).
Individuals in young adulthood face developmental and life changes that
may make them particularly susceptible to drug use for several reasons: a
desire to explore their identity, a response to the instability and disruption
associated with life changes, or because of a tendency to focus on the pos-
63

sible positive consequences of drug use rather than negative consequences.

Additionally, this is a time period when experimentation with risky behav-
ior is most tolerated (IOM and NRC, 2014).
The unique psychosocial maturation of the adolescent and young adult
developmental period, coupled with various environmental and social influ-
ences, results in a milieu that increases the desire for engaging in health-risk
behaviors, including tobacco use. Furthermore, brain function and height-
ened sensitivity to nicotine characteristic of this period of development pro-
vides the biological context underlying the psychosocial and environmental
influences related to adolescents’ and young adults’ decisions to start and
continue to use tobacco.
The chapter begins with a review of the complex and layered cognitive,
psychosocial, and biological aspects of adolescent and young adult develop-
ment, with a focus on factors most likely to explain the heightened likeli-
hood of tobacco initiation, continued use, and dependence. The chapter
then ties these factors into the decision-making capabilities of adolescents
and young adults. The chapter concludes with a discussion of the environ-
mental context of tobacco use, including salient residential, school, and
work changes and the role of tobacco marketing on adolescent and young
adult tobacco use.
COGNITIVE, PSYCHOSOCIAL, AND BIOLOGICAL

DEVELOPMENT IN ADOLESCENTS AND YOUNG ADULTS
Adolescence and young adulthood is a period of change with respect to
cognitive, psychosocial, neurobiological, and physical development. These
changes often result in increased vulnerabilities to using tobacco. These
factors are reviewed next.
Cognitive Development
During adolescence, thinking becomes less concrete and more abstract,
giving adolescents the ability to consider many components necessary for
competent decision making at one time, consider potential positive and
negative outcomes associated with each decision, and plan for the future.
Studies have shown that by the time adolescents reach age 16, their general
cognitive abilities, such as the ability to understand consequences—including
the risks and benefits of their decisions—to process information, and to
reason, are essentially identical to those of adults (Albert and S teinberg,
2011; Halpern-Felsher and Cauffman, 2001; Steinberg et al., 2009a). For
example, in a study of 935 individuals ranging from age 10 to 30, Steinberg
and colleagues (2009a) found no significant differences in cognitive skills
between older adolescents (as young as ages 15–16) and adults.

DEVELOPMENTAL AND ENVIRONMENTAL CONTEXT 65
Although there are individual differences and within-age-group varia-

tion, most adolescents reach a level of cognitive maturity comparable to
adults by age 16. Despite the fact that cognitive maturity is reached by mid-
adolescence, other aspects of psychosocial maturity, such as peer influence,
sensation seeking, reward seeking, and impulse control, are still developing
(as discussed later in this chapter). These different developmental systems
explain in part why adolescents and young adults may have the cognitive
ability to make safe and healthy decisions, yet are more prone than adults
to make risky decisions. As shown below, even though adolescents have the
ability to think abstractly and judge risks, they do not always adequately
employ these abilities. Instead, adolescents are often seeking rewards and
pleasures and therefore may decide to use tobacco despite knowing and
understanding both the short-term and long-term risks.
Perceptions of Risks and Benefits

A hallmark of cognitive development is the ability to identify and un-
derstand consequences associated with a particular behavior. Perceptions
of social, physical, and health risks associated with any given behavior as
well as the perceived benefits, including both social and physical benefits,
are key components of any competent decision. Research has shown that
such perceptions actually predict the onset of behavior (Song et al., 2009b).
Adolescents, young adults, and adults are generally similar in their
ability to identify and consider positive and negative consequences of their
decisions. In some cases, adolescents actually perceive greater risks than
do adults (e.g., Millstein and Halpern-Felsher, 2002). Several studies have
shown that adolescents and young adults consider risks, benefits, and the
value of behavior-related outcomes just prior to deciding on a particular
behavior and that adolescents and young adults are keenly aware of risks
(e.g., Halpern-Felsher and Cauffman, 2001; Lewis, 1981; Michels et al.,
2005). In a review article, Albert and Steinberg (2011) concluded that
there are few differences between the evaluations that adolescents (with
ages varying depending on the study sample) and adults make of the risks
inherent in various risky behaviors and few differences in their perceptions
of the seriousness of these consequences (see also Kuther, 2003). Despite
adolescents’ general understanding—and often overestimation—of risks,
the perceptions of risks are only one part of the equation that adolescents
and young adults use to make decisions. Adolescents naturally consider
the importance of the social and physical benefits that they perceive they
will gain from any given behavior (Song et al., 2009b). Furthermore, ado-
lescents’ emotional immaturity and psychosocial factors influencing their
behavior, such as impulsivity and peer pressure, often override the cognitive
understanding of a risk.

Perceptions of Tobacco-Related Risks and Benefits Associated with

Tobacco Use
Many studies have examined risk and benefit perceptions related to
tobacco use. In general, studies show that people who smoke perceive less
harm and greater benefits from cigarettes than do nonsmokers (Chassin et
al., 2000; Fischhoff et al., 2010; Halpern-Felsher et al., 2004; Morrell et
al., 2010; Soldz and Cui, 2002; Song et al., 2009b). Compared to nonsmok-
ers, those who have smoked believe that they are less likely to experience
long-term risks, such as lung cancer, heart attack, addiction, and death,
and less likely to experience short-term consequences, such as smelling
bad or having trouble breathing (Halpern-Felsher et al., 2004; Morrell
et al., 2010; Song et al., 2009a). Smokers also believe that they are more
likely to experience pleasure, feel relaxed, and “look cool” from smoking
when compared to nonsmokers (Halpern-Felsher et al., 2004; Morrell et
al., 2010; Song et al., 2009b). A prospective study of adolescents 14 to 16
years old demonstrated that perceptions of low long- and short-term risk
and greater benefits predict the onset of tobacco use (Song et al., 2009b).
A much smaller body of work has examined whether perceptions of
risks and benefits vary by type, brand, or packaging of the tobacco product.
Historically, this research has focused on light and ultra-light cigarettes,
with studies showing that most adults and adolescents incorrectly perceive
that light cigarettes deliver less tar and nicotine, produce milder sensations,
result in less health risk, and can make cessation easier (Etter et al., 2003;
Gilpin et al., 2002; Kozlowski et al., 1998; Kropp and Halpern-Felsher,
2004; Shiffman et al., 2001; Tindle et al., 2006). More recent research has
shown that consumers perceive that menthol-flavored cigarettes are less
harmful than non-menthol-flavored cigarettes (Anderson, 2011; Klausner,
2011). Similarly, perceptions of the harms associated with snus (Choi et al.,
2012; Øverland et al., 2008), smokeless tobacco (Callery et al., 2011), and
cigars (Nyman et al., 2002) are lower compared to the perceived harms of
cigarettes, and people perceive differences in risk based on type and color
of product packaging (Bansal-Travers et al., 2011).
Psychosocial Development
In addition to developing the ability to consider the possible conse-
quences of actions, including the likelihood and value of each consequence,
adolescents and young adults are also maturing with respect to their psy-
chosocial abilities. Psychosocial components relevant to tobacco decision
making include social and peer comparison, sensation seeking and impulsiv-
ity, peer affiliation, susceptibility to peer pressure, the ability to understand
and plan for the future, and perceived social norms.

While individuals vary even within the same age range, generally speak-
ing most adolescents are on par with adults by age 16 with respect to
thinking about the future (e.g., Albert and Steinberg, 2011; Halpern-Felsher
and Cauffman, 2001; Steinberg et al., 2009b). However, other critical
aspects of psychosocial development, such as those associated with peer
pressure, sensation seeking, reward seeking, and impulse control, are much
less developed during adolescence than during adulthood (Halpern-Felsher
and Cauffman, 2001; Steinberg, 2008; Steinberg et al., 2008, 2009a;
Zuckerman, 1979). “Dynamic accounts of factors that predict adolescent
decisions” take into consideration the social, emotional, and self-regulatory
factors that help explain why adolescents can make decisions just as ra-
tionally as adults, but often do not (Albert and Steinberg, 2011, p. 211).
These areas of immaturity help explain why adolescents and young adults
are more susceptible than older adults to initiating tobacco use.
Future Perspective Taking

Future perspective taking includes the ability to project into the fu-
ture, to consider possible positive and negative outcomes associated with
choices, and to plan for the future (Steinberg et al., 2009b), and is a hall-
mark of decision-making competence. Without an adequate understanding
of future consequences and without the ability to have the future be part
of present planning, it is more difficult to make decisions about behavior,
including whether or not to use tobacco. It is not enough to have a work-
ing understanding of the possible risks and benefits that might come from
using tobacco; it is equally important to be able to apply that information
to making decisions about behaviors that could have an effect in the future.
Steinberg and colleagues (2009b) found that the ability to plan for the
future and to anticipate future consequences continues to develop through
the mid-20s (see also Halpern-Felsher and Cauffman, 2001).
Sensation Seeking and Impulsivity

Sensation seeking refers to the drive to seek out experiences that are
new, different, exciting, and highly stimulating as well as the willingness to
take risks in order to have these experiences (Steinberg, 2008; Zuckerman,
1979). Higher sensation seeking is associated with drug use in early and
middle adolescence (e.g., ages 12–16) (Kosten et al., 1994; Teichman et al.,
1989) and with pubertal development; early maturers tend to rate higher on
sensation-seeking scales and also on drug-seeking behavior (Martin et al.,
2001; Steinberg, 2008). While sensation seeking follows a developmental
trajectory, it is also viewed as a stable trait that is associated with risky
behavior (Zuckerman, 2007).

Impulsivity refers to a tendency to make decisions in a quick fashion,

without much thought or information. Impulsivity steadily declines from
age 10 on (Steinberg et al., 2008). Becoming competent to make decisions
requires that adolescents be able to control their desires and resist impulsive
actions. Recent studies have highlighted the complex relationship among
impulsivity, peer pressure, and delinquent behavior. Vitulano and colleagues
(2010) have found that individuals with low impulsivity are actually more
vulnerable to delinquent peer influences than those with high impulsivity.
Thus, adolescents find themselves in a bit of a quagmire in that those with
high impulsivity are likely to engage in risky behavior and those with low
impulsivity are particularly sensitive to peer pressure that may also lead
them to engage in risky behavior.
While impulsivity and sensation seeking are related, they are distinct
features of decision making. Impulsive behavior may lead to experiences
that are neither stimulating nor rewarding, and individuals may make the
decision to engage in sensation-seeking behavior in a deliberate and non-
impulsive manner (Steinberg et al., 2008). Additionally, while impulsive
behavior decreases in a linear fashion from age 10 on, sensation-seeking
patterns of development follow a curvilinear pattern in which sensation
seeking increases between childhood and early adolescence and then either
declines or remains stable in late adolescence and adulthood (Steinberg et
al., 2008). For example, Steinberg and colleagues found that while 16- to
17-year-olds and 18- to 21-year-olds exhibit more impulse control than 10-
to 15-year-olds, they exhibit significantly less impulse control than 22- to
25-year-olds and 26- to 30-year-olds.
Thus, adolescence and young adulthood is a time of low impulse con-
trol coupled with high rates of sensation seeking, which results in a greater
likelihood that individuals in these development periods will engage in risky
behavior. The coupling of low impulse control and high sensation seeking is
especially harmful in more emotionally charged situations, in which adoles-
cents are seeking rewards and pleasure yet do not have the ability to control
these desires. Hence, adolescents are more likely to seek rewards such as
those associated with tobacco use than they will be later in life, once the
connections between their rewards pathways and impulse control are more
in sync, which occurs in their mid-20s (Steinberg, 2013).
Social Norms
Social norms refer to common codes of behavior for a social group. The
construct is used in a number of disciplines and theories of health behavior,
including the Theory of Planned Behavior (Ajzen, 1985), Social Cognitive
Theory (Bandura, 2001), and the Theory of Normative Social Behavior
(Rimal and Real, 2005). Social norms are often classified as either descrip-

tive norms, which are perceptions of how people actually behave (which are
often operationalized as perceived prevalence rates), and injunctive norms,
which are perceptions of how people should behave (and are often opera-
tionalized by asking who would approve or disapprove of you engaging in
a behavior) (Cialdini et al., 1990; Kallgren et al., 2000).
Both injunctive and descriptive norms are associated with smoking
behaviors among adolescents and young adults. Alexander and colleagues
(2001) analyzed data from the National Longitudinal Study of Adolescent
Health and found that among 7th through 12th graders, adolescents in peer
groups where 50 percent or more members smoked, or whose best friends
smoked, were two times more likely to also smoke than those in peer groups
in which fewer than half of the members smoked. Additionally, popular stu-
dents who went to schools with higher smoking rates were more likely to
smoke than non-popular students, while popular students in schools with
low smoking rates were less likely to smoke. Etcheverry and Agnew (2008)
found that among college students, friends, and romantic partners, smoking
and injunctive norms were predictive of smoking behavior.
Peer Affiliation and Susceptibility to Peer Pressure

The ability to make rational decisions is mediated by a number of fac-
tors and, for adolescents, social factors in particular play a very large role
in behavioral decision making. The transition to adolescence is marked by
a decrease in time spent with parents and an increase in time spent either
alone or with peers (Steinberg and Morris, 2001). This is a time period in
which the opinions and actions of peers become increasingly important in
influencing behavior (Crone and Dahl, 2012). Observational studies show
that adolescents who engage in delinquent behavior are more likely to do
so in groups (as opposed to adults, who are more likely to engage in delin-
quent behavior alone) (Albert et al., 2013; IOM and NRC, 2011; Zimring,
2000). Experimental studies have also shown that adolescents are more
likely to make riskier decisions when they are told that they are being ob-
served by peers than when they believe they are working alone (Albert et al.,
2013). Compared with adults, adolescents exhibit exaggerated responses
to positive social cues, and this reaction is coupled with more impulsive
responses to stimuli (Albert et al., 2013; Gardner and Steinberg, 2005).
Generally, susceptibility to peer pressure that is undesirable or that goes
against an individual’s goals decreases steadily from age 14 to 18 (Steinberg
and Monahan, 2007). In order to make competent decisions, individuals
must have the ability to resist undue pressure from others. That being said,
studies also show that peers remain powerful influences and reinforcers
of behavior even in late adolescence and young adulthood. For example,
Duncan and colleagues found that males entering college with a history of

binge drinking were more likely to binge drink if they were paired with
roommates who also binge drank in high school than they were if they were
paired with a roommate who did not binge drink (Duncan et al., 2005).
Furthermore, an experimental study assessing the differences in how peers
influence risky behavior in adolescents (ages 13–16), young adults (ages
18–22) and adults (ages 24 and older) found that all three age groups
made safe decisions when alone. However, in the presence of peers, both
adolescents and young adults made risky decisions, with adolescents mak-
ing riskier decisions than young adults, while adults, on average, made the
safest decisions (Gardner and Steinberg, 2005).
Importance of Experience
Behavioral decisions and the perceptions of related outcomes are in-
fluenced by the extent to which a person has knowledge of and experi-
ence with the behavior or behavior-linked outcomes (Albert and Steinberg,
2011; IOM and NRC, 2004). Knowledge varies not only across ages but
also within age groups. Adolescents and, to a lesser extent, young adults
experience greater motivation to seek external rewards compared to adults,
which results in this age group being more likely to exhibit approach be-
haviors (i.e., those driven by positive or desirable events or outcomes) than
avoidance behaviors (i.e., those driven by negative or undesirable events or
outcomes) (Elliot, 1999). Risk taking and sensation seeking can be viewed
as part of this drive to experience potential rewards; thus, adolescence is a
period in which individuals are particularly likely to initiate behaviors such
as smoking (Lydon et al., 2014). This is particularly troubling because indi-
viduals who initiate smoking during adolescence are more likely to have a
pleasurable first experience than individuals who initiate smoking in adult-
hood. Furthermore, studies show that pleasurable initial experiences are
associated with rapid progression to regular smoking as well as continued
smoking (DiFranza et al., 2007; Sartor et al., 2010).
In addition to the impact of having (or not having) direct personal
experiences with particular consequences of behaviors, research has also in-
vestigated the effect of vicarious experiences, or knowledge about behaviors
and related positive and negative outcomes experienced by others (Morrell
et al., 2010). Applied to tobacco, adolescents and young adults rarely have
knowledge of peers who have experienced tobacco-related disease, which
lowers their perceptions of the likelihood of negative outcomes occurring
after using tobacco (Morrell et al., 2010). When adolescents and young
adults have had experience with tobacco-related illness, it is often in those
much senior to them. Given their immature sense of the future and their
ease at discounting the idea that what happens to others may also apply

to themselves, adolescents often do not apply the experiences of others to

themselves (Morrell et al., 2010).
Additional Psychosocial Aspects of Young Adulthood

Later adolescence through young adulthood is a time of great demo-
graphic change and instability, including changes involving place of resi-
dence, employment, school attendance, and family formation, all of which
play a substantial role in influencing tobacco use. Around age 18, most
young people have moved away from home, and young adults continue to
change residences more than any other age group (Arnett, 2000). In 2012,
among adults ages 18 to 31, 23 percent were married and living in their
own residence, 27 percent lived independently with others (i.e., cohabitat-
ing, living with a roommate or as a boarder, or in single parenthood), and
fewer than 10 percent were living on their own (Fry, 2014). Young adults
are also experiencing changes in their employment status, as they obtain
various part-time or full-time jobs to earn money for school and living
expenses, move, change colleges, and so on. Although employment rates
among young adults have declined considerably over the past few decades,
this decline has been largely offset by increases in educational attainment
(IOM and NRC, 2014). Indeed, young adults are significantly more edu-
cated now compared with previous generations, with twice as many adults
ages 18 to 31 having attained some education beyond high school in 2012
than in 1968 (Fry, 2014). However, while 85 percent of young adults enroll
in college within 1 year of their 18th birthday, a majority of these young
adults have not completed their degrees before age 25 (IOM and NRC,
2014). Young adults who do complete college often continue their educa-
tion in graduate or professional school (Arnett, 2000). Finally, in terms of
family formation, by age 25 nearly half of all young adults report having
cohabitated with a romantic partner, roughly one-third have become a par-
ent, and more than a quarter have married, with nearly two-thirds of young
adults having engaged in at least one of these family formation transitions
(IOM and NRC, 2014). These demographic changes and instabilities are
likely to play a role in young adults’ initiation with tobacco.
Young adulthood is also an intense time of personal change and growth,
which occur as the young adult is less subject to parental and societal re-
strictions, while simultaneously not being bound by the restrictions and
responsibilities that typically characterize adulthood. Given that delay in
assuming adult roles and responsibilities, the young adult period is ripe for
exploration and experimentation. Young adult exploration is not so much
to prepare for adult roles, but for the sake of exploration itself; it is a time
of exploration prior to settling into adult roles and responsibilities. This is
a time with very little expectation for marriage, parenthood, or permanent

employment, coupled with few, if any, parental restrictions, which creates

a near-perfect atmosphere for identity development (Arnett, 2000, 2004).
Furthermore, as young adults explore new living situations, including be-
ing away from parental restrictions and opportunities for cohabitation,
and also affiliate more with peers who use tobacco, it stands to reason that
opportunities for greater substance use will emerge.
Self-reflection is an important characteristic of young adult identity
exploration, and three areas of self-reflection that often occur during young
adulthood may affect decisions regarding tobacco use. The first is the extent
to which young adults feel as if they have reached adulthood. In addition
to demographic shifts, studies show that a large number of young adults do
not consider themselves to have achieved adult status, as defined by finan-
cial independence and family formation; they can be characterized as still
being in the “age of feeling in-between” (Arnett, 2004; Nelson and Barry,
2005). Second, given the vast amount of exploration desired by young
adults and the limited restrictions and accountability that results from work
or family obligations, young adults are likely to feel less accountable and
therefore less vulnerable to risks during this “age of possibilities” (Arnett,
2004). Finally, young adults’ perceptions of the extent to which their peers
are using tobacco, as well as whether tobacco use is viewed as acceptable,
are likely to influence patterns of tobacco use (Simons-Morton et al., 2001).
A number of important findings can be drawn from the above review:
Finding 3-1: The period from adolescence through young adulthood is

one of continuous development that involves increasing cognitive skills
and psychosocial maturity. There are no specific age markers.
Finding 3-2: The development of some cognitive abilities, such as un-

derstanding risks and benefits, is achieved by age 16. However, many
areas of psychosocial maturity, including sensation seeking, impulsivity,
and future perspective taking continue to develop and change through
late adolescence and into young adulthood.
Finding 3-3: Adolescence is a period of greatest peer affiliation and

susceptibility to peer influence.
Biological Development of Adolescents and Young Adults
Physical Development
Physical development, including the development of secondary sexual
characteristics, is one of the most important and noticeable hallmarks of
adolescence. The emergence of these newly developed physical features

occurs on average between the ages of 10 and 15 for both girls and boys
(Susman et al., 2010), leading adolescents to begin to have more of an
adult-like appearance, which then often results in their own and others’
beliefs that they can and should adopt more adult roles. However, as noted
below, looking like an adult does not equate to having the cognitive, social,
or emotional readiness to make adult-like decisions.
Physically maturing either earlier (the lowest 10 to 15 percent of the
adolescent population) or later (the highest 10 to 15 percent of the adoles-
cent population) is associated with an individual’s likelihood of engaging
in risky behavior, including tobacco use (Cance et al., 2013; Mendle and
Ferrero, 2012; Mendle et al., 2007). For males, being either an early or late
maturer can have negative outcomes on psychosocial adjustment (Mendle
and Ferrero, 2012) and can lead to increased substance use (Cance et al.,
2013). For females, being an earlier physical maturer can result in adjust-
ment problems and, most relevant to tobacco use, to problems with body
image that can lead to eating disorders (Mendle et al., 2007). Studies have
shown that girls, primarily white girls, sometimes initiate tobacco use in
order to lose weight (HHS, 2012). By contrast, for females, having a later
physical maturation can be protective against risky behaviors and adjust-
ment issues, including tobacco use, as compared with males, who have
more adjustment difficulty if they mature late (Crockett and Petersen, 1987;
Mendle and Ferrero, 2012; Mendle et al., 2007; Siegel et al., 1999).
Neurobiological Development
Physical maturation, including brain development, occurs throughout
the adolescent and young adult years. Neuroscience research provides in-
sights that show how brain maturation affects the social and emotional
development of adolescents and young adults and helps explain why they
are more susceptible to using tobacco than are adults.
The majority of the recent research on adolescent and young adult
brain development has found that both structural and functional changes
occur during adolescence, continuing into young adulthood (e.g., Giedd,
2008; Luna et al., 2004). There are four lobes in the brain: the parietal
lobe, occipital lobe, temporal lobe, and frontal lobe. The frontal lobe, the
largest part of the brain, contains the prefrontal cortex, which is located in
the front of the brain, behind the forehead. The prefrontal cortex is respon-
sible for executive functioning, including cognition, thought, imagination,
abstract thinking, planning, and impulse control. Brain development begins
at the back and progresses to the front of the brain, with the prefrontal
cortex being one of the last areas to mature (Gogtay et al., 2004; Sowell
et al., 1999).
The prefrontal regions of the brain, which regulate executive function-

ing and oversee critical abilities for decision making, show gradual changes
in structure and function during adolescence (Casey et al., 2000) and are
not fully developed until later in young adulthood (Steinberg, 2007). As a
result, self-regulatory and self-control skills are not yet fully developed. In
contrast, the neural network responsible for social and emotional develop-
ment matures earlier, closer to the onset of puberty, and may well drive
much of adolescent decision making (Steinberg, 2007). This imbalance
between impulsive and reflective neural systems is normal in adolescents
(Steinberg, 2007).
Throughout childhood and early adolescence, the brain undergoes syn-
aptic overproduction, in which connections between neurons proliferate in
the brain. Since this leads to more neural connections than can survive, the
brain then undergoes a selective synaptic “pruning” process in adolescence
into young adulthood, in which unused synapses are selectively eliminated.
The synapses that survive this pruning process become more efficient and
adept at transmitting information between neurons. For the prefrontal
regions of the brain, which is responsible for individuals’ ability to think,
this pruning process results in greater cognitive abilities (Casey et al., 2008;
Giedd, 2008; IOM and NRC, 2011; Johnson et al., 2009; Weinberger et
al., 2005).
At around the same time, a process of myelination occurs, whereby the
amount of white matter—the part of the brain that modulates the signals
between nerves—increases in the prefrontal cortex. In this myelination
process, nerve fibers become coated or sheathed in myelin, a white fatty
substance. Myelin accelerates the velocity at which signals travel along
nerves, making nerve-to-nerve communication faster and more efficient.
This process continues until young adulthood and results in more efficient
neural connections, which in turn results in improvements in higher-order
cognitive functioning, planning, understanding of positive and negative
consequences, and decision making.
During adolescence and through young adulthood, there is also an
increase in the number of dopamine transmitters in the brain. These re-
ceptors connect to the limbic system, which is the part of the brain most
responsible for emotions, rewards, and punishment. This increase in dopa-
mine receptors during this period results in an increased desire for rewards
and increased sensation seeking in order to feed these desires for reward
(Counotte et al., 2011).
Finally, during adolescence and into young adulthood, more and more
efficient connections develop between the prefrontal cortex and the limbic
system. With greater connectivity, there is more likelihood of self-regulation
and impulse control. During adolescence, there is less communication
among the various centers of the brain and, hence, less likelihood to con-
trol impulses associated with rewards (Steinberg, 2013).

These processes of neurodevelopment have been shown to continue

through the mid-20s, with large individual differences in the rate and
amount of brain maturation over time. As such, the portions of the brain
believed to be most responsible for decision making, impulse control, peer
susceptibility, and other aspects of psychosocial maturity are not fully
developed until young adulthood, with males developing more slowly than
females (Casey et al., 2008; Giedd, 2008; Luna et al., 2004).
Research on the brain helps explain why adolescents and young adults
are more likely to act impulsively and to make emotionally based deci-
sions. This pattern is due in part to the fact that the amygdala—a part of
the limbic system—rather than the prefrontal cortex is used in many deci-
sion tasks during adolescence and young adulthood (Smith et al., 2013;
Steinberg, 2007). Brain imaging research shows that the prefrontal cortex,
which controls self-regulation, impulse control, and sensation seeking, is
less mature and less effectively used in adolescents than in adults (Casey et
al., 2008; Luna et al., 2010; Smith et al., 2013). Of particular importance
is that the limbic and paralimbic areas of the brain (amygdala, orbito-
frontal cortex, medial prefrontal cortex, superior temporal sulcus, and
ventral striatum) are developing during adolescence. Given that the areas
of the brain particularly responsible for processing social and emotional
information and reward pathways develop earlier, it stands to reason that
adolescents are particularly focused on engaging in activities for which they
receive rewards and acceptance from their peers and others (Smith et al.,
2013). This reward seeking and focus on peer acceptance is responsible in
part for greater risk taking during adolescence. Using functional magnetic
resonance imaging (fMRI), Chein and colleagues (2011) examined activity
within the brains of adolescents (14 to 18 years old), young adults (19 to
22 years old) and adults (24 to 29 years old) to determine which parts of the
brain are more active when an individual is making simple driving decisions
that are observed by peers. The researchers found that, compared to adults,
adolescents used those areas of the brain most responsible for cognitive
control less. Furthermore, there was more activation in the reward areas of
the brain among adolescents than among adults.
As adolescents age into young adulthood, the part of the brain used
to make decisions and understand information changes, with gradual im-
provements and shifts to the brain areas more responsible for higher-level
cognitive control. Furthermore, the ability to process information and to do
so without or with limited influence from others and with little emotional
influences is not fully developed until the mid-20s (Giedd, 2008; Luna et
al., 2004).

Implications of Tobacco Use for the Neurobiology of the Adolescent and

Young Adult Brain
The developing adolescent brain is vulnerable to tobacco use not only
because of its biological immaturity but also because some of the brain
areas that are critical to the emergence of nicotine dependence may not
be fully developed until late adolescence or young adulthood. The ongo-
ing changes in both brain structure and function are likely to heighten an
adolescent’s vulnerability to tobacco use. The neurobiological stages and
changes characteristic of adolescence, as described above, may translate di-
rectly into challenges adolescents will have in competently planning and ex-
ecuting the complex array of coping skills that are needed to resist prompts
to use tobacco. Although most logical reasoning abilities are developed by
age 16 (Steinberg et al., 2009a), the fact that some psychosocial capacities
of adolescents are still immature, including delay of gratification, impulse
control, emotional regulation, and the ability to resist social influences, may
undermine the plans and efforts needed to resist tobacco use in the presence
of cues to use. Steinberg (2007) suggests that when adolescents are emotion-
ally aroused, their cognitive control mechanisms are further compromised.
Casey and Jones (2010) outlined how the imbalance in adolescents’
developing neurobiological systems makes them particularly susceptible
to the motivational properties of substances. Smoking-specific models of
adolescent smoking initiation and brain development (Lydon et al., 2014)
show that the adolescent developmental period is particularly critical with
regards to smoking initiation. Nicotine exposure also affects the adoles-
cent brain differently than the adult brain. Individuals exposed to nicotine
during adolescence are more likely to experience the symptoms of a pro-
tracted abstinence syndrome than are individuals exposed to nicotine only
in adulthood; thus, adolescents who use tobacco products are more at risk
for continuation and relapse than individuals who started to use tobacco
products in adulthood (Lydon et al., 2014).
In addition to the imbalance in the maturational stages of different
brain regions, the adolescent brain may be especially primed to be recep-
tive to the rewarding effects of nicotine. Adolescent brain development is
characterized by a dynamic combination of changes, including increased
innervations of fibers with modulatory neurotransmitters, synaptic pruning,
increased myelination of higher-order associative areas (notably the pre-
frontal cortex), and adaptations of various receptor levels (Counotte et al.,
2011). The levels of different receptor types follow a pattern of peaking in
adolescence and then declining to adult levels; thus, adolescent brains may
be especially sensitive to the effects of nicotine. Some of the receptor level
changes that occur during adolescence include those that play important
roles in modulating the circuitry of the prefrontal cortex and in mediating

nicotine reward signals (e.g., glutamate receptors, dopamine, and nicotinic

acetylcholine receptors). Adolescents also have greater brain reactivity to
rewards in general than do young adults, which may also be related to
novelty seeking such as tobacco use (Chein et al., 2011).
Most of the evidence about the vulnerability of the adolescent brain
to nicotine comes from animal studies because of the ethical challenges of
conducting this type of research in humans. Substantial evidence from these
animal studies suggests that the adolescent brain has heightened sensitivity
to the reinforcing effects of nicotine compared to the adult brain (Jamner
et al., 2003; Slotkin, 2002), as demonstrated by both conditioned place
preference paradigms and self-administration of nicotine (Belluzzi et al.,
2004; Chen et al., 2008; Shram et al., 2006; Torres et al., 2008). While
both animal and human studies indicate that adolescents experience fewer
nicotine withdrawal symptoms than adults (Counotte et al., 2011), studies
show that the reinforcing effects of nicotine are greater in adolescent rats
than in adult rats, and additives to cigarettes, such as acetaldehyde, may
also enhance the rate of the self-administration of nicotine in adolescent
but not adult rats (Belluzzi et al., 2004). Animal models also suggest that
exposure to nicotine during adolescence may increase the potential for
dependence in adulthood, as adolescent rats exposed to nicotine increase
their intravenous self-administration of nicotine when they reach adulthood
(Adriani et al., 2003). In contrast, when rats are exposed to nicotine only
after adolescence, the rewarding properties are reduced in conditioned place
preference paradigms (Adriani et al., 2006).
The c-Fos gene is a marker of neuronal activation during brain develop-
ment whose expression in response to nicotine is known to vary with age,
with discrete periods of sensitivity in adolescence. The cingulate cortex,
which is important for attention, and the retrosplenial cortex, which is ac-
tivated by emotionally salient stimuli, show increased nicotine c-Fos mRNA
in adolescence than in adulthood (Goldstein and Volkow, 2002; Jamner et
al., 2003). These brain areas are connected with the primary visual cortex,
where visual stimuli are processed initially. The visual cortex c-Fos mRNA
is activated by nicotine in adolescence but is not similarly activated in adult
brains, suggesting that even occasional tobacco use during adolescence may
prime receptivity to the visual cues in tobacco advertising (Jamner et al.,
2003).
Both the cingulate cortex and the retrosplenial cortex also influence ar-
eas of the amygdala, which are important in regulating attention, memory,
and emotional response to sensory stimuli (Jamner et al., 2003; Swanson
and Petrovich, 1998). Even low doses of nicotine in adolescence cause
increases in c-Fos mRNA in the medial extended amygdala. This pathway
also is critical to regulation of two other areas, the shell of the nucleus
acumbens and the paraventricular nucleus of the hypothalamus, which reg-

ulate pathways for endocrine and behavioral outputs (Jamner et al., 2003;
Swanson, 2000). c-Fos mRNA expression in the paraventricular nucleus is
extremely sensitive to nicotine in the adolescent brain, and only during late
adolescence (not adulthood) does nicotine-induced c-Fos expression appear
in the shell of the nucleus acumbens.
In short, there are multiple brain regions that are highly activated
during adolescence, and these regions form interconnected circuits that
are critical to attention and motivational behavior. It is worth noting that
brain development varies by sex, and these developmental differences may
provide clues to differential rates of tobacco use seen in adolescent boys and
girls. For example, in both animal and human studies, males are often more
responsive to the rewarding effects of nicotine than are females (Donny et
al., 2000; Perkins et al., 1999). There may also be sex differences in the
effects of nicotine withdrawal. In animal models, nicotine administration
in adolescents produces changes in brain circuitry, cell damage, and loss
related to learning and memory, but these effects may be greater in the
female hippocampus than in the male (Slotkin, 2002). To date, however,
it has proved difficult to determine whether sex differences in patterns of
brain development influence differences in the developmental trajectories
of tobacco use.
In sum, brain development continues beyond adolescence into young
adulthood. Individuals continue to undergo normal neurobiological
changes, including developmental transformation of the prefrontal cortex
and limbic brain regions, and myelination of the intracortical and meso-
limbic dopamine systems continues (Benes, 1989; Thompson and Nelson,
2001). These patterns reflect growing executive function control, improved
decision making, and decreases in behavioral impulsivity (Casey and Jones,
2010; Smith et al., 2013; Steinberg, 2004, 2013). The reward centers of the
brain are most activated during adolescence (Chein et al., 2011; Steinberg,
2013).
The literature implies critical findings concerning adolescent and young
adult brain development and its application to tobacco use. Most germane
to this report are the following findings:
Finding 3-4: Brain development continues until about age 25.
Finding 3-5: While the development of some cognitive abilities is

achieved by age 16, the parts of the brain most responsible for decision
making, impulse control, sensation seeking, future perspective taking,
and peer susceptibility and conformity continue to develop and change
through young adulthood.

Finding 3-6: Animal studies suggest that adolescent brains, because

of their level of development, are uniquely vulnerable to the effects of
nicotine and nicotine addiction.
TOBACCO-RELATED DECISION MAKING BY

ADOLESCENTS AND YOUNG ADULTS
Traditional models of decision making—for example, the Theory of
Reasoned Action (see Fishbein, 1979), the Theory of Planned Behavior
(Ajzen, 1985), and the Health Belief Model (Rosenstock, 1974)—describe
decision making as taking place through a deliberate, analytic process; a
process that involves many of the cognitive abilities discussed previously.
According to these theories, decisions are based on cognitive processes that
involve: (1) an assessment of the potential positive and negative outcomes
associated with the behavior in question; (2) an assessment of the likelihood
of experiencing personal harm from engaging in the behavior, including the
likelihood that each positive (benefit) and negative (risk) outcome can and
would occur; (3) consideration of one’s desire to engage in the behavior,
given the potential positive and negative consequences; (4) perceptions of
the extent to which similar others are engaging in the behavior; (5) percep-
tions of the extent to which others would accept or not accept engagement
in the behavior; and (6) intention to engage in the behavior.
The understanding that adolescent cognitive abilities are largely forged
by about age 16 while psychosocial maturation is still continuing has led to
the development of new decision-making models that include both cognitive
and noncognitive components. These dual-process models are especially
relevant to tobacco use, which involves a deliberate decision process in a
developmental context strongly affected by psychosocial influences that
adolescents are not always equipped to process.
The dual-process models include, first, the cognitive path involving the
more traditional, deliberate, reasoned, and informed aspects of the decision
process. In this path, decisions rely on cognitive skills such as weighing
risks and benefits and social norms, and these attitudes are expected to
predict intentions and ultimately behavior. This is the path sometimes used
by adolescents when making decisions that are less emotional, and it is the
path most often used by adults.
The second path, which is used more often by adolescents during emo-
tional decisions such as whether to use tobacco, involves the noncognitive
aspects of decision making, such as impulsiveness, sensation seeking, and
reward seeking. The influence of this path is rooted in the asynchrony ob-
served in the adolescent and young adult brain structure and function. This
path involves the more hypersensitive affective system, which leads to deci-
sions that are more affectively based and influenced by psychosocial fac-

tors, such as peers, lower impulse control, increased sensation seeking, and
self-regulation (Smith et al., 2013), particularly in an emotionally charged
situation such as develops when an adolescent is faced with the dilemma
of whether or not to use tobacco.
Given these two paths to decision making, it is evident that adoles-
cents and, in some cases, young adults are strongly susceptible to devel-
opmentally grounded social and emotional influences in making decisions
concerning tobacco use. Delaying the socially sanctioned opportunity for
this decision, and strengthening the social disincentives to use tobacco,
can reasonably be expected to reduce the likelihood that adolescents and
young adults will affiliate with peers who are using tobacco and reduce the
chances that they will be induced or pressured to use tobacco while their
brains continue to mature.
Finding 3-7: The developmental trajectories in adolescents and young

adults may be altered by social and environmental contextual influ-
ences. Such changes are commonly observed because of normative
developmental transitions into and out of school or work or because
of changes in living arrangements or relationships.
TOBACCO INDUSTRY TARGETING

ADOLESCENTS AND YOUNG ADULTS
Tobacco industry influence is an important environmental factor that
increases adolescents’ and young adults’ susceptibility to using tobacco use.
Tobacco companies have historically targeted children and young adults,
recognizing that they needed the “youth market” to perpetuate the sales
of their products (Teague, 1973). Since the 1998 Master Settlement Agree-
ment, tobacco companies are legally prohibited from marketing to indi-
viduals younger than 18 years of age (NAAG, 1998). But while traditional
cigarette advertisements are no longer allowed in broadcast television or
radio, tobacco companies have responded to these restrictions by increasing
their advertising and promotion at points of purchase (Feighery et al., 2001;
Henriksen, 2012) and vigorously marketing to young adults via promotions
at venues such as bars or events such as concerts (Ling and Glantz, 2002).
The aggressive marketing of tobacco products at points of purchase and
popular venues as well as the heavy exposure to images of tobacco use that
individuals receive via television and movies is troubling, as studies show
that adolescents and young adults may be particularly vulnerable to such
marketing practices (e.g., Scull et al., 2010; Ward et al., 2006). Accord-
ing to recent theories of media exposure, such as the “super peer” theory
(Brown et al., 2005), the media exerts a distinct influence on adolescents’
perceptions of what is normal, acceptable, and expected of them, and it

may actually exert more influence than either parents or peers (Ward et al.,
2006). For example, Scull and colleagues (2010) found that adolescents’ be-
liefs regarding the attractiveness of advertisements for alcohol and tobacco,
how realistic they felt the ads were, and how similar they felt they were to
individuals in the ads predicted current use and intentions to use alcohol
and tobacco over and above variables of peer and parental influence.
As described in previously secret tobacco industry documents, tobacco
companies use marketing strategies to shape consumers’ and potential
consumers’ perceptions of risk and to increase beliefs in the acceptability
of tobacco products (Anderson, 2011). For example, the tobacco industry
used terms such as “light” and “mild” to encourage tobacco use as the
awareness of the health dangers of smoking grew (Etter et al., 2003; Gilpin
et al., 2002; Kropp and Halpern-Felsher, 2004; Shiffman et al., 2001;
Tindle et al., 2006). Similarly, a review of more than 900 tobacco industry
documents revealed that menthol cigarettes were marketed as healthier than
non-menthol cigarettes; such marketing was related to adolescents’ and
young adults’ perceptions that menthol-flavored cigarettes were a healthier
alternative (Anderson, 2011; Klausner, 2011). In addition, the tobacco
industry has used aspirational visual imagery (e.g., sexy women smoking,
baseball players using smokeless tobacco) to motivate tobacco use (Cortese
et al., 2009; Mejia and Ling, 2010; Toll and Ling, 2005). The prominent
use of the Internet and social media to market new products such as elec-
tronic nicotine delivery systems (ENDS) and smokeless tobacco further
facilitates these marketing strategies and increases the tobacco companies’
reach, relevance, and opportunities to interact with young consumers,
which results in perceptions of reduced risk, greater benefits, and greater
social acceptability of marketed tobacco products. These messages are espe-
cially effective when the marketing messages appear to come from peers and
other tobacco consumers rather than the manufacturer (Sepe et al., 2002).
Numerous longitudinal studies have found a significant relationship
between exposure to cigarette marketing and subsequent smoking behavior.
Hanewinkel and colleagues (2011), for example, found that adolescents
with high levels of exposure to cigarette advertising were significantly more
likely to smoke than adolescents who had been exposed to low levels of
cigarette advertising, while exposure to other types of advertising did not
affect smoking initiation rates (Hanewinkel et al., 2011). Anti-tobacco
counter-marketing campaigns such as the truth® campaign have also been
shown to be successful at reducing tobacco initiation and use among ado-
lescents and young adults (Davis et al., 2009; Emery et al., 2012; Farrelly
et al., 2005, 2009; Richardson et al., 2010; Sly et al., 2001).
Point-of-sale marketing is also associated with adolescent initiation
of smoking (Slater et al., 2007). In a longitudinal study showing that
adolescents who frequently visit liquor stores, convenience stores, and

markets with high concentrations of point-of-sale advertising for cigarettes

are significantly more likely to initiate smoking (Henriksen et al., 2010).
Among young adults, it has been shown that both smokers and nonsmokers
ages 18 to 30 are twice as likely to attend bars and clubs than their older
counterparts, and they are also more attracted and susceptible to tobacco
advertising (Biener and Albers, 2004). Ling and Glantz have also shown
that marketing targeted at young adults has the consequence of promoting
smoking in older teens as well (Ling and Glantz, 2002).
Influence of Seeing Smoking in the Movies

Overall, between 1950 and 1990 there was a decrease in depictions of
smoking in the movies, but this was followed by a rapid increase so that by
2002 depictions of smoking were comparable in scale to what had existed
in 1950 (Charlesworth and Glantz, 2005). This trend has continued, and
between 2011 and 2012 there was a 45 percent increase in the number of
tobacco incidents displayed per movie (Glantz et al., 2013). Exposure to
smoking images in movies as well as in other sources such as newspapers
and television has been found to be associated with positive assessments
related to the social acceptability of smoking, smoking as a means of stress
and emotional control (Watson et al., 2003), and assessments of smoking
being “sexy” and “stylish” (McCool et al., 2004). Experimental studies
and cross-sectional surveys have found a relationship between exposure
to smoking images in the movies and smoking initiation, and longitudinal
studies have found that adolescents with higher exposure to smoking in the
movies were more likely to initiate smoking than peers who reported low
levels of exposure (Dal Cin et al., 2012).
The tobacco industry’s efforts to manipulate tobacco-related percep-
tions and acceptability are more concerning as new tobacco products come
to market (Ganz et al., 2015; Grana and Ling, 2014; Kornfield et al., 2015;
Pokhrel et al., 2015). The tobacco industry has continued to market tobacco
products aggressively. The impact of this marketing will depend on Food and
Drug Administration regulation of marketing and promotional materials.
IMPLICATIONS
It is clear that the juxtaposition of numerous risk factors during the
adolescent and young adult years is likely to increase the probability that
first trials of tobacco use will turn into persistent use. These factors include
the sequence of neurodevelopment in the adolescent years, the unique sen-
sitivity of the adolescent brain to the rewarding properties of nicotine, the
early development of symptoms of dependence in an adolescent’s smoking
experience (well before reaching the 100-cigarette lifetime threshold), and

the difficulties that adolescents have in stopping smoking. Delaying the

onset of any tobacco use beyond adolescence will likely decrease the prob-
ability that early trials of tobacco will be experienced as rewarding and to
increase an individual’s ability to discontinue tobacco use after initial trials.
REFERENCES
Adriani, W., S. Spijker, V. Deroche-Gamonet, G. Laviola, M. Le Moal, A. B. Smit, and P. V.
Piazza. 2003. Evidence for enhanced neurobehavioral vulnerability to nicotine during
periadolescence in rats. Journal of Neuroscience 23(11):4712–4716.
Adriani, W., V. Deroche-Gamonet, M. Le Moal, G. Laviola, and P. V. Piazza. 2006. Preexpo-
sure during or following adolescence differently affects nicotine-rewarding properties in
adult rats. Psychopharmacology 184(3–4):382–390.
Ajzen, I. 1985. From intentions to actions: A theory of planned behavior. Berlin: Springer.
Albert, D., and L. Steinberg. 2011. Judgment and decision making in adolescence. Journal of
Research on Adolescence 21(1):211–224.
Albert, D., J. Chein, and L. Steinberg. 2013. The teenage brain: Peer influences on adolescent
decision making. Current Directions in Psychological Science 22(2):114–120.
Alexander, C., M. Piazza, D. Mekos, and T. Valente. 2001. Peers, schools, and adolescent
cigarette smoking. Journal of Adolescent Health 29(1):22–30.
Anderson, S. J. 2011. Marketing of menthol cigarettes and consumer perceptions: A review of
tobacco industry documents. Tobacco Control 20(Suppl 2):ii20–ii28.
Arnett, J. J. 2000. Emerging adulthood—A theory of development from the late teens through
the twenties. American Psychologist 55(5):469–480.
———. 2004. Emerging adulthood: The winding road from the late teens through the twen-
ties. New York: Oxford University Press.
Bandura, A. 2001. Social cognitive theory: An agentic perspective. Annual Review of Psychol-
ogy 52:1–26.
Bansal-Travers, M., D. Hammond, P. Smith, and K. M. Cummings. 2011. The impact of ciga-
rette pack design, descriptors, and warning labels on risk perception in the U.S. American
Belluzzi, J. D., R. Wang, and F. M. Leslie. 2004. Acetaldehyde enhances acquisition of nicotine
self-administration in adolescent rats. Neuropsychopharmacology 30(4):705–712.
Benes, F. M. 1989. Myelination of cortical-hippocampal relays during late adolescence. Schizo-
phrenia Bulletin 15(4):585–593.
Biener, L., and A. B. Albers. 2004. Young adults: Vulnerable new targets of tobacco marketing.
American Journal of Public Health 94(2):326–330.
Brown, J. D., C. T. Halpern, and K. L. L’Engle. 2005. Mass media as a sexual super peer for
early maturing girls. Journal of Adolescent Health 36(5):420–427.
Callery, W. E., D. Hammond, R. J. O’Connor, and G. T. Fong. 2011. The appeal of smokeless
tobacco products among young Canadian smokers: The impact of pictorial health warn-
ings and relative risk messages. Nicotine & Tobacco Research 13(5):373–383.
Cance, J. D., S. T. Ennett, A. A. Morgan-Lopez, V. A. Foshee, and A. E. Talley. 2013. Perceived
pubertal timing and recent substance use among adolescents: A longitudinal perspective.
Addiction 108(10):1845–1854.
Casey, B., and R. M. Jones. 2010. Neurobiology of the adolescent brain and behavior: Im-
plications for substance use disorders. Journal of the American Academy of Child and
Adolescent Psychiatry 49(12):1189–1201.
Casey, B., J. N. Giedd, and K. M. Thomas. 2000. Structural and functional brain develop-
ment and its relation to cognitive development. Biological Psychology 54(1-3):241–257.

Casey, B., R. M. Jones, and T. A. Hare. 2008. The adolescent brain. Annals of the New York
Academy of Sciences 1124:111–126.
Charlesworth, A., and S. A. Glantz. 2005. Smoking in the movies increases adolescent smok-
ing: A review. Pediatrics 116(6):1516–1528.
Chassin, L., C. C. Presson, S. C. Pitts, and S. J. Sherman. 2000. The natural history of cigarette
smoking from adolescence to adulthood in a Midwestern community sample: Multiple
trajectories and their psychosocial correlates. Health Psychology 19(3):223–231.
Chein, J., D. Albert, L. O’Brien, K. Uckert, and L. Steinberg. 2011. Peers increase adolescent
risk taking by enhancing activity in the brain’s reward circuitry. Developmental Science
14(2):F1–F10.
Chen, H., Y. Fu, and B. M. Sharp. 2008. Chronic nicotine self-administration augments hy-
pothalamic–pituitary–adrenal responses to mild acute stress. Neuropsychopharmacology
33(4):721–730.
Choi, K., L. Fabian, N. Mottey, A. Corbett, and J. Forster. 2012. Young adults’ favorable per-
ceptions of snus, dissolvable tobacco products, and electronic cigarettes: Findings from a
focus group study. American Journal of Public Health 102(11):2088–2093.
Cialdini, R. B., R. R. Reno, and C. A. Kallgren. 1990. A focus theory of normative conduct:
Recycling the concept of norms to reduce littering in public places. Journal of Personality
and Social Psychology 58(6):1015–1026.
Cortese, D. K., M. J. Lewis, and P. M. Ling. 2009. Tobacco industry lifestyle magazines tar-
geted to young adults. Journal of Adolescent Health 45(3):268–280.
Counotte, D. S., A. B. Smit, T. Pattij, and S. Spijker. 2011. Development of the motivational
system during adolescence, and its sensitivity to disruption by nicotine. Developmental
Cognitive Neuroscience 1(4):430–443.
Crockett, L. J., and A. C. Petersen. 1987. Pubertal status and psychosocial development: Find-
ings from the Early Adolescence Study. In R. M. Lerner and T. T. Foch, eds., Biological–
Psychosocial Interactions in Early Adolescence (pp. 173–188). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Crone, E. A., and R. E. Dahl. 2012. Understanding adolescence as a period of social–affective
engagement and goal flexibility. Nature Reviews Neuroscience 13(9):636–650.
Dal Cin, S., M. Stoolmiller, and J. D. Sargent. 2012. When movies matter: Exposure to
smoking in movies and changes in smoking behavior. Journal of Health Communication
17(1):76–89.
Davis, K. C., M. C. Farrelly, P. Messeri, and J. Duke. 2009. The impact of national smok-
ing prevention campaigns on tobacco-related beliefs, intentions to smoke and smoking
initiation: Results from a longitudinal survey of youth in the united states. International
Journal of Environmental Research and Public Health 6(2):722–740.
DiFranza, J. R., J. A. Savageau, K. Fletcher, L. Pbert, J. O’Loughlin, A. D. McNeill, J. K.
Ockene, K. Friedman, J. Hazelton, and C. Wood. 2007. Susceptibility to nicotine de-
pendence: The Development and Assessment of Nicotine Dependence in Youth 2 study.
Pediatrics 120(4):e974–e983.
Donny, E., A. Caggiula, P. Rowell, M. Gharib, V. Maldovan, S. Booth, M. Mielke, A. Hoffman,
and S. McCallum. 2000. Nicotine self-administration in rats: Estrous cycle effects, sex
differences and nicotinic receptor binding. Psychopharmacology 151(4):392–405.
Duncan, G. J., J. Boisjoly, M. Kremer, D. M. Levy, and J. Eccles. 2005. Peer effects in drug use
and sex among college students. Journal of Abnormal Child Psychology 33(3):375–385.
Elliot, A. J. 1999. Approach and avoidance motivation and achievement goals. Educational
Psychologist 34(3):169–189.

Emery, S., Y. Kim, Y. K. Choi, G. Szczypka, M. Wakefield, and F. J. Chaloupka. 2012. The
effects of smoking-related television advertising on smoking and intentions to quit
among adults in the United States: 1999–2007. American Journal of Public Health
102(4):751–757.
Etcheverry, P. E., and C. R. Agnew. 2008. Romantic partner and friend influences on young
adult cigarette smoking: Comparing close others’ smoking and injunctive norms over
time. Psychology of Addictive Behaviors 22(3):313–325.
Etter, J. F., L. T. Kozlowski, and T. V. Perneger. 2003. What smokers believe about light and
ultralight cigarettes. Preventive Medicine 36(1):92–98.
Farrelly, M. C., K. C. Davis, M. L. Haviland, P. Messeri, and C. G. Healton. 2005. Evidence
of a dose–response relationship between “truth” antismoking ads and youth smoking
prevalence. American Journal of Public Health 95(3):425–431.
Farrelly, M. C., J. Nonnemaker, K. C. Davis, and A. Hussin. 2009. The influence of the na-
tional truth campaign on smoking initiation. American Journal of Preventive Medicine
36(5):379–384.
Feighery, E. C., K. M. Ribisl, N. Schleicher, R. E. Lee, and S. Halvorson. 2001. Cigarette
advertising and promotional strategies in retail outlets: Results of a statewide survey in
California. Tobacco Control 10(2):184–188.
Fischhoff, B., W. Bruine de Bruin, A. M. Parker, S. G. Millstein, and B. L. Halpern-Felsher.
2010. Adolescents’ perceived risk of dying. Journal of Adolescent Health 46(3):265–269.
Fishbein, M. 1979. A theory of reasoned action: Some applications and implications. Nebraska
Symposium on Motivation 27:65–116.
Fry, R. 2014. A rising share of young adults live in their parents’ home. Pew Research Centers
Social Demographic Trends Project. http://www.pewsocialtrends.org/2013/08/01/a-rising-
share-of-young-adults-live-in-their-parents-home (accessed February 7, 2015).
Ganz, O., J. Cantrell, J. Moon-Howard, A. Aidala, T. R. Kirchner, and D. Vallone. 2015.
Electronic cigarette advertising at the point-of-sale: A gap in tobacco control research.
Tobacco Control 24:e110–e112.
Gardner, M., and L. Steinberg. 2005. Peer influence on risk taking, risk preference, and risky
decision making in adolescence and adulthood: An experimental study. Developmental
Psychology 41(4):625–635.
Giedd, J. N. 2008. The teen brain: Insights from neuroimaging. Journal of Adolescent Health
42(4):335–343.
Gilpin, E. A., S. Emery, M. M. White, and J. P. Pierce. 2002. Does tobacco industry market-
ing of “light” cigarettes give smokers a rationale for postponing quitting? Nicotine &
Tobacco Research 4(Suppl 2):S147–S155.
Glantz, S. A., A. Iaccopucci, K. Titus, and J. R. Polansky. 2013. Smoking in top-grossing
U.S. movies, 2012. Preventing Chronic Disease 9:120170. http://www.cdc.gov/pcd/
issues/2012/12_0170.htm (accessed February 7, 2015).
Gogtay, N., J. N. Giedd, L. Lusk, K. M. Hayashi, D. Greenstein, A. C. Vaituzis, T. F. Nugent,
D. H. Herman, L. S. Clasen, and A. W. Toga. 2004. Dynamic mapping of human corti-
cal development during childhood through early adulthood. Proceedings of the National
Academy of Sciences of the United States of America 101(21):8174–8179.
Goldstein, R. Z., and N. D. Volkow. 2002. Drug addiction and its underlying neurobiological
basis: Neuroimaging evidence for the involvement of the frontal cortex. American Journal
of Psychiatry 159(10):1642–1652.
Grana, R. A., and P. M. Ling. 2014. “Smoking revolution”: A content analysis of electronic
cigarette retail websites. American Journal of Preventive Medicine 46(4):395–403.
Halpern-Felsher, B. L., and E. Cauffman. 2001. Costs and benefits of a decision: Decision-
making competence in adolescents and adults. Journal of Applied Developmental Psy-
chology 22(3):257–273.

Halpern-Felsher, B. L., M. Biehl, R. Y. Kropp, and M. L. Rubinstein. 2004. Perceived risks and
benefits of smoking: Differences among adolescents with different smoking experiences
and intentions. Preventive Medicine 39(3):559–567.
Hanewinkel, R., B. Isensee, J. D. Sargent, and M. Morgenstern. 2011. Cigarette advertising
and teen smoking initiation. Pediatrics 127(2):e271–e278.
Henriksen, L. 2012. Comprehensive tobacco marketing restrictions: Promotion, packaging,
price and place. Tobacco Control 21(2):147–153.
Henriksen, L., N. C. Schleicher, E. C. Feighery, and S. P. Fortmann. 2010. A longitudi-
nal study of exposure to retail cigarette advertising and smoking initiation. Pediatrics
126(2):232–238.
Health.
———. 2011. The science of adolescent risk-taking: Workshop report. Washington, DC: The
———. 2014. Investing in the health and well-being of young adults. Edited by R. J. Bonnie,
C. Stroud, and H. Breiner. Washington, DC: The National Academies Press.
5(Suppl 1):S71–S87.
Johnson, S. B., R. W. Blum, and J. N. Giedd. 2009. Adolescent maturity and the brain: The
promise and pitfalls of neuroscience research in adolescent health policy. Journal of
Adolescent Health 45(3):216–221.
Kallgren, C. A., R. R. Reno, and R. B. Cialdini. 2000. A focus theory of normative conduct:
When norms do and do not affect behavior. Personality and Social Psychology Bulletin
26(8):1002–1012.
Klausner, K. 2011. Menthol cigarettes and smoking initiation: A tobacco industry perspective.
Tobacco Control 20(Suppl 2):ii12–ii19.
Kornfield, R., J. Huang, L. Vera, and S. L. Emery. 2015. Rapidly increasing promotional
expenditures for e-cigarettes. Tobacco Control 24(2):110–111.
Kosten, T. A., S. A. Ball, and B. J. Rounsaville. 1994. A sibling study of sensation seeking and
opiate addiction. Journal of Nervous and Mental Disease 182(5):284–289.
Kozlowski, L. T., M. E. Goldberg, B. A. Yost, E. L. White, C. T. Sweeney, and J. L. Pillitteri.
1998. Smokers’ misperceptions of light and ultra-light cigarettes may keep them smoking.
American Journal of Preventive Medicine 15(1):9–16.
Kropp, R. Y., and B. L. Halpern-Felsher. 2004. Adolescents’ beliefs about the risks involved
in smoking “light” cigarettes. Pediatrics 114(4):e445–e451.
Kuther, T. L. 2003. Medical decision-making and minors: Issues of consent and assent. Ado-
lescence 38(150):343–358.
Lewis, C. C. 1981. How adolescents approach decisions: Changes over grades seven to twelve
and policy implications. Child Development 52(2):538–544.
92(6):908–916.

Luna, B., K. E. Garver, T. A. Urban, N. A. Lazar, and J. A. Sweeney. 2004. Maturation of cog-
nitive processes from late childhood to adulthood. Child Development 75(5):1357–1372.
Luna, B., A. Padmanabhan, and K. O’Hearn. 2010. What has fMRI told us about the devel-
opment of cognitive control through adolescence? Brain and Cognition 72(1):101–113.
Lydon, D. M., S. J. Wilson, A. Child, and C. F. Geier. 2014. Adolescent brain maturation
and smoking: What we know and where we’re headed. Neuroscience and Biobehavioral
Reviews 45:323–342.
Martin, C., T. Logan, C. Leukefeld, R. Milich, H. Omar, and R. Clayton. 2001. Adolescent
and young adult substance use: Association with sensation seeking, self esteem and retro-
spective report of early pubertal onset. A preliminary examination. International Journal
of Adolescent Medicine and Health 13(3):211–219.
McCool, J., L. Cameron, and K. Petrie. 2004. Stereotyping the smoker: Adolescents’ appraisals
of smokers in film. Tobacco Control 13(3):308–314.
Mejia, A. B., and P. M. Ling. 2010. Tobacco industry consumer research on smokeless tobacco
users and product development. American Journal of Public Health 100(1):78–87.
Mendle, J., and J. Ferrero. 2012. Detrimental psychological outcomes associated with pubertal
timing in adolescent boys. Developmental Review 32(1):49–66.
Mendle, J., E. Turkheimer, and R. E. Emery. 2007. Detrimental psychological outcomes associ-
ated with early pubertal timing in adolescent girls. Developmental Review 27(2):151–171.
Michels, T. M., R. Y. Kropp, S. L. Eyre, and B. L. Halpern-Felsher. 2005. Initiating sexual
experiences: How do young adolescents make decisions regarding early sexual activity?
Journal of Research on Adolescence 15(4):583–607.
Millstein, S. G., and B. L. Halpern-Felsher. 2002. Judgments about risk and perceived in-
vulnerability in adolescents and young adults. Journal of Research on Adolescence
12(4):399–422.
Morrell, H. E., A. V. Song, and B. L. Halpern-Felsher. 2010. Predicting adolescent percep-
tions of the risks and benefits of cigarette smoking: A longitudinal investigation. Health
Nelson, L. J., and C. M. Barry. 2005. Distinguishing features of emerging adulthood: The role
of self-classification as an adult. Journal of Adolescent Research 20(2):242–262.
Nyman, A., T. Taylor, and L. Biener. 2002. Trends in cigar smoking and perceptions of health
risks among Massachusetts adults. Tobacco Control 11(Suppl 2):ii25–ii28.
Øverland, S., J. Hetland, and L. E. Aarø. 2008. Relative harm of snus and cigarettes: What
do Norwegian adolescents say? Tobacco Control 17(6):422–425.
Perkins, K. A., E. Donny, and A. R. Caggiula. 1999. Sex differences in nicotine effects and self-
administration: Review of human and animal evidence. Nicotine & Tobacco Research
1(4):301–315.
Pokhrel, P., P. Fagan, L. Kehl, and T. A. Herzog. 2015. Receptivity to e-cigarette mar-
keting, harm perceptions, and e-cigarette use. American Journal of Health Behavior
39(1):121–131.
Richardson, A. K., M. Green, H. Xiao, N. Sokol, and D. Vallone. 2010. Evidence for truth®:
The young adult response to a youth-focused anti-smoking media campaign. American
Rimal, R. N., and K. Real. 2005. How behaviors are influenced by perceived norms: A test
of the theory of normative social behavior. Communication Research 32(3):389–414.
Rosenstock, I. M. 1974. Historical origins of the health belief model. Health Education and
Behavior 2(4):328–335.

Sartor, C. E., C. N. Lessov-Schlaggar, J. F. Scherrer, K. K. Bucholz, P. A. Madden, M. L.

Pergadia, J. D. Grant, T. Jacob, and H. Xian. 2010. Initial response to cigarettes predicts
rate of progression to regular smoking: Findings from an offspring-of-twins design. Ad-
dictive Behaviors 35(8):771–778.
Scull, T. M., J. B. Kupersmidt, A. E. Parker, K. C. Elmore, and J. W. Benson. 2010. Adoles-
cents’ media-related cognitions and substance use in the context of parental and peer
influences. Journal of Youth and Adolescence 39(9):981–998.
Sepe, E., P. M. Ling, and S. A. Glantz. 2002. Smooth moves: Bar and nightclub tobacco pro-
motions that target young adults. American Journal of Public Health 92(3):414–419.
Settersten, R., and B. E. Ray. 2010. Not quite adults. New York: Bantam Books.
Shiffman, S., J. L. Pillitteri, S. L. Burton, J. M. Rohay, and J. G. Gitchell. 2001. Effect of
health messages about “light” and “ultra light” cigarettes on beliefs and quitting intent.
Tobacco Control 10(Suppl 1):i24–i32.
Shram, M. J., D. Funk, Z. Li, and A. D. Lê. 2006. Periadolescent and adult rats respond dif-
ferently in tests measuring the rewarding and aversive effects of nicotine. Psychophar-
macology 186(2):201–208.
Siegel, J. M., A. K. Yancey, C. S. Aneshensel, and R. Schuler. 1999. Body image, perceived pu-
bertal timing, and adolescent mental health. Journal of Adolescent Health 25(2):155–165.
Simons-Morton, B., D. L. Haynie, A. D. Crump, P. Eitel, and K. E. Saylor. 2001. Peer and
parent influences on smoking and drinking among early adolescents. Health Education
and Behavior 28(1):95–107.
Slater, S. J., F. J. Chaloupka, M. Wakefield, L. D. Johnston, and P. M. O’Malley. 2007. The
impact of retail cigarette marketing practices on youth smoking uptake. Archives of
Pediatrics and Adolescent Medicine 161(5):440–445.
Slotkin, T. A. 2002. Nicotine and the adolescent brain: Insights from an animal model. Neu-
rotoxicology and Teratology 24(3):369–384.
Sly, D. F., R. S. Hopkins, E. Trapido, and S. Ray. 2001. Influence of a counteradvertising media
campaign on initiation of smoking: The Florida “truth” campaign. American Journal of
Smith, A. R., J. Chein, and L. Steinberg. 2013. Impact of socio-emotional context, brain de-
velopment, and pubertal maturation on adolescent risk-taking. Hormones and Behavior
64(2):323–332.
Soldz, S., and X. Cui. 2002. Pathways through adolescent smoking: A 7-year longitudinal
grouping analysis. Health Psychology 21(5):495–504.
Song, A. V., S. A. Glantz, and B. L. Halpern-Felsher. 2009a. Perceptions of second-hand
smoke risks predict future adolescent smoking initiation. Journal of Adolescent Health
45(6):618–625.
Song, A. V., H. E. Morrell, J. L. Cornell, M. E. Ramos, M. Biehl, R. Y. Kropp, and B. L.
Halpern-Felsher. 2009b. Perceptions of smoking-related risks and benefits as predictors
of adolescent smoking initiation. American Journal of Public Health 99(3):487–492.
Sowell, E. R., P. M. Thompson, C. J. Holmes, T. L. Jernigan, and A. W. Toga. 1999. In vivo
evidence for post-adolescent brain maturation in frontal and striatal regions. Nature
Neuroscience 2(10):859–861.
Steinberg, L. 2004. Risk taking in adolescence: What changes, and why? Annals of the New
York Academy of Sciences 1021:51–58.
———. 2007. Risk taking in adolescence: New perspectives from brain and behavioral science.
———. 2008. A social neuroscience perspective on adolescent risk-taking. Developmental
Review 28(1):78–106.
———. 2013. The influence of neuroscience on U.S. Supreme Court decisions about adoles-
cents’ criminal culpability. Nature Reviews Neuroscience 14(7):513–518.

Steinberg, L., and K. C. Monahan. 2007. Age differences in resistance to peer influence. De-
velopmental Psychology 43(6):1531–1543.
Steinberg, L., and A. S. Morris. 2001. Adolescent development. Annual Review of Psychol-
ogy 52:83–110.
Steinberg, L., D. Albert, E. Cauffman, M. Banich, S. Graham, and J. Woolard. 2008. Age
differences in sensation seeking and impulsivity as indexed by behavior and self-report:
Evidence for a dual systems model. Developmental Psychology 44(6):1764–1778.
Steinberg, L., E. Cauffman, J. Woolard, S. Graham, and M. Banich. 2009a. Are adolescents
less mature than adults?: Minors’ access to abortion, the juvenile death penalty, and the
alleged APA “flip-flop.” American Psychologist 64(7):583–594.
Steinberg, L., S. Graham, L. O’Brien, J. Woolard, E. Cauffman, and M. Banich. 2009b. Age
differences in future orientation and delay discounting. Child Development 80(1):28–44.
Susman, E. J., R. M. Houts, L. Steinberg, J. Belsky, E. Cauffman, G. Dehart, S. L. Friedman,
G. I. Roisman, B. L. Halpern-Felsher, and Eunice Kennedy Shriver NICHD Early Child
Care Research Network. 2010. Longitudinal development of secondary sexual charac-
teristics in girls and boys between ages 9½ and 15½ years. Archives of Pediatrics and
Adolescent Medicine 164(2):166–173.
Swanson, L. W. 2000. Cerebral hemisphere regulation of motivated behavior. Brain Research
886(1–2):113–164.
Swanson, L. W., and G. D. Petrovich. 1998. What is the amygdala? Trends in Neurosciences
21(8):323–331.
Teague, C. 1973. Research planning memorandum on some thoughts about new brands of
cigarettes for the youth market. http://legacy.library.ucsf.edu/tid/iiu91f00 (accessed Janu-
ary 28, 2015).
Teichman, M., Z. Barnea, and G. Ravav. 1989. Personality and substance use among adoles-
cents: A longitudinal study. British Journal of Addiction 84(2):181–190.
Thompson, R. A., and C. A. Nelson. 2001. Developmental science and the media: Early brain
development. American Psychologist 56(1):5–15.
Tindle, H. A., N. A. Rigotti, R. B. Davis, E. M. Barbeau, I. Kawachi, and S. Shiffman. 2006.
Cessation among smokers of “light” cigarettes: Results from the 2000 National Health
Interview Survey. American Journal of Public Health 96(8):1498–1504.
Toll, B., and P. Ling. 2005. The Virginia Slims identity crisis: An inside look at tobacco indus-
try marketing to women. Tobacco Control 14(3):172–180.
Torres, O. V., H. A. Tejeda, L. A. Natividad, and L. E. O’Dell. 2008. Enhanced vulnerability
to the rewarding effects of nicotine during the adolescent period of development. Phar-
macology, Biochemistry and Behavior 90(4):658–663.
Vitulano, M. L., P. J. Fite, and J. L. Rathert. 2010. Delinquent peer influence on childhood
delinquency: The moderating effect of impulsivity. Journal of Psychopathology and Be-
havioral Assessment 32(3):315–322.
Ward, L. M., K. M. Day, and M. Epstein. 2006. Uncommonly good: Exploring how mass
media may be a positive influence on young women’s sexual health and development.
New Directions for Child and Adolescent Development 2006(112):57–70.
Watson, N., J. Clarkson, R. Donovan, and B. Giles-Corti. 2003. Filthy or fashionable? Young
people’s perceptions of smoking in the media. Health Education Research 18(5):554–567.
Weinberger, D. R., B. Elvevåg, and J. N. Giedd. 2005. The adolescent brain: A work in prog-
ress. Washington, DC: The National Campaign to Prevent Teen Pregnancy.
Zimring, F. E. 2000. American youth violence. New York: Oxford University Press.
Zuckerman, M. 1979. Sensation seeking: Beyond the optimal level of arousal. Hillsdale, NJ:
Lawrence Erlbaum Associates.
———. 2007. Sensation seeking and risk. Washington, DC: American Psychological
Association.


The Effects of Tobacco Use on Health
T
he scope of the burden of disease and death that cigarette smoking
imposes on the public’s health is extensive. Cigarette smoking is the
major focus of this chapter because it is the central public health
problem, but the topics of secondhand smoke exposure, smoking of other
combustible tobacco products, smokeless tobacco, and electronic nicotine
delivery systems (ENDS) are also considered. The magnitude of the public
health threat posed by cigarette smoking stems from two factors: (1) the
prevalence of cigarette smoking is so high, and (2) smoking causes so many
deleterious health effects. A policy change that reduces the prevalence of
cigarette smoking will result in a commensurate reduction in the population
burden of disease and death caused by cigarette smoking. The associations
between cigarette smoking and the adverse health effects caused by smok-
ing are dose-dependent (HHS, 2014). Thus, a public health benefit would
be realized if a policy change led to reduced exposure to cigarette smoke
via means other than reducing the prevalence of smoking. For example,
additional reduction in the population burden of smoking-caused disease
and death will be generated if the policy also results in delayed initiation of
cigarette smoking. The population health benefit from delayed initiation, al-
though potentially large, will be less than the benefit from a commensurate
reduction in smoking prevalence because delayed initiation is associated
with reduced exposure to cigarette smoking rather than with the complete
prevention of the exposure. A decrease in the prevalence of cigarette smok-
ing will have additional downstream benefits by reducing the potential for
nonsmokers to be exposed to secondhand tobacco smoke.
91

TIME HORIZON FOR THE HEALTH EFFECTS

OF CIGARETTE SMOKING
Cigarette smoking causes chronic diseases that appear at older ages,
such as lung cancer, as well as adverse health effects that occur in the short
run. The immediate and short-term adverse health effects of cigarette smok-
ing are less likely to be directly fatal than the long-term health effects. Nev-
ertheless, they are important public health indicators because they lead to
suboptimal health status throughout the life course in smokers and because
many of the short-term physiologic effects mechanistically contribute to the
etiology of smoking-caused diseases that usually do not become clinically
apparent until later adulthood.
The short-term adverse health effects caused by cigarette smoking can
be observed in smokers immediately or soon after they begin smoking. The
health effects of cigarette smoking thus begin at or near the age of initia-
tion of cigarette smoking, which is usually in adolescence. To highlight the
immediacy of the adverse impact of smoking on health, this report uses a
life-course perspective by considering health effects of smoking according
to the various stages of life, which include childhood, adolescence, and
young adulthood as well as middle and late adulthood, when most of the
chronic disease burden imposed by smoking occurs. A particularly vulner-
able time during the life course is pregnancy (for both mother and fetus)
and the months following birth (for the infant); for this reason, this stage
of life is considered separately. In this report, the term “immediate health
effects” refers to effects that occur within days of cigarette smoking, while
“long-term health effects” refers to the clinical morbidity and mortality that
occur primarily in middle and late adulthood, and the term “intermediate
health effects” is used to refer broadly to the health outcomes that occur
between the immediate and long-term health effects.
SPECTRUM OF HEALTH EFFECTS

Cigarette smoke contains more than 7,000 chemicals (HHS, 2010).
Inhaling cigarette smoke exposes the cigarette smoker to these numerous
toxins, which include the various tobacco constituents and the products of
pyrolysis. As summarized below, exposure to this complex chemical mix-
ture causes immediate adverse physiologic effects shortly after the exposure
occurs (HHS, 2010).
The ultimate harm caused by exposure to the toxic agents in cigarette
smoke is determined in large part by the extent of the exposure, and most
adult cigarette smokers tend to smoke many cigarettes per day for decades
(HHS, 2014). This repeated inhalation of the complex mixture of cigarette
smoke toxicants at high daily doses, often sustained over the course of

THE EFFECTS OF TOBACCO USE ON HEALTH 93
TABLE 4-1 Immediate Adverse Health Outcomes Causally Associated

with Cigarette Smoking Based on Surgeon General’s Reports
Stage of Life
Childhood/ Young Middle Older
Health Outcome Adolescence Adulthood Adulthood Adulthood
Oxidative Stress ü ü ü ü
Depletion of ü ü ü ü
Antioxidant
Micronutrients
Increased ü ü ü ü
Inflammation
Compromised ü ü ü ü
Immune Status
Altered Lipid ü ü ü ü
Metabolism
Lower Self-Rated ü ü ü ü
Health Status
Respiratory ü ü ü ü
Symptoms (coughing,
phlegm, wheezing,
dyspnea)
Nicotine Addiction ü ü ü ü
NOTE: The health outcomes are organized in Tables 4-1 through 4-3 according to whether
they are immediate, intermediate, or long term and by the stages of life affected.
many years, causes a broad spectrum of short-term and long-term health

effects that affect most major organ systems (see Tables 4-1 through 4-3).
In the short run, cigarette smoking causes the smoker to have overall di-
minished health status as measured by a diverse array of indices, including
biomarkers of physiologic disadvantage, lower self-reported health, sus-
ceptibility to acute illnesses and respiratory symptoms, and absence from
school and work. Among the long-term health effects are smoking-caused
diseases that are the major causes of death in middle- and upper-income
nations: coronary heart disease, cancer, and chronic obstructive pulmonary
disease, or COPD (HHS, 2014).
The net result of the broad spectrum of short-term and long-term
deleterious health effects caused by cigarette smoking and the substantial
prevalence of smoking is that cigarette smoking is the single most important
cause of preventable disease and premature mortality in the United States

TABLE 4-2 Intermediate Adverse Health Outcomes Causally Associated

Stage of Life
Increased Absence ü ü ü
from Schoola/Work
Increased Use of ü ü ü
Medical Services
Subclinical ü ü ü
Atherosclerosis
Impaired Lung
Development/
Function
Impaired lung ü
growth
Accelerated lung ü ü ü
function decline
Increased Risk of ü ü ü
Lung Infections
(tuberculosis,
pneumonia)
Diabetes ü ü ü
Periodontitis ü ü ü
Exacerbation of ü ü ü
Asthma
Subclinical Organ ü ü ü
Injury
Adverse Surgical ü ü ü
Outcomes
aHealth outcome not included in the 2014 Surgeon General’s report.
and in many other high-income nations (Thun et al., 2012). For example,
in the United States cigarette smoking is estimated to account for at least
480,000 deaths per year (HHS, 2014). The magnitude of this burden is a
direct function of two key facts: (1) cigarette smoking causes an incredibly
broad spectrum of short-term and long-term deleterious health effects, and
(2) a large proportion of the population is exposed (i.e., the prevalence of
smoking is very high).

TABLE 4-3 Long-Term Adverse Health Outcomes Causally Associated

Stage of Life
Cancer (colorectal, ü
liver, lung, bladder,
cervical, esophageal,
kidney, laryngeal,
pancreatic, gastric,
oral, and pharynx;
acute myeloid
leukemia)
Precancerous ü
Lesions (colorectal
adenomatous polyps)
Cardiovascular ü
Disease (coronary
heart disease, stroke,
abdominal aortic
aneurysm)
Respiratory Diseases ü
(COPD)
Eye Disease (age- ü
related macular
degeneration, nuclear
cataracts)
Rheumatoid Arthritis ü
Reduced Effectiveness ü
of Tumor Necrosis
Factor-Alpha
Inhibitors
Bone Health (hip ü
fractures, low
bone density in
postmenopausal
women)
In assessing the potential public health impact of enacting a new to-

bacco policy such as raising the minimum age of legal access to tobacco
products (MLA), it is worth keeping in mind that this lengthy catalogue of
well-established consequences of cigarette smoking will continue to expand
as scientific knowledge advances and more definitive evidence is generated

concerning additional health outcomes. Thus, the characterization of the

potential impact of a policy change that reduces exposure to cigarette
smoke is a conservative estimate of the true public health impact. For
example, in addition to the many adverse health outcomes established as
causally related to tobacco smoke and summarized in Tables 4-1, 4-2, and
4-3, Tables 4-4 and 4-5 summarize health outcomes for which the evidence
summarized in the 2014 Surgeon General’s report is currently considered
strong enough to be considered suggestive of a causal association but not
yet strong enough to be rated as causal. These are outcomes for which the
currently existing body of evidence falls short of being definitive, but the
association between cigarette smoking and these outcomes remains under
active investigation.
MORBIDITY
Tables 4-1 through 4-3 summarize the preclinical health effects and
morbidity caused by cigarette smoking, organized according to whether the
effects occur in the immediate, intermediate, or long-term time horizon and
by the stages of life usually affected by the health outcome.
Immediate Health Effects

Cigarette smoking causes a constellation of subclinical health effects
that occur shortly after initiation of smoking. As described below, these
immediate adverse health effects include increased oxidative stress; deple-
tion of selected bioavailable antioxidant micronutrients; increased inflam-
mation; impaired immune status; altered lipid profiles; poorer self-rated
health status; respiratory symptoms, including coughing, phlegm, wheezing,
and dyspnea; and nicotine addiction. Taken in combination, these detri-
mental effects detract from a smoker’s overall health status and lead to
what has been referred to as “diminished health status” (HHS, 2004).
Physiologic markers of diminished health status include subclinical out-
comes such as increased oxidative stress, reduced antioxidant defenses,
increased inflammation, impaired immune status, and altered lipid profiles
(see Tables 4-1 through 4-3). Smoking’s impacts on such short-term physi-
ologic outcomes impair the smoker’s overall health status, which in turn
renders the smoker more susceptible to various adverse health outcomes,
such as developing acute illnesses, respiratory symptoms, and a lessened
capacity to heal wounds. One downstream marker of the diminished health
status induced by cigarette smoking is that smokers are more likely to miss
school and work. In short, soon after the initiation of smoking, an array
of smoking-induced short-term deleterious health effects sets in motion a
lifelong t rajectory that leaves persistent smokers highly disadvantaged com-

TABLE 4-4 Intermediate Adverse Health Outcomes with Evidence

Suggestive of a Causal Association with Cigarette Smoking Based on
Surgeon General’s Reports
Stage of Life
Behavioral
Substance use (risk ü
factor for use of
marijuana and other
substances)
Behavioral and ü
learning disorders
(disruptive behavioral
disorders, attention
deficit hyperactivity
disorder)
Dental
Dental caries ü ü ü ü
Root-surface caries ü ü ü ü
Failure of dental ü
implants
Respiratory
Incidence of asthma ü ü ü ü
Exacerbation of ü
asthma
Recurrent ü
tuberculosis infectiona
Idiopathic pulmonary ü
fibrosis
Nonspecific bronchial ü ü
hyper-responsiveness
aHealth outcome not included in the 2014 Surgeon General’s report.
pared to their counterparts who never smoked. By looking at the immedi-

ate and intermediate adverse health effects of cigarette smoking, it is clear
that cigarette smoking contributes in important ways to suboptimal health
beginning shortly after smoking initiation—long before the chronic diseases
that smoking causes at older ages become clinically apparent (HHS, 2004).

TABLE 4-5 Long-Term Adverse Health Outcomes with Evidence

Suggestive of a Causal Association with Cigarette Smoking Based on
Surgeon General’s Reports
Stage of Life
Health Outcomes Adolescence Adulthood Adulthood Adulthood
Cancer (fatal prostate ü
cancer, higher risk
of advanced stage
cancer, and disease
progression in men
who have prostate
cancer; noncardia
gastric cancers; breast
cancer)
Bone Health (low ü
bone density in men)
Eye Disease ü
(opthalmopathy
associated with
Graves’ disease)
Peptic Ulcer ü
Complications
Physiologic Markers of Diminished Health Status

Increased oxidative stress Cigarette smoke contains free radicals and other
oxidants in abundance. A single puff of a cigarette exposes the smoker to
more than 1015 free radicals in the gas phase and additional radicals and
oxidants in the tar phase (Pryor and Stone, 1993).
The biological impacts of the oxidative stress induced by cigarette
smoking have been extensively documented in humans (HHS, 2004). These
include oxidative injury to proteins, DNA, and lipids. Assaying protein car-
bonyls is one method of measuring oxidative damage to proteins, and pro-
tein carbonyl concentrations have been observed to be significantly higher
in smokers than in nonsmokers (Kapaki et al., 2007; Marangon et al.,
1999; Padmavathi et al., 2010). One way of quantifying the oxidative dam-
age to DNA is to measure the DNA damage in peripheral white blood cells
induced by the hydroxyl radical at the C8 position of guanine, 8-hydroxy-
deoxyguanosine (8-OH-dG). Most of the available evidence indicates that
current smokers have concentrations of 8-OH-dG in peripheral leukocytes
that are at least 20 percent higher than nonsmokers (HHS, 2004). Studies

of 8-OH-dG in DNA extracted from urine provide corroborative evidence,

with 8-OH-dG concentrations that are 6 to 50 percent higher in smokers
than in nonsmokers (Campos et al., 2011; HHS, 2004; Lowe et al., 2009;
Seet et al., 2011). Measures of lipid peroxidation include F2-isoprostanes
and malondialdehyde (MDA). Many studies have demonstrated that cur-
rent smokers have substantially higher concentrations of isoprostanes in
both plasma and urine than nonsmokers (Bloomer et al., 2008; HHS, 2004;
Kocyigit et al., 2011; Ozguner et al., 2005; Seet et al., 2011; Taylor et al.,
2008). The results of several studies indicate that MDA concentrations
are 30 percent more abundant in current-versus-nonsmokers, suggesting
cigarette smoking directly increases MDA concentrations (Bloomer et al.,
2008; Jain et al., 2009; Kocyigit et al., 2005; Ozguner et al., 2005). This
is further corroborated by evidence from several studies that have found
concentrations of thiobarbituric acid–reactive substances (TBARS) found in
MDA range from 6 percent to 118 percent more in smokers than in people
who have never smoked (HHS, 2004).
Cigarette smoking clearly generates substantial quantities of oxidative
stress, as indicated by a consistent body of evidence indicating that cigarette
smoking significantly increases biomarkers of oxidative damage to proteins,
DNA, and lipids. Cigarette smokers experience measurable and immediate
oxidative damage. This oxidative damage, experienced over long periods
of time, is one pathway contributing to smoking-caused disease and death
(HHS, 2010).
Depletion of circulating antioxidant micronutrient concentrations Ciga-

rette smoking exposes the smoker to potential oxidative damage not expe-
rienced by the nonsmoker. One direct result of the exposure to oxidative
stress is the depletion of the body’s defenses against oxidative stress. For
example, the antioxidant defense system is partly comprised of antioxidant
micronutrients (Evans and Halliwell, 2001). Antioxidant status provides a
biomarker of health status because oxidative damage is thought to be cen-
trally involved in the aging process as well as in enhanced susceptibility to
a wide range of specific diseases. Evidence from a number of studies firmly
establishes that smokers have circulating concentrations of ascorbic acid
and provitamin A carotenoids such as a-carotene, b-carotene, and cryp-
toxanthin that are more than 25 percent lower than nonsmokers (Alberg,
2002). Considered in total, a strong and diverse body of evidence consis-
tently implicates oxidative stress from cigarette smoking in the depletion of
antioxidant micronutrients in circulation. Furthermore, the results across
studies are consistent with a dose–response relationship, with the amount of
smoking being inversely related to the circulating concentrations of vitamin
C and provitamin A carotenoids (HHS, 2004).
The immediate effects of cigarette smoking on these concentrations

have been examined with measurements of circulating micronutrient con-

centrations taken before and after a smoker stops smoking. One such study,
for example, found substantially increased concentrations of vitamin C
and provitamin A carotenoids after 84 hours without a cigarette (Brown,
1996). In another study, the exposure of plasma to the equivalent of six
puffs of cigarette smoke completely depleted the ascorbic acid present in the
serum (Eiserich et al., 1995). In yet another, measurements taken at base-
line and 20 minutes after smoking a cigarette found decreased circulating
micronutrient concentrations (Yeung, 1976). Results such as these highlight
the immediate impact that smoking a cigarette can have on health status.
Cigarette smoking causes depletion of antioxidant micronutrients, leading
smokers to have lower circulating concentrations of these antioxidant mi-
cronutrients than nonsmokers. The direct immediate result on the smoker’s
lower concentrations of antioxidant micronutrients such as vitamin C is
to reduce the smoker’s antioxidant defenses, and thus the smoker’s cells
throughout the body are more prone to the damaging effects of oxidative
stress. Oxidative stress is hypothesized to be associated with premature ag-
ing and greater risk of disease (Laher, 2014).
Increased inflammation The direct pro-oxidant effects of cigarette smoke

are further exacerbated by additional endogenous oxidant formation via
the smoking-induced inflammatory-immune response (van der Vaart et al.,
2004; Yao and Rahman, 2011). Another measure of smokers’ poorer health
is the chronically higher level of inflammatory response experienced by
smokers compared to nonsmokers. Chronic inflammation is hypothesized
to play a role in the pathogenesis of numerous chronic diseases (Pawelec et
al., 2014; Prasad et al., 2012).
For example, cigarette smoking is strongly and consistently associ-
ated with higher leukocyte concentrations (HHS, 2004); this suggests that
smoking induces a sustained, long-term inflammatory response. Compared
to nonsmokers, current smokers have been uniformly found, across many
studies, to have approximately 20 percent higher leucocyte counts. Fur-
thermore, leucocyte counts increase with a greater degree of smoking,
measured either by the number of cigarettes smoked per day or the depth
of inhalation (HHS, 2004). Prospective cohort studies that evaluate how
changes in smoking status relate to changes in leucocyte counts provide
evidence that eliminating cigarette smoking leads to reductions in leucocyte
counts (HHS, 2004). Leucocytes are a marker of chronic inflammation, but
cigarette smoking is also associated with markers of the acute inflammatory
response, such as C-reactive protein (HHS, 2014).
Impaired immune status The 2014 Surgeon General’s report was the first
report of the Surgeon General to review thoroughly the contribution of

cigarette smoking to impaired immune status. Cigarette smoking was found

to adversely impact the two major immune pathways, innate immunity and
adaptive immunity. Recognizing the extreme complexity of the immune
system, with its built-in compensatory mechanisms, the conclusion of the
Surgeon General’s report was that the evidence is sufficient to infer that cig-
arette smoking compromises the immune system and compromises immune
homeostasis by diminishing both innate and adaptive immunity (HHS,
2014). The impact of the adverse effects on immune status would be to
make smokers more susceptible to disease, which in turn contributes to the
etiology of acute infectious and chronic diseases above and beyond the way
in which cigarette smoking contributes to acute and chronic inflammation.
Altered lipid profiles Cigarette smoking causes altered lipid metabolism

(HHS, 2010). The alterations in the lipid profile induced by cigarette smok-
ing create a higher risk profile: Compared with nonsmokers, cigarette
smokers have significantly higher serum cholesterol, triglyceride, and low-
density lipoprotein (LDL) levels and lower high-density lipoprotein (HDL)
levels (Ambrose and Barua, 2004). In a meta-analysis of 54 epidemiologic
studies, smokers were found to have serum concentrations of cholesterol,
triglycerides, and very low density lipoprotein (VLDL) cholesterol that
were 3 percent, 9 percent, and 10 percent higher, respectively, and HDL
cholesterol concentrations that were 6 percent lower than nonsmokers
(Craig et al., 1989). Furthermore, clear dose–response associations were ob-
served, with these associations growing stronger as the number of cigarettes
smoked per day increased. The alteration of the lipid profile in the direction
of increased cardiovascular disease risk has been extensively documented
not only in adults but also in children and adolescents. In a meta-analysis of
studies in which study participants ranged from 8 to 19 years of age, ado-
lescents who smoked cigarettes had serum LDL c holesterol and t riglyceride
concentrations that were significantly higher than in nonsmokers, whereas
smokers had lower serum concentrations of HDL cholesterol than non-
smokers (Craig et al., 1990). These differences are likely due to a direct ef-
fect of cigarette smoking. In a cohort of middle school students in Germany,
those who initiated smoking had significantly lower HDL cholesterol levels
than nonsmokers after 2 years of follow-up despite there having been
similar baseline levels of HDL cholesterol in the two groups—those who
would remain nonsmokers and those who would go on to begin smoking
(Dwyer et al., 1988).
Poorer Self-Rated Health Status

The adverse impact of smoking on health status has been directly mea-
sured by comparing self-rated health in smokers versus nonsmokers. Studies

of varying design have uniformly shown that smokers tend to rate their
overall health status lower than nonsmokers do (HHS, 2004, 2014). The
consistent reporting of poorer self-rated health among smokers compared
to nonsmokers across numerous dimensions of health status provides direct
evidence that smoking impairs the health of cigarette smokers in ways that
are perceptible to the smoker even in the absence of clinical disease.
Respiratory Symptoms: Coughing, Phlegm, Wheezing, Dyspnea

The immediate adverse health effects of cigarette smoking are not lim-
ited to subclinical measures; they can also result in physical symptoms. In re-
viewing the evidence separately for children/adolescents and adults, the 2004
Surgeon General’s report concluded that cigarette smoking was causally
associated with all major respiratory symptoms in both age groups (HHS,
2004). The specific symptoms caused by cigarette smoking are coughing,
phlegm, wheezing, and dyspnea. The consistent presence of the causal as-
sociation across the life course supports the classification of these symptoms
as an immediate health effect based on the definition used in this report.
Nicotine Addiction
Another clinical, immediate adverse health effect of cigarette smoking
is nicotine addiction. The 2012 Surgeon General’s report concluded that
cigarette smoking was causally associated with nicotine addiction, begin-
ning in adolescence (HHS, 2012). The onset of nicotine addiction begins
soon after smoking initiation.
The importance of nicotine addiction as an immediate adverse health
effect cannot be underestimated. Nicotine addiction, via its role in propa-
gating sustained smoking, assumes a role as a central determinant of the
entire catalogue of downstream health effects of cigarette smoking. The
often long-term, sustained addiction to nicotine is the underlying factor
driving the long-term, sustained exposure to the toxins in tobacco smoke
that drive the adverse health effects of cigarette smoking.
Finding 4-1: Cigarette smoking is causally associated with a broad

spectrum of adverse health effects that begin soon after the onset of
regular smoking and that, in total, significantly diminish the health
status of the smoker compared to nonsmokers.
Intermediate-Term Effects on Morbidity

The health effects included in the category of “intermediate adverse
health effects” consist largely of health outcomes that are not dependent on

having smoked a cigarette in the immediate past but rather require a more
extensive smoking history for the adverse outcome to become manifest.
For example, intermediate adverse health effects are often direct sequellae
of some of the immediate health effects of smoking, such as absenteeism
and medical care utilization, or else they are diagnoses that are precursors
of subsequent, more severe disease endpoints, such as type 2 diabetes and
subclinical atherosclerosis. Cigarette smoking cessation diminishes the risk
of experiencing these intermediate adverse health effects, but individuals
with a past history of cigarette smoking still have greater risks than those
who never smoked.
Absenteeism
Another indicator of diminished health status is absence from work.
Among the many factors that contribute to attendance, health status is
clearly a major determinant. Thus, attendance patterns are potential mark-
ers of health status (Alberg et al., 2003).
Cigarette smoking is a determinant of absence. A substantial body of
evidence on the association in adults between cigarette smoking and ab-
sence from work consistently demonstrates that smokers are significantly
more likely to have greater workplace absenteeism (HHS, 2004). The likeli-
hood of workplace absence increases with the number of cigarettes smoked
per day (HHS, 2004). Furthermore, smoking cessation is associated with
reduced absence rates (HHS, 2004). In addition to smokers having more
episodes of absence than nonsmokers, smokers tend to stay out longer when
they are sick than nonsmokers. Thus, smokers miss more cumulative work
time than nonsmokers (HHS, 2004).
A strong and consistent body of evidence demonstrates that cigarette
smoking is associated with a greater likelihood of absence from work.
This association could be at least partially due to smoking being a marker
for other causes of absenteeism, such as mental illness and abuse of other
substances. In considering the societal toll of cigarette smoking, attendance
is not only a useful marker of diminished health status, but also a marker
of other downstream costs. On the individual level, workplace absentee-
ism can lead to problems on the job and even result in unemployment. At
the societal level, absenteeism decreases productivity and is a drain on the
economy.
Increased Utilization of Medical Services

Utilization of medical services provides an additional indicator of health
status. Despite the complexities inherent in studying the association be-
tween cigarette smoking and use of medical services, the evidence reviewed

in the 2004 and 2014 Surgeon General’s reports yields a clear signal indi-
cating that cigarette smokers generate higher medical care costs and have
more inpatients and outpatient visits than those who do not smoke (HHS,
2004, 2014). Among patients admitted to the hospital, smokers have longer
lengths of stay and incur greater expenses per admission than nonsmokers.
Subclinical Atherosclerosis
Atherosclerosis is a cardiovascular disease precursor that begins early
in life; it is the underlying pathogenic mechanism that ultimately leads to
many cardiovascular disease endpoints. The epidemiologic evidence has
been consistent in demonstrating a strong, dose-dependent association
between cigarette smoking and subclinical atherosclerosis as measured by
carotid intimal–medial thickness. Consequently, cigarette smoking has been
established as a cause of atherosclerosis (HHS, 2004). Establishing the link
between cigarette smoking and atherosclerosis provides a strong, biologi-
cally plausible rationale for the role of cigarette smoking in the pathogen-
esis of clinical cardiovascular endpoints that occur as a consequence of
atherosclerosis.
Impaired Lung Development and Accelerated Decline in Function

In addition to smoking’s long-term health effects on the respiratory sys-
tem from diseases such as lung cancer and COPD, some adverse respiratory
effects experienced by adolescent cigarette smokers manifest themselves
shortly after smoking initiation. Compared to nonsmokers, adolescents
who smoke cigarettes are more likely to experience impaired lung growth,
early onset in the decline of lung function, and asthma-related symptoms
(HHS, 2004). Among adults who smoke cigarettes, lung function begins to
decline at younger ages, and the age-related decline in lung function occurs
faster (HHS, 2004).
Increased Susceptibility to Infectious Lung Diseases

Due at least in part to its adverse impact on immune status, cigarette
smoking predisposes the smoker to developing acute infectious respiratory
illnesses such as pneumonia. Established effects of cigarette smoking on the
immune system provide a clear biological basis for the increased likelihood
that has been observed among smokers of developing an infection after ex-
posure to microbes that cause respiratory infections and also of developing
a clinically apparent disease once infected (HHS, 2004). Further, impaired
cilia function in the trachea and bronchi also contributes to the increased
risk of respiratory infections in smokers (Simet et al., 2010). Thus, it is no

surprise that cigarette smokers have an increased susceptibility to respira-

tory infections.
Cigarette smoking is causally associated with an increased risk of
pneumonia (HHS, 2004). The 2014 Surgeon General’s report was the first
to review the evidence on the association between cigarette smoking and
tuberculosis. A strong statistical association has been observed between
cigarette smoking and risk of M. tuberculosis infection and also the risk,
once infected, of progressing to tuberculosis disease, but showing a clear
causal connection between smoking and risk of tuberculosis has been chal-
lenging because cigarette smokers often have a much higher risk profile than
nonsmokers for these outcomes because of other social determinants of
health. These challenges notwithstanding, the evidence has now coalesced
to the point that cigarette smoking is causally associated with tuberculosis
disease and tuberculosis mortality (HHS, 2014).
Diabetes
Type 2 diabetes mellitus is a leading underlying cause of mortality from
cardiovascular disease, and it also leads to other adverse consequences
such as kidney failure and blindness (HHS, 2014). Obesity has long been
established as a major risk factor for diabetes, but the association between
cigarette smoking and diabetes has only more recently been elucidated.
The results of a meta-analysis of 51 prospective cohort studies in the 2014
Surgeon General’s report demonstrated that cigarette smokers have a 30–40
percent greater risk of diabetes than nonsmokers and that there is a strong
dose–response relationship, with the risk increasing with the number of
cigarettes smoked per day (HHS, 2014). In addition to having an increased
risk of developing diabetes, evidence also indicates that, among patients
with diabetes, cigarette smokers are more likely to suffer cardiovascular
complications and to have higher mortality rates. Based on this body of
evidence, the 2014 Surgeon General’s report concluded that cigarette smok-
ing is a cause of diabetes (HHS, 2014).
Periodontitis
A synthesis of the evidence in the 2004 Surgeon General’s report re-
vealed a strong, consistent, and dose-dependent relationship between ciga-
rette smoking and the risk of periodontitis. Based on this evidence, cigarette
smoking was judged to be causally associated with periodontitis. Approxi-
mately one-half of all diagnoses of adult periodontitis are attributable to
cigarette smoking (HHS, 2004).

Asthma Exacerbation
The fact that cigarette smoking is causally associated with so many
outcomes that are relevant to asthma has long raised suspicions that ciga-
rette smoking is a risk factor for asthma. Examples of these asthma-relevant
factors are persistent inflammation, diminished immune status, and the
respiratory symptoms of coughing, phlegm, wheezing, and dyspnea. At the
present time, the evidence is considered suggestive but not sufficient to infer
a causal association between cigarette smoking and the risk of developing
asthma in adolescents or adults or between smoking and the risk of asthma
exacerbations in adolescents (HHS, 2014). However, the 2014 Surgeon
General’s report did conclude that cigarette smoking is causally associated
with asthma exacerbation in adults (HHS, 2014).
Adverse Surgical Outcomes: Wound Healing and Respiratory

Complications
The fact that smoking causes diminished health status by impairing fac-
tors such as immune response and lung function provides a strong reason to
believe that cigarette smoking could be associated with a worse prognosis
after surgical procedures. Based on a large and diverse body of evidence
with outcomes that ranged from short- and long-term complications of sur-
gery to survival, the 2004 Surgeon General’s report concluded that cigarette
smoking is a cause of adverse surgical outcomes (HHS, 2004).
Finding 4-2: Cigarette smoking causes many adverse health effects

classified as “intermediate,” which include increased absence from
work, the increased use of medical services, subclinical atherosclerosis,
impaired lung development and function, an increased risk of lung in-
fections, diabetes, periodontitis, the exacerbation of asthma in adults,
subclinical organ injury, and adverse surgical outcomes.
Long-Term Morbidity
Cigarette smoking contributes to a major portion of the population
burden of many of the chronic diseases that typically occur in middle and
late adulthood, such as cancer, cardiovascular disease, and COPD (HHS,
2004). As noted below, the full scope of long-term morbidity attributable
to cigarette smoking also extends to numerous other disease endpoints.
Cessation of cigarette smoking diminishes the risk of experiencing these
long-term adverse health effects, but a past history of cigarette smoking
is still associated with increased risk compared to never having smoked
(HHS, 2014).

Cancer
Cigarette smoking is causally associated with 12 different types of
malignancy and is responsible for approximately 30 percent of all cancer
deaths in the United States (ACS, 2007; HHS, 2014). Cigarette smoking has
been known for many years to be a cause of cancers of the lung, oral cavity,
larynx, esophagus, bladder, pancreas, kidney, uterine cervix, and stomach,
and of acute myeloid leukemia. The conclusions of the 2014 report of the
Surgeon General indicate that cigarette smoking is also causally associated
with colorectal cancer and liver cancer. Furthermore, cigarette smoking
is causally associated with clinical precursors of cancer lesions, such as
colorectal adenomatous polyps (HHS, 2014).
Vascular Disease
Cigarette smoking is associated with numerous clinical cardiovascular
disease endpoints, including coronary heart disease, stroke, and abdomi-
nal aortic aneurism. Coronary heart disease is a leading cause of death in
the United States and most high-income countries. Cigarette smoking has
been established as a major cause of coronary heart disease for decades.
The impact of cigarette smoking is particularly strong among younger age
groups, as it causes 40 percent of ischemic heart disease deaths in 35- to
64-year-olds (HHS, 2004).
Cigarette smoking has long been identified as a major cause of cere-
brovascular disease. As with coronary heart disease, the impact of cigarette
smoking is proportionally larger in relatively younger adults. Among 35- to
64-year-olds, more than 40 percent of all cerebrovascular disease deaths are
attributable to cigarette smoking (HHS, 2004).
Cigarette smoking is an established cause of abdominal aortic aneu-
rysm (HHS, 2004). This condition is often fatal and accounts for more than
10,000 deaths per year in the United States.
COPD
The process of inhaling cigarette smoke brings the smoker’s respiratory
system into direct contact with heavy doses of tobacco toxins. Given these
profound levels of exposure, it is not surprising that cigarette smoking’s del-
eterious effects on the respiratory system extend well beyond lung cancer.
Cigarette smoking is estimated to have caused 7.5 million prevalent cases
of COPD in the United States in 2009 (Rostron et al., 2014). More than
138,000 Americans died from COPD in 2010, making it the third leading
cause of death in the United States (Heron, 2013). As the predominant

cause of COPD, cigarette smoking is responsible for approximately 80

percent of the mortality burden from COPD (HHS, 2004).
Eye Disease: Age-Related Macular Degeneration and Nuclear Cataracts

Cigarette smoking also adversely affects eye health, causing nuclear
cataracts (HHS, 2004). The body of evidence linking cigarette smoking
with age-related macular degeneration that was accumulated over the past
two decades has now been judged to be strong and consistent enough to
prove a causal association between the two (HHS, 2014).
Rheumatoid Arthritis
Cigarette smoking also causes joint disease. More than 1 million Ameri-
cans have been diagnosed with rheumatoid arthritis, a disease linked to
immune dysregulation. Enough supportive evidence has been accumulated
to indicate a clear link between cigarette smoking and rheumatoid arthri-
tis. The conclusions of the 2014 Surgeon General’s report contained the
conclusion that a causal association has been established between cigarette
smoking and rheumatoid arthritis (HHS, 2014).
Bone Health: Hip Fractures and Bone Density

Cigarette smoking has adverse consequences for bone health. Ciga-
rette smoking is causally associated with hip fractures. In postmenopausal
women, a causal association has been established between cigarette smok-
ing and low bone density (HHS, 2004).
Finding 4-3: Cigarette smoking is causally associated with a broad

spectrum of adverse long-term health effects which cause suffering,
impaired quality of life, and death.
Maternal/Fetal and Infancy Health Effects

Pregnancy represents a particularly vulnerable time of life for both the
mother and the developing fetus, and this critical time window extends into
the neonatal period and infancy. Because of the unique features of this pe-
riod of enhanced vulnerability and its critical public health importance, the
topic is considered separately. Cigarette smoking is an established cause of a
broad spectrum of health effects to the mother, fetus, and infant, including
decreased likelihood of becoming pregnant, increased risk of experienc-
ing adverse pregnancy outcomes, and adverse effects on the newborn that
can range from organ impairment to congenital malformations to death,

as summarized in Table 4-6. Table 4-6 also includes the immediate physi-
ologic effects of smoking from Table 4-1 to emphasize the point that preg-
nant women who smoke incur the same short-term adverse health effects
incurred by all cigarette smokers. It is estimated that more than 400,000
infants are exposed each year to maternal smoking in utero. Furthermore,
recent data indicate that more than 1.2 million births each year in the
TABLE 4-6 Maternal, Fetal, and Infant Adverse Health Outcomes

Causally Associated with Cigarette Smoking Based on Surgeon General’s
Reports
Health Outcome Maternal Fetal Infant/Child
Immediate Health Effects on All Smokers, Including During Pregnancy (selected)
Oxidative Stress ü
Depletion of Antioxidant ü
Micronutrients
Increased Inflammation ü
Compromised Immune Status ü
Altered Lipid Metabolism ü
Lower Self-Rated Health Status ü
Likelihood of Becoming Pregnant
Reduced Fertility (maternal and ü
paternal)
Pregnancy Complications
Complications of Pregnancy ü
(ectopic pregnancy, premature
rupture of the membranes,
placenta previa, and placental
abruption)
Shortened Pregnancy (pre-term ü
delivery and shortened gestation)
Outcomes of Childbirth and Survival
Impaired Fetal Growth (fetal ü ü
growth restriction or low birth
weight)
Congenital Malformations ü
(orofacial clefts)
Impaired Organ Function ü ü
(reduced lung function)
Death (stillbirth, infant mortality, ü ü
sudden infant death syndrome)

United States occur among mothers under 25 years of age. In the United
States in 2012, 31 percent of all births were to mothers less than 25 years
old (1,225,871/3,952,841); of these, 90,095 were to mothers less than 18
years old, 85,310 were to mothers who were 18 years old, and 1,050,466
were to mothers who were 19–24 years old (Martin et al., 2013).
Decreased Likelihood of Conception

Cigarette smoking is associated with a decreased likelihood of preg-
nancy because of smoking’s adverse effects on the female and the male
reproductive systems. Cigarette smoking is causally associated with reduced
fertility in women (HHS, 2004). Further, the 2014 Surgeon General’s re-
port pointed to a diverse body of research evidence supported by a strong
biologic rationale to conclude that cigarette smoking is a cause of erectile
dysfunction in men.
Pregnancy Complications
Maternal smoking during pregnancy reduces the likelihood of a full-
term gestational period with optimal fetal growth. Cigarette smoking by
pregnant women adversely affects pregnancy by making it more likely they
will experience ectopic pregnancies, complications of pregnancy such as
premature rupture of the membranes, placenta previa, and placental abrup-
tion. Furthermore, cigarette smoking in expectant mothers causes preterm
delivery and shortened gestation (HHS, 2004).
Outcomes: Childbirth, Infancy, and Survival

Maternal cigarette smoking during pregnancy directly harms the fetus
and, later, the infant in several ways (HHS, 2004). Cigarette smoking is
causally associated with stunted fetal growth and is an important cause
of shortened gestation. In combination, stunted fetal growth and prema-
ture delivery are major determinants of low birth weight. Cigarette smok-
ing causes congenital malformations, specifically orofacial clefts. Cigarette
smoking is also associated with impaired organ function, specifically re-
duced lung function (HHS, 2014).
Based on these many severe effects, it is logical to infer that cigarette
smoking negatively affects the viability of the fetus and child. Specifically,
smoking is causally associated with fetal deaths, or stillbirths; furthermore,
among live births smoking is an established cause of overall infant mortal-
ity. That is, compared with infants of mothers who do not smoke, infants
with mothers who smoke during or after pregnancy experience higher rates
of death before reaching 1 year of age. One specific cause of increased mor-

tality of infants whose mothers smoke is sudden infant death syndrome,

which is more likely to strike those infants than infants whose mothers do
not smoke (HHS, 2004).
After birth, children who are exposed to secondhand smoke (SHS) via
parental smoking suffer numerous adverse health effects as a consequence.
In infants, symptoms associated with SHS exposure include increased lower
respiratory illnesses, otitis media, middle ear effusion, reduced lung func-
tion, and the respiratory symptoms of coughing, phlegm, wheezing, and
dyspnea (HHS, 2006). In addition to the increased risk of symptoms, in-
fants of smoking mothers are more likely to experience subclinical immedi-
ate adverse health effects of cigarette smoke exposure as well. For example,
evidence indicates that infant exposure to parental smoking is associated
with physiologic markers of diminished health status, such as increased oxi-
dative damage to DNA and lipids. As noted above, 8-OH-dG can be used
as a measure of oxidative damage to DNA, and neonatal levels of urinary
8-OH-dG have been found to be significantly associated with exposure to
the toxicants from tobacco smoke due to the mother’s smoking (Hong et
al., 2001). Newborns with mothers who smoked had concentrations of
8-OH-dG that were 333 percent higher than newborns whose mothers did
not smoke (Hong et al., 2001).
Finding 4-4: Maternal smoking during pregnancy and secondhand

smoke exposure during infancy are causally associated with many ad-
verse health outcomes. This not only leaves exposed infants prone to
short- and long-term health risks but also can result in death.
Age of Initiation and Health Outcomes

The following four factors were used to assess the effects that the age
of initiation had on an individual’s cigarette smoking trajectory and subse-
quent health effects: (1) nicotine dependence, (2) the number of cigarettes
smoked per day (smoking intensity), (3) the likelihood of smoking cessation
(or, conversely, the likelihood of remaining a smoker), and (4) health out-
comes. These four factors are closely interrelated. Nicotine dependence is
associated with smoking intensity (Hu et al., 2006), and both of these mea-
sures are in turn associated with the likelihood of remaining a smoker in the
long term. The interrelationships among the factors involve both smoking
intensity (number of cigarettes per day) and smoking duration (number of
years smoked) and hence also the effects of the lifetime cumulative exposure
to cigarette smoking. Many of the established deleterious health effects of
cigarette smoking are dose-dependent, thus providing a mechanistic expla-
nation for how earlier age of initiation could exert a powerful contribution

on smoking-caused health effects that is mediated by leading to increased

doses of exposure to cigarette smoke.
In particular, the mechanistic basis for a powerful influence of the age
of initiation on smoking-caused adverse health outcomes is grounded in
the evidence, reviewed in Chapter 3, that those who start smoking earlier
are more likely to (1) have a greater degree of nicotine dependence (Breslau
and Peterson, 1996; Buchmann et al., 2013; HHS, 2012; Hu et al., 2006;
Lando et al., 1999; Park et al., 2004), (2) smoke cigarettes more frequently
(Breslau, 1993; Buchmann et al., 2013; Chen and Millar, 1998; D’Avanzo
et al., 1994; Escobedo et al., 1993; Everett et al., 1999; Fernandez et al.,
1999; Hu et al., 2006; Lando et al., 1999; Reidpath et al., 2014; Taioli and
Wynder, 1991), and (3) remain smokers for longer periods of time (Breslau
and Peterson, 1996; Chen and Millar, 1998; D’Avanzo et al., 1994; Eisner
et al., 2000; Everett et al., 1999; Khuder et al., 1999). These associations
all point toward an association between a younger age of initiation and
greater exposure to the toxicants in cigarette smoke, which because of
well-established dose–response relationships would therefore be expected to
lead to higher risk of smoking-caused disease and death. A further negative
consequence of starting to smoke at younger ages is that tissues and organ
systems that are still in the growth and maturation phase may be particu-
larly vulnerable to the toxicants in smoke, so that even a given exposure
dose to cigarette smoke may be more harmful when exposure occurs during
childhood and adolescence than during adulthood.
Younger age of initiation has been found to be associated with one
short-term health effect in particular: an increased risk of hospital inpatient
stay during the previous year (Lando et al., 1999). Concerning long-term
health effects, the lung is exquisitely sensitive to the adverse consequences
of cigarette smoke because it is directly exposed to inhaled cigarette smoke
and is further exposed to harmful smoke toxicants via the circulation of
those toxicants in the blood. In a prospective cohort study, a strong as-
sociation was observed between an earlier age of smoking initiation and
an increased risk of respiratory disease (Kenfield et al., 2008). Compared
to people who have never smoked, the relative odds (and 95 percent con-
fidence intervals) of contracting respiratory disease were 7.0 (3.9–12.4)
for those who started smoking at 26 years old or older; 8.1 (5.5–11.9)
for those who started between 22 and 25; 10.2 (9.9–13.2) for smoking
initiation between 18 and 21; and 13.4 (9.8–18.2) for those who started
smoking at 17 or younger; the age trend is highly statistically significant (a
p-value of 0.001). The same study also observed a statistically significant
trend for the risk of lung cancer, which was not grouped under respiratory
disease (Kenfield et al., 2008); this finding was also observed in another
population-based cohort study (Prizment et al., 2014). The strong associa-
tion between an earlier age of starting to smoke and increased lung cancer

risk was summarized in a meta-analysis of 69 studies, which estimated that

the summary odds ratio for lung cancer was 10.3 (95 percent confidence
interval of 8.0–13.3) for starting to smoke around the age of 14 years;
7.5 (5.9–9.4) for starting to smoke at approximately 18 years; and 3.9
(3.3–4.6) for starting to smoke at age 26 years (Lee et al., 2012). Thus,
an earlier age of initiation is strongly associated with an increased risk of
respiratory diseases (primarily COPD) and lung cancer.
The evidence for cardiovascular disease has been mixed. The risk of
cardiovascular disease increased significantly with younger age of initiation
in the ARIC prospective cohort study (Huxley et al., 2012), but the results
of the Nurses’ Health Study did not find a significant effect (Kenfield et al.,
2008). In another study, younger age of initiation was significantly associ-
ated with peripheral artery disease (Planas et al., 2002).
Overall, the evidence is consistent in finding that the younger the age
of initiation, the greater the risk of nicotine dependence, smoking intensity,
and persistent smoking/reduced likelihood of cessation. The associations
between a younger age of initiation and these outcomes holds true even
after accounting for time from first cigarette to first daily smoking. The
findings consistently show a dose–response trend, with younger ages of ini-
tiation associated with a higher likelihood of nicotine dependence, greater
smoking intensity, and reduced likelihood of cessation. The absence of any
apparent age threshold on these associations or any diminution of the as-
sociations across the age continuum indicates that any delay in initiation,
regardless of the ages affected (e.g., late childhood to early adolescence,
early to mid-adolescence, or adolescence to young adulthood) would be
expected to have measurable benefits in reducing the lifetime consumption
of cigarettes and hence in reducing the risk for smoking-caused disease and
death. The adverse consequences of a younger age of initiation appear to
manifest at young ages and be sustained over the life course.
Finding 4-5: A younger age of initiation is associated with an increased

risk of many adverse health outcomes, such as a hospital inpatient
stay in the past year and lifetime risk of respiratory disease, especially
chronic obstructive pulmonary disease and lung cancer.
Other Tobacco Products and Sources of Exposure

So far, the discussion has focused specifically on cigarette smoking.
SHS exposure and other tobacco products and nicotine delivery devices
are discussed below.

Secondhand Smoke Exposure

The health effects of cigarette smoking are not limited to the adverse
health effects on the smoker; they also include the health consequences that
exposure to SHS has on nonsmokers (HHS, 2014). SHS exposure has now
been linked with a host of adverse health effects in addition to the long-
established causal associations with lung cancer and heart disease.
As cigarette smokers, parents who smoke cigarettes increase their per-
sonal risk for all of the adverse health outcomes described above. If par-
ents smoke in the presence of their children, they also negatively affect the
health of their children by exposing them to SHS. The health effects of SHS
exposure are not limited to long-term enhanced susceptibility to chronic
diseases, but, as in the case of cigarette smoking, they also include imme-
diate and substantial effects that leave SHS-exposed individuals prone to
short-term health risks (see Table 4-7).
Thus, as is the case with cigarette smoking, SHS exposure is associated
with diminished health status. Exposure to SHS is associated with increased
oxidative damage to DNA and lipids. As noted above, MDA can be used
as a measure of lipid peroxidation, and children exposed to SHS have
been found to have significantly higher circulating levels of MDA and also
significantly lower levels of glutathione peroxidase (Zalata et al., 2007).
Concerning antioxidant micronutrients, the evidence for SHS exposure
mirrors the evidence for smoking. Compared to nonsmokers not exposed
to SHS, nonsmokers exposed to SHS have significantly reduced circulating
concentrations of vitamin C and provitamin A carotenoids, indicating that
even low-dose cigarette smoke exposures lower circulating antioxidant
micronutrient concentrations. Evidence of lowered circulating antioxidant
micronutrient concentrations has also been observed in children of smok-
ers (Wilson et al., 2011; Yilmaz et al., 2009; Zalata et al., 2007). Children
whose mothers were smokers had 29 percent and 26 percent lower circulat-
ing concentrations of vitamin E and vitamin A, respectively, than children
whose mothers did not smoke (Yilmaz et al., 2009).
Nonsmokers exposed to SHS have also been found to have lessened
immune status (HHS, 2010). The body of evidence firmly indicates that
among nonsmokers, SHS exposure is associated with greater oxidative
damage, lower circulating antioxidant micronutrient concentrations, and
lessened immune status. Given the consistent body of evidence and the clear
biological rationale based on the causal associations seen with cigarette
smoking these associations are likely to be rated as causal in the future, but
the evidence base has not yet reached the standard for these associations to
be judged as causal in the Surgeon General’s report.
Consistent with the health effects observed for cigarette smoking, the
health effects of SHS exposure also include reduced lung function and the

respiratory symptoms of coughing, phlegm, wheezing, and dyspnea. SHS

exposure in children causes numerous adverse health effects, including
lower respiratory illnesses, otitis media, and middle ear effusion (HHS,
2006).
In adults, SHS exposure is also causally associated with increased risk
of long-term chronic diseases, just as in the case of cigarette smoking. These
diseases include lung cancer, coronary heart disease, stroke, and inflamma-
tory bowel disease.
As expected, based on the lower-exposure doses of exposure to tobacco
toxins that result from secondhand smoke, the health risks of SHS exposure
for most health outcomes tend to be less than the risks of cigarette smoking.
Nevertheless, the fact that these risks are incurred even at very low doses
indicates that there is no safe threshold for exposure to cigarette smoke.
The importance of this public health challenge is accentuated by the fact
that these health risks are incurred as the result of smoking by others rather
by the affected individuals themselves.
Finding 4-6: Secondhand smoke exposure is causally associated with

adverse health effects.
It is worth keeping in mind that this lengthy catalogue of well-

established consequences of SHS exposure will continue to grow as more
definitive evidence coalesces for additional health outcomes. For example,
Table 4-8 summarizes health outcomes for which the evidence summarized
in the 2014 Surgeon General’s report is currently considered strong enough
to be considered suggestive of a causal association but not yet strong
enough to be rated as causal.
Smoking of Pipes, Cigars, and Other Combustible Tobacco Products

Combustible tobacco products other than cigarettes are also associated
with the same sort of chronic disease outcomes associated with cigarette
smoking, such as cancer and cardiovascular disease. Pipe and cigar smoke
contain similar profiles of harmful toxins to those found in cigarette smoke
(HHS, 2014). A key distinction in the health risks is that the doses of toxins
delivered to the smoker are often less for pipes and cigars than for ciga-
rettes because pipes and cigars are usually smoked less frequently and the
smoke tends to be inhaled less deeply (HHS, 1998). For example, pipe and
cigar smoking pose risks for malignancies of the larynx, oral cavity, and
esophagus that are similar to the risks associated with smoking cigarettes
(HHS, 1998). Pipes and cigars are causally associated with lung cancer,
even though the risks are less than observed for cigarette smoking because
compared to cigarette smoking pipes and cigars are smoked on average

116
TABLE 4-7 Adverse Health Outcomes Causally Associated with Secondhand Smoke Exposure Based on Surgeon
General’s Reports
Stage of Life
Health Outcome Infancy Adolescence Adulthood Adulthood Adulthood
Short-Term and Intermediate-Term Health Effects
Maternal/Fetal Development (low birth weight) ü
Ear
Ear Problems ü ü
Middle Ear Disease ü

Respiratory
Acute Respiratory Infections ü ü

Slower Lung Growth ü ü
Respiratory Tract Injury ü ü ü ü ü
Coughing, Phlegm, Wheezing, Breathlessness ü
Lower Respiratory Illness ü ü
Wheeze Illnesses ü
Lower Level of Lung Function ü
Odor Annoyance ü ü ü ü
Nasal Irritation ü ü ü ü
Long-Term Health Effects
Inflammatory Bowel Disease (Crohn’s disease) ü ü ü
Cancer (lung) ü
Cardiovascular ü
Stroke, Coronary Heart Disease ü
Endothelial Cell Dysfunctions ü ü ü ü ü

117
TABLE 4-8 Adverse Health Outcomes with Evidence Suggestive of a Causal Association with Secondhand Smoke
118
Exposure Based on Surgeon General’s Reports

Stage of Life
Health Outcome Pregnancy Infancy Adolescence Adulthood Adulthood Adulthood
Diminished Immune Function (immune activating and ü
suppressive effects)
Maternal/Fetal Development (pre-term delivery) ü
Dental Carries ü
Respiratory
Incidence of asthma ü ü ü ü
Worsening of asthma control and symptoms ü ü ü ü
Coughing, wheezing, chest tightness, and breathlessness ü ü ü ü
COPD ü
Chronic respiratory symptoms ü ü ü ü
Small decrement in lung function ü ü ü ü
Cardiovascular (angina, sudden coronary death, stroke, ü
atherosclerosis)
Cancer
Breast cancer ü

Childhood leukemias ü
Childhood lymphomas ü
Childhood brain tumors ü
Nasal sinus cancer ü
less frequently and the smoke is inhaled less deeply (Alberg et al., 2013).
The available evidence indicating that pipe and cigar smoking have similar
adverse health effects to cigarette smoking thus supports the conclusion
that the impact of a policy change that resulted in lower uptake or delayed
initiation of pipes or cigars would have a significant impact on public health
but would be expected to be less than a similar reduction in cigarette smok-
ing because of the lower exposure to tobacco toxins due to the manner in
which pipes and cigars are smoked.
Another way to smoke tobacco is with a hookah, or waterpipe. From
an exposure assessment perspective, the distinctive features of this tobacco
smoke delivery system are that the tobacco is sometimes indirectly heated
and that the smoke passes through a water column prior to inhalation (Akl
et al., 2010). Hookah use is becoming more common throughout the world,
including in the United States (Cobb et al., 2010; Jawad et al., 2013). In a
study comparing the urinary concentrations of the tobacco-specific nitro-
samine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) in cigarette
smokers, hookah smokers, and nonsmokers, it was found that hookah
smokers had significantly higher NNAL concentrations than nonsmokers
but significantly lower concentrations than cigarette smokers (Radwan et
al., 2013). In a study in which urine samples were collected from hookah
smokers before and after they smoked from the hookah, significant post-
smoking increases were noted in the urinary concentrations of nicotine,
cotinine, NNAL, and volatile organic compounds (St. Helen et al., 2014).
Expired carbon monoxide concentrations (Jacob et al., 2011) and benzene
exposure (Jacob et al., 2013) tend to be much higher for hookah smoking
than for cigarette smoking. Studies have assessed the association between
hookah smoking and selected health outcomes, but there is a paucity of
evidence available on this topic, and the body of evidence is generally of
low quality (Akl et al., 2010). In a meta-analysis of data from four studies,
hookah smoking was significantly associated with an increased lung cancer
risk (odds ratio, 2.1; 95 percent confidence interval, 1.3–3.4) (Akl et al.,
2010). In this same systematic review, only one study each was identified
to assess the association between hookah smoking and cancers of the blad-
der, esophagus, and nasopharynx, and none of the observed associations
were statistically significant (Akl et al., 2010). With respect to pregnancy
outcomes, three studies found hookah smoking to be associated with a
significantly increased risk of low birth weight (2.1; 1.1–4.2) (Akl et al.,
2010). In one study, hookah smoking was found to be associated with a
significantly increased risk of respiratory illness (2.3; 1.1–5.1) (Akl et al.,
2010). Definitive conclusions on the risks associated with hookah smok-
ing versus cigarette smoking are not possible with the limited quality and
quantity of the evidence currently available.
Little evidence on the health effects of newer combustible tobacco prod-

ucts has been generated. In attempting to estimate risks, it is important to

account for the specific product features. For example, the 2014 Surgeon
General’s report points out that when considering the emergence of small
cigarette-like cigars, the health risks may more closely parallel those of ciga-
rettes than of the traditional cigar because of the way that small cigarette-
like cigars are used (HHS, 2014). This line of reasoning emphasizes that
the health risks of tobacco use are directly linked to doses of exposure to
disease-causing toxins, which is a function not only of the tobacco product
but also of the frequency and duration of and the manner in which the
product is smoked, when factors such as depth of inhalation are accounted
for. This concept is also critical to thinking about the health risks of dual
use or poly-use of combustible tobacco products and ENDS, an exposure
pattern that will likely increase in the future but for which data on health
risks are needed.
Finding 4-7: Smoking of combustible tobacco products other than

cigarettes, such as pipes and cigars, is causally associated with a broad
spectrum of adverse health effects.
Smokeless Tobacco Products

The marketplace for smokeless tobacco products has diversified con-
siderably in recent years. In addition to the traditional smokeless tobacco
products of chewing tobacco and snuff, a number of new products have
been introduced, such as snus and dissolvable tobacco products.
The 1986 Surgeon General’s report examined the evidence concerning
smokeless tobacco and concluded that it was a cause of cancer of the oral
cavity. Smokeless tobacco use can also lead to oral leukoplakia, gingival
recession, and nicotine addiction. A 2007 monograph of the International
Agency for Research on Cancer (IARC) that focused on smokeless tobacco
concluded that smokeless tobacco is a Group 1 carcinogen, meaning that it
is a human carcinogen (IARC, 2007). The IARC review of the evidence led
to the conclusion, “Smokeless tobacco causes cancers of the oral cavity and
pancreas” (IARC, 2007, p. 370). Smokeless tobacco may also be linked to
an increased risk of esophageal cancer (IARC, 2012).
These earlier reviews of the evidence concerning the health effects of
smokeless tobacco use were primarily based on evidence related to tradi-
tional smokeless tobacco products and did not take into account the newer
products. A more recent review of the epidemiologic evidence for Swedish-
type snus, a moist snuff, suggests that the use of snus may be less harmful
than cigarette smoking (Lee, 2011). How the health risks of Swedish-type
snus differ from the more traditional smokeless tobacco products has yet
to be precisely characterized; furthermore, direct epidemiologic evidence is

not yet available on the health effects of the Swedish-type snus products
presently marketed in the United States.
Finding 4-8: The use of smokeless tobacco products is causally associ-

ated with oral cancer.
ENDS
The marketplace for tobacco products and devices that deliver nicotine
has recently expanded in response to the smoking bans that have increas-
ingly limited the locations where traditional cigarette smoking is allowed
(Jawad et al., 2013; Kamerow, 2013; Popova and Ling, 2013; Schuster et
al., 2013). Electronic nicotine delivery systems, or ENDS, have experienced
a rapid upsurge in use and are now marketed by the major U.S. tobacco
companies (Dockrell et al., 2013; Kamerow, 2013; Li et al., 2013; Popova
and Ling, 2013).
Monitoring this expansion in products and how the products are used
is important to tobacco control. An ENDS product that decreases the de-
livery of tobacco toxins would ostensibly also reduce the risk of developing
smoking-caused disease if current cigarette smokers were to switch from
cigarettes to exclusive use of the ENDS. On the other hand, the risk of
smoking-caused disease could be increased if the ENDS maintained nicotine
addiction and its users continued to smoke cigarettes and to use multiple
products that deliver nicotine. Furthermore, these alternative products, par-
ticularly those that involve flavorings attractive to adolescents, may serve as
a gateway for adolescents to initiate smoking and thus start on a path that
eventually leads to tobacco addiction. Currently there is a paucity of data
on issues such as these; along with the direct adverse health effects associ-
ated with use of these alternative products, these remain important lines
of inquiry for future research. Definitive evidence on the long-term health
effects of ENDS products will not be available for many years because any
long-term health effects associated with these products will take decades to
emerge. Furthermore, generating the needed evidence base will be compli-
cated by the facts that there are so many different ENDS products and the
products and their contents are evolving.
IMPACT OF CIGARETTE SMOKING ON MORTALITY

Cigarette smoking contributes significantly to the population burden
of many of the leading causes of chronic disease deaths that typically occur
in middle and late adulthood, such as cancer, cardiovascular disease, and
COPD (HHS, 2004).
The combined death toll linked to cigarette smoking is staggering. Cig-

arette smoking is estimated to account for approximately 480,000 deaths

per year in the United States (HHS, 2014). In 2010 the four leading causes
of death in the United States were heart disease (597,700 deaths), cancer
(574,700 deaths), chronic lower respiratory diseases (138,100 deaths), and
stroke and cerebrovascular disease (129,500) (Heron, 2013). Cigarette
smoking is a major cause of all four of these diseases. Furthermore, smok-
ing is also a cause of the seventh (diabetes, 69,000 deaths) and eighth (in-
fluenza/pneumonia, 50,100 deaths) leading causes of death (Heron, 2013).
Cancer
As a cause of 12 different types of malignancy, cigarette smoking is
responsible for 163,700 cancer deaths per year in the United States (HHS,
2014; NCHS, 2013). Most of this mortality burden (130,700 deaths) is due
to lung cancer, but cigarette smoking also caused 36,000 deaths from other
malignancies (HHS, 2014).
Cardiovascular Disease
Cigarette smoking is estimated to cause 160,600 cardiovascular dis-
ease deaths per year in the United States (HHS, 2014). The majority of the
smoking-caused cardiovascular deaths (99,300 deaths) are due to coronary
heart disease, but smoking also causes 25,500 deaths from other forms of
heart disease. Furthermore, cigarette smoking causes 15,300 deaths from
cerebrovascular disease and 11,500 deaths from other forms of vascular
disease.
Diabetes
Type 2 diabetes mellitus is a leading underlying cause of mortality from
cardiovascular disease, and it also leads to other adverse consequences such
as kidney failure and blindness (HHS, 2014). It is the seventh leading cause
of death in the United States (Heron, 2013). Cigarette smoking is estimated
to cause 9,000 deaths from type 2 diabetes per year in the United States
(HHS, 2014).
COPD
More than 138,000 Americans died from COPD in 2010 (Heron,
2013), making it the third leading cause of death in the United States.
Cigarette smoking is the predominant cause of COPD. Estimates indicate
that 100,600 COPD deaths per year in the United States are attributable
to cigarette smoking (HHS, 2014).

Increased Susceptibility to Infectious Lung Diseases

Cigarette smoking is causally associated with an increased risk of
pneumonia (HHS, 2004) and tuberculosis mortality (HHS, 2014). Cigarette
smoking is estimated to cause 12,500 deaths from these infectious diseases
per year.
IMPACT OF EXPOSURE TO SECONDHAND

SMOKE ON MORTALITY
Due to its causal associations with coronary heart disease and lung
cancer, secondhand smoke exposure is estimated to cause more than 41,300
deaths per year in the United States (HHS, 2014). The majority of these
(almost 34,000 deaths) are due to coronary heart disease, while more than
7,000 deaths per year are from lung cancer (HHS, 2014). Furthermore,
parental smoking is estimated to cause approximately 600 deaths per year
from prenatal conditions and 400 deaths per year from sudden infant death
syndrome (HHS, 2014).
Finding 4-9: Tobacco use is causally associated with premature mortal-

ity from a variety of causes, such as lung infections, chronic obstructive
pulmonary disease, coronary heart disease, and a variety of cancers.
REFERENCES
ACS (American Cancer Society). 2007. Cancer facts & figures 2007. Atlanta, GA: American
Cancer Society.
Akl, E. A., S. Gaddam, S. K. Gunukula, R. Honeine, P. A. Jaoude, and J. Irani. 2010. The
effects of waterpipe tobacco smoking on health outcomes: A systematic review. Interna-
tional Journal of Epidemiology 39(3):834–857.
Alberg, A. J. 2002. The influence of cigarette smoking on circulating concentrations of anti-
oxidant micronutrients. Toxicology 180(2):121–137.
Alberg, A. J., G. B. Diette, and J. G. Ford. 2003. Attendance and absence as markers of health
status—The example of active and passive cigarette smoking. American Journal of Epi-
demiology 157(10):870–873.
Alberg, A. J., M. V. Brock, J. G. Ford, J. M. Samet, and S. D. Spivack. 2013. Epidemiol-
ogy of lung cancer: Diagnosis and management of lung cancer, 3rd ed.: American
College of Chest Physicians evidence-based clinical practice guidelines. Chest 143(5
Suppl):e1S–e29S.
Ambrose, J. A., and R. S. Barua. 2004. The pathophysiology of cigarette smoking and
cardiovascular disease: An update. Journal of the American College of Cardiology
43(10):1731–1737.
Bloomer, R. J., A. D. Solis, K. H. Fisher-Wellman, and W. A. Smith. 2008. Postprandial
oxidative stress is exacerbated in cigarette smokers. British Journal of Nutrition 99(5):
1055–1060.
Breslau, N. 1993. Daily cigarette consumption in early adulthood: Age of smoking initiation
and duration of smoking. Drug and Alcohol Dependence 33(3):287–291.

Breslau, N., and E. L. Peterson. 1996. Smoking cessation in young adults: Age at initiation
of cigarette smoking and other suspected influences. American Journal of Public Health
86(2):214–220.
Brown, A. J. 1996. Acute effect of smoking cessation on antioxidant status. Journal of Nutri-
tional Biochemistry 7(1):29–39.
Buchmann, A. F., D. Blomeyer, C. Jennen-Steinmetz, M. H. Schmidt, G. Esser, T. Banaschewski,
and M. Laucht. 2013. Early smoking onset may promise initial pleasurable sensations
and later addiction. Addiction Biology 18(6):947–954.
Campos, C., R. Guzman, E. Lopez-Fernandez, and A. Casado. 2011. Urinary biomarkers of
oxidative/nitrosative stress in healthy smokers. Inhalation Toxicology 23(3):148–156.
Chen, J., and W. J. Millar. 1998. Age of smoking initiation: Implications for quitting. Health
Reports 9(4):39–46.
Cobb, C., K. D. Ward, W. Maziak, A. L. Shihadeh, and T. Eissenberg. 2010. Waterpipe
tobacco smoking: An emerging health crisis in the United States. American Journal of
Health Behavior 34(3):275–285.
Craig, W. Y., G. E. Palomaki, and J. E. Haddow. 1989. Cigarette smoking and serum lipid
and lipoprotein concentrations: An analysis of published data. BMJ 298(6676):784–788.
Craig, W. Y., G. E. Palomaki, A. M. Johnson, and J. E. Haddow. 1990. Cigarette smoking-
associated changes in blood lipid and lipoprotein levels in the 8- to 19-year-old age group:
A meta-analysis. Pediatrics 85(2):155–158.
D’Avanzo, B., C. La Vecchia, and E. Negri. 1994. Age at starting smoking and number of
cigarettes smoked. Annals of Epidemiology 4(6):455–459.
Dockrell, M., R. Morison, L. Bauld, and A. McNeill. 2013. E-cigarettes: Prevalence and at-
titudes in Great Britain. Nicotine & Tobacco Research 15(10):1737–1744.
Dwyer, J. H., G. E. Rieger-Ndakorerwa, N. K. Semmer, R. Fuchs, and P. Lippert. 1988. Low-
level cigarette smoking and longitudinal change in serum cholesterol among adolescents.
The Berlin-Bremen study. JAMA 259(19):2857–2862.
Eiserich, J. P., A. van der Vliet, G. J. Handelman, B. Halliwell, and C. E. Cross. 1995. Dietary
antioxidants and cigarette smoke-induced biomolecular damage: A complex interaction.
American Journal of Clinical Nutrition 62(6 Suppl):1490S–1500S.
Eisner, M. D., E. H. Yelin, P. P. Katz, S. C. Shiboski, J. Henke, and P. D. Blanc. 2000. Predic-
tors of cigarette smoking and smoking cessation among adults with asthma. American
Escobedo, L. G., S. E. Marcus, D. Holtzman, and G. A. Giovino. 1993. Sports participation,
age at smoking initiation, and the risk of smoking among U.S. high school students.
JAMA 269(11):1391–1395.
Evans, P., and B. Halliwell. 2001. Micronutrients: Oxidant/antioxidant status. British Journal
of Nutrition 85(Suppl 2):S67–S74.
Everett, S. A., C. W. Warren, D. Sharp, L. Kann, C. G. Husten, and L. S. Crossett. 1999. Ini-
tiation of cigarette smoking and subsequent smoking behavior among U.S. high school
students. Preventive Medicine 29(5):327–333.
Fernandez, E., A. Schiaffino, C. La Vecchia, J. M. Borras, M. Nebot, E. Salto, R. Tresserras,
L. Rajmil, J. R. Villalbi, and A. Segura. 1999. Age at starting smoking and number of
cigarettes smoked in Catalonia, Spain. Preventive Medicine 28(4):361–366.
Heron, M. 2013. Deaths: Leading causes for 2010. National Vital Statistics Reports 62(6):1–96.
HHS (Department of Health and Human Services). 1998. Smoking and tobacco control
monograph 9: Cigars: Health effects and trends. Bethesda, MD: U.S. Department of
Health and Human Services; Office on Smoking and Health; National Cancer Institute.
Original edition, Bethesda, MD.

———. 2004. The health consequences of smoking: A report of the Surgeon General. Atlanta,
Prevention, National Center for Chronic Disease Prevention and Health Promotion, Of-
fice on Smoking and Health.
———. 2006. The health consequences of involuntary exposure to tobacco smoke: A report
of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services,
Centers for Disease Control and Prevention, Coordinating Center for Health Promo-
tion, National Center for Chronic Disease Prevention and Health Promotion, Office on
Smoking and Health.
———. 2010. How tobacco smoke causes disease: The biology and behavioral basis for
smoking-attributable disease: A report of the Surgeon General. Atlanta, GA: U.S. De-
partment of Health and Human Services, Centers for Disease Control and Prevention,
National Center for Chronic Disease Prevention and Health Promotion, Office on Smok-
ing and Health.
———. 2012. Preventing tobacco use among youth and young adults: A report of the Sur-
geon General. Atlanta GA: U.S. Department of Health and Human Services, Centers for
Disease Control and Prevention, National Center for Chronic Disease Prevention and
Hong, Y. C., H. Kim, M. W. Im, K. H. Lee, B. H. Woo, and D. C. Christiani. 2001. Mater-
nal genetic effects on neonatal susceptibility to oxidative damage from environmental
tobacco smoke. Journal of the National Cancer Institute 93(8):645–647.
Huxley, R. R., H. Yatsuya, P. L. Lutsey, M. Woodward, A. Alonso, and A. R. Folsom. 2012.
Impact of age at smoking initiation, dosage, and time since quitting on cardiovascular
disease in African Americans and whites: The Atherosclerosis Risk in Communities Study.
American Journal of Epidemiology 175(8):816–826.
IARC (International Agency for Research on Cancer). 2007. Smokeless tobacco and some
tobacco-specific n-nitrosamines. IARC Monographs on the Evaluation of Carcinogenic
Risks to Humans Vol. 89.
———. 2012. Personal habits and indoor combustions. Volume 100e. A review of human
carcinogens. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans
100(Pt E).
Jacob, P., III, A. H. Abu Raddaha, D. Dempsey, C. Havel, M. Peng, L. Yu, and N. L. Benowitz.
2011. Nicotine, carbon monoxide, and carcinogen exposure after a single use of a water
pipe. Cancer Epidemiology, Biomarkers and Prevention 20(11):2345–2353.
———. 2013. Comparison of nicotine and carcinogen exposure with water pipe and cigarette
smoking. Cancer Epidemiology, Biomarkers and Prevention 22(5):765–772.
Jain, A., B. K. Agrawal, M. Varma, and A. A. Jadhav. 2009. Antioxidant status and smoking
habits: Relationship with diet. Singapore Medical Journal 50(6):624–627.
Jawad, M., A. McEwen, A. McNeill, and L. Shahab. 2013. To what extent should waterpipe
tobacco smoking become a public health priority? Addiction 108(11):1873–1884.
Kamerow, D. 2013. Big tobacco lights up e-cigarettes. BMJ 346:f3418.
Kapaki, E., I. Liappas, L. Lyras, G. P. Paraskevas, I. Mamali, I. Theotoka, N. Bourboulis, I.
Liosis, O. Petropoulou, and K. Soldatos. 2007. Oxidative damage to plasma proteins in
patients with chronic alcohol dependence: The effect of smoking. In Vivo 21(3):523–528.

Kenfield, S. A., M. J. Stampfer, B. A. Rosner, and G. A. Colditz. 2008. Smoking and smoking
cessation in relation to mortality in women. JAMA 299(17):2037–2047.
Khuder, S. A., H. H. Dayal, and A. B. Mutgi. 1999. Age at smoking onset and its effect on
smoking cessation. Addictive Behaviors 24(5):673–677.
Kocyigit, A., H. Keles, S. Selek, S. Guzel, H. Celik, and O. Erel. 2005. Increased DNA dam-
age and oxidative stress in patients with cutaneous leishmaniasis. Mutation Research
585(1–2):71–78.
Kocyigit, A., S. Selek, H. Celik, and M. Dikilitas. 2011. Mononuclear leukocyte DNA damage
and oxidative stress: The association with smoking of hand-rolled and filter cigarettes.
Mutation Research 721(2):136–141.
Laher, I. 2014. Systems biology of free radicals and antioxidants. Berlin: Springer Reference.
Lando, H. A., D. T. Thai, D. M. Murray, L. A. Robinson, R. W. Jeffery, N. E. Sherwood, and
D. J. Hennrikus. 1999. Age of initiation, smoking patterns, and risk in a population of
working adults. Preventive Medicine 29(6 Part 1):590–598.
Lee, P. N. 2011. Summary of the epidemiological evidence relating snus to health. Regulatory
Toxicology and Pharmacology 59(2):197–214.
Lee, P. N., B. A. Forey, and K. J. Coombs. 2012. Systematic review with meta-analysis of the
epidemiological evidence in the 1900s relating smoking to lung cancer. BMC Cancer
12:385.
Li, J., C. Bullen, R. Newcombe, N. Walker, and D. Walton. 2013. The use and acceptability
of electronic cigarettes among New Zealand smokers. New Zealand Medical Journal
126(1375):48–57.
Lowe, F. J., E. O. Gregg, and M. McEwan. 2009. Evaluation of biomarkers of exposure and
potential harm in smokers, former smokers and never-smokers. Clinical Chemistry and
Laboratory Medicine 47(3):311–320.
Marangon, K., S. Devaraj, and I. Jialal. 1999. Measurement of protein carbonyls in plasma of
smokers and in oxidized LDL by an ELISA. Clinical Chemistry 45(4):577–578.
Martin, J. A., B. E. Hamilton, M. J. K. Osterman, S. C. Curtin, and T. J. Mathews. 2013.
Births: Final data for 2012. National Vital Statistics Reports 64(1):1–65.
NCHS (National Center for Health Statistics). 2013. Faststats: Leading causes of death. http://
www.cdc.gov/nchs/fastats/leading-causes-of-death.htm (accessed January 7, 2015).
Ozguner, F., A. Koyu, and G. Cesur. 2005. Active smoking causes oxidative stress and de-
creases blood melatonin levels. Toxicology and Industrial Health 21(1–2):21–26.
Padmavathi, P., V. D. Reddy, G. Kavitha, M. Paramahamsa, and N. Varadacharyulu. 2010.
Chronic cigarette smoking alters erythrocyte membrane lipid composition and properties
in male human volunteers. Nitric Oxide 23(3):181–186.
Park, S. M., K. Y. Son, Y. J. Lee, H. C. Lee, J. H. Kang, Y. J. Lee, Y. J. Chang, and Y. H. Yun.
2004. A preliminary investigation of early smoking initiation and nicotine dependence
in Korean adults. Drug and Alcohol Dependence 74(2):197–203.
Pawelec, G., D. Goldeck, and E. Derhovanessian. 2014. Inflammation, ageing and chronic
disease. Current Opinion in Immunology 29:23–28.
Planas, A., A. Clara, J. Marrugat, J. M. Pou, A. Gasol, A. de Moner, C. Contreras, and F.
Vidal-Barraquer. 2002. Age at onset of smoking is an independent risk factor in periph-
eral artery disease development. Journal of Vascular Surgery 35(3):506–509.
Popova, L., and P. M. Ling. 2013. Alternative tobacco product use and smoking cessation: A
national study. American Journal of Public Health 103(5):923–930.
Prasad, S., B. Sung, and B. B. Aggarwal. 2012. Age-associated chronic diseases require age-
old medicine: Role of chronic inflammation. Preventive Medicine 54(Suppl):S29–S37.
Prizment, A. E., H. Yatsuya, P. L. Lutsey, J. H. Lubin, M. Woodward, A. R. Folsom, and R. R.
Huxley. 2014. Smoking behavior and lung cancer in a biracial cohort: The Atherosclero-
sis Risk in Communities Study. American Journal of Preventive Medicine 46(6):624–632.

Pryor, W. A., and K. Stone. 1993. Oxidants in cigarette smoke: Radicals, hydrogen perox-
ide, peroxynitrate, and peroxynitrite. Annals of the New York Academy of Sciences
686:12–27.
Radwan, G., S. S. Hecht, S. G. Carmella, and C. A. Loffredo. 2013. Tobacco-specific nitrosa-
mine exposures in smokers and nonsmokers exposed to cigarette or waterpipe tobacco
smoke. Nicotine & Tobacco Research 15(1):130–138.
Reidpath, D. D., T. M. Davey, A. Kadirvelu, I. N. Soyiri, and P. Allotey. 2014. Does one
cigarette make an adolescent smoker, and is it influenced by age and age of smoking ini-
tiation? Evidence of association from the U.S. Youth Risk Behavior Surveillance System
(2011). Preventive Medicine 59:37–41.
Rostron, B. L., C. M. Chang, and T. F. Pechacek. 2014. Estimation of cigarette smoking–
attributable morbidity in the United States. JAMA Internal Medicine 174(12):1922–1928.
Schuster, R. M., A. W. Hertel, and R. Mermelstein. 2013. Cigar, cigarillo, and little ci-
gar use among current cigarette-smoking adolescents. Nicotine & Tobacco Research
15(5):925–931.
Seet, R. C. S., C.Y. Lee, W. M. Loke, S. H. Huang, H. Huang, W. F. Looi, E. S. Chew, A. M.
Quek, E. C. Lim, and B. Halliwell. 2011. Biomarkers of oxidative damage in cigarette
smokers: Which biomarkers might reflect acute versus chronic oxidative stress? Free
Radical Biology and Medicine 50(12):1787–1793.
Simet, S. M., J. H. Sisson, J. A. Pavlik, J. M. DeVasure, C. Boyer, X. Liu, S. Kawasaki, J. G.
Sharp, S. I. Rennard, and T. A. Wyatt. 2010. Long-term cigarette smoke exposure in a
mouse model of ciliated epithelial cell function. American Journal of Respiratory Cell
and Molecular Biology 43(6):635–640.
St. Helen, G., N. L. Benowitz, K. M. Dains, C. Havel, M. Peng, and P. Jacob III. 2014.
Nicotine and carcinogen exposure after water pipe smoking in hookah bars. Cancer
Epidemiology, Biomarkers & Prevention 23(6):1055–1066.
Taioli, E., and E. L. Wynder. 1991. Effect of the age at which smoking begins on frequency of
smoking in adulthood. New England Journal of Medicine 325(13):968–969.
Taylor, A. W., R. S. Bruno, and M. G. Traber. 2008. Women and smokers have elevated uri-
nary F(2)-isoprostane metabolites: A novel extraction and LC-MS methodology. Lipids
43(10):925–936.
Thun, M., R. Peto, J. Boreham, and A. D. Lopez. 2012. Stages of the cigarette epidemic on
entering its second century. Tobacco Control 21(2):96–101.
van der Vaart, H., D. S. Postma, W. Timens, and N. H. T. Ten Hacken. 2004. Acute effects of
cigarette smoke on inflammation and oxidative stress: A review. Thorax 59(8):713–721.
Wilson, K. M., J. N. Finkelstein, A. K. Blumkin, D. Best, and J. D. Klein. 2011. Micronutrient
levels in children exposed to secondhand tobacco smoke. Nicotine & Tobacco Research
13(9):800–808.
Yao, H., and I. Rahman. 2011. Current concepts on oxidative/carbonyl stress, inflammation
and epigenetics in pathogenesis of chronic obstructive pulmonary disease. Toxicology
and Applied Pharmacology 254(2):72–85.
Yeung, D. L. 1976. Relationships between cigarette smoking, oral contraceptives, and plasma
vitamins A, E, C, and plasma triglycerides and cholesterol. American Journal of Clinical
Nutrition 29(11):1216–1221.
Yilmaz, G., P. Isik Agras, S. Hizli, C. Karacan, H. T. Besler, K. Yurdakok, and T. Coskun.
2009. The effect of passive smoking and breast feeding on serum antioxidant vitamin (A,
C, E) levels in infants. Acta Paediatrica 98(3):531–536.
Zalata, A., S. Yahia, A. El–Bakary, and H. M. Elsheikha. 2007. Increased DNA damage in
children caused by passive smoking as assessed by comet assay and oxidative stress.
Mutation Research 629(2):140–147.


Restrictions on Youth Access

to Tobacco Products
L
aws aiming to reduce underage access to tobacco include restrictions
on both distribution of tobacco products to and purchase of tobacco
products by underage individuals. Laws limiting distribution apply
both to commercial tobacco sales and to other methods of provision, such
as giving tobacco to a minor or buying tobacco on behalf of a minor (i.e.,
proxy sales). Restrictions on purchase are distinguished from the restric-
tions on distribution by the fact that they punish the underage buyer. Pur-
chase laws are commonly accompanied by restrictions on underage tobacco
use and possession and are therefore frequently referred to as p urchase–
use–possession (PUP) laws. There is vast variation and inconsistency across
the United States in youth access laws and how they are implemented and
enforced. Despite the profusion and complexity of these laws, there is a
common thread, which is that the enforcement of these restrictions has
focused primarily on curtailing youth access to tobacco from commercial
sources. Accordingly, that is the focus of this analysis. This chapter sum-
marizes youth access restrictions in the United States and their enforcement,
and it describes survey data regarding where underage users obtain their
tobacco products.
YOUTH TOBACCO ACCESS LAWS IN THE UNITED STATES
Federal Youth Tobacco Access Laws

In 1992 Congress enacted the Synar Amendment to reduce the avail-
ability of tobacco to underage individuals. This law requires states to
129

enact and enforce laws prohibiting the sale and distribution of tobacco to
underage persons or face the loss of federal block grant funding for sub-
stance abuse prevention and treatment programs. In 1996 the Food and
Drug Administration (FDA) issued its Tobacco Rule, in which it asserted
its authority to regulate tobacco products (HHS, 1996). As part of this ef-
fort, FDA issued regulations on the advertising and marketing of tobacco
products to reduce the appeal of tobacco to children and adolescents and
also issued restrictions on retail sales to underage persons to reduce youth
access to tobacco. The tobacco industry challenged FDA’s authority to
regulate tobacco in court, and in 2000 the Supreme Court nullified FDA’s
rule on the grounds that Congress had not granted FDA explicit jurisdiction
over tobacco (IOM, 2007).
At the same time that it was fighting federal efforts to regulate tobacco,
the tobacco industry was also battling legal challenges brought by the
attorneys general of individual states. In 1994 Mississippi Attorney Gen-
eral Michael Moore filed a lawsuit against the major tobacco companies
to recoup state Medicaid expenditures on residents with tobacco-related
diseases (IOM, 2007). Attorneys general from every state soon followed
suit, and on November 23, 1998, the attorneys general from 46 states, the
District of Columbia, and several U.S. territories signed the Master Settle-
ment Agreement (MSA) with the major tobacco companies (NAAG, 1998).
(Four states had previously reached a separate settlement with the tobacco
companies, which awarded them $40 billion.) Although the primary aim of
these suits and the resulting agreement focused on the tobacco companies’
payment of $206 billion to the states, distributed from 2000 to 2025, as
a reimbursement for health care costs that the states had incurred because
of tobacco-related health issues, the terms of the agreement also included
the establishment of a national charitable foundation (now known as the
American Legacy Foundation) devoted to reducing adolescent and young
adult smoking and to preventing tobacco-related diseases. The agreement
also included tobacco sales and marketing provisions aimed at reducing
youth access to tobacco. These provisions included bans on gifts to under-
age individuals in exchange for proof of purchase of tobacco products,
gifts through the mail without proof of the recipient’s age, and distribu-
tion of free samples except in locations restricted to adults. The MSA also
restricted cigarette pack size to a minimum of 20 cigarettes and prohibited
tobacco companies from opposing legislation restricting cigarette pack size
through 2001. The MSA further prohibited tobacco companies from legally
challenging the enforceability or constitutionality of state and local tobacco
control laws enacted before June 1, 1998, including state and local youth
access laws that may have been enacted in compliance with Synar.
In 2009 President Barack Obama signed the Family Smoking Preven-
tion and Tobacco Control Act (hereafter referred to as the Tobacco Control

RESTRICTIONS ON YOUTH ACCESS TO TOBACCO PRODUCTS 131
Act) into law, granting FDA broad authority to regulate the manufacture,
marketing, and sales of tobacco products to protect the public’s health and
to reduce adolescent tobacco use.1 In pursuit of these goals, the act directs
FDA to reissue its 1996 Tobacco Rule along with its advertising and ac-
cess regulations. FDA regulations issued under the act currently apply to
cigarettes, cigarette tobacco, and smokeless tobacco. They do not yet cover
other tobacco and nicotine products, such as electronic nicotine delivery
systems (ENDS), or e-cigarettes; cigars; snus; etc. However, FDA has for-
mally proposed to “deem” e-cigarettes, little cigars, and other products
to be “tobacco products” subject to its regulatory jurisdiction under the
Tobacco Control Act (FDA, 2014a). When the final rule is issued and goes
into effect, it will almost certainly extend federal youth access restrictions
to these other products.
The advertising and marketing regulations issued under the Tobacco
Control Act include federal bans complementing the MSA provisions at
the state level: banning the sale of cigarette packs containing fewer than 20
cigarettes and prohibiting the distribution of free samples. The act autho-
rizes FDA to restrict tobacco sales to minors, including requiring face-to-
face sales, with exceptions for vending machines and self-service displays in
adult-only facilities, and requiring age verification for all over-the-counter
sales by checking a driver’s license or other form of photographic identifi-
cation of anyone under age 27. The Tobacco Control Act also grants FDA
the authority to enforce these restrictions, provides a set of sanctions for
violations, and directs FDA to contract with states to assist with retailer
compliance checks—random, unannounced inspections of tobacco retail-
ers—to determine whether retailers are illegally selling tobacco to underage
individuals. In compliance with the congressional direction, FDA reissued
its 1996 Tobacco Rule in 2010 (FDA, 2010).
The Tobacco Control Act also sets limits on FDA’s authority. Limits
relevant to youth access include prohibiting FDA from banning face-to-
face sales by any specific type of tobacco retailer (i.e., FDA cannot ban all
pharmacies or convenience stores from selling tobacco) and from raising
the federal minimum age of legal access to tobacco products (MLA). The
act does, however, establish a federal MLA of 18 without preempting exist-
ing state laws or penalties while allowing states and localities to establish
a higher MLA.
1Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31 111th
Cong. (June 22, 2009).

State and Local Youth Access Laws

Although federal law requires an MLA of 18, some states and localities
have experimented with higher MLAs. Currently, 46 states have an MLA of
18, while 4 states (Alabama, Alaska, New Jersey, and Utah) have an MLA
of 19. (See Appendix A for a list of select U.S. jurisdictions with an MLA of
19.) In the past decade, a number of localities have also adopted an MLA
over 18. In 2005, Needham, Massachusetts, became the first location in the
United States to establish an MLA of 21. The Needham Board of Health
enacted a town regulation raising the age under a Massachusetts state
provision that allows local boards of health to make “reasonable health
regulations.”2 Since 2005 numerous Massachusetts towns have followed
suit; as of November 2014, 6 towns had an MLA of 19, 22 had an MLA
of 21, and another 9 towns were considering proposals to raise the MLA
to ages higher than 18. Outside of Massachusetts, Nassau, Onondaga, and
Westchester counties in New York State have an MLA of 19, and Hawaii
County (the big island in Hawaii), Suffolk County in New York State,
and, most notably, New York City have also recently raised the MLA to
21. (See Appendix A for a list of select U.S. jurisdictions with an MLA of
21.) A number of states and localities, including Colorado, Maryland, New
Jersey, Texas, Utah, and several localities in California, New Jersey, and
Washington State have also considered proposals to raise the MLA to 21.
These differing MLAs have not been in place long enough, however, for
any differential effects on tobacco use to be detected. (See Appendix A for
select states and localities with either proposed or enacted MLAs over 18.)
In compliance with the Synar Amendment, all 50 states and the District
of Columbia (51 jurisdictions total) have enacted laws prohibiting the sale
or distribution of tobacco products to underage persons. All 51 jurisdic-
tions prohibit commercial transfers, while 48 states and the District of
Columbia also prohibit noncommercial transfers (e.g., giving, exchanging,
bartering, furnishing, or otherwise distributing tobacco). At least 18 states
explicitly differentiate between commercial and noncommercial tobacco
transfers for penalty purposes. Penalties vary significantly: 28 jurisdictions
authorize license revocation or suspension for sales to minors; about two-
thirds of the jurisdictions classify the offense as a criminal offense; and, of
the 37 jurisdictions that increase the penalty for repeat violations, 25 autho-
rize substantial fines of $1,000 or more. Currently, all 51 jurisdictions cover
cigarettes and smokeless and roll-your-own tobacco, while 31 jurisdictions
prohibit the distribution of ENDS. Appendix B provides full details on the
laws regarding commercial and noncommercial tobacco transfers to under-
age individuals for the 50 states and the District of Columbia.
2 MASS. GEN. LAWS ch. 111 § 31.

Currently, the youth access laws of 44 states and the District of

Columbia penalize underage individuals for the purchase, use, or posses-
sion of tobacco. (Maryland, Massachusetts, Nevada, New Jersey, and New
York have no PUP prohibitions.) In the vast majority of states, the offense
is punishable as a civil infraction. Sanctions for violations of PUP laws
include confiscation of the tobacco product, notifying parents of the viola-
tion, community service, participation in a tobacco prevention education
program, and fines ranging from $5 to $300. In addition, in nine states
underage users caught in violation of PUP laws may be subject to having
their driver’s license suspended or revoked or to having limits placed on
their driving privileges (e.g., only from home to work or school and back).
Appendix C provides more details on PUP laws for tobacco in the 50 states
and the District of Columbia.
Finding 5-1: Although most states currently set the minimum age of
legal access to tobacco products at 18, 4 states set it at 19, and New
York City and several other localities around the country have raised
the minimum legal access age to 21.
Finding 5-2: All 51 jurisdictions prohibit commercial transfers, while

48 states and the District of Columbia also prohibit noncommercial
transfers (e.g., giving, exchanging, bartering, furnishing, or otherwise
distributing tobacco).
Finding 5-3: All 51 jurisdictions cover cigarettes, smokeless tobacco,

and roll-your-own tobacco, while 31 jurisdictions currently prohibit the
distribution of electronic nicotine delivery systems.
Finding 5-4: The great majority of jurisdictions (47) prohibit underage

individuals from purchasing, attempting to purchase, possessing, or
using covered tobacco products. Sanctions typically include a fine or
community service.
ENFORCEMENT OF YOUTH ACCESS LAWS

This section summarizes current enforcement policies and practices at
the federal, state, and local levels pertaining to youth access restrictions.
Because enforcement of these restrictions is largely focused on assuring
compliance by licensed tobacco retailers, the committee’s review and analy-
sis is also focused here. However, this section also summarizes what little is
known about the enforcement of MLA restrictions against Internet vendors
and black market sellers as well as the noncommercial distribution of to-
bacco by so-called social sources.

Enforcing Restrictions Against Licensed Retailers

States and localities in the United States did not seriously enforce
youth access laws in the early 1990s, when these laws were first being
implemented (IOM, 1994). Evidence from the United States and abroad
further suggests that retailers are not likely to comply with MLA laws if
there is no meaningful enforcement (e.g., compliance checks and sanctions
for violations) (CDC, 1993; Cismoski and Sheridan, 1993; DiFranza, 1999,
2000; DiFranza and Coleman, 2001; Erickson et al., 1993; Kuendig, 2011;
Rigotti et al., 1997; Schensky et al., 1996; Verdonk-Kleinjan et al., 2008).
Federal Support for Retailer Enforcement

The federal government oversees two comprehensive programs to en-
force the MLA for tobacco products: the Synar program of the Substance
Abuse and Mental Health Services Administration (SAMHSA) and FDA’s
tobacco retail compliance inspection contracts, which are implemented by
states and localities.
The Synar program The 1992 Synar Amendment requires states to enact
and enforce laws prohibiting the sale or distribution of tobacco to persons
under age 18 or face the loss of 40 percent of federal Substance Abuse
Prevention and Treatment Block Grants. SAMHSA, charged with imple-
menting the amendment, issued regulations in 1996 to provide further
guidance to states. These regulations stipulate that, in addition to enacting
laws restricting underage access to tobacco, states must also enforce these
laws “in a manner that can reasonably be expected to reduce the extent to
which tobacco products are available to individuals under the age of 18”
(SAMHSA, 1996, p. 1492); must develop a strategy to reduce the rate of
illegal tobacco sales to underage persons to 20 percent or less by 2003;
and must conduct annual compliance checks of retailers selling tobacco
both over the counter and from vending machines to ensure compliance
with the law. Moreover, because Synar primarily aims to survey the rate of
illegal tobacco sales to underage persons, it requires states to demonstrate
that their compliance checks include a statistically representative sample of
tobacco retail outlets accessible to children and adolescents. These compli-
ance checks may (but are not required to) include a state-level enforcement
component. Thus, some programs may not have sanctions for violations
and may instead use other measures, such as education programs targeted
at retailers and mass media campaigns, to ensure high levels of compliance.
Despite the lack of a regulatory requirement, a study of Synar implementa-
tion (DiFranza and Dussault, 2005) found that the Department of Health
and Human Services pressured some states to adopt compliance checks as

an instrument of enforcement instead of using those checks as a basis for

retailer education alone. Furthermore, although federal block grants are
conditioned on state enforcement of their youth access laws, states are
explicitly prohibited from using the block grant funds to finance Synar
compliance checks. The regulation also requires states to submit an annual
report to SAMHSA detailing activities they conducted to reduce illegal
sales of tobacco to underage persons (including methods used to conduct
compliance checks), progress achieved, and plans for enforcing the youth
tobacco access law in the next year. The Synar regulation applies to all 50
states, the District of Columbia, and 8 U.S. territories. It is not applicable
to American Indian tribes.
FDA tobacco retail inspection contracts FDA’s tobacco retail inspection

contracts provide funding to state partners to conduct compliance check
inspections specifically for enforcement purposes. As such, violations may
lead to escalating fines, from warning letters to civil monetary penalties to
suspension or revocation of retailers’ licenses to sell tobacco. Unlike Synar,
FDA only requires inspection of over-the-counter tobacco retailers because
the Tobacco Control Act restricts vending machines to adult-only facilities,
to which underage persons should not have access. Additionally, since the
program is not intended for comprehensive surveillance, FDA contracts
neither require a statistically valid survey of tobacco retailers nor set a
performance target. However, FDA requires inspections using older decoys
(ages 16–17) in neighborhoods considered to be at higher risk for viola-
tions, including neighborhoods with greater concentrations of populations
with low socioeconomic status or of racial/ethnic minorities; these com-
munities tend to have a greater density of tobacco retailers or have tradi
tionally been targeted by the tobacco industry (CTP, 2014). Furthermore,
states and territories may use FDA inspection contract funds to support
Synar compliance checks so long as compliance check protocols and grant
recipients meet the requirements of both programs. Moreover, because FDA
contracts are narrowly restricted to enforcement activities, it is likely that
states and localities will need to continue to conduct other youth tobacco
access prevention activities, such as mass media campaigns and community
and retailer education programs, to meet the Synar performance target
(i.e., an 80 percent rate of compliance). FDA is authorized to contract with
all states, the District of Columbia, five U.S. territories (American Samoa,
Commonwealth of the Northern Mariana Islands, Guam, Puerto Rico, and
the U.S. Virgin Islands), and—unlike Synar—also with American Indian
tribes.

State and Local Enforcement Strategies Against Licensed Retailers

Under the two federal programs, enforcement activities are required
to include compliance check inspections. However, because they are imple-
mented at the state and local levels, enforcement activities and penalties
for violations range considerably. A standard compliance check protocol
involves sending supervised underage individuals into tobacco retailers
to attempt to purchase tobacco. The underage decoys are typically non
smokers who have no visible tattoos or piercings and are sent alone or in
pairs. The decoys range in age (from 13 to 17), gender, and race. Most
decoys ask to purchase cigarettes, but some are instructed to ask for smoke-
less or other tobacco products. Some carry and are instructed to present
their own genuine photographic identification, while others are instructed
not to present identification and to tell clerks that they have forgotten it.
Some decoys carry out purchases, while others refuse the sale once it is
verified that vendors were willing to sell to underage users. In some inspec-
tions, supervisors are stationed discreetly in the store to observe and record
details of the transaction. In others, supervisors wait outside for the decoys
to report a list of details about the store and their transactions immediately
following each purchase attempt. Each of these variations in compliance
check protocol may influence a state’s compliance rate.
In addition to the variation in the compliance check protocol, there is
significant variation in the frequency of inspections, whether and how often
violators are reinspected, how and when violators are prosecuted, which (if
any) agency has authority over enforcement, how much funding is available
for enforcement, and the penalties for violations. Thus, although there is
general agreement that youth tobacco access laws must be actively enforced
to reduce illegal tobacco sales to minors, there remains a profusion of en-
forcement strategies and little evidence about the relative effectiveness and
efficiency of these various activities. In an effort to identify best practices,
a study by DiFranza (2005) examined 26 enforcement strategies in the 10
states with the highest retailer compliance rates and the 10 states with the
lowest retailer compliance rates that had been reported to Synar. DiFranza
concluded that the strategies essential for achieving high compliance include
having a plan to enforce the state’s MLA law, designating a single state
agency to oversee and coordinate enforcement, conducting ongoing com-
pliance check inspections, allocating state funding for enforcement inspec-
tions, prosecuting violators, setting penalties for violations, and practicing
effective merchant education. He also identified a number of strategies that
were recommended, but not essential, and also listed strategies that were
not recommended because they waste resources or hinder enforcement.
There were also a number of other strategies that could not be rated due
to insufficient evidence. Indeed, despite the multitude of enforcement prac-

tices, relatively few of these practices have been evaluated, and there is little
evidence about which specific enforcement practices successfully reduce the
availability of tobacco to underage individuals.
Trends in Illegal Tobacco Retail Sales to Minors

Both the Synar Amendment and FDA’s compliance testing program
have resulted in considerable strengthening of state and local enforcement
practices. Since the 1990s, all states have adopted youth access laws and
have seen significant improvements in retailer compliance. In 1997, im-
mediately following the implementation of Synar, the national average
rate of illegal tobacco sales to minors reported to Synar was greater than
40 percent, with a high of 72.7 percent in one state (SAMHSA, 2014). By
2006 all states and the District of Columbia achieved compliance with the
Synar requirements, including achieving the target sales rate of 20 percent
or less (within the 3 percent margin of error), and they have continued to
be in compliance since then (SAMHSA, 2014). In 2013 the national aver-
age rate of tobacco sales to minors for all states and the District of Colum-
bia was 9.6 percent, and it ranged from the highest reported rate of 22.5
percent in Oregon to 1.0 percent in Minnesota and Nevada (SAMHSA,
2014). Since establishing its tobacco retail inspection contracts in 2010,
FDA has granted more than $100 million in contracts in all 50 states, the
District of Columbia, and five American territories (FDA, 2014b), which
has resulted in more than 249,000 inspections of tobacco retailers, 12,600
warning letters, and 1,160 civil money penalties (Lindblom, 2014). Since
FDA’s program does not measure program performance, the degree to
which these contracts have increased retailer compliance with MLA laws is
unknown. Finally, although most data on illegal tobacco sales to underage
individuals come from cigarette sales, some evidence suggests that rates of
illegal sales of other tobacco products (e.g., smokeless tobacco, snus, and
snuff) to underage adolescents are comparable to, if not higher than, those
for cigarettes (Choi et al., 2014; Clark et al., 2000; Hanson et al., 2000).
Finding 5-5: Although the intensity of retailer enforcement continues

to vary widely among the states, federal support has strengthened state
and local enforcement practices across the country.
Finding 5-6: According to data collected by the federally supported

compliance testing program, the average rate of tobacco sales to mi-
nors (i.e., noncompliance in all of the states) in 2013 was 9.6 percent
nationally and ranged from 1 percent to more than 20 percent in the
individual states.

Penalties for Violations

Youth access laws vary widely in the range of sanctions prescribed
for tobacco retailers who sell tobacco to underage individuals. Violations
can be designated as either civil or criminal offenses. Minimum penalties
for first offenses range from a warning letter to fines up to $1,000, while
maximum penalties for subsequent offenses range from license revocation
to fines up to $15,000. As noted above, criminal penalties are prescribed for
violations in about two-thirds of the jurisdictions. However, the committee
has been unable to identify systematic information about the nature and
severity of the sanctions actually imposed in practice.
Enforcing Restrictions Against Internet Sellers

Although the evidence suggests that very few underage persons ob-
tain tobacco from the Internet (Johnston et al., 2014b), Internet tobacco
vendors are a new and growing potential source of tobacco for under-
age individuals, especially among the youngest smokers (Johnston et al.,
2014b). Accordingly, there have been some efforts to curtail Internet sales
to minors. A survey of Internet cigarette vendors (Ribisl et al., 2002) found
that, while the majority of vendors had minimum age warnings on some
part of their website, age verification procedures were generally weak, the
most common being to ask users to check a box affirming that they were of
legal age or to type their birth date. In 2002, California passed legislation
requiring Internet cigarette vendors to verify the age of purchasers upon
both purchase transaction and delivery (Williams et al., 2006). Unfortu-
nately, an evaluation of this law found zero compliance (Williams et al.,
2006). Although it was targeted at reducing illicit sales of untaxed ciga-
rettes and only incidentally affected underage access, in 2005 the Bureau
of Alcohol, Tobacco, Firearms and Explosives, in conjunction with several
state attorneys general, entered into a voluntary agreement with major
credit card and private shipping companies to ban payment transfers and
the delivery of cigarettes purchased on the Internet. As with the California
legislation, this effort was unsuccessful, and a study of Internet cigarette
vendors and sales following these agreements found that despite increases
in the proportion of vendors complying with these agreements, the overall
number of Internet cigarette vendors increased, leading to a net increase in
Internet cigarette sales (Ribisl et al., 2011). FDA’s authority under the Fam-
ily Smoking Prevention and Tobacco Control Act extends to online tobacco
retailers, but as of September 2014 FDA’s Center for Tobacco Products has
issued only four warning letters to Internet vendors found selling to under-
age customers, and it is unclear to what extent the center will pursue these
violations (FDA, 2014c).

Finding 5-7: Limited evidence suggests that youth access restrictions

against Internet sellers are weakly enforced, and that tobacco products
are relatively easily available to underage individuals.
Enforcing Restrictions Against Non-Licensed

Sellers and Social Distributors
An expected effect of restricting the retail sale of tobacco to minors is
that underage persons will seek tobacco from alternative sources. These
would include both alternative commercial sources (e.g., non-licensed deal-
ers in illegal markets) and so-called social sources, such as proxy sales (i.e.,
tobacco purchases on behalf of an underage person) and gifts from peers,
relatives, and strangers (Fichtenberg and Glantz, 2002; Glantz, 1996; Ling
et al., 2002). Indeed, there is some evidence that when access from retail
sources is restricted, there is a corresponding increase in recourse to the
use of non-retail sources (Cummings et al., 2003; DiFranza and Coleman,
2001; Rigotti et al., 1997; Rimpela and Rainio, 2004). It is to curb these
transactions that almost all state youth access laws prohibit non-retail
sources of tobacco to underage individuals. These laws restrict other com-
mercial sales, such as illegal suppliers (i.e., street vendors and those selling
untaxed cigarettes), as well as noncommercial distribution. Unfortunately,
there is little information on the enforcement of laws against sales and dis-
tribution by these other sources, much less the effects of such enforcement.
Black Market Sellers

Aside from the occasional study on the purchase of single cigarettes,
or “loosies,” from street vendors (e.g., Smith et al., 2007), there is little
information available on the frequency of youth purchases on the illegal
market (i.e., from commercial sellers other than retail stores) or on enforce-
ment activities aiming to curtail sales to underage individuals. At the same
time, there is little evidence that underage individuals are obtaining tobacco
from the illegal commercial market. A recent report (NRC, 2015) estimates
that underage individuals constitute at most 1 percent of the illicit market.
Finding 5-8: Although there is an illicit market for tobacco products di-
verted from legal channels, there is little evidence that underage persons
are obtaining tobacco from the illegal commercial market.
Social Sources
Despite the facts that underage persons obtain most of their tobacco
products from “social sources” (see next section) and that most state laws

prohibit noncommercial distribution, there is no evidence indicating that

youth access restrictions against noncommercial distributors are enforced.
As is discussed in Chapter 6, similar prohibitions against the noncommer-
cial distribution of alcohol are sometimes enforced—for example, against
parents who facilitate underage drinking and against adults who agree to
purchase alcohol for underage persons who recruit them to do so outside
liquor stores (the so-called shoulder taps) (IOM and NRC, 2004). However,
equivalent restrictions against tobacco transfers appear to be unenforced.
Finding 5-9: There is no evidence indicating that bans on noncommer-

cial distribution of tobacco by friends, proxy purchasers, and other
“social sources” are enforced.
Summary
Although the intensity of retailer enforcement continues to vary widely
among the states, federal support has strengthened state and local enforce-
ment practices across the country. According to data collected by the
federally supported compliance testing program, the national average rate
of tobacco sales to minors (i.e., noncompliance) was 9.6 percent in 2013
and ranged from 1 percent to 20 percent in the individual states. Limited
evidence suggests that youth access restrictions against Internet sellers are
weakly enforced and that tobacco products are relatively easily available
to underage individuals who have credit cards. Although there is an illicit
market for tobacco products diverted from legal channels, there is little evi-
dence that underage individuals are obtaining tobacco from the illegal com-
mercial market. Although almost all states ban noncommercial distribution
to minors, there is no indication in the literature that these restrictions are
being enforced, and the committee strongly suspects that these restrictions
are essentially unenforced throughout the country. As discussed in Chapter
9, the committee does not expect that situation to change, whether or not
the legal purchase age is raised.
SOURCES OF CIGARETTES FOR UNDERAGE INDIVIDUALS

Having described the scope and enforcement of underage access restric-
tions, this section reviews survey data indicating where underage persons
obtain tobacco, whether use of these sources varies by age of the user,
whether these sources have changed over time, and whether inferences can
be drawn from these data regarding the effects of enforcing MLA restric-
tions on the availability of tobacco to underage users. As discussed above,
underage users obtain tobacco from both commercial and social sources.
Table 5-1 lists the primary sources considered in this report from which

TABLE 5-1 Sources of Tobacco for Underage Persons

Source Definition Common Indicators
Licensed tobacco retailers Licensed commercial I bought a pack of cigarettes

dealers on legal markets myself …
Face-to-face Stores (e.g., gas station, … in a store where the clerk
convenience store, has to hand you the pack or
supermarket) with tobacco carton.
located behind the counter
Self-service Stores where one can pick … in a store where you pick
up a pack or carton and up the pack and bring it to the
bring it to a checkout checkout counter.
counter
Vending machine Stores and other facilities … from vending machines.
(e.g., sports arenas, music
venues) with tobacco
located in vending
machines
Internet vendors Online vendors who mail I bought a pack (or carton) of
tobacco to an individual’s cigarettes myself: … from a
home (or other physical website; … over the Internet.
location)
Social sources Non-licensed non-
commercial distributors
Casual distributors Relatives, friends, and I asked someone to give me a
strangers who give tobacco cigarette; Someone offered me
to underage users a cigarette
Proxy sources Relatives, friends, and I had someone else buy a pack
strangers who purchase of cigarettes for me; I bought
tobacco for underage users cigarettes from another person
and are paid a small fee
(e.g., a few dollars or a
portion of tobacco); gray
market
Illicit tobacco dealers/Black Non-licensed dealers on

market sellers illegal markets (e.g., sellers
of untaxed cigarettes, $5
man, single or “loosie”
cigarettes); black market

adolescents obtain tobacco. Although new patterns of tobacco use suggest

that adolescents and young adults are increasingly using new and other
types of tobacco products (Arrazola et al., 2013; Eaton et al., 2012), most
empirical data about underage acquisition behaviors are largely restricted
to cigarettes. Two national surveys of adolescent tobacco use provide some
detailed information on how and where adolescents obtain their cigarettes:
the National Youth Tobacco Survey (NYTS) and the Monitoring the Future
(MTF) survey.
Table 5-2 provides the responses by current smokers in the 2012 NYTS
survey, broken down by age group and gender, to the question: “During the
past 30 days, how did you get your own cigarettes?”
Among all high school students, the most commonly reported answers
to the question of where they got their tobacco were: Someone offered me
a cigarette (40.2 percent); I asked someone to give me a cigarette (32.0
percent); I had someone else buy a pack of cigarettes for me (30.6 percent);
and I bought a pack of cigarettes myself (27.9 percent). Less common re-
sponses were: I bought cigarettes from another person (8.3 percent); and I
took cigarettes from a store or another person (9.9 percent). Sixteen percent
said they got cigarettes some other way.
Responses varied considerably by age. The youngest age group (9 to
14 years old) was the most varied in the types of methods used to obtain
cigarettes and included 40.1 percent who answered “some other way.” The
oldest age group (18 and older), who can legally purchase cigarettes, cited
fewer methods, with 71.1 percent saying they bought their own cigarettes.
Responses did not vary greatly by gender.
Table 5-3 summarizes the responses by current smokers in NYTS, by
age group and gender, to the question, “During the past 30 days, where did
you buy your own cigarettes?” Among the students who said that they had
purchased their own cigarettes, the most commonly cited specific source
was “a gas station or convenience store” (45.8 percent of high school stu-
dents). This was true even among the youngest age group (9 to 14 years
old), although relatively few of these students (10.8 percent, as indicated
in Table 5-2) actually purchased their own cigarettes. Responses did not
vary greatly by gender.
MTF surveys ask two questions about the sources of cigarettes for cur-
rent smokers: (1) “During the last 30 days, about how many times (if any)
have you bought cigarettes?” with a list of possible methods for purchas-
ing offered as potential answers, and (2) “During the last 30 days, about
how many times (if any) did you buy cigarettes for your own use?” with a
list of possible places for purchasing offered as potential answers. Tables
5-4a and 5-4b provide the responses by grade and by year group. Among
12th graders, the responses are provided separately for those under age
18, who cannot legally purchase cigarettes, versus those 18 and older who

TABLE 5-2 Methods for Obtaining Cigarettes Among High School Students, by Age and Gender, National Youth
Tobacco Survey, 2012
Percent of Smokers
Age group Gender
All ages 9–14 15–17 18+ Female Male
19. During the past 30 days, how did you get your own cigarettes?
(CHOOSE ALL THAT APPLY)
a. I did not get cigarettes during the past 30 days — — — — — —
b. I bought a pack of cigarettes myself 27.9 10.8 20.3 71.1 21.7 32.4
c. I had someone else buy a pack of cigarettes for me 30.6 31.4 36.6 9.9 35.3 27.3
d. I asked someone to give me a cigarette 32.0 32.3 34.4 23.8 38.3 27.5
e. Someone offered me a cigarette 40.2 46.9 41.0 30.8 46.8 35.5
f. I bought cigarettes from another person 8.3 14.2 7.6 4.8 8.4 8.2
g. I took cigarettes from a store or another person 9.9 26.2 6.2 5.2 10.8 9.2
h. I got cigarettes some other way 16.0 40.1 12.0 4.8 14.0 17.5
SOURCE: Committee analysis of CDC, 2014.

143
144
TABLE 5-3 Sources of Purchased Cigarettes Among High School Students, by Age and Gender, National Youth
Tobacco Survey, 2012
Percent of Smokers
Age group Gender
Total 9–14 15–17 18+ Female Male
20. During the past 30 days, where did you buy your own cigarettes?
a. I did not buy cigarettes during the past 30 days — — — — — —
b. A gas station or convenience store 45.8 28.5 43.1 72.3 45.6 46.0
c. A grocery store 7.2 9.5 4.7 13.4 6.6 7.7
d. A drugstore 7.6 8.5 6.2 11.3 7.6 7.6
e. A vending machine 3.6 8.2 2.2 3.8 2.3 4.6
f. Over the Internet 2.1 4.5 1.0 3.5 1.8 2.3
g. Through the mail 1.1 1.4 0.7 2.3 0.7 1.4
h. Some other place not mentioned here 19.3 39.8 17.0 5.9 16.5 21.2

TABLE 5-4a Trends in Sources of Cigarettes Among Current Smokers, 8th and 10th Grades, MTF, 1997–2013
8th Grade 10th Grade
1997– 2002– 2006– 2010– 1997– 2002– 2006– 2010–
2001 2005 2009 2013 2001 2005 2009 2013
During the past 30 days, about how many times (if any) have you bought cigarettes …
a. By having a friend or relative buy them for you 61.9 60.8 59.6 58.3 66.9 63.3 65.0 61.9
b. On your own from vending machines 16.6 12.4 12.3 10.2 11.0 6.5 5.4 4.5
c. Through the mail 4.0 4.5 6.0 6.5 1.9 2.6 2.2 2.5
d. In a store where you pick up the pack (or carton) and bring 16.6 15.3 14.3 12.0 25.9 15.3 13.2 14.8
it to the checkout counter
e. In a store where the clerk has to hand you the pack or 20.4 17.1 18.1 14.6 35.9 27.8 23.4 24.2
carton
f. Bought them in some other way — 52.9 53.3 53.2 — 37.1 38.8 36.6
During the past 30 days, about how many times (if any) did YOU buy cigarettes for your own use …
a. At a big supermarket 10.0 9.3 9.3 6.9 10.9 8.3 7.4 6.1
b. At a small grocery store 22.2 19.6 16.8 15.2 25.3 20.1 16.2 16.0
c. At a drugstore 13.2 12.4 13.8 13.4 12.1 10.4 11.3 13.1
d. At a convenience store (like a Hop-In or 7-Eleven) or a gas 36.0 31.0 30.3 24.8 48.1 40.2 38.3 37.5

station
e. From a website — 4.4 5.2 6.0 — 2.1 1.9 2.6
NOTE: Entries are percentages of current smokers reporting source of cigarettes.
SOURCE: Committee analysis of Johnston et al., 2014a.
145
TABLE 5-4b Trends in Sources of Cigarettes Among Current Smokers, 12th Grade, MTF, 1997–2013
146
12th Grade, <18 12th Grade, 18+

1997– 2002– 2006– 2010– 1997– 2002– 2006– 2010–
2001 2005 2009 2013 2001 2005 2009 2013
During the past 30 days, about how many times (if any) have you bought cigarettes …
a. By having a friend or relative buy them for you 59.9 59.7 56.3 55.8 20.4 21.8 17.7 21.4
b. On your own from vending machines 10.3 7.1 3.6 4.1 7.1 5.8 5.8 4.4
c. Through the mail 1.3 1.9 1.5 2.5 0.5 2.4 1.8 1.8
d. In a store where you pick up the pack (or carton) and bring 32.5 21.9 17.0 17.1 60.4 35.4 31.9 27.5
it to the checkout counter
e. In a store where the clerk has to hand you the pack or 52.7 47.3 42.8 47.4 82.6 80.6 79.8 75.6
carton
f. Bought them in some other way — 20.2 21.1 25.3 — 13.3 12.7 11.7
During the past 30 days, about how many times (if any) did YOU buy cigarettes for your own use …
a. At a big supermarket 16.3 14.2 9.7 9.3 35.3 28.8 19.7 17.9
b. At a small grocery store 29.1 26.0 17.7 15.4 46.8 39.6 33.7 27.9
c. At a drugstore 14.8 13.6 12.9 15.5 23.8 22.1 19.1 19.9
d. At a convenience store (like a Hop-In or 7-Eleven) or a gas 58.7 57.2 49.2 53.3 81.1 77.3 77.6 72.5

station
e. From a website — 2.5 2.2 2.8 — 2.2 1.7 2.5

NOTE: Entries are percentages of current smokers reporting source of cigarettes.
SOURCE: Committee analysis of Johnston et al., 2014a.
can. “Having a friend or relative buy them for you” was the most often
cited method of access for 12th-grade smokers under the age of 18. For
12th graders who were 18 or older, the most cited method was to purchase
cigarettes for themselves. Twelfth graders under the age of 18 were less
likely than those 18 or older to say they purchased their own cigarettes, but
a considerable proportion did say they purchased their own cigarettes. For
example, in 2010–2013, 47.4 percent purchased cigarettes in a store where
the clerk had to hand them the pack or carton. With respect to the places
where students purchased their own cigarettes, convenience stores and gas
stations were clearly the most common, particularly for those 18 and older.
In both NYTS and MTF, a considerable portion of younger adolescents
reported obtaining cigarettes in “some other way.” These responses likely
include adolescents who are given cigarettes by family members but who
are reluctant to disclose this and thereby inculpate their relatives (CDC,
2014; Johnston et al., 2014b). Among adolescents who reported buying
cigarettes and were asked where (e.g., Table 5-3), the high rates of obtain-
ing cigarettes “from some other place not mentioned here” likely refers to
superstores (e.g., Kmart, Target, Walmart), which have proliferated recently
and do not fall into the other survey response categories (CDC, 2014).
With respect to trends, Tables 5-4a and 5-4b show that self-service
(i.e., purchasing cigarettes in a store where one can pick up a pack or
carton and bring it to a checkout counter) has declined considerably since
1997–2001 among all three grade levels. Purchases from vending machines
are also down, by about half in all groups. Purchasing cigarettes at a big
supermarket or at a small grocery store has declined considerably over time.
Purchases from a website have not changed noticeably and remain at very
low levels.
In addition to trends observed in survey data, limited empirical evi-
dence suggests that the relative reliance on different types of sources has
also changed over time. An analysis of access to cigarettes in the Minnesota
Adolescent Community Cohort (Widome et al., 2007) found that between
2000 and 2003 the likelihood of having obtained cigarettes from a com-
mercial source in the past month declined, while the likelihood of having
obtained cigarettes from a social source in the past month increased. A
New Zealand study (Gendall et al., 2014) of adolescents’ main source of
tobacco supply between 2006 and 2011, which further differentiated by
type of social source (friend, caregiver, or other), found a significant decline
in the percentage of adolescents ages 14 to 15 reporting friends as a main
source of cigarettes, significant increases in the percent reporting caregivers
and others as a main source, and no significant change in the percentage
who reported purchasing from a shop. This shift in sources likely reflects
the success of youth access restrictions at decreasing adolescents’ access to

tobacco from commercial retailers. The mechanism by which this may be

occurring is elaborated on in Chapter 6.
Unlike the case with cigarettes, adolescents were most likely to pur-
chase other tobacco products for themselves, followed by someone else
offering tobacco to adolescents. Table 5-5 shows the responses to the ques-
tion in the 2012 NYTS: “During the past 30 days, how did you get your
own cigars, cigarillos, or little cigars?” Among high school students who
reported smoking cigars, cigarillos, or little cigars in the past 30 days, the
most commonly reported answers were: I bought them myself (31.4 per-
cent); someone offered it to me (26.6 percent); I had someone else buy them
for me (25.5 percent); and I asked someone to give me one (14.4 percent).
Very few younger students reported smoking cigars, cigarillos, or little ci-
gars in the past 30 days, so data are not reported by age.
Table 5-6 summarizes the responses by current smokers in the 2012
NYTS, by gender, to the question, “During the past 30 days, where did
you buy your own cigars, cigarillos, or little cigars?” As was the case with
cigarettes, by far the most commonly cited specific source among all the
current smokers was a gas station or convenience store (44.0 percent). No
other specific source was cited by more than 6 percent of respondents. Re-
sponses did not vary much by gender.
TABLE 5-5 Methods for Obtaining Cigars, Cigarillos, or Little Cigars

Among High School Students, by Gender, NYTS, 2012
Percent of Smokers
Total Female Male
26. During the past 30 days, how did you

get your own cigars, cigarillos, or little
cigars?
a. I did not get cigars, cigarillos, or little — — —
cigars during the past 30 days
b. I bought them myself 31.4 25.1 34.7
c. I had someone else buy them for me 25.5 25.6 25.4
d. I asked someone to give me one 14.4 20.2 11.4
e. Someone offered it to me 26.6 33.9 22.7
f. I bought them from another person 4.9 5.8 4.5
g. I took them from a store or another person 4.8 3.5 5.5
h. I got them some other way 8.6 7.5 9.2

TABLE 5-6 Sources of Purchased Cigars, Cigarillos, or Little Cigars

Percent of Smokers
Total Female Male
27. During the past 30 days, where did you
buy your own cigars, cigarillos, or little
cigars?
a. I did not buy cigars, cigarillos, or little — — —
cigars during the past 30 days
b. A gas station or convenience store 44.0 39.5 46.3

c. A grocery store 5.1 4.8 5.3
d. A drugstore 5.6 5.1 5.9
e. A vending machine 1.9 1.2 2.3
f. Over the Internet 2.1 1.7 2.3
g. Through the mail 1.1 1.5 0.9
h. Some other place not listed here 17.3 18.3 16.7
Table 5-7 shows the responses to the question in the 2012 NYTS:
“During the past 30 days, how did you get your own chewing tobacco,
snuff, or dip?” Among high school students who reported using chewing
tobacco, snuff, or dip in the past 30 days, the most commonly reported
answers were: I bought it myself (32.0 percent); someone offered it to
me (27.7 percent); I had someone else buy it for me (26.4 percent); and I
asked someone to give me some (23.9 percent). Very few younger students
reported using chewing tobacco, snuff, or dip in the past 30 days, so the
data are not reported by age.
Table 5-8 summarizes the responses by current users of chewing to-
bacco, snuff, or dip in the 2012 NYTS, by gender, to the question, “During
the past 30 days, where did you buy your own chewing tobacco, snuff, or
dip?” Among all the current users, and similar to the case for both cigarette
and cigar purchases, by far the most commonly cited specific source was a
gas station or convenience store (43.4 percent).

TABLE 5-7 Methods for Obtaining Chewing Tobacco, Snuff, or Dip

Percent of Users
Total Female Male
32. During the past 30 days, how did you get
your own chewing tobacco, snuff, or dip?
a. I did not get chewing tobacco, snuff, or — — —
dip during the past 30 days
b. I bought it myself 32.0 23.9 33.5

c. I had someone else buy it for me 26.4 20.9 27.4
d. I asked someone to give me some 23.9 28.0 23.2
e. Someone offered it to me 27.7 30.4 27.2
f. I bought it from another person 8.6 12.2 7.9
g. I took it from a store or another person 7.0 12.5 6.0
h. I got it some other way 11.7 16.5 10.8
TABLE 5-8 Sources of Purchased Chewing Tobacco, Snuff, or Dip

Percent of Users
Total Female Male
33. During the past 30 days, where did you buy
your own chewing tobacco, snuff, or dip?
a. I did not buy chewing tobacco, snuff, or dip — — —

during the past 30 days
b. A gas station or convenience store 43.4 31.2 45.6
c. A grocery store 7.9 9.9 7.5
d. A drugstore 7.3 14.5 6.0
e. A vending machine 5.5 11.2 4.4
f. Over the Internet 3.4 7.4 2.7
g. Through the mail 3.2 5.1 2.8
h. Some other place not listed here 20.1 33.0 17.6

Finding 5-10: The proportion of underage youth reporting that they

obtained cigarettes from vending machines and from self-service dis-
plays has declined substantially since these practices were outlawed.
Finding 5-11: The proportion of underage youth reporting that they

obtained cigarettes in a face-to-face retail transaction has declined sig-
nificantly since 1997, while the proportion of underage users relying
primarily on social sources has increased since 1997, probably reflect-
ing increased retailer compliance with age verification requirements and
sales prohibitions.
Finding 5-12: Although twelfth graders and 16- to 17-year-olds find it

easier than younger teenagers to obtain cigarettes from a commercial
retailer, the proportion who are able to do so has steadily declined in
all age groups since 1997.
REFERENCES
62(45):893–897.
CDC (Centers for Disease Control and Prevention). 1993. Minors’ access to tobacco—Missouri,
1992, and Texas, 1993. Morbidity and Mortality Weekly Report 42(7):125–128.
———. 2014. National Youth Tobacco Survey. CDC, Office on Smoking and Health, National
Center for Chronic Disease Prevention and Health Promotion. http://www.cdc.gov/
tobacco/data_statistics/surveys/nyts (accessed February 7, 2015).
Choi, K., L. E. Fabian, B. Brock, K. H. Engman, J. Jansen, and J. L. Forster. 2014. Availability
of snus and its sale to minors in a large Minnesota city. Tobacco Control 23(5):449–451.
Cismoski, J., and M. Sheridan. 1993. Availability of cigarettes to under-age youth in Fond du
Lac, Wisconsin. Wisconsin Medical Journal 92(11):626–630.
Clark, P. I., S. L. Natanblut, C. L. Schmitt, C. Wolters, and R. Iachan. 2000. Factors associ-
ated with tobacco sales to minors: Lessons learned from the FDA compliance checks.
JAMA 284(6):729–734.
CTP (Center for Tobacco Products). 2014. Compliance and enforcement report. http://www.
fda.gov/TobaccoProducts/GuidanceComplianceRegulatoryInformation/ucm396171.htm
(accessed January 16, 2015).
Cummings, K. M., A. Hyland, J. Perla, and G. A. Giovino. 2003. Is the prevalence of youth
smoking affected by efforts to increase retailer compliance with a minors’ access law?
DiFranza, J. R. 1999. Are the federal and state governments complying with the Synar Amend-
ment? Archives of Pediatrics and Adolescent Medicine 153(10):1089–1097.
———. 2000. State and federal compliance with the Synar Amendment: Federal fiscal year
1997. Archives of Pediatrics and Adolescent Medicine 154(9):936–942.
———. 2005. Best practices for enforcing state laws prohibiting the sale of tobacco to minors.
Journal of Public Health Management and Practice 11(6):559–565.

DiFranza, J. R., and M. Coleman. 2001. Sources of tobacco for youths in communities with
strong enforcement of youth access laws. Tobacco Control 10(4):323–328.
DiFranza, J. R., and G. F. Dussault. 2005. The federal initiative to halt the sale of tobacco
to children—The Synar Amendment, 1992–2000: Lessons learned. Tobacco Control
14(2):93–98.
Eaton, D. K., L. Kann, S. Kinchen, S. L. Shanklin, K. H. Flint, J. Hawkins, W. A. Harris, R.
Lowry, T. McManus, D. Chyen, L. Whittle, C. Lim, and H. Wechsler. 2012. Youth risk
behavior surveillance—United States, 2011. Morbidity and Mortality Weekly Report
61(4):1–162.
Erickson, A. D., S. I. Woodruff, M. B. Wildey, and E. Kenney. 1993. A baseline assessment
of cigarette sales to minors in San Diego, California. Journal of Community Health
18(4):213–224.
FDA (Food and Drug Administration). 2010. 21 CFR, part 1140. Final rule. Federal Register
75(53):13225–13232.
———. 2014a. 21 CFR, parts 1100, 1140, and 1143. Proposed rule. Federal Register 79(80):
23142–23207.
———. 2014b. FDA tobacco retail inspection contracts. http://www.fda.gov/TobaccoProducts/
ResourcesforYou/ucm228914.htm (accessed October 31, 2014).
———. 2014c. Inspections, compliance, enforcement and criminal investigations. http://www.
fda.gov/ICECI/EnforcementActions/WarningLetters/default.htm (accessed October 31,
2014).
Fichtenberg, C. M., and S. A. Glantz. 2002. Youth access interventions do not affect youth
smoking. Pediatrics 109(6):1088–1092.
Gendall, P., J. Hoek, L. Marsh, R. Edwards, and B. Healey. 2014. Youth tobacco access:
Trends and policy implications. BMJ Open 4(4):e004631. http://bmjopen.bmj.com/
content/4/4/e004631.full (accessed February 8, 2015).
Glantz, S. A. 1996. Preventing tobacco use—The youth access trap. American Journal of
Hanson, K., D. Hatsukami, R. Boyle, and S. Brown. 2000. Minors’ access to snuff: Purchase
outcomes at corporate owned and independently owned convenience stores. Addictive
Behaviors 25(2):289–293.
HHS (Department of Health and Human Services). 1996. 21 CFR, parts 801, 803, 804, 807,
820, and 897: Regulations restricting the sale and distribution of cigarettes and smoke-
less tobacco to protect children and adolescents; final rule. Federal Register 61(168):
44396–44618.
Academy Press.
———. 2007. Ending the tobacco problem: A blueprint for the nation. Edited by R. J. Bonnie.
Monitoring the Future: National survey results on drug use 1975–2013. Volume 1: Sec-
ondary school students. Ann Arbor: Institute for Social Research, University of Michigan.
Monitoring the Future: National survey results on drug use: 1975–2013: Overview, key
Michigan.

Kuendig, H. 2011. Enforcement of regional bans on tobacco sales to minors in Switzerland—

Example of missed opportunity? European Journal of Public Health 21(6):750–752.
Lindblom, E. 2014. The health implications of raising the minimum age for purchasing to-
bacco products. Presentation at Committee on the Health Implications of Raising the
Minimum Age for Purchasing Tobacco Products, February 4, 2014. Washington, DC.
http://www.iom.edu/Activities/PublicHealth/TobaccoMinimumAge/2014-FEB-04.aspx
(accessed February 25, 2015).
Ling, P. M., A. Landman, and S. A. Glantz. 2002. It is time to abandon youth access tobacco
programmes. Tobacco Control 11(1):3–6.
NRC (National Research Council). 2015. Understanding the U.S. illicit tobacco market:
Characteristics, policy context, and lessons from international experiences. Washington,
DC: The National Academies Press.
Ribisl, K. M., A. E. Kim, and R. S. Williams. 2002. Are the sales practices of Internet cigarette
vendors good enough to prevent sales to minors? American Journal of Public Health
92(6):940–941.
Ribisl, K. M., R. S. Williams, Z. Gizlice, and A. H. Herring. 2011. Effectiveness of state and
federal government agreements with major credit card and shipping companies to block
illegal Internet cigarette sales. PLoS ONE 6(2):e16754.
Rigotti, N. A., J. R. DiFranza, Y. Chang, T. Tisdale, B. Kemp, and D. E. Singer. 1997. The
effect of enforcing tobacco-sales laws on adolescents’ access to tobacco and smoking
behavior. New England Journal of Medicine 337(15):1044–1051.
Rimpela, A. H., and S. U. Rainio. 2004. The effectiveness of tobacco sales ban to minors: The
case of Finland. Tobacco Control 13(2):167–174.
SAMHSA (Substance Abuse and Mental Health Services Administration). 1996. 45 CFR, part
96. Final rule. Federal Register 61(13):1492–1509.
———. 2014. FFY 2013 annual Synar reports: Tobacco sales to youth. http://www.samhsa.
gov/sites/default/files/synar-annual-report-2013.pdf (accessed October 30, 2014).
Schensky, A. E., S. S. Smith, D. L. Icenogle, and M. C. Fiore. 1996. Youth tobacco sale com-
pliance checks: Impact on vendor practices and community policy. Wisconsin Medical
Journal 95(11):775–778.
Smith, K. C., F. Stillman, L. Bone, N. Yancey, E. Price, P. Belin, and E. E. Kromm. 2007. Buying
and selling “loosies” in Baltimore: The informal exchange of cigarettes in the community
context. Journal of Urban Health 84(4):494–507.
Verdonk-Kleinjan, W. M. I., R. A. Knibbe, B. Bieleman, H. N. De Groot, and H. De Vries.
2008. The tobacco sales ban and tobacco purchases by adolescents: A general population
study in the Netherlands. European Journal of Public Health 18(5):498–503.
Widome, R., J. L. Forster, P. J. Hannan, and C. L. Perry. 2007. Longitudinal patterns of youth
access to cigarettes and smoking progression: Minnesota Adolescent Community Cohort
(MACC) study (2000–2003). Preventive Medicine 45(6):442–446.
Williams, R. S., K. M. Ribisl, and E. C. Feighery. 2006. Internet cigarette vendors’ lack of
compliance with a California state law designed to prevent tobacco sales to minors.
Archives of Pediatrics and Adolescent Medicine 160(9):988–989.


Evidence on the Effects of

Youth Access Restrictions
U
ltimately, the salient policy question concerning the minimum
age of legal access to tobacco products (MLA) is whether and to
what e xtent raising the MLA would reduce underage tobacco use.
Although several U.S. localities have raised the MLA to 19 and 21 years,
most of these actions have been done only very recently, and to date none
has been systematically evaluated.1 Furthermore, there have been only a
handful of natural experiments in which the MLA for tobacco has been
raised to 16 or 18, and they have taken place in other countries. Indeed,
most of the relevant literature pertains not to raising the MLA but rather to
enforcing an existing MLA more stringently. Therefore, conclusions about
raising the MLA to ages higher than 18 must be extrapolated from review
of other evidence on MLA laws and their enforcement as well as from
analogous policy interventions.
To address the question whether and to what extent raising the MLA
would reduce underage tobacco use, this chapter first reviews the limited
international studies investigating the effect of raising the MLA for tobacco
and then reviews evidence relating to the effects of raising the minimum
legal drinking age for alcohol as an analogous policy intervention in a par-
allel domain. The remainder of the chapter reviews the body of literature
1 Although Needham, Massachusetts, the first jurisdiction in the United States to raise the
MLA to 21, has been cited as having seen significant declines in tobacco use and tobacco-
related disease, there are no published data on these outcomes. In addition, the little available
data that exist (EDC, 2010a,b; NPHD, 2008, 2012) have no baseline measurements and are
confounded by the presence of other tobacco control measures that occurred in the town and
throughout the state of Massachusetts at the same time the MLA was increased.
155

investigating the effects of enforcing current youth access restrictions in

the United States. Although these studies are beset by many challenges and
limitations, they enabled the committee to reach some general conclusions
about the nature and direction of the effects of enacting and enforcing a
tobacco MLA, even though they do not provide a basis for estimating the
precise magnitude of such effects. As an aid to interpreting this body of
research, the committee developed a logic model identifying the behavioral
mechanisms through which an MLA policy and its enforcement against
commercial retailers would be expected to affect underage tobacco use. The
committee believes that this body of scientific literature provides a reason-
able predicate for policy making in the absence of direct evidence regarding
the effectiveness of raising the MLA. It is used in Chapter 7 to inform the
committee’s judgment about the probable effects of raising the MLA on
the initiation of tobacco use by underage youth.
THE IMPACT OF ENACTING OR RAISING THE MINIMUM

LEGAL AGE TO PURCHASE TOBACCO PRODUCTS
Only a small number of studies have examined the effects of enacting
or raising an MLA on underage tobacco use. All of these studies have come
from international experience: one from Finland (Rimpela and Rainio,
2004) and two from the United Kingdom (Fidler and West, 2010; Millett
et al., 2011).2 All of the studies that investigated the effect of the policy on
tobacco use reported decreases in underage smoking prevalence.
Rimpela and Rainio (2004) examined the effect in Finland of enact-
ing an MLA of 16 in 1977 and increasing it to 18 in 1995. Adolescent
tobacco outcomes were assessed using a biennial, nationally representative
postal survey of adolescents (ages 12, 14, 16, and 18) for 1977–2003 as
well as an annual postal survey of eighth and ninth graders (ages 14–16)
for 1996–2003. Following implementation of the original MLA legisla-
tion in 1977, there was a significant—but small and short-term—decrease
both in tobacco purchases from commercial sources and in tobacco use.
After the MLA increased to 18 in 1995, there was no immediate effect
on tobacco use. However, after a 2000 revision of the MLA policy re-
quiring tobacco retailers to develop and implement an enforcement plan
to prevent sales to underage persons, experimental smoking and later daily
smoking decreased significantly among adolescents ages 14 and 16 (i.e.,
2 Another small qualitative study (Borland and Amos, 2009) examined attitudes about
raising the MLA from 16 to 18 in Scotland among 16- to 17-year-old regular smokers who
had dropped out of high school and were attending a work skills program. However, given
the small sample size and sample characteristics, these findings are likely not generalizable to
larger or different populations. This study also provides no findings on the effect of raising the
MLA on either reducing sales to adolescents or reducing underage tobacco use.

EVIDENCE ON THE EFFECTS OF YOUTH ACCESS RESTRICTIONS 157
those targeted by the policy). Only experimental smoking decreased among

12-year-old adolescents, but the sample was too small for the daily smok-
ing category. There was no change in either experimental or daily s moking
observed among those ages 18 and older (i.e., those not targeted by the
policy). In addition, Rimpela and Rainio (2004) found that purchases
from commercial sources decreased, while obtaining tobacco from social
sources (i.e., purchasing or being given tobacco from relatives, friends, or
strangers) increased. Consistent with that finding, the frequency with which
18-year-olds, of legal age, also reported purchasing tobacco for friends was
greater in 1999 than it had been in the 1970s. There were also changes in
perceived access to tobacco: The proportion of adolescents reporting that
it was rather difficult or very difficult to purchase tobacco was higher after
the MLA increase, but the proportion of students reporting that it was very
easy or fairly easy to purchase tobacco from commercial sources never
theless remained high (72 percent in 2002–2003). In sum, these findings
suggest that, among adolescents, raising the MLA decreased the amount of
tobacco available from commercial sources, increased difficulty of obtain-
ing tobacco, and reduced tobacco use despite adolescents having continued
access to social sources.
Fidler and West (2010) assessed the effects on smoking prevalence of
an increase in MLA from 16 to 18 in 2007 in England and Wales. Smoking
outcomes were assessed using data from monthly cross-sectional household
surveys of a representative sample of adults ages 16 and older. Following
the 2007 increase in the MLA, smoking prevalence decreased significantly
among all ages. This decline occurred against the background of a larger
societal trend of an overall decrease in smoking prevalence, but the great-
est percentage decrease during this period was seen among those of ages
16–17 (a 7.1 percent decrease) compared to those 18 and older (2.4 percent
decrease), suggesting that raising the MLA did indeed decrease smoking
prevalence beyond secular trends. Moreover, smoking prevalence was sig-
nificantly higher among those 18 and over, and this difference in prevalence
by age was significantly greater after the MLA increase than it had been
before, suggesting that the MLA increase was successful in at least delaying
initiation.
Millett and colleagues (2011) examined the effects of the same 2007
legislation that Fidler and West studied, but they looked at it in England,
Scotland, and Wales, among younger ages (11–15) and by socioeconomic
status (SES). Smoking outcomes were assessed using data from a national
school-based survey of students in grades 7 to 11 from 2003 to 2008,
excluding 2007, the year of the MLA increase. The effect of the policy on
socioeconomic smoking disparities was assessed by comparing students
who were eligible for free school meals (a proxy measure for low SES be-
cause eligibility for free school meals is assessed using parental employment

and income) with those not eligible. The study found that the MLA increase
was associated with a significant reduction in regular smoking among all
adolescents, with no differences found between those eligible and ineligible
for free school meals. The study also assessed the effects of raising the MLA
on perceived access to tobacco from retailers. After the MLA increase, the
proportion of students who smoked regularly and perceived that purchasing
cigarettes from a shop was difficult did not increase among those eligible for
free school meals, but it did increase significantly among those not eligible.
At the same time, the percentage of students reporting that purchasing ciga-
rettes from a shop was easy did not change from before the MLA increase
to afterwards. These findings suggest that increasing the MLA decreased
tobacco use overall and that the decrease was neutral with respect to SES.
LESSONS FROM ALCOHOL

Given the paucity of directly relevant data from prior experience with
raising the minimum age for tobacco, the committee recognized the op-
portunity to look at similar domains, most obviously alcohol, to see what
lessons might be learned. The United States had direct experience with
raising the minimum legal drinking age (MLDA) for alcohol from roughly
18 (with some variation across states) to a national standard of 21 years
of age, and that experience came recently enough that the country has not
changed dramatically in the interim but long enough ago for there to be
an extensive literature evaluating its consequences. Furthermore, different
states implemented the change at different times, resulting in a stronger
basis for causal inference than if all had acted simultaneously.
Tobacco is, of course, different from alcohol in myriad ways. T obacco
products are psychoactive, but they are not intoxicants. Alcohol has been
embedded within human culture for millennia, whereas modern, mass-
produced tobacco products (namely, cigarettes) are, comparatively speak-
ing, a relatively recent phenomenon. And, of course, the mechanism of
consumption, the neural pathways triggered, the patterns of use and ces-
sation, and various other specific details differ in a variety of ways. So
one could hardly observe a point estimate of the reduction in alcohol use
following the raising of the MLDA for alcohol and imagine that same num-
ber would necessarily be a best estimate for the corresponding reduction
one might expect from increasing the MLA for tobacco products.
Nevertheless, there are obvious similarities between the two products,
their legal status, and their industries’ practices. Both are dependence-
inducing substances that are legal for adults but subject to legal and social
constraints on underage use. Both are relatively inexpensive and widely
used by both adults and underage users. Both cause very large numbers of
premature deaths. Both are marketed aggressively by industries that have,

at least in some product classes, a high degree of market concentration.

(Beer and cigarettes have higher degrees of market concentration than do
wine or cigars.3) And, as is being considered with tobacco, the MLDA for
alcohol was increased to 21—not to other ages and, particularly, not to
higher ages.
In brief, the experience with raising the MLDA for alcohol is highly
suggestive with respect to the prospects that raising the MLA for tobacco
will appreciably reduce smoking rates. Kypri et al. (2006, p. 126) go so far
as to say, “No traffic safety policy, with the possible exception of motor-
cycle safety helmet laws, has more evidence for its effectiveness than do the
minimum legal drinking age laws.”
Of course, underage drinking still occurs, and it seems clear that if
the MLA for tobacco is increased, there will still be some tobacco use by
those under the legal age. Indeed, it would be unreasonable to expect that
raising the MLA could completely eliminate all underage use. However,
if the question is simply whether raising the MLA will noticeably reduce
the use and use-related harms of tobacco among youth, then the academic
literature evaluating the alcohol experience indicates that there will indeed
be substantial benefits (e.g., Dejong and Blanchette, 2014; McCartt et al.,
2010; Wagenaar and Toomey, 2002).
It is worth briefly mentioning the historical context. Following the re-
peal of national alcohol prohibition in 1933, MLDAs were set by the states,
typically at 21. In the early 1970s, 29 states lowered their MLDAs to 18,
19, or 20. In response to increasing highway traffic fatalities, some states
reversed course, and then in 1984 Congress passed the National Minimum
Drinking Age Act (NMDAA). The NMDAA does not prescribe an MLDA
of 21. Rather it encourages states to raise their MLDA to 21 by withhold-
ing a percentage of federal highway dollars if they fail to do so. By 1988
all states and the District of Columbia had an MLDA of 21.
By some measures, alcohol MLDAs are enforced fairly aggressively.
For example, it is common for mere possession of alcohol to be an offense
(known as a “minor in possession”). The severity of sanction for such of-
fenses varies by state but can include fines and the loss of one’s driver’s
license. Likewise, social host ordinances can lead to severe penalties for
other individuals (not just businesses) who provide alcohol to underage
drinkers. On the other hand, in many states parents are allowed to pro-
vide alcohol to their children. So comparisons between the intensity of the
enforcement of alcohol MLDAs and the intensity of enforcement of either
3 The Centers for Disease Control and Prevention reports that the market share of the
dominant cigar firms is mostly below 20 percent even for specific types of cigars, and different
firms dominate those different segments, whereas three companies control nearly 85 percent
of the cigarette market (CDC, 2014a).

current or potential future tobacco MLAs are not straightforward, but on

the whole, the severity of sanctions for the alcohol equivalent of purchase–
use–possession (PUP) laws for tobacco may be greater than what might be
contemplated under at least some scenarios involving raising the MLA for
tobacco products.
The literature evaluating the effects of changing the MLDA for alcohol
is large. DeJong and Blanchette (2014), McCartt et al. (2010), and Wagenaar
and Toomey (2002) offer useful reviews. The trends that are observed in
aggregate descriptive statistics are consistent with the idea that raising the
MLDA has an effect on alcohol use. Specifically, the rates of drinking and
binge drinking among those under 21 have been in sustained long-term
decline since the MLA was raised, the death rates of 18- to 20-year-olds in
nighttime driving accidents have fallen, and the rates of problem alcohol
use are lower in the United States than they are in Europe, where drinking
ages are lower. However, such correlations could be coincidental. The more
persuasive comparisons involve looking at neighboring birth cohorts who
reached an MLDA just before versus after the MLDA changed and looking
at patterns of use by people who are just a little younger versus just a little
older than a given MLDA (Carpenter and Dobkin, 2011). For example,
the rates of binge drinking are appreciably higher for 21-year-olds than for
20-year-olds (SAMHSA, 2009).
A number of these studies have found that raising the MLDA for al-
cohol reduced consumption and consumption-related harms, with the esti-
mate of nearly 1,000 premature deaths prevented per year being a typical
number. Other studies have found no statistically significant effect (perhaps
from a lack of statistical power), and a few outliers have found that con-
sumption increased. For example, Wagenaar and Toomey (2002) reported
that of 33 high-quality empirical analyses for which consumption was the
outcome measure, 11 found that raising the MLDA decreased consump-
tion, and only one found the opposite. The proportion of studies finding a
favorable effect on traffic crashes was even greater (DeJong and Blanchette,
2014). An illustrative study, conducted by Shults et al. (2001), found that
raising the MLDA reduced fatal and nonfatal crashes by 16 percent. Other
studies identified less obvious outcomes. For example, Birckmayer and
Hemenway (1999) estimated that raising the MLDA reduces teen suicide
and, conversely, that lowering it from 21 back to 18 could lead to approxi-
mately 125 additional suicides per year among 18- to 20-year-olds.
DeJong and Blanchette’s (2014) review includes international compari-
sons. Notably, New Zealand reduced its MLDA from 20 to 18 in 1999,
and Huckle and Parker (2014) and Kypri et al. (2006) reported that this
led to significantly more alcohol-related crashes among 15- to 19-year-olds.
Conover and Scrimgeour (2013) found similar effects on alcohol-related
hospitalizations among those newly eligible to purchase alcohol. Interna-

tional studies on the direct effect of MLDA on alcohol consumption come

to comparable conclusions. The 2004 report by the National Academies
on underage drinking (IOM and NRC, 2004) found that lower drinking
ages in European countries were associated with higher rates of drinking,
problem drinking, and drinking by underage individuals, despite common
conceptions that underage users drink less in Europe.
The experience with alcohol also suggests that raising the MLDA
may even affect patterns of consumption for people who are over the new
MLDA (Norberg et al., 2009). For example, Plunk et al. (2013) argue that
the ability to purchase alcohol before age 21 increases rates of binge drink-
ing later in life, although the overall drinking frequency is not changed
because the increase in binge drinking is accompanied by a reduction in
non-heavy drinking. O’Malley and Wagenaar (1991), as reported in DeJong
and Blanchette (2014), found that high school seniors and recent high
school graduates drank less when the MLDA was 21 and that they also
drank less throughout their early 20s, after they had reached the legal age.
Summary
Although alcohol and tobacco have considerable differences, they are
similar products in many respects. As such, U.S. and international experi-
ence with enacting and raising the minimum legal drinking age may provide
insights into the potential effects of raising the minimum age of legal access
to tobacco products. In particular, experience with alcohol suggests that
raising the MLDA has reduced consumption behaviors among adolescents
and adults as well as reducing alcohol-related adverse events.
Finding 6-1: Evidence from U.S. experience with alcohol has shown
that raising the minimum legal drinking age for alcohol, coupled with
rigorous enforcement and penalties for violations, has been associated
with lowered rates of alcohol consumption among adolescents and
adults as well as with reduced rates of alcohol-related adverse events
(e.g., traffic crashes and hospitalizations).
A LOGIC MODEL FOR PREDICTING THE EFFECTS OF AN MLA

In light of the dearth of literature on the question of interest—whether
raising the MLA for tobacco would reduce underage use—and acknowledg-
ing the indirect analogy of the U.S. experience with alcohol, the committee
next focused on the scientific literature bearing on the effects of enforcing
the existing MLA and other retailer interventions on underage access to
and use of tobacco. In order to organize and interpret this literature, the
committee developed a logic model to examine whether and to what extent

laws restricting the commercial retail availability of tobacco products to

underage persons, and MLA laws in particular, have the potential to reduce
underage tobacco use. This logic model (see Figure 6-1), which draws on
bodies of research on legal deterrence and behavioral economics, details
the behavioral mechanisms by which an MLA policy is expected to reduce
underage tobacco use. According to the most simplified form of the model,
an MLA policy is expected to affect the behavior of potential users and
distributors by declaring social norms and by deterring illegal behavior.
Deterrence depends on an expectation among sellers that the law will be
enforced. Enforcement is expected to increase retailer compliance with the
MLA law. High levels of retailer compliance are expected to reduce retail
tobacco availability to underage individuals, which in turn is expected to
reduce underage tobacco use. The logic model is described in more detail
below.
Declarative Effects and Deterrent Effects of Legal Restrictions

An MLA law can affect the behavior of tobacco retailers (and other
sellers) in two ways. First, it may have a “declarative” effect on both re-
tailers and potential underage users because they are disposed to comply
with legal norms or because enactment of the law affects their beliefs and
attitudes toward tobacco use by minors (Bonnie, 1982; IOM and NRC,
2004). Second, the law and its anticipated enforcement may deter potential
violators from using or selling tobacco by communicating a credible threat
of detection and punishment for violations. The variables that are expected
to affect the likelihood of a violation by the targeted population are the
probability of detection, the severity of the expected punishment (a func-
tion of the prescribed punishment and the probability of its imposition),
and the swiftness with which the penalty is imposed. “General deterrence”
refers to the effect of the perceived threat of enforcement and punishment
on the target population of potential sellers or users. “Specific deterrence”
refers to the effect of the imposition of sanctions on detected violators.
Figure 6-2 shows a somewhat expanded view of the logic model detailing
these enforcement mechanisms.
Decreased Decreased
Minimum Increased
Active retail tobacco underage
legal age retailer
enforcement availability to tobacco
policy compliance
underage youth use
FIGURE 6-1 Simplified logic model of the effects of prescribing and enforcing a
minimum age of legal access to tobacco products.

Declarative effect
Specific deterrence (retailer

detected by enforcement and
sanctioned)
Enforcement of
MLA law
• Compliance checks Decreased
• Penalties Decreased
Minimum retail tobacco
Retailer underage
legal age availability to
compliance tobacco use
policy underage
Threat of penalty youth
from enforcement
of MLA law
General deterrence (NOT

detected or sanctioned)
FIGURE 6-2 Expanded view of the logic model detailing enforcement mechanisms.
Deterrence will not occur unless potential violators perceive a credible

threat of detection and punishment (hereafter, “enforcement”). Accord-
ingly, any MLA law will need to be actively enforced using random compli-
ance checks to maintain the retailers’ perception that there is a significant
risk that an illegal sale will be detected. The compliance checks are expected
to have a “specific” deterrent effect on tobacco retailers caught selling to
underage individuals, increasing the likelihood of future compliance by
these retailers. In addition, an awareness of the possibility of such checks
is expected to deter violations by the entire population of retailers who be-
lieve themselves to be at risk of compliance checks (“general deterrence”).
If MLA laws achieve high rates of compliance (through the combined ef-
fect of declaring the legal norm and enforcing it), tobacco availability to
underage consumers from commercial retail sources will likely be reduced.
If these effects were complete, underage users would not be able to obtain
tobacco from retailers. However, the more likely scenario is that enforce-
ment increases the number of compliant retailers (or clerks) and increases
the “search-time” costs incurred by underage users who are looking for a
noncompliant retailer (or clerk).
Reducing Availability by Increasing Search-Time Costs

To fully understand the effects of the MLA enforcement on search-time
costs, it is first necessary to consider that commercial retailers are only one
among a range of sources from which underage users obtain tobacco. As

described in Chapter 5, other sources include commercial sellers such as

Internet vendors and non-licensed retail sellers (i.e., black market traffick-
ing of untaxed tobacco and single/unpackaged/loose cigarettes) and also
social sources. “Social sources” are primarily relatives and peers who give
tobacco to underage users or else proxy purchasers (including strangers as
well as relatives and peers) who purchase cigarettes on behalf of underage
users and are paid a small fee (e.g., a few dollars or a portion of tobacco)
for their service. If the law applies only to retailers or is not enforced against
noncommercial providers (i.e., social sources), it is likely that any decrease
in retail tobacco availability will result in a corresponding increase in access
from social sources, although this shift is likely to be incomplete. Neverthe-
less, if overall tobacco supply to underage users is successfully reduced, it
is likely that the overall cost of tobacco will increase to the underage users
who purchase tobacco outside the retail market. Together with the increase
in “search-time” costs, this increase in monetary cost will make tobacco
products more expensive and will likely reduce the demand for the products
by underage users, thereby reducing consumption.
One of the most basic and widely documented empirical regularities
in economics is the so-called law of demand, which is typically stated
as, “All else being equal, when the price of a good goes up, consumers
demand less of it.” There is ample literature documenting that the law
of demand applies to tobacco products (Chaloupka and Warner, 2000),
including for adolescents in the United States (Carpenter and Cook, 2008;
Ross and Chaloupka, 2003) and abroad (e.g., Kostova et al., 2011; Nikaj
and Chaloupka, 2014; Sen and Wirjanto, 2010). Indeed, there is consider-
able, although not unanimous, evidence that adolescents are more price
responsive than are older smokers (e.g., Ding, 2003; Franz, 2008; Harris
and Chan, 1999).
Although the law of demand is often stated informally in terms of
“price,” which would connote the monetary price paid by the customer to
the seller, the proper interpretation is broader. The underlying behavioral
model is that whenever the total opportunity cost of obtaining the good
goes up, then the quantity demanded will go down. This total cost includes
the monetary price, of course, but it also includes other costs such as the
time and inconvenience of locating the seller and consummating the trans-
action, which is sometimes referred to as search-time costs.
The modern American economy often offers low search-time costs; the
very term “convenience store” derives precisely from the idea that those
stores enable customers to obtain their goods quickly and easily. However,
search-time costs can dramatically affect market outcomes both in general
economic models (e.g., Stahl, 1989) and, specifically, for drugs whose pur-
chase is banned. Indeed, these costs can be important even for illegal drugs,
such as heroin, whose monetary price is so high that one might expect it

to dominate other considerations because sellers have difficulty advertising

directly to banned customers, and parties to a transaction seek to avoid
being detected by the authorities (Moore, 1973; Rocheleau and Boyum,
1994).
Penalties for Users

Raising the MLA can be understood as an attempt to raise these non-
monetary costs for individuals who were not underage under the previous
policy but are underage under the new policy and who are trying to obtain
tobacco.4 The costs are not infinite, so purchase and acquisition by those
under the legal age will not go to zero. But it is more convenient simply
to walk into a store and purchase what one wants than it is to enlist the
assistance of a proxy purchaser, and the counterfeit brands available from
the Internet and black markets are not always equivalent in the qualities
that smokers value.
Similarly, at least in theory, the demand for illegal drugs could be tem-
pered by increasing the risk that users will be apprehended and punished
for possession. There is some debate as to how effective that particular
deterrent is, however. Specifically, while some find that decriminalization
affects use (e.g., Model, 1993), others have argued that decriminalizing
prohibited drugs will not meaningfully increase demand (Bonnie, 1982;
Hughes and Stevens, 2010; MacCoun and Reuter, 2001), and still others
argue that the term “decriminalization” covers such a wide range of ac-
tions that generalizations concerning its effects are suspect (Pacula et al.,
2005). Presumably, however large the deterrent effect for illegal drugs,
it could well be smaller for underage tobacco use because the sanctions
imposed under purchase–use–possession laws tend to be much less severe
than the maximum sentences permitted for possession of illegal drugs (see
IOM, 2007). Nevertheless, to the extent that PUP provisions exist and are
enforced, raising the MLA could also be seen as increasing that aspect of
the total cost of underage smokers acquiring cigarettes.
Measures of Availability
To assess the overall effect of MLA laws and their enforcement on use,
tobacco availability as a mediating variable can be assessed in two ways.
First, it can be assessed as the “observed availability” measured as the num-
4 In certain circumstances they may also increase the monetary cost (e.g., if only a subset of
retail stores are willing to sell to underage customers, and that restriction makes it harder for
youth to shop for the best prices, or when a youth enlists a proxy purchaser and the proxy
purchase charges a fee for that service).

ber of noncompliant retailers (or rates of violations) within a specified area.

Second, it can be measured as “perceived availability,” a subjective measure
of how easy or difficult it is for underage users to get cigarettes, which is
also a reflection of an underage individual’s willingness to seek out tobacco
products and take up smoking. Presumably, “perceived availability” bears
some relationship to the perceived difficulty of accessing tobacco from retail
sources, which might be a function of the observed availability; whatever
its relationship with observed availability, however, reducing the perceived
availability itself may also serve as a mechanism that decreases demand,
deterring underage users from purchasing and using tobacco. This includes
deterring underage individuals from taking up or escalating smoking as well
as increasing likelihood of quitting. In addition to this pathway through
perceived availability, there is also likely to be a direct effect of the MLA
policy on underage tobacco use through the declarative effect of enacting
and enforcing the higher MLA as well as through any effects of enforcement
against users. Because social sources are another principal means of access-
ing tobacco, the impact of restricting retail access on the use of social sources
and the corresponding implications for the success of an MLA policy also
must be considered. Figure 6-3 illustrates the complete logic model, includ-
ing pathways through these various measures of availability.
Declarative effect
Specific deterrence (retailer detected by

enforcement and sanctioned) Concentration (density &
proximity) of licensed
tobacco retail outlets
Enforcement of MLA Underage tobacco
law use behaviors:
• Compliance checks • Status (never/
• Penalties Actual/measured access
Minimum current/ former
legal age use)
Retailer Concentration (density • Transitions
policy
compliance & proximity) of non- (initiation,
compliant licensed progression to
Threat of penalty tobacco retail outlets established use,
from enforcement Perceived
availability cessation)
of MLA law • Intensity
of retail
tobacco to (frequency &
Other (non-licensed) quantity)
General deterrence (NOT underage
commercial/retail sales
detected or sanctioned)
• Illicit (wholesale)
trafficking
Laws against other
commercial sales to
underage
Internet sales Demand
Laws against Internet
sales to underage
Other sources:
• Social sources
Laws against other (family & peers)
• Proxy sales Other individual
distribution to underage (e.g. and social
proxy sellers, social sources) Threat of
determinants
enforcement of
penalty for
minor
FIGURE 6-3 Complete logic model of the effects of prescribing and enforcing a
minimum age of legal access to tobacco products.

The Tobacco Control Context

Youth access restrictions are implemented in the context of other
t obacco control programs. However, community-level natural experiments
and controlled experiments may not take into account the potential contri-
butions of existing tobacco control programs that may enhance or mitigate
the potential effects of increasing the MLA. Thus, this chapter also reviews
literature on the effects of an MLA policy and its enforcement in the context
of other tobacco control programs aimed at preventing or reducing tobacco
use among adolescents and young adults and across the population at large.
EFFECTS OF RETAILER INTERVENTIONS ON

ACCESS TO AND USE OF TOBACCO
Within the framework of the logic model, this section reviews the sci-
entific literature bearing on the effects of retailer interventions on underage
access to and use of tobacco. The first subsection summarizes studies that
assess the impact of enforcement activities on retailer compliance. Even if
the number of noncompliant stores (or unsuccessful purchase attempts) is
reduced, the question remains whether increased retailer compliance has a
discernible impact on the availability of tobacco to underage users and, if
so, whether it reduces underage use. The next two subsections summarize
studies addressing these two questions. The section closes with method-
ological observations.
Effect of Retail Enforcement and

Other Interventions on Retailer Compliance
In the previous chapter (Chapter 5), the committee reviewed the current
status of federal, state, and local youth tobacco access laws in the United
States as well as their enforcement under the Synar and the Food and
Drug Administration (FDA) inspection contract programs. As discussed,
the rates of illegal sales to minors under the Synar program have decreased
significantly over the past 20 years. However, these data are challenging to
assess because of a number of factors. For one, these data are derived from
compliance protocols that can vary significantly by locality in terms of the
frequency of inspections, the number of reinspections of a particular re-
tailer, the characteristics of the sales clerk and underage decoy, and the time
of day of purchase, among other factors. In addition, a number of other
factors aside from inspection protocol, such as the total number of inspec-
tions in a region, whether neighboring retailers have been inspected, and
whether a retailer has previously been cited for violations, may also influ-
ence compliance rates. Variations in each of these factors may influence a

state’s or locality’s compliance rate. As such, it is difficult to compare data

collected under different enforcement programs and data collected over
time as well as to assess the impact of such compliance data on underage
tobacco use. It is also difficult to compare data across geographic regions
of both the same and different scales (e.g., across states or from the local
to state or state to national levels). These variations are equally a factor in
observational studies on the effects of youth access restrictions.
Notwithstanding these limitations, it is possible to draw out several
general findings from this body of research regarding the nature and direc-
tion of the effects of enforcing youth access restrictions against retailers, if
not their magnitude. It is clear, first of all, that restrictions on youth tobacco
access are much more seriously and consistently enforced and complied
with now than they were two decades ago, when they were first imple-
mented. Early studies (CDC, 1993; Cismoski and Sheridan, 1993; Erickson
et al., 1993) examining the effects of enacting an MLA law reported high
rates of sales, suggesting that tobacco retailers will not comply with MLA
laws absent of active enforcement. Studies of experiences in other countries
(Kuendig, 2011; Sanson-Fisher et al., 1992; Sundh and Hagquist, 2004,
2006, 2007) report similar findings. However, both the sales rates reported
to Synar and the limited scientific evidence suggest that active enforcement
of youth access restrictions using compliance checks paired with penalties
for violations are effective at increasing retailer compliance with youth
access laws. However, evidence bearing on the relationship between the
intensity of enforcement and the rate of compliance is inconsistent.
General Deterrence
Most studies evaluating enforcement programs investigate the effect of
these programs on the rates of illegal sales by retailers to underage buyers.
These studies support the existence of general deterrence stemming from
the threat of compliance checks. The studies are typically conducted at the
town level and evaluate sales rates before and after the implementation of
an active enforcement program. Most of these studies reported some reduc-
tion in sales rates following the implementation of enforcement activities,
but the reported declines in underage purchases varied, ranging from less
than 10 percent to as high as 68 percent (e.g., DiFranza et al., 2001a; Jason
et al., 1991, 1996, 1999a; Junck et al., 1997; Ma et al., 2001; Mawkes
et al., 1997; Pokorny et al., 2008; Rigotti et al., 1997; Tangirala et al.,
2006). In addition to looking at the rates of illegal sales, some studies (e.g.,
CDC, 1996; Cummings et al., 1998; DiFranza et al., 2001a,b; Schofield et
al., 1997) examined the effect of enforcement activities on other measures
of retailer compliance (e.g., more frequent and consistent age verification
using photographic identification, displaying requisite warning signs, and

adherence to other point-of-sale marketing and advertising restrictions)

and typically reported that enforcement increased compliance with these
other requirements as well. A small number of studies (e.g., Bagott et al.,
1998; Cummings et al., 1998; Gemson et al., 1998) compared compliance
rates in jurisdictions with active enforcement to those without, and findings
were mixed.
Specific Deterrence
Several studies examined the effect of multiple or repeated inspections
on the same vendor. Each of these found that prior checks increased future
compliance, typically measured by reduced likelihood of future illegal sales
(Jason et al., 1996; Pearson et al., 2007; Schensky et al., 1996), while one
found increased age verification but no effect on sales (Cummings et al.,
1998). Taken together, these studies suggest that active enforcement using
compliance inspections may have the specific deterrent effect of increasing
compliance among retailers who have been detected and sanctioned for
illegally selling tobacco to minors as well as a general deterrent effect of
increasing retailers’ perceived threat of enforcement.
Retailer Education
Targeted retailer education has also been employed as a strategy to
increase retailer compliance with the MLA laws, either in lieu of or in
addition to active enforcement. Such education may include direct mailings
with information about the MLA law and potential penalties for violations,
personal visits delivering education kits and other resources, phone calls
presenting information, and letters from senior government officials (e.g.,
the mayor or police chief). Studies of retailer education are mixed. Many
(e.g., Abernathy, 1994; Altman et al., 1989, 1991, 1999; Dovell et al., 1998;
Feighery et al., 1991; Gemson et al., 1998; Keay et al., 1993; Naidoo and
Platts, 1985; Wildey et al., 1995; Woodruff et al., 1993) have found educa-
tion effective at increasing compliance as measured by decreases in the rates
of illegal sales, although some (e.g., Forster et al., 1992; McDermott et al.,
1998; Schofield et al., 1997) have found no effect. Other studies have found
that education increases compliance with other requirements—for instance,
age verification (Krevor et al., 2011) and warning signs (Skretny et al.,
1990). One study that specifically investigated an education intervention
alone compared with the education intervention combined with enforcement
(Feighery et al., 1991) observed a slight reduction in sales following the
education-only intervention and a much larger reduction when enforcement
was added. As such, retailer education programs appear to be more effective
when reinforced by enforcement activities than when implemented alone.

Comprehensive Youth Sales Interventions

Comprehensive youth sales interventions comprise the active enforce-
ment of MLA laws, retailer and community education programs, and mass
media campaigns. For example, Forster and colleagues (1998) conducted
the Tobacco Policy Options for Prevention study, a controlled experiment
that mobilized treatment communities to introduce, pass, and enforce a
youth tobacco access ordinance. This and other studies examining the ef-
fect of such comprehensive MLA interventions on illegal tobacco sales to
underage users (e.g., Altman et al., 1999; Biglan et al., 1995, 1996; CDC,
1996; Cook et al., 2000; Glanz et al., 2007; Kan and Lau, 2010; Landrine
et al., 2000; Tutt et al., 2009; Watson and Grove, 1999) all found that
comprehensive interventions are effective at decreasing sales.
Summary
Limited evidence suggests that the active enforcement of MLA laws
sing random, unannounced compliance checks of tobacco retailers and
u
sanctions for violations tend to reduce underage sales and, as a result,
probably reduces the availability of tobacco to underage individuals from
commercial tobacco retailers. Furthermore, additional measures such as
targeted retailer education about sales laws, community education and
mobilization, and mass media campaigns appear to bolster the effect of
enforcement activities on increasing retailer compliance. However, evidence
on the relationship between the intensity of the enforcement of the tobacco
MLA restrictions and retailer compliance is slim.
Finding 6-2: Active enforcement of restrictions on the minimum age

of legal access to tobacco products, including meaningful penalties for
violations, increases retailer compliance, and a reasonable inference can
be drawn that enforcement decreases the availability of retail tobacco
to underage persons. These effects can be increased by coupling en-
forcement with retailer and community education programs and media
campaigns about the minimum age policy.
Relationship Between Retail Interventions and Underage Tobacco Use

While the evidence concerning the effects of enforcement of the MLA
policies on retailer compliance inferentially supports the effectiveness of the
MLA policy, this finding does not directly answer the ultimate question of
interest: whether increased retailer compliance is associated with reduced
underage use. Three types of studies bear on this question: those investi-
gating whether the intensity of retailer enforcement is related to the levels

of underage use; those investigating whether levels of retailer compliance

are related to levels of use; and those investigating whether comprehensive
youth access programs, including retailer interventions, have an impact on
underage use.
Studies attempting to ascertain the effects of retail enforcement on under
age tobacco use primarily examine town-level interventions. W oodridge,
Illinois, was one of the first jurisdictions to restrict youth access to tobacco
from retailers; in 1989 it passed a cigarette ordinance with licensing, enforce-
ment using compliance checks, and possession provisions. An observational
study (Jason et al., 1991) assessing the impact of this ordinance on middle
school smoking rates found significant reductions in experimental smoking
(from 46 percent to 23 percent) and in smoking (from 16 percent to 5 per-
cent) nearly 2 years later. Follow-up studies nearly a decade later (Jason et
al., 1999a,b) found that low rates of regular smoking among middle school
students had been maintained (5.3 percent in 1997) as well as similarly
low rates of experimental (15.4 percent), social (19.5 percent), and regu-
lar (8.1 percent) smoking among high schoolers. Moreover, this rate of regular
smoking—8.1 percent—in W oodridge, where there was active enforcement,
was significantly lower than in towns in the same region that lacked active
enforcement (15.5 percent). Similar results were observed elsewhere (e.g.,
Cook et al., 2000; DiFranza et al., 1992; Levinson and Mickiewicz, 2007),
although some studies (e.g., Bagott et al., 1997, 1998) saw no effect. Interest-
ingly, Rigotti and colleagues (1997) found an increase in adolescent smoking
in communities that received the enforcement intervention, but not in the
control communities, despite increasing retailer compliance.
Studies of retailer compliance are similar to those evaluating active
enforcement, but rather than investigating whether any enforcement ef-
forts affect underage tobacco use, studies of retailer compliance typically
examine the relationship between tobacco sales rates or retailer compliance
rates (as well as changes in those rates) and underage tobacco use. While
analyses of town-level interventions have found high retail compliance to be
associated with a number of reduced smoking outcomes (Cummings et al.,
2003; Dent and Biglan, 2004; DiFranza et al., 2009; Pokorny et al., 2003),
a meta-analysis that pooled studies of active enforcement into a single com-
pliance measure (compliance rate) (Fichtenberg and Glantz, 2002) found
no relationship between the level of retailer compliance and either 30-day
or regular smoking prevalence. While some (e.g., Cummings et al., 2003;
Rigotti et al., 1997) have hypothesized that sales restrictions and their en-
forcement must achieve high rates of compliance before they begin to affect
underage tobacco use, findings are mixed (see, e.g., Cummings et al., 2003;
Dent and Biglan, 2004; and Fichtenberg and Glantz, 2002).
Finally, findings about the effects of comprehensive interventions incor-
porating such actions as retailer and community education programs and

mass media campaigns on underage smoking are mixed. While two studies
(Chen and Forster, 2006; Cook et al., 2000) found that comprehensive
programs decreased smoking prevalence, one study (Altman et al., 1999)
reported mixed results. A fourth study investigating effects of a compre-
hensive program on both adolescent and adult smoking (Rohrbach et al.,
2002) found that multicomponent exposure was associated with reductions
in adult but not adolescent smoking prevalence.
Overestimation of Retail Compliance

Some of the observed discrepancies in the effects of enforcement and
compliance on underage tobacco use may be due to methodological errors
that result in an inaccurate measurement of the true rate of illegal tobacco
sales to minors. Specifically, the standard compliance check protocol re-
quires the use of underage nonsmokers who have no experience purchasing
tobacco, whereas underage smokers deploy a wide range of strategies to
obtain tobacco from retail stores, including knowing and sharing knowl-
edge of specific stores and clerks that are more likely to sell to underage
persons and strategies to appear older (Crawford et al., 2002; DiFranza and
Coleman, 2001; Robinson and Amos, 2010). Studies comparing inexperi-
enced nonsmokers following the standard compliance inspection protocol
with underage smokers behaving as they normally do (Croghan et al.,
2005; DiFranza et al., 2001b) found that more realistic smoker protocols
substantially increased the likelihood of sale. These methodological issues
suggest that the standardized protocols may be too artificial and may cue
retailers that the purchase attempts are not sincere attempts but, in fact, are
enforcement inspections. Consequently, the rates of tobacco sales to under-
age persons reported through Synar and observed in enforcement interven-
tions may underestimate the true rates of sales to minors. Furthermore, if
enforcement interventions are unlikely to reduce commercial availability,
they are also unlikely to reduce overall tobacco availability to underage
individuals or the actual use of tobacco products. Indeed, in a recent review
of the literature on interventions to reduce the sale of tobacco to minors,
DiFranza (2012) argued that previous reviews of literature on MLA laws
and their enforcement may have failed to find an association between the
MLA laws and adolescent smoking because they did not distinguish in-
terventions that successfully reduce retail tobacco availability from those
that did not. Thus, in his review and analysis, DiFranza (2012) concluded
that active enforcement programs that disrupt the sale of tobacco to minors
will reduce adolescent smoking.

A Key Variable: Reliance on Social Sources

While it is certainly likely that some of the inconsistencies in the find-
ings are due to measurement errors, some of the inconsistencies may also
be genuine. Some critics of youth access policies have suggested that high
rates of compliance may not affect use because of the shift to reliance on
social sources (e.g., Craig and Boris, 2007; Etter, 2006; Ling et al., 2002).
Given that only approximately 50 percent of underage tobacco users report
obtaining tobacco from commercial retailers (see Tables 5-2 through 5-8
in Chapter 5), even a complete cut-off of retail tobacco to underage users
will contain, but not eliminate, overall tobacco availability to them unless
there is a major crackdown on social distribution.
Although the evidence is slim, a handful of studies (Dent and Biglan,
2004; Kim et al., 2013; Levinson and Mickiewicz, 2007; Millett et al., 2011;
Rigotti et al., 1997; Rimpela and Rainio, 2004) suggest that the successful
restriction of retail tobacco will effectively decrease adolescent purchases of
tobacco from retail sources. At the same time, such a restriction is likely to
increase reliance on social sources, including both proxy purchases and be-
ing given tobacco (DiFranza and Coleman, 2001; Levinson and Mickiewicz,
2007; Millett et al., 2011; Rigotti et al., 1997; Rimpela and Rainio, 2004).
Interestingly, a study of Oregon adolescents (Dent and Biglan, 2004) found
that increased compliance was associated with an increased reliance on s ocial
sources and a decreased use of commercial sources among 11th graders, but
that the opposite was true for 8th grade students. It is possible that the
younger students’ social networks were restricted to underage persons so
that increased retail compliance reduced access from these social sources,
leading to an increased need for the 8th graders to try to purchase tobacco
for themselves. On the other hand, the older students may have been more
likely to have social networks that included those who were old enough to
buy tobacco products on their behalf.
It seems clear that curtailing retail access will lead to greater use of
social sources. Whether the reduction in retail access has an effect on
underage use depends on whether the substitution of the social sources for
the commercial sources is complete. To the extent that this substitution of
social sources for commercial sources is incomplete, the search-time costs
for underage users to obtain tobacco will likely increase, and tobacco con-
sumption among underage users will likely decrease. All of the evidence
reviewed above is consistent with incomplete substitution.
Relationship Between Retail Interventions and Perceived Availability

Given the mixed findings regarding the relationship between retailer
interventions and levels of tobacco use in adolescents, it is instructive to

consider whether the intensity of retailer intervention is related to sub-

jective measures of reduced “access” by underage users. Two subjective
factors reported by underage youth—the perceived availability of tobacco
and self-reported decreases in the use of retail sources—can be considered
to be proxy measures of actual underage access. Moreover, as intervening
variables, both measures may moderate the effect of an MLA restriction
and its enforcement on underage tobacco use.
Perceived Availability
One notable trend in adolescents’ access to cigarettes is that the perceived
ease of access has declined considerably in recent years. The Monitoring the
Future surveys ask 8th and 10th graders how difficult they think it would
be for them to get cigarettes, if they wanted to. Among 8th graders in 1996,
77 percent said they could get cigarettes “fairly easily” or “very easily,” while
in 2013 that figure had declined to 50 percent. Among 10th graders, the
corresponding decline was from 90 percent to 70 percent. Thus, although
most adolescents still believe they could easily obtain cigarettes, reports of
easy access have declined considerably over time (Johnston et al., 2014).
This finding is also consistent with reduced retail availability and incomplete
substitution by social sources.
Impact of Enforcement on Perceived Availability

The perceived availability of tobacco represents a subjective assessment
of an underage person’s actual opportunities to obtain tobacco (i.e., supply)
and can be assessed either in reference to specific sources or location types
(e.g., availability from home, school, or stores) or globally. Findings on the
relation between MLA laws and their enforcement and perceived avail-
ability are mixed. However, while these studies typically assess perceived
tobacco availability globally (e.g., Borland and Amos, 2009; Cummings
et al., 2003; Jason et al., 1999a; Rigotti et al., 1997; Rimpela and Rainio,
2004; Staff et al., 1998; Thomson et al., 2004), Forster and colleagues
(1997) assessed perceived availability in reference to specific sources and
found that the intervention decreased the perceived availability from com-
mercial but not social sources. This suggests that the MLA laws and their
enforcement, as expected, may increase the difficulty of obtaining tobacco
from commercial sources, but they do not have an impact on social sources.
It is likely that the inconsistent findings concerning the impact of the MLA
and its enforcement on perceived access may be due to the conflation of
sources.

Relationship Between Perceived Availability and Underage Tobacco Use

The evidence on the relationship between perceived availability and
underage tobacco use is challenging to interpret because this relationship
is dependent on the relative availability of and reliance on tobacco from
social sources. For example, Doubeni and colleagues (2008) found that high
perceived availability increases the risk for multiple smoking outcomes and
that high perceived availability and peer smoking together increased the risk
of regular smoking and of smoking progression among initiators more than
either variable alone. This is logical given that adolescents with more peers
who smoke will likely have greater access to tobacco from these peers and
also more positive attitudes toward tobacco use.
Perceived Availability as a Reflection of Social Norms

Perceived tobacco availability may also reflect perceptions of the social
environment about tobacco use (e.g., social norms) as well as an under-
age individual’s willingness or intentions to attempt to get tobacco (i.e.,
demand). By bridging the interface between tobacco supply and demand,
perceived availability can be understood as a psychosocial mechanism by
which youth tobacco restrictions affect underage tobacco use. Interpreting
the impact of perceived availability on consumption is even more challeng-
ing precisely because it may reflect changes in both tobacco supply and
demand. For example, Gilpin et al. (2004) examined neighboring birth
cohorts before and after implementation of a comprehensive, statewide
tobacco control program in California and found that adolescents who
perceived cigarettes easy to access were more likely to initiate smoking than
those who perceived cigarettes hard to obtain, but only in the cohort under
higher enforcement conditions. The authors therefore suggest that perceived
availability was less a reflection of opportunities to obtain tobacco than
of the declarative effect of the tobacco control program changing social
norms and thereby decreasing demand to take up tobacco use. Finally, a
cross-sectional study (Speizer et al., 2008) examined perceived availability
from different sources and found that current and ever smokers were more
likely to perceive easy access to cigarettes from all sources (home, school,
and stores) than those who never smoked, which suggests that perceived
ease of access reflects both a greater demand for tobacco and opportunities
to access tobacco.
Summary
Findings about the effects of retail enforcement, retail compliance,
and comprehensive interventions on underage tobacco use are difficult to

interpret. The difficulty is reflected in the discrepancy between observed

commercial availability (as measured by the rates of retail compliance) and
perceived tobacco availability, as self-reported by adolescents and young
adults. Whereas all states are currently in compliance with Synar and have
achieved average compliance rates of 90 percent, 50 to 70 percent of ado-
lescents report fair or very easy access to cigarettes (Johnston et al., 2014).
The apparent inconsistency may be partially attributable to an overestimate
of compliance rates in compliance checks. In addition, changes in perceived
availability may reflect changes not only in opportunities to obtain tobacco
but also in social norms and demand for tobacco.
Overall, this body of evidence suggests that the enforcement of MLA
laws increases the perceived difficulty of obtaining tobacco from commer-
cial sources. Additionally, MLA laws are likely to change social norms,
and thereby indirectly affect perceived ease of access from social sources,
especially among younger adolescents. Insofar as the substitution of non-
retail sources for commercial retail sources is incomplete, the total tobacco
available to underage individuals is probably reduced. However, reduced
access does not have a robust and easily measurable impact on use because
of the youths’ increased reliance on social and other non-retail sources,
especially by older adolescents and youth who are already daily smokers.
Finding 6-3: While increasing retailer compliance reduces the availabil-

ity of retail tobacco to underage persons, the magnitude of this effect
and its impact on underage consumption are highly uncertain due to
the continued availability of tobacco from noncommercial sources.
However, the level of substitution by social sources is likely to be lowest
for the youngest underage users.
UNDERAGE ACCESS RESTRICTIONS IN THE CONTEXT

OF OTHER TOBACCO CONTROL POLICIES
It is unlikely that any revised MLA laws will be aggressively enforced in
isolation, so examining the MLA laws and their enforcement in the context
of other tobacco control policies can help elucidate their likely effects in
circumstances that more closely resemble the likely real world scenarios in
which an MLA increase would be implemented. In particular, investigating
the effect of the MLA laws in this way may help explain some of the ob-
served variations in community-level natural experiments. Studies in these
small localities may not account for the contributions of other concurrent
tobacco control programs at the state and national levels (e.g., smoke-free
policies, excise taxes and price, mass media campaigns), and they also may
be subject to spillover effects from neighboring jurisdictions, in particular,

smuggling.5 The following section reviews the evidence on MLA retailer

interventions in the context of other youth access restrictions, followed by
other, general tobacco control programs.
Multiple Statewide Retailer Interventions and

Underage Tobacco Consumption
In addition to MLA laws and their enforcement, youth access policies
also include licensing requirements (i.e., requiring a license to sell tobacco
products), signage requirements at the point of sale (i.e., posting warning
signs about the MLA), vending machine restrictions, inspection require-
ments, clerk intervention policies (i.e., retailer and clerk education), penal-
ties for retailers found to be in violation (especially graduated penalties),
identification requirements, packaging restrictions (e.g., minimum pack
size, labeling standards), restrictions on free distribution (i.e., bans on free
samples), and establishing or designating a statewide enforcement authority.
These policies are typically examined at the state level and can be examined
both individually and in an aggregate measure of overall “extensiveness.”
When examined in the aggregate, having more policies constitutes a higher
score and is considered to be more extensive. In this context, it is imperative
to control for the impact of other policies in order to isolate the indepen-
dent effect of MLA on underage tobacco use and also to identify possible
interactions.
Of the studies that examined multiple youth access policies, includ-
ing an MLA and its enforcement, Chaloupka and Pacula (1998), Luke et
al. (2000), and Powell et al. (2003), found that more extensive policies
were associated with decreased teen current smoking prevalence. Further,
Chaloupka and Pacula (1998) also showed that, when measured individu-
ally, the use of compliance inspections versus only observing retailers and
the use of statewide sampling to measure compliance versus local or no
sampling were both associated with significantly lower adolescent smoking
prevalence. On the other hand, Thomson and colleagues (2004) examined
six types of youth access ordinances (licensing, fines for merchants who
5 Indeed, adolescents achieve increasing mobility as they begin to drive, and implementation
at the town or county level may have a smaller effect than state- or national-level implementa-
tion due to the potential smuggling of tobacco from neighboring jurisdictions where tobacco
availability is higher. Lessons from the alcohol experience suggest precisely this: Lovenheim
and Slemrod (2010) and Dejong and Blanchette (2014) examined the effect of a minimum legal
drinking age on fatal traffic accidents when states were implementing an MLDA of 21 in a
patchwork while the national MLDA remained 18. Their analysis of county-level data found
no reduction in fatal traffic crashes involving youth in counties with an MLDA of 21 that were
within 25 miles of a state with a lower minimum drinking age, but significant reductions in
fatal traffic crashes involving youth in counties further from the state borders.

sell to minors, vending machine restrictions, self-service bans, bans of the

sale of single cigarettes, and bans on distributing free samples) in several
Massachusetts towns and statewide and found that none of the youth ac-
cess ordinances were associated with any measure of adolescent smoking.
One study examined how state youth access policies in effect when study
participants were age 17 affected their smoking after they became adults
(Grucza et al., 2013). The researchers examined the effects of the poli-
cies both individually and in multi-policy models and found that multiple
youth access policies together were associated with significant reductions
in prevalence of both ever and current smokers among females (although
not among males) despite the individual policies having only small effects in
isolation. These findings suggest that individual youth access policies alone
may have small, additive effects that contribute to more substantial impacts
when implemented together.
Comprehensive Tobacco Control Policies and

Underage Tobacco Consumption
Other policies that have an effect on tobacco use in addition to youth
access initiatives are smoke-free laws, state-level expenditures, excise taxes,
and minimum cigarette prices. As with studies of multiple youth access poli-
cies, studies of multiple tobacco control policies can examine the policies
individually or in aggregate.
Of the studies attempting to isolate the independent effects of MLA
laws in the context of other tobacco control policies, two (Botello-Harbaum
et al., 2009; Farrelly et al., 2013) found no association between youth ac-
cess policies and any adolescent smoking outcome after controlling for the
other policies. However, Ross and Chaloupka (2001) found that the deci-
sion to smoke and smoking intensity were each negatively associated with
retailer compliance in models that both included all policies together and
controlled for the effects from the other tobacco control measures. Tworek
et al. (2010) examined the effects of tobacco control policies, including
an index of the strength of youth tobacco sales restrictions on adolescent
smoking cessation, and found that youth access restrictions slightly in-
creased the odds of non-continuation of smoking, but they were not associ-
ated with any other cessation measure.
Other studies have investigated the effects of comprehensive tobacco
control programs; these can be considered to be studies of multiple to-
bacco control policies in aggregate. The multi-pronged tobacco-control
approaches integrate educational, clinical, regulatory, economic, and social
strategies to prevent or reduce tobacco use and to reduce tobacco-related
diseases (CDC, 2014b; HHS, 2000). For example, Helakorpi et al. (2008)
investigated the effects of the 1976 Tobacco Control Act in Sweden, which

prohibited smoking in public places, including public transit; prohibited to-

bacco sales to those under 16; required health warnings on cigarette packs;
and established a tobacco tax whose revenue was earmarked for health
education and tobacco-related research on adult smoking by gender and
socioeconomic status. Among men, the researchers found that after the pas-
sage of the act, the prevalence of ever-daily smoking declined beyond secu-
lar trends for all SES groups, with the strongest declines observed among
the higher SES group (white-collar employees). Among women, there was
an increasing secular trend for all SES groups prior to the legislation, but
after the act women’s ever-daily smoking prevalence reversed in all groups.
A comprehensive tobacco control program in New Zealand was similarly
effective at reducing tobacco use in adolescents and adults and also reduc-
ing tobacco-related death and disease (Laugesen and Swinburn, 2000).
Comprehensive programs in the United States have been shown to
effectively reduce tobacco use among adolescents and adults (e.g., Farrelly
et al., 2008, 2013, 2014; Kuiper et al., 2005; Pierce et al., 2009; Stillman
et al., 2003; Tauras et al., 2005; Wakefield and Chaloupka, 2000; Zaza et
al., 2005) as well as to reduce tobacco-related death and disease (e.g., Jemal
et al., 2003; Kuiper et al., 2005). However, they frequently do not specify
the inclusion of youth access policies (e.g, because comprehensive tobacco
control efforts are frequently measured using state-level expenditures). De-
spite the lack of explicitly identified youth access program components, it
is reasonable to assume that studies of state-level comprehensive tobacco
control programs within the past two decades would have included youth
access restrictions conducted in compliance with Synar. Moreover, the inclu-
sion of youth access restrictions in comprehensive approaches is considered
best practice, and stronger state-level tobacco control programs are likely
to include extensive youth access measures (CDC, 2014b). Thus, it is likely
that these comprehensive approaches to tobacco control that have proved
effective at reducing tobacco use and tobacco-related morbidity and mortal-
ity include some youth access provisions.
Summary
Evidence on the independent effect of youth access policies in the
context of other tobacco control policies is mixed. However, studies of
multiple statewide retailer interventions that include active enforcement of
the MLA restrictions suggest that these interventions are effective in reduc-
ing underage use. Moreover, there is some evidence that comprehensive
tobacco programs that include youth access restrictions are effective at
reducing underage tobacco use, although it is difficult to isolate the relative
contributions of youth access restrictions in these comprehensive programs.

Finding 6-4: Underage tobacco use is most substantially reduced when

the jurisdiction adopts a strong array of tobacco control measures,
including strongly enforced youth access restrictions.
TOBACCO PURCHASE, USE, AND POSSESSION LAWS

Although the focus of this analysis is on sales restrictions, it is also
important to consider the effects of supplementing bans on distribution
and sales with laws targeting underage tobacco PUP. As noted in Chapter 5,
the laws of 44 states and the District of Columbia penalize underage indi
viduals for purchasing, using, or possessing tobacco products, typically by
civil fines or community service. Proponents of the laws argue that PUP
laws are another effective strategy for deterring underage tobacco use (e.g.,
Jason et al., 2009b; Lazovich et al., 2007; Livingood et al., 2001), while
critics argue that PUP laws shift blame from retailers and tobacco industry
marketing and advertising practices toward adolescents and young adults
and, furthermore, that PUP laws may actually increase the desirability of
tobacco as an aspirational, adult product, further enticing adolescents to
use tobacco (e.g., Wakefield and Giovino, 2003). Opponents of PUP laws
also suggest that enforcement would be difficult, expensive, and therefore
realistically infeasible (Tworek et al., 2011). The IOM’s report Growing
Up Tobacco Free (1994) elaborated on this, arguing that PUP laws lacking
enforcement would only serve as a symbolic deterrent, which would be
unlikely to deter tobacco use any more than laws punishing sellers, while
also undermining respect for the law.
There is currently no systematic surveillance of PUP laws, and thus
there is little information about either the extent to which they are enforced
or their efficacy. The only available data on statewide enforcement that the
committee was able to locate (Rogers et al., 2008) come from California,
an aggressive tobacco control state, and the data show that, in 2007, 76
percent of youth access enforcement agencies across the state indicated
that they did not “often” or “very often” issue citations to minors for PUP
violations. Additionally, the average number of citations issued in the past
12 months across all 249 enforcement agencies statewide was 24.1 cita-
tions, or an average of two per agency per month. Similarly, qualitative
studies also suggest that PUP laws are seldom enforced. Two studies using
key informant interviews with individuals responsible for enforcing PUP
laws (e.g., mayors, police officers, and school officials) found that PUP
laws are poorly enforced and that only a small number of citations are
issued (Hrywna et al., 2004); they also found that there was little knowl-
edge about PUP enforcement, that active enforcement of PUP laws was not
a priority, and that even when they were enforced, the enforcement was
inconsistent (Hahn et al., 2007). Indeed, any widely violated and under-

enforced prohibition is likely to be plagued by selective enforcement, and

PUP laws are no exception. For example, Gottlieb et al. (2004) found differ-
ential enforcement of PUP laws, such that African-American and Hispanic
students had a significantly greater probability of receiving a citation than
white students.
Despite lax enforcement efforts, limited empirical data suggest that
active enforcement of PUP laws in addition to active enforcement of youth
tobacco sales restrictions may be effective at reducing tobacco sales to
underage persons and, ultimately, at reducing underage tobacco consump-
tion. Most of these findings come from a series of studies conducted by
Jason and colleagues in a convenience sample of small, suburban towns in
Illinois, which may not be representative of the rest of the state or the coun-
try as a whole. They found that active enforcement of PUP laws in addition
to sales restrictions is associated with reduced tobacco sales to underage
users (Jason et al., 2003); decreased observed and perceived adolescent
tobacco use (Jason et al., 2009a); slower increases in the rate of smoking
compared to enforcing sales restrictions alone (Jason et al., 2008); reduced
smoking, both among whites only (Jason et al., 2003) and, alternately,
among all groups (Jason et al., 2007c); lowered rates of heavy smoking
(Jason et al., 2009b); reduced use of other drugs (Jason et al., 2010), and
reduced crime rates (Jason et al., 2000). They also found that underage
individuals who were fined for PUP violations were more likely to reduce
tobacco use or quit than those who participated in tobacco prevention
education programs (Jason et al., 2007b). Additionally, fines had a bigger
effect than education on changing parental and adolescent attitudes toward
tobacco use (Jason et al., 2007b). Finally, Jason and colleagues also found
that the presence of PUP laws facilitated the uptake of smoke-free policies
(Jason et al., 2007a). Studies by other researchers have further supported
these findings, including studies demonstrating that actively enforcing PUP
laws may be effective at reducing underage tobacco use (Gottlieb et al.,
2004; Lazovich et al., 2007; Livingood et al., 2001) and increasing adoles-
cent smoking cessation (Langer and Warheit, 2000). Moreover, Gottlieb et
al. (2004) found that the threat of driver’s license suspension as a penalty
for PUP violations reduced smoking intentions among adolescent ever-daily
smokers (but not ever or experimental smokers), suggesting a general de-
terrent effect. At the same time, having received a citation was associated
with reduced smoking intentions in only some of the schools sampled, thus
showing mixed findings with respect to specific deterrence.
On the other hand, two analyses also examined PUP laws in the context
of other youth access restrictions, and neither found that they decreased ad-
olescent smoking. Ross and Chaloupka (2001) found that punishing minors
for the use or possession of cigarettes increased the number of cigarettes

adolescents smoked, while Tworek and colleagues (2010) found no associa-

tion between PUP laws and any measure of adolescent smoking cessation.
In sum, there continues to be some controversy about the relative
advantages and disadvantages of implementing PUP laws for tobacco.
Although a small number of observational studies of PUP interventions
suggest that they can contribute to the reduction of underage tobacco use if
enforced, there is scant evidence of enforcement. Moreover, the few existing
studies also suggest that, when enforced, the laws are selectively applied
and that minority populations may carry a disproportionate burden of PUP
violations.
Finding 6-5: Enforcement of purchase–use–possession laws is a contro-

versial strategy for reducing underage tobacco use. Although a small
number of studies suggest that enforcing these laws, in combination
with strategies that limit retail tobacco sales, can reduce use, they also
raise concerns about fair enforcement.
SUMMARY
This chapter reviewed the existing evidence on the effects of raising the
minimum legal age to purchase tobacco products, in particular the effect on
underage tobacco use. No published evidence is yet available on the effects
of raising the MLA to 21 in any of the localities in the United States that
have done so. Limited international evidence suggests that raising the MLA
from 16 to 18 in countries that already had an actively enforced MLA can
be implemented successfully to reduce the availability of retail tobacco to
newly underage persons and thereby reduce underage tobacco use. Experi-
ence with raising the minimum legal drinking age for alcohol in the United
States from 18 to 21 is instructive for tobacco control, in that it has led
to reductions in the use of alcohol and concomitant harms, such as motor
vehicle accidents in the underage population, although it also demonstrates
that the prevalence of underage drinking remains high.
In light of the dearth of direct evidence on the effects of raising the
MLA for tobacco, the committee focused its attention on the substantial
body of literature on the effects of enforcing the MLA restrictions that have
already existed in the United States for more than two decades. This litera-
ture suggests that the MLA policies that are actively enforced and supported
by other retailer interventions will likely increase retailer compliance and
thereby reduce retail tobacco availability to underage individuals. Further-
more, although increased retailer compliance is predictably accompanied
by a corresponding increase in the use of social sources to obtain tobacco,
this substitution of sources is likely to be incomplete, leading to decreased

use, especially if the youth access policy is implemented in a robust com-

prehensive tobacco control context.
REFERENCES
Abernathy, T. J. 1994. Compliance for kids: A community-based tobacco prevention project.
Canadian Journal of Public Health 85(2):82–84.
Altman, D. G., V. Foster, L. Rasenick-Douss, and J. B. Tye. 1989. Reducing the illegal sale of
cigarettes to minors. JAMA 261(1):80–83.
Altman, D. G., L. Rasenick-Douss, V. Foster, and J. B. Tye. 1991. Sustained effects of an
educational program to reduce sales of cigarettes to minors. American Journal of Public
Health 81(7):891–893.
Altman, D. G., A. Y. Wheelis, M. McFarlane, H. Lee, and S. P. Fortmann. 1999. The relation-
ship between tobacco access and use among adolescents: A four community study. Social
Science and Medicine 48(6):759–775.
Bagott, M., C. Jordan, C. Wright, and S. Jarvis. 1997. Test sales do not have impact on preva-
lence of smoking by children. BMJ 315(7106):491.
———.1998. How easy is it for young people to obtain cigarettes, and do test sales by trading
standards have any effect? A survey of two schools in Gateshead. Child: Care, Health
and Development 24(3):207–216.
Biglan, A., J. Henderson, D. Humphrey, M. Yasui, R. Whisman, C. Black, and L. James. 1995.
Mobilising positive reinforcement to reduce youth access to tobacco. Tobacco Control
4(1):42–48.
Biglan, A., D. Ary, V. Koehn, D. Levings, S. Smith, Z. Wright, L. James, and J. Henderson.
1996. Mobilizing positive reinforcement in communities to reduce youth access to
tobacco. American Journal of Community Psychology 24(5):625–638.
Birckmayer, J., and D. Hemenway. 1999. Minimum-age drinking laws and youth suicide,
1970–1990. American Journal of Public Health 89(9):1365–1368.
Bonnie, R. 1982. Discouraging the use of alcohol, tobacco and other drugs: The effects of
legal controls and restrictions. In N. K. Mello, ed., Advances in substance abuse research,
Volume II (pp. 145–184). Greenwich, CT: JAI Press.
Borland, T., and A. Amos. 2009. An exploratory study of the perceived impact of raising the
age of cigarette purchase on young smokers in Scotland. Public Health 123(10):673–679.
Botello-Harbaum, M. T., D. L. Haynie, R. J. Iannotti, J. Wang, L. Gase, and B. Simons-
Morton. 2009. Tobacco control policy and adolescent cigarette smoking status in the
United States. Nicotine & Tobacco Research 11(7):875–885.
Carpenter, C., and P. J. Cook. 2008. Cigarette taxes and youth smoking: New evidence from
national, state, and local youth risk behavior surveys. Journal of Health Economics
27(2):287–299.
Carpenter, C., and C. Dobkin. 2011. The minimum legal drinking age and public health.
Journal of Economic Perspectives 25(2):133–156.
CDC (Centers for Disease Control and Prevention). 1993. Minors’ access to tobacco—
Missouri, 1992, and Texas, 1993. Morbidity & Mortality Weekly Report 42(7):125–128.
———. 1996. Estimates of retailers willing to sell tobacco to minors—California, August–
September 1995 and June–July 1996. Morbidity & Mortality Weekly Report 45(50):1095–
1099. [Erratum appears in Morbidity and Mortality Weekly Report 45(51–52):1129.]
———. 2014a. Smoking and tobacco use: Economic facts about U.S. tobacco production
and use. http://www.cdc.gov/tobacco/data_statistics/fact_sheets/economics/econ_facts

———. 2014b. Best practices for comprehensive tobacco control programs—2014. Atlanta,
Prevention, National Center for Chronic Disease Prevention and Health Promotion,
Office on Smoking and Health.
Chaloupka, F. J., and R. L. Pacula. 1998. Limiting youth access to tobacco: The early impact
of the Synar Amendment on youth smoking. Paper presented at Biennial Pacific Rim
Allied Economic Organizations Conference, 1997, Bangkok, Thailand. http://tigger.uic.
edu/~fjc/Presentations/Papers/Synar94b1.pdf (accessed February 25, 2015).
Chaloupka, F. J., and K. E. Warner. 2000. The economics of smoking. Handbook of Health
Economics 1(B):1539–1627.
Chen, V., and J. L. Forster. 2006. The long-term effect of local policies to restrict retail sale of
tobacco to youth. Nicotine & Tobacco Research 8(3):371–377.
Cismoski, J., and M. Sheridan. 1993. Availability of cigarettes to under-age youth in Fond du
Lac, Wisconsin. Wisconsin Medical Journal 92(11):626–630.
Conover, E., and D. Scrimgeour. 2013. Health consequences of easier access to alcohol: New
Zealand evidence. Journal of Health Economics 32(3):570–585.
Cook, D., C. Edwards, L. Bauer, and D. Tutt. 2000. Reducing adolescent smoking rates.
Maintaining high retail compliance results in substantial improvements. Health Promo-
tion Journal of Australia 10(1):20–24.
Crawford, M., G. Balch, and R. Mermelstein. 2002. Responses to tobacco control policies
among youth. Tobacco Control 11(1):14–19.
Croghan, E., P. Aveyard, and C. Johnson. 2005. Is it as easy as young people claim for them to
buy cigarettes? Comparing the results of realistic test purchases with those from trading
standards test purchases. Health Education 105(2):103–108.
Cummings, K. M., A. Hyland, T. Saunders-Martin, J. Perla, P. R. Coppola, and T. F. Pechacek.
1998. Evaluation of an enforcement program to reduce tobacco sales to minors. Ameri-
can Journal of Public Health 88(6):932–936.
Cummings, K. M., A. Hyland, J. Perla, and G. A. Giovino. 2003. Is the prevalence of youth
smoking affected by efforts to increase retailer compliance with a minors’ access law?
DeJong, W., and J. Blanchette. 2014. Case closed: Research evidence on the positive public
health impact of the age 21 minimum legal drinking age in the United States. Journal of
Studies on Alcohol and Drugs 75(Suppl 17):108–115.
Dent, C., and A. Biglan. 2004. Relation between access to tobacco and adolescent smoking.
Tobacco Control 13(4):334–338.
DiFranza, J. R. 2012. Which interventions against the sale of tobacco to minors can be
expected to reduce smoking? Tobacco Control 21(4):436–442.
DiFranza, J. R., and M. Coleman. 2001. Sources of tobacco for youths in communities with
strong enforcement of youth access laws. Tobacco Control 10(4):323–328.
DiFranza, J. R., R. P. Carlson, and R. E. Caisse. 1992. Reducing youth access to tobacco.
Tobacco Control 1(1):58.
DiFranza, J. R., C. C. Celebucki, and P. D. Mowery. 2001a. Measuring statewide merchant
compliance with tobacco minimum age laws: The Massachusetts experience. American
DiFranza, J. R., J. A. Savageau, and J. Bouchard. 2001b. Is the standard compliance check
protocol a valid measure of the accessibility of tobacco to underage smokers? Tobacco
Control 10(3):227–232.

DiFranza, J. R., J. A. Savageau, and K. E. Fletcher. 2009. Enforcement of underage sales

laws as a predictor of daily smoking among adolescents—A national study. BMC Public
Health 9:107.
Ding, A. 2003. Youth are more sensitive to price changes in cigarettes than adults. Yale Journal
of Biology and Medicine 76(3):115–124.
Doubeni, C. A., W. Li, H. Rouayzi, and J. R. DiFranza. 2008. Perceived accessibility as a
predictor of youth smoking. Annals of Family Medicine 6(4):323–330.
Dovell, R. A., D. L. Mowat, J. Dorland, and M. Lam. 1998. Tobacco access to youth: Beliefs
and attitudes of retailers. Canadian Journal of Public Health. 89(1):17–21.
EDC (Education Development Center). 2010a. Highlights from the Metrowest Adolescent
Health Survey: Needham High School report. Newton, MA: MetroWest Community
Health Care Foundation.
———. 2010b. Highlights from the Metrowest Adolescent Health Survey: Needham Middle
School report. Newton, MA: MetroWest Community Health Care Foundation.
Erickson, A. D., S. I. Woodruff, M. B. Wildey, and E. Kenney. 1993. A baseline assessment
of cigarette sales to minors in San Diego, California. Journal of Community Health
18(4):213–224.
Etter, J. F. 2006. Laws prohibiting the sale of tobacco to minors: Impact and adverse conse-
quences. American Journal of Preventive Medicine 31(1):47–51.
Farrelly, M. C., T. F. Pechacek, K. Y. Thomas, and D. Nelson. 2008. The impact of tobacco
control programs on adult smoking. American Journal of Public Health 98(2):304–309.
103(3):549–555.
Feighery, E., D. G. Altman, and G. Shaffer. 1991. The effects of combining education and
enforcement to reduce tobacco sales to minors. A study of four Northern California
communities. JAMA 266(22):3168–3171.
Fichtenberg, C. M., and S. A. Glantz. 2002. Youth access interventions do not affect youth
smoking. Pediatrics 109(6):1088–1092.
Fidler, J. A., and R. West. 2010. Changes in smoking prevalence in 16–17-year-old versus older
adults following a rise in legal age of sale: Findings from an English population study.
Addiction 105(11):1984–1988.
Forster, J. L., M. Hourigan, and P. McGovern. 1992. Availability of cigarettes to underage
youth in three communities. Preventive Medicine 21(3):320–328.
Forster, J. L., M. Wolfson, D. M. Murray, A. C. Wagenaar, and A. J. Claxton. 1997. Perceived
and measured availability of tobacco to youths in 14 Minnesota communities: The TPOP
Study. American Journal of Preventive Medicine 13(3):167–174.
Forster, J. L., D. M. Murray, M. Wolfson, T. M. Blaine, A. C. Wagenaar, and D. J. Hennrikus.
1998. The effects of community policies to reduce youth access to tobacco. American
Franz, G. A. 2008. Price effects on the smoking behaviour of adult age groups. Public Health
122(12):1343–1348.
Gemson, D. H., H. L. Moats, B. X. Watkins, M. L. Ganz, S. Robinson, and E. Healton.
1998. Laying down the law: Reducing illegal tobacco sales to minors in central Harlem.
Gilpin, E. A., L. Lee, and J. P. Pierce. 2004. Does adolescent perception of difficulty in getting
cigarettes deter experimentation? Preventive Medicine 38(4):485–491.

Glanz, K., A. D. Jarrette, E. A. Wilson, D. L. O’Riordan, and K. R. Jacob Arriola. 2007. Re-
ducing minors’ access to tobacco: Eight years’ experience in Hawaii. Preventive Medicine
44(1):55–58.
Gottlieb, N., A. Loukas, M. Corrao, A. McAlister, C. Snell, and P. Huang. 2004. Minors’
tobacco possession law violations and intentions to smoke: Implications for tobacco
control. Tobacco Control 13(3):237–243.
Grucza, R. A., A. D. Plunk, P. R. Hipp, P. Cavazos-Rehg, M. J. Krauss, R. C. Brownson, and
L. J. Bierut. 2013. Long-term effects of laws governing youth access to tobacco. American
Hahn, E. J., C. Riker, K. M. Butler, S. Cavendish, P. Lewis, L. W. Greathouse Maggio, and
V. Nunley. 2007. Enforcement of tobacco purchase, use, and possession laws in four
Kentucky communities. Policy, Politics, & Nursing Practice 8(2):140–147.
Harris, J. E., and S. W. Chan. 1999. The continuum-of-addiction: Cigarette smoking in rela-
tion to price among Americans aged 15–29. Health Economics 8(1):81–86.
Helakorpi, S., T. Martelin, J. Torppa, E. Vartiainen, A. Uutela, and K. Patja. 2008. Impact
of the 1976 tobacco control act in Finland on the proportion of ever daily smokers by
socioeconomic status. Preventive Medicine 46(4):340–345.
HHS (Department of Health and Human Services). 2000. Reducing tobacco use: A report
of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services;
Centers for Disease Control and Prevention; National Center for Chronic Disease Preven-
tion and Health Promotion; Office on Smoking and Health.
Hrywna, M., R. K. Adler, C. D. Delnevo, and J. D. Slade. 2004. Content analysis and key
informant interviews to examine community response to the purchase, possession, and/
or use of tobacco by minors. Journal of Community Health 29(3):209–216.
Huckle, T., and K. Parker. 2014. Long-term impact on alcohol-involved crashes of lower-
ing the minimum purchase age in New Zealand. American Journal of Public Health
104(6):1087–1091.
Hughes, C. E., and A. Stevens. 2010. What can we learn from the Portuguese decriminalization
of illicit drugs? British Journal of Criminology 50(6):999–1022.
Academy Press.
Jason, L. A., P. Y. Ji, M. D. Anes, and S. H. Birkhead. 1991. Active enforcement of cigarette
control laws in the prevention of cigarette sales to minors. JAMA 266(22):3159–3161.
Jason, L., W. Billows, D. Schnopp-Wyatt, and C. King. 1996. Reducing the illegal sales of
cigarettes to minors: Analysis of alternative enforcement schedules. Journal of Applied
Behavior Analysis 29(3):333–344.
Jason, L. A., M. Berk, D. L. Schnopp-Wyatt, and B. Talbot. 1999a. Effects of enforcement of
youth access laws on smoking prevalence. American Journal of Community Psychology
27(2):143–160.
Jason, L. A., R. Katz, J. Vavra, D. L. Schnopp-Wyatt, and B. Talbot. 1999b. Long-term follow-
up of youth access to tobacco laws’ impact on smoking prevalence. Journal of Human
Behavior in the Social Environment 2(3):1–13.
Jason, L. A., R. Katz, S. B. Pokorny, M. Engstrom, G. Tegart, and C. Curie. 2000. The rela-
tionship between youth tobacco control enforcement and crime rates in a Midwestern
county. American Journal of Health Promotion 14(4):229–231.

Jason, L. A., S. B. Pokorny, and M. E. Schoeny. 2003. Evaluating the effects of enforcements
and fines on youth smoking. Critical Public Health 13(1):33–45.
Jason, L. A., Y. M. Hunt, M. L. Adams, S. B. Pokorny, and P. B. Gadiraju. 2007a. Strengthen-
ing communities’ youth access policies may facilitate clean indoor air action. Preventing
Chronic Disease 4(4):A113.
Jason, L. A., S. B. Pokorny, M. Adams, Y. Hunt, P. Gadiraju, T. Morello, M. Schoeny, and C.
Dinwiddie. 2007b. Youth caught in violation of tobacco purchase, use, and possession
laws: Education versus fines. Behavior Modification 31(6):713–731.
Jason, L. A., S. B. Pokorny, M. Adams, Y. Hunt, P. Gadiraju, and M. Schoeny. 2007c. Do fines
for violating possession–use–purchase laws reduce youth tobacco use? Journal of Drug
Education 37(4):393–400.
Jason, L. A., S. B. Pokorny, and M. Adams. 2008. A randomized trial evaluating tobacco
possession–use–purchase laws in the USA. Social Science and Medicine 67(11):1700–1707.
Jason, L. A., S. B. Pokorny, M. Adams, A. Topliff, C. Harris, and Y. Hunt. 2009a. Youth
tobacco access and possession policy interventions: Effects on observed and perceived
tobacco use. American Journal on Addictions 18(5):367–374.
———. 2009b. E ffects of youth tobacco access and possession policy interventions on heavy
adolescent smokers. International Journal of Environmental Research and Public Health
6(1):1–9.
Jason, L. A., S. B. Pokorny, M. Adams, A. Nihls, H. Y. Kim, and Y. Hunt. 2010. Cracking
down on youth tobacco may influence drug use. Journal of Community Psychology
38(1):1–15.
Jemal, A., V. E. Cokkinides, O. Shafey, and M. J. Thun. 2003. Lung cancer trends in young
adults: An early indicator of progress in tobacco control (United States). Cancer Causes
and Control 14(6):579–585.
Johnston, L. D., P. M. O’Malley, J. G. Bachman, J. E. Schulenberg, and R. A. Miech. 2014.
Monitoring the Future: National survey results on drug use, 1975–2013. Volume 1:
Secondary school students. Ann Arbor, MI: Institute for Social Research, The University
of Michigan.
Junck, E., J. Humphries, and C. Rissel. 1997. Reducing tobacco sales to minors in Manly:
10 months follow-up. Health Promotion Journal of Australia 7(1):29–34.
Kan, M. Y., and M. Lau. 2010. Minor access control of Hong Kong under the Framework
Convention on Tobacco Control. Health Policy 95(2–3):204–210.
Keay, D., S. I. Woodruff, M. B. Wildey, and E. M. Kenney. 1993. Effect of a retailer inter-
vention on cigarette sales to minors in San Diego County, California. Tobacco Control
2(2):145–151.
Kim, A. E., B. R. Loomis, A. H. Busey, M. C. Farrelly, J. G. Willett, and H. R. Juster. 2013.
Influence of retail cigarette advertising, price promotions, and retailer compliance on
youth smoking-related attitudes and behaviors. Journal of Public Health Management
and Practice 19(6):E1–E9.
Kostova, D., H. Ross, E. Blecher, and S. Markowitz. 2011. Is youth smoking responsive to
cigarette prices? Evidence from low- and middle-income countries. Tobacco Control
20(6):419–424.
Krevor, B. S., W. R. Ponicki, J. W. Grube, and W. DeJong. 2011. The effect of mystery shopper
reports on age verification for tobacco purchases. Journal of Health Communications
16(8):820–830.
Kuendig, H. 2011. Enforcement of regional bans on tobacco sales to minors in Switzerland—
Example of missed opportunity? European Journal of Public Health 21(6):750–752.

Kuiper, N. M., D. E. Nelson, and M. Schooley. 2005. Evidence of effectiveness: A summary

of state tobacco control program evaluation literature. Atlanta, GA: U.S. Department
of Health and Human Services, Public Health Service, Centers for Disease Control and
Kypri, K., R. B. Voas, J. D. Langley, S. C. R. Stephenson, D. J. Begg, A. S. Tippetts, and G. S.
Davie. 2006. Minimum purchasing age for alcohol and traffic crash injuries among 15-
to 19-year-olds in New Zealand. American Journal of Public Health 96(5):126–131.
Landrine, H., E. A. Klonoff, and A. Reina-Patton. 2000. Minors’ access to tobacco before and
after the California STAKE Act. Tobacco Control 9(Suppl 2):II15–II17.
Langer, L. M., and G. J. Warheit. 2000. Teen tobacco court: A determination of the short-term
outcomes of judicial processes with teens engaging in tobacco possession. Adolescent &
Family Health 1(1):5–10.
Laugesen, M., and B. Swinburn. 2000. New Zealand’s tobacco control programme, 1985–
1998. Tobacco Control 9(2):155–162.
Lazovich, D., J. Forster, R. Widome, and P. VanCoevering. 2007. Tobacco possession, use, and
purchase laws and penalties in Minnesota: Enforcement, tobacco diversion programs,
and youth awareness. Nicotine & Tobacco Research 9(Suppl 1):S57–S64.
Levinson, A. H., and T. Mickiewicz. 2007. Reducing underage cigarette sales in an iso-
lated community: The effect on adolescent cigarette supplies. Preventive Medicine
45(6):447–453.
Ling, P. M., A. Landman, and S. A. Glantz. 2002. It is time to abandon youth access tobacco
programmes. Tobacco Control 11(1):3–6.
Livingood, W. C., C. D. Woodhouse, J. J. Sayre, and P. Wludyka. 2001. Impact study
of tobacco possession law enforcement in Florida. Health Education and Behavior
28(6):733–748.
Lovenheim, M. F., and J. Slemrod. 2010. The fatal toll of driving to drink: The effect of
minimum legal drinking age evasion on traffic fatalities. Journal of Health Economics
29(1):62–77.
Luke, D. A., K. A. Stamatakis, and R. C. Brownson. 2000. State youth-access tobacco control
policies and youth smoking behavior in the United States. American Journal of Preventive
Medicine 19(3):180–187.
Ma, G. X., S. Shive, and M. Tracy. 2001. The effects of licensing and inspection enforcement
to reduce tobacco sales to minors in greater Philadelphia, 1994–1998. Addictive Behav-
iors 26(5):677–687.
MacCoun, R., and P. Reuter. 2001. Evaluating alternative cannabis regimes. British Journal
of Psychiatry 178:123–128.
Mawkes, L., M. Swanson, L. Wood, P. Markham, N. De Klerk, and N. Walker. 1997. C hoking
the supply: Restricting the sale of cigarettes to children in western Australia. Health
Promotion Journal of Australia 7(1):22–28.
McCartt, A. T., L. A. Hellinga, and B. B. Kirley. 2010. The effects of minimum legal drinking
age 21 laws on alcohol-related driving in the United States. Journal of Safety Research
41(2):173–181.
McDermott, S. R., K. L. Scott, and M. P. Frintner. 1998. Accessibility of cigarettes to minors
in suburban Cook County, Illinois. Journal of Community Health 23(2):153–160.
Millett, C., J. T. Lee, D. C. Gibbons, and S. A. Glantz. 2011. Increasing the age for the legal
purchase of tobacco in England: Impacts on socio-economic disparities in youth smoking.
Thorax 66(10):862–865.
Model, K. E. 1993. The effect of marijuana decriminalization on hospital emergency room drug
episodes, 1975–1978. Journal of the American Statistical Association 88(423):737–747.

Moore, M. H. 1973. Policies to achieve discrimination on the effective price of heroin. Ameri-
can Economic Review 63(2):270–277.
Naidoo, J., and C. Platts. 1985. Smoking prevention in Bristol: Getting maximum results using
minimum resources. Health Education Journal 44(1):39–42.
Nikaj, S., and F. J. Chaloupka. 2014. The effect of prices on cigarette use among youths in
the Global Youth Tobacco Survey. Nicotine & Tobacco Research 16(Suppl 1):S16–S23.
Norberg, K. E., L. J. Bierut, and R. A. Grucza. 2009. Long-term effects of minimum drinking
age laws on past-year alcohol and drug use disorders. Alcoholism: Clinical and Experi-
mental Research 33(12):2180–2190.
NPHD (Needham Public Health Department). 2008. Thirteen years of tobacco efforts in
Needham. http://www.needhamma.gov/DocumentCenter/Home/View/1868 (accessed
November 20, 2014).
———. 2012. Metrowest adolescent health survey—Highlights from Needham High School,
2012. http://www.needhamma.gov/documentcenter/view/8554 (accessed January 17,
2014).
O’Malley, P. M., and A. C. Wagenaar. 1991. Effects of minimum drinking age laws on alcohol
use, related behaviors and traffic crash involvement among American youth: 1976–1987.
Journal of Studies on Alcohol and Drugs 52(5):478–491.
Pacula, R. L., R. MacCoun, P. Reuter, J. Chriqui, B. Kilmer, K. Harris, L. Paoli, and C.
Schafer. 2005. What does it mean to decriminalize marijuana? A cross-national empirical
examination. Advances in Health Economics and Health Services Research 16:347–369.
Pearson, D. C., L. Song, R. B. Valdez, and A. S. Angulo. 2007. Youth tobacco sales in a
metropolitan county: Factors associated with compliance. American Journal of Preven-
tive Medicine 33(2):91–97.
Plunk, A. D., P. Cavazaos-Rehg, L. J. Bierut, and R. A. Grucza. 2013. The persistent effects of
minimum legal drinking age laws on drinking patterns later in life. Alcoholism: Clinical
and Experimental Research 37(3):463–469.
Pokorny, S. B., L. A. Jason, and M. E. Schoeny. 2003. The relation of retail tobacco avail-
ability to initiation and continued smoking. Journal of Clinical Child and Adolescent
Pokorny, S. B., S. Corbin, M. Driscoll, and L. A. Jason. 2008. The effect of enforcement
history and ordinance provisions on merchant compliance with tobacco-control laws.
Journal of Social Work Practice in the Addictions 8(2):228–243.
Powell, L. M., J. A. Tauras, and H. Ross. 2003. Peer effects, tobacco control policies, and
youth smoking behavior. Chicago, IL: ImpacTeen.
Rigotti, N. A., J. R. DiFranza, Y. Chang, T. Tisdale, B. Kemp, and D. E. Singer. 1997. The
effect of enforcing tobacco-sales laws on adolescents’ access to tobacco and smoking
behavior. New England Journal of Medicine 337(15):1044–1051.
Rimpela, A. H., and S. U. Rainio. 2004. The effectiveness of tobacco sales ban to minors: The
case of Finland. Tobacco Control 13(2):167–174.
Robinson, J., and A. Amos. 2010. A qualitative study of young people’s sources of cigarettes
and attempts to circumvent underage sales laws. Addiction 105(10):1835–1843.
Rocheleau, A. M., and D. Boyum. 1994. Measuring heroin availability in three cities. Wash-
ington, DC: Executive Office of the President, Office of National Drug Control Policy.
Rogers, T., E. C. Feighery, and H. H. Haladijan. 2008. Current practices in enforcement of
California laws regarding youth access to tobacco products and exposure to secondhand
smoke. Sacramento, CA: California Department of Public Health, California Tobacco
Control Program.

Rohrbach, L. A., B. Howard-Pitney, J. B. Unger, C. W. Dent, K. A. Howard, T. B. Cruz, K. M.

Ribisl, G. J. Norman, H. Fishbein, and C. A. Johnson. 2002. Independent evaluation of
the California Tobacco Control Program: Relationships between program exposure and
outcomes, 1996–1998. American Journal of Public Health 92(6):975–983.
Ross, H., and F. J. Chaloupka. 2001. The effect of public policies and prices on youth smok-
ing. Chicago, IL: ImpacTeen.
———. 2003. The effect of cigarette prices on youth smoking. Health Economics
12(3):217–230.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2009. The NSDUH
report: Alcohol use before and after the 21st birthday. Rockville, MD: Office of Applied
Studies.
Sanson-Fisher, R. W., M. J. Schofield, and M. See. 1992. Availability of cigarettes to minors.
Australian Journal of Public Health 16(4):354–359.
Schensky, A. E., S. S. Smith, D. L. Icenogle, and M. C. Fiore. 1996. Youth tobacco sale com-
pliance checks: Impact on vendor practices and community policy. Wisconsin Medical
Journal 95(11):775–778.
Schofield, M. J., R. W. Sanson-Fisher, and S. Gulliver. 1997. Interventions with retailers to
reduce cigarette sales to minors: A randomised controlled trial. Australian and New
Zealand Journal of Public Health 21(6):590–596.
Sen, A., and T. Wirjanto. 2010. Estimating the impacts of cigarette taxes on youth smoking
participation, initiation, and persistence: Empirical evidence from Canada. Health Eco-
nomics 19(11):1264–1280.
Shults, R. A., R. W. Elder, D. A. Sleet, J. L. Nichols, M. O. Alao, V. G. Carande-Kulis, S. Zaza,
D. M. Sosin, and R. S. Thompson. 2001. Reviews of evidence regarding interventions to
reduce alcohol-impaired driving. American Journal of Preventive Medicine 21(4):66–88.
Skretny, M. T., K. M. Cummings, R. Sciandra, and J. Marshall. 1990. An intervention to
reduce the sale of cigarettes to minors. New York State Journal of Medicine 90(2):54–55.
Speizer, I. S., M. K. Bean, C. P. Obando, and E. Fries. 2008. Middle school students’ perceived
access to cigarettes in Virginia. American Journal of Health Behavior 32(4):399–410.
Staff, M., L. March, A. Brnabic, K. Hort, J. Alcock, S. Coles, and R. Baxter. 1998. Can non-
prosecutory enforcement of public health legislation reduce smoking among high school
students? Australian and New Zealand Journal of Public Health 22(3 Suppl):332–335.
Stahl, D. O. 1989. Oligopolistic pricing with sequential consumer search. American Economic
Review 79(4):700–712.
Stillman, F. A., A. M. Hartman, B. I. Graubard, E. A. Gilpin, D. M. Murray, and J. T. Gibson.
2003. Evaluation of the American Stop Smoking Intervention Study (ASSIST): A report
of outcomes. CancerSpectrum Knowledge Environment 95(22):1681–1691.
Sundh, M., and C. Hagquist. 2004. The importance of a minimum age law for the possibility
of purchase of tobacco by adolescents: A study based on Swedish experiences. Scandina-
vian Journal of Public Health 32(1):68–74.
———. 2006. Compliance with a minimum-age law of 18 for the purchase of tobacco—The
case of Sweden. Health Education Research 21(3):378–385.
———. 2007. Does a minimum-age law for purchasing tobacco make any difference? Swedish
experiences over eight years. European Journal of Public Health 17(2):171–177.
Tangirala, M. K., E. L. J. McKyer, D. D. Goetze, and J. McCarthy–Jean. 2006. Use of tobacco
retailer inspections to reduce tobacco sales to youth: Do inspections increase retailer
compliance? International Journal of Consumer Studies 30(3):278–283.
Tauras, J. A., S. Markowitz, and J. Cawley. 2005. Tobacco control policies and youth smoking:
Evidence from a new era. Advances in Health Economics and Health Services Research
16:277–291.

Thomson, C. C., M. Gokhale, L. Biener, M. B. Siegel, and N. A. Rigotti. 2004. Statewide

evaluation of youth access ordinances in practice: Effects of the implementation of
community-level regulations in Massachusetts. Journal of Public Health Management
and Practice 10(6):481–489.
Tutt, D., L. Bauer, and J. Difranza. 2009. Restricting the retail supply of tobacco to minors.
Journal of Public Health Policy 30(1):68–82.
Tworek, C., R. Yamaguchi, D. D. Kloska, S. Emery, D. C. Barker, G. A. Giovino, P. M.
O’Malley, and F. J. Chaloupka. 2010. State-level tobacco control policies and youth
smoking cessation measures. Health Policy 97(2–3):136–144.
Tworek, C., G. A. Giovino, K. M. Cummings, A. Hyland, and F. Chaloupka. 2011. Youth
access tobacco possession, use and purchase laws: Measures of state and local enforce-
ment. Chicago, IL: ImpacTeen.
Wagenaar, A. C., and T. L. Toomey. 2002. Effects of minimum drinking age laws: Review and
analyses of the literature from 1960 to 2000. Journal of Studies on Alcohol and Drugs
(Suppl 14):206–225.
Wakefield, M., and F. Chaloupka. 2000. Effectiveness of comprehensive tobacco control
programmes in reducing teenage smoking in the USA. Tobacco Control 9(2):177–186.
Wakefield, M., and G. Giovino. 2003. Teen penalties for tobacco possession, use, and pur-
chase: Evidence and issues. Tobacco Control 12(Suppl 1):i6–i13.
Watson, A., and N. Grove. 1999. Larimer County Tobacco and Youth Project. American
Wildey, M. B., S. I. Woodruff, A. Agro, K. D. Keay, E. Kenney, and T. L. Conway. 1995.
Sustained effects of educating retailers to reduce cigarette sales to minors. Public Health
Reports 110(5):625–629.
Woodruff, S. I., A. D. Erickson, M. B. Wildey, and E. M. Kenney. 1993. Changing retailer
knowledge, attitudes, and behaviors related to cigarette sales to minors. Journal of Com-
munity Psychology 21(3):234–245.
Zaza, S., P. A. Briss, and K. W. Harris. 2005. The guide to community preventive services:
What works to promote health. New York: Task Force on Community Preventive
Services.


The Effect on Tobacco Use of

Raising the Minimum Age of
Legal Access to Tobacco Products
T
he charge to the committee, as discussed in Chapter 1, was to assess
the public health implications of raising the minimum age of legal
access to tobacco products (MLA) through a review of the literature
on tobacco initiation, modeling, and other approaches, as appropriate.
This chapter provides the rationale for the committee’s consensus conclu-
sions about the likely effects of raising the MLA on tobacco initiation. The
committee’s conclusions serve as inputs to the two commissioned models,
which provide quantification of the likely effects of increases in the MLA
on smoking prevalence in the United States. The two simulation models
used for the findings presented in both this chapter and the next (the Cancer
Intervention and Surveillance Modeling Network [CISNET] and SimSmoke
models) are well established approaches for estimating the likely impact of
changes in tobacco control policies on population-level smoking initiation
and prevalence, and on population health outcomes. The next chapter
(Chapter 8) uses the results presented in this chapter (i.e., the estimates of
the effects of different MLA policies on smoking initiation) as inputs for
modeling several important public health outcomes, smoking-related mor-
bidity and mortality. Chapter 8 concludes with a discussion of the likely
effect of a change in the MLA on the many tobacco-related health effects
not modeled.
METHODS
The committee followed a principled and evidence-based process to
arrive at its estimates of the potential impact of a change in the MLA on
193

tobacco initiation behavior. First, the committee conducted a review of rel-

evant literature and synthesized the background evidence relevant to under-
standing age-related effects of tobacco use. This review included attention
to the effects of MLA enforcement and the sources through which underage
adolescents and young adults get tobacco as well as the development of an
understanding of the biological, psychological, social, and environmental
influences on tobacco use during adolescence and young adulthood.
Second, through an iterative and consensus-driven process, the commit-
tee considered how these age-related effects would translate into potential
changes in rates of initiation across different age segments in adolescence
and young adulthood. The committee considered the likely magnitude of
changes in initiation effects that each of the three policy options under
consideration (raising the MLA to 19, 21, or 25 years of age) would have
on the different age segments and arrived at ordered, categorical estimates
labeled as “small,” “medium,” or “large.”
Third, once consensus about the magnitude of the effect at each age
segment and policy option combination was reached, the committee at-
tached numeric ranges to each of the magnitude estimate descriptors. These
ranges were developed through consideration of reasonable and conserva-
tive estimates of effects from a variety of public health interventions, includ-
ing prior tobacco-related policy changes and data from the experience of
changing the minimum age for alcohol purchase and use. These estimates
were well vetted over a series of discussions in the committee. The commit-
tee selected ranges that showed increasing relationships with the ordered
categories and which ranged in increments of 5 percent (to avoid implying
an unrealistic precision in the estimates) from 5 to 30 percent for potential
changes in initiation.
Finally, the committee discussed how to deal with the fact that there
is scant direct evidence about how raising the MLA would affect tobacco
use at different ages and thus that there is necessarily less confidence about
some effects than others. The committee has most confidence about the
estimates for the effects of raising the MLA to 19 and 21, and it is much
less confident in estimates for an MLA of 25 because of the greater degree
of extrapolation needed for estimating change. Thus, the inputs of esti-
mates for the simulation models in Chapters 7 and 8 also include a range
otential values, with a broader range for the MLA of 25.
of p
In assessing the possible impact of raising the MLA, the committee
made a number of assumptions that affect the conclusions about the mag-
nitude of the estimates and the inputs into the simulation models. Some
of these assumptions are discussed in this chapter; others are discussed in
Chapter 9. These assumptions include

THE EFFECT ON TOBACCO USE OF RAISING THE MINIMUM AGE 195
• While policies implemented in the past will continue to have effects

in the future, existing tobacco policies will remain in effect at their
current rates and no new policies will be implemented. This as-
sumption was made to isolate the effects of the new MLA.
• Levels of enforcement and of retailers’ compliance with age-based
laws will remain at levels similar to those that currently exist.
• Noncommercial or social sources (e.g., through social networks of
peers, families, and coworkers) of tobacco products will remain
essentially as they are now, and no new efforts will be made to en-
force the MLA restrictions against noncommercial or social sources
who provide tobacco products to underage users.
• Sanctions will be directed as they are now, primarily toward retailers
and not toward individual users.
• The new MLA will be applied to all tobacco products. However,
the estimates of the magnitudes of effects are based on decades of
research on cigarette use and not on other products. The committee
acknowledges that the tobacco use landscape is changing rapidly
with the introduction of a variety of tobacco products, including
electronic nicotine delivery systems (ENDS; e.g., “e-cigarettes”).
How these products may change existing patterns of cigarette use is
not yet known, however, and thus these potential sources of influ-
ence could not be reasonably considered in arriving at estimates.
• The estimate of effects will be constant over time.
• Subpopulation differences in tobacco use that currently exist and
that go into current initiation and prevalence rates will continue
into the future. The committee did not consider whether there
would be differential effects in subpopulations over future years.
• Rates of use of alcohol and other illicit drugs will not change in
response to a change in the MLA for tobacco.
RATIONALE FOR EXPECTED IMPACT OF

RAISING THE MINIMUM AGE OF LEGAL ACCESS
ON INITIATION OF TOBACCO USE
Estimating changes in the prevalence of tobacco use following a change
in the MLA requires consideration of how each of the dynamic and inter-
acting biological, psychosocial, and environmental factors contributing to
use may vary by age. At any age, prevalence is a function of the rates of
initiation (defined here as reaching a minimum of 100 cigarettes/lifetime)
and the rates of cessation (defined in models as no use for 2 years after
achieving the threshold for initiation). Both initiation and cessation are
strongly related to age: Initiation decreases dramatically after young adult-

hood, and cessation rates start to increase during that same period. Thus,
changes in the MLA will have their strongest impact on rates of initiation.
The committee based its estimates on an understanding of factors that
are most relevant to achieving the threshold of 100 cigarettes for initiation,
the data about which come from the National Health Interview Survey
and are used in the simulation models in Chapter 8. The committee con-
sidered factors that operate both directly on tobacco use, such as access to
tobacco products, and factors that operate more indirectly or distally, such
as changes in biological vulnerability to the effects of nicotine with age or
changes in social norms that indirectly affect motives for use. Smoking at
least 100 cigarettes in one’s lifetime goes beyond occasionally trying ciga-
rettes or “experimentation.” To achieve the benchmark of 100 cigarettes,
an individual must have access to cigarettes over a period of time and is
also likely to have developed symptoms of dependence and stronger mo-
tives for use beyond any perceived peer or social group pressure (Dierker
and Mermelstein, 2010). Thus, the factors that influence vulnerabilities to
developing dependence are more central to achieving the initiation thresh-
old than factors related to the vulnerabilities to trying just one cigarette.
A critical component in the development of dependence and in contin-
ued tobacco use is the reinforcing effects of nicotine. As reviewed in Chap-
ter 3, adolescents are at a heightened sensitivity to the rewarding effects
of nicotine, and this sensitivity diminishes with age (Adriani et al., 2006;
Jamner et al., 2003). Thus, the probability of use escalating to dependence
after the first few trials is likely to decrease as one moves further away
from adolescence. In addition, better developed executive functions provide
young adults with increased decision-making capacity compared to younger
adolescents, especially during times of emotional arousal (Steinberg, 2007),
and, as a consequence, young adults may be less susceptible to cues to use
tobacco than adolescents. These changes in biological vulnerabilities with
age provide good support for suggesting that initiation rates and overall
prevalence will decline with each increase in the MLA.
However, changes in the prevalence of tobacco use may not necessarily
be linear with increases in the MLA or equal for all segments of under-
age individuals. Consider, for example, the declarative effect of raising
the MLA. Changing the MLA has an indirect effect of helping to change
norms about the acceptability of tobacco use, but this effect may take
time to build. In addition, norms about the acceptability of tobacco use
are also likely to vary by age, with a more stringent perceived unaccept-
ability the farther away one is in age from the MLA. For example, if the
MLA increases to 21, the social unacceptability of smoking is greater for a
16-year-old than it is for a 20-year-old.
Given this assumption that changes in the MLA will have different
effects on adolescents at different ages, the committee considered possible

changes in initiation rates for three age divisions: (1) children and adoles-
cents under age 15, (2) adolescents between 15 and 17, and (3) individu-
als at age 18 (for estimates of MLA of 19) or individuals at ages 18–20
or 21–24, for an MLA change to either 21 or 25. These age groupings
reflect not just differences in years from the MLA but several important
life developmental transitions that play a role in tobacco use as well. These
developmental transitions include the increased mobility that comes with
driving privileges, changes in social networks as adolescents enter and leave
high school or transition to higher education, changes in employment lev-
els and venues, leaving home, and potential changes in relationship status
(e.g., marriage) and parenthood (Arnett, 2000, 2004; IOM and NRC,
2014; S ettersten and Ray, 2010). Each of these life transitions and markers
changes both potential sources of access to tobacco as well as motives for
use (Bachman et al., 2002).
Adolescents Less Than 15 Years of Age

A substantial percentage of adolescents under age 15 are not yet in
high school and, importantly, not yet of driving age. Adolescents under 15
are less likely to have peer networks that include individuals who are over
the MLA (and the distance increases as the MLA increases), and these ado-
lescents are also unlikely to be working in established work environments
where they have coworkers who are over the MLA. Thus, social network
sources and mobility are most restricted for adolescents under age 15. So-
cial sources are the greatest access point for tobacco for underage youth
(see Tables 5-2 and 5-3 in Chapter 5), and changes to the MLA affect the
relative ease of availability of tobacco through social sources. For adoles-
cents under 15 years of age, raising the MLA from 18 to 19 may have only
a modest impact on reducing social sources, given the closeness in age. If
adolescents already have networks with 18-year-olds, then these networks
may also include 19-year-olds who have access to tobacco. Increasing the
MLA to 21, however, provides a greater distancing of social sources. Al-
though 19-year-olds may still be in high schools and thus potentially influ-
ence those under 15, it is far less likely that 21-year-olds are in the same
social networks. Increasing the MLA from 21 to 25, however, is not likely
to achieve any additional notable reductions in social sources for those
under 15 than what is achieved with the 21-year-old MLA policy.
Although social sources play a central role in establishing adolescent
tobacco use patterns, other factors that contribute to early adolescent to-
bacco use (for those who initiate before age 15) may place a limit on the
reductions that would be achieved with increases in the MLA. Adolescents
who reach a level of 100 cigarettes prior to age 15 may be those who are
most susceptible to the reinforcing effects of nicotine, given their neuro

developmental stage and the brain’s increased sensitivity to nicotine during

those prime adolescent years (Jamner et al., 2003; Slotkin, 2002; Spear,
2000). These adolescents are also likely to be ones who have higher levels of
psychological or substance use comorbidities, and ones for whom tobacco
use has critical mood management properties (Kassel et al., 2003). These
adolescents are the ones most likely to have a combination of problem be-
haviors, of which tobacco use is one manifestation (Ellickson et al., 2001;
Silk et al., 2003). Their early use may be accelerated by these reinforcing
and functional properties of nicotine, and these benefits of tobacco use may
outweigh perceived consequences (Baker et al., 2004). These more vulner-
able adolescents may also have social networks within which tobacco and
other substances are more readily available, regardless of age, or they may
have more contact with older individuals (Kobus, 2003). Thus, the com-
mittee expects that there may be a limit to the effect that changes in the
MLA have on this subset of adolescents who initiate prior to age 15 and
that changes in the MLA will not totally eliminate initiation at this young
age. Thus, considering the balance of these factors, the committee estimates
that, for adolescents under age 15, reductions in initiation will be small for
an MLA of 19 and medium for MLAs of 21 and 25.
Adolescents 15 to 17 Years of Age

The committee expects that the greatest gains in reducing tobacco use
will be achieved for adolescents between the ages of 15 and 17. This is
a critical period in which to intervene to prevent not only an initial trial
of tobacco use but also escalation to reach a threshold for initiation. A
substantial proportion of adolescents try tobacco during these high school
years, but most of these adolescents do not escalate beyond initial “experi-
mentation,” and continued access to tobacco products is a major factor in
this progression of use (Widome et al., 2007). Initial trials are often mo-
tivated by opportunity, social influences, as well as in-the-moment image
enhancement, curiosity, and emotional arousals (both positive and negative)
(Sarason et al., 1992). For this age group, negative consequences for to-
bacco use, through parental or school controls, are still relevant (IOM and
NRC, 2011), and changes in the MLA are likely to increase these negative
consequences as social norms adjust. Yet access to tobacco will still exist.
Adolescents in this age group are still most likely to get tobacco through
social sources (CDC, 2014; Johnston et al., 2014b). Between the ages of
15 and 17, youth mobility increases with the arrival driving privileges.
Adolescents’ social networks and potential social sources of tobacco start
to grow as some take on formal, part-time jobs with coworkers who may
be over the MLA. Changing the MLA for tobacco to 19 may not change
social sources substantially for these adolescents, but the committee expects

that raising the MLA to 21 will have a substantial impact in reducing the
prevalence of tobacco use. Raising the MLA further to 25 may provide only
an additional modest reduction in prevalence over that achieved with an
MLA of 21, given that the resulting changes to social network sources may
not be substantially different. However, boosting the MLA to 25 does have
the additional benefit of social norm change.
Balancing these factors, the committee estimates that, for all the policy
options considered, the reduction in initiation in the 15- to 17-year-old age
group will likely be greater than the reduction in initiation among adoles-
cents less than 15 years of age. Furthermore, the committee estimates that
the higher the MLA is raised, the greater the effect on initiation rates is
likely to be.

By age 18, many adolescents graduate from high school and have
numerous transitions, including entering higher education, being exposed
to more adults in the workforce, leaving home, and often experiencing
significant changes in social networks (Arnett, 2000, 2004; Bachman et al.,
2002; IOM and NRC, 2014; Settersten and Ray, 2010). To date, patterns
of initiation have shown a tailing off in initiation by age 18 (see Table 2-8
in Chapter 2). The committee considered an estimate for this age group
specifically only in the case of a raise in the MLA to 19, when individuals at
age 18 are most directly affected by this policy change. Given that the social
networks of 18-year-olds overlap more with the 19-year-olds, the commit-
tee expected a small reduction in initiation for 18-year-olds under an MLA
of 19. The expected effect on initiation rates is higher for an MLA of 21 and
higher still for an MLA of 25. The committee expects the effects of increas-
ing the MLA to 21 or 25 on the initiation rates of 19- and 20-year-olds will
be similar to the effects on 18-year-olds. This expectation of increased effect
is due primarily to the increased social distancing expected when the MLA
is raised to 21 or 25, but it also takes into account the benefits of the ad-
ditional maturing of executive functions, the decreased sensitivity to the
rewarding properties of nicotine, the additional social norms proscribing
tobacco use, and the decreased social value of tobacco and motives for its
use as individuals enter the workforce or parenthood.

Changes in initiation rates for young adults in the 21–24 age group
were considered only for the case of raising the MLA to 25. The probability
of initiation at these ages is substantially less than at earlier ages, given the
developmental changes in life settings and milestones which are likely to

reduce the various motives for smoking (Bachman et al., 2002). In addi-
tion, young adults who have not yet begun smoking have likely developed
coping strategies other than smoking for dealing with mood management
and life stressors. Thus, the overall probability of new initiation is substan-
tially lower. However, current patterns of tobacco marketing suggest that
young adults are increasingly being targeted in tobacco promotions (Ling
and Glantz, 2002), and tobacco promotions are frequently linked with bar
settings and alcohol consumption, which may also keep this age group
susceptible to initiation (Ling and Glantz, 2002). In addition, the commit-
tee considered that there may be more lax enforcement for an MLA of 25.
Considering the balance of factors, the committee anticipates that some
reduction in initiation would still occur with a raise in the MLA to 25 but
that this reduction will be small.
Rebound
Changes in the MLA for tobacco may also create some rebound
e ffects—that is, delaying initiation to a later age. Rebound will result in
increases in initiation over what has been seen historically in a given age
group. The committee anticipates that most of the potential rebound from
delayed initiation will occur in the first year after the new MLA and that
this effect is likely to be modest. The changes in the MLA are likely to have
an effect of further moving social norms and attitudes toward discouraging
tobacco use and making it less appealing. These social normative changes
may help to reduce rebound effects. Rebound may be most likely at the
lower end of young adulthood (18–21) and very unlikely after age 25, when
decision making has matured, individuals have established other coping
strategies, and normative developmental life changes often further push
individuals away from tobacco use.
Intensity
The models commissioned by the committee considered only changes
in initiation and not changes in intensity of smoking. Intensity is important
to consider both because of its strong association with nicotine dependence
(Fagerstrom et al., 1990; Prokhorov et al., 1996), and thus difficulties in
quitting tobacco use and also because of the strong dose–response rela-
tionship between smoking intensity and morbidity (Hu et al., 2006). It is
reasonable to expect that changes in the MLA for tobacco will also change
the intensity of smoking for underage individuals, given the likely resulting
changes in the ease of access to tobacco. At the same time, though, there is
a background of ongoing historical changes, with overall consumption and
intensity both decreasing among smokers, as there are more environmental

restrictions on tobacco use and increasing prices (HHS, 2014). Thus, it is

difficult to estimate the independent gains in reducing intensity that will
result from changes in the MLA, although the committee expects that these
additional gains may be modest. Given this level of uncertainty and the lack
of data about potential reductions in intensity, changes in intensity are not
included in the modeling. For that reason the overall model estimates may
ultimately underrepresent the potential health gains of changes in the MLA
for tobacco.
Summary of Committee Estimates and Conclusions of the

Likely Effects of Raising the MLA on Tobacco Use Initiation
Table 7-1 summarizes the committee’s ordered, categorical estimates
of the effects that changes in the MLA will have in reducing initiation for
the different age groups. The committee has more confidence in its esti-
mates pertaining to the raising of the MLA to 19 or 21 than it does for
raising the MLA to 25 because of the greater level of extrapolation needed
for estimating change and other factors with increasing age. There are a
variety of reasons for the uncertainty in these estimates. One is the lack of
empirical evidence directly linking changes in the MLA and levels of MLA
enforcement with changes in tobacco use. Another is the changing array of
available tobacco products and uncertainty about how these new products
may change patterns of tobacco use. The estimates being used as inputs for
the simulation models include a range of potential values, with a broader
range for the MLA of 25.
Conclusion 7-1: Increasing the minimum age of legal access to tobacco

products will likely prevent or delay initiation of tobacco use by ado-
Conclusion 7-2: Although changes in the minimum age of legal access

to tobacco products will directly pertain to individuals who are age 18
or older, the largest proportionate reduction in the initiation of tobacco
use will likely occur among adolescents 15 to 17 years old.
Conclusion 7-3: The impact on initiation of tobacco use of raising

the minimum age of legal access to tobacco products (MLA) to 21
will likely be substantially higher than raising it to 19, but the added
effect of raising the MLA beyond age 21 to age 25 will likely be con-
siderably smaller.
The previous section outlined, in qualitative terms, the expected effects

of raising the MLA on initiation of tobacco use. The modeling exercise

TABLE 7-1 Committee Estimates Regarding Effects on Initiation
Reduction in Initiation by Age Group MLA 19 MLA 21 MLA 25

Reduction in initiation for adolescents under age 15 small medium medium
Reduction in initiation for adolescents ages 15–17 small large large
Reduction in initiation for young adults age 18 small medium medium
Reduction in initiation for young adults ages 19–20 n/a medium medium
Reduction in initiation for young adults ages 21–24 n/a n/a small

Characteristics of the Age Group That

Might Influence Responsiveness to an Factors Related to Effects on This Age Group
MLA Increase as the MLA Increase Is Larger
The youngest adolescents who have Younger adolescents are less likely to be
access to tobacco products and who in social groups with older adolescents
persist in using tobacco beyond the first or young adults. Their mobility is most
experimentation with them are ones who restricted, depending on parents and other
may be most susceptible to the reinforcing adults for transportation, thus reducing ease
effects of nicotine or who have higher of access. Social sources remain the greatest
levels of psychological or substance use access point for tobacco products, so the
comorbidities, placing them at greater risk characteristics of social networks and the
for escalation beyond 100 cigarettes and problem behaviors in those networks matter.
into established initiation and smoking. Thus, the effect of MLA 19 will be less than
These youth may be less affected by that of MLA 21 or MLA 25.
increases in the MLA than even slightly
older youth.
Changes in the MLA will increase the MLA 21 will begin to change access to
negative social consequences of tobacco tobacco products from social sources, much
use; adolescents at this age are most more so than MLA 19. MLA 25 will have
likely to get tobacco from social sources, only modest additional changes to social
including from coworkers above the MLA. network, but includes benefits of social norm
change.
Most graduate from high school and Networks of 18-year-olds overlap with
experience life transitions. However, some 19-year-olds but less so with those 21 or over.
18-year-olds are in high school with, or
are friends with, 19-year-olds who could
purchase tobacco products.
Young adults benefit from increased 19- and 20-year-olds are often in college or
executive functioning, as well as decreased the workforce, and their network of friends
sensitivity to rewarding properties of includes those age 21 and older. MLA 25 will
nicotine and decreased social value of have only modest additional changes to their
tobacco and motives for use as individuals social network, but includes benefits of social
enter workforce or parenthood. norm change.
The probability of initiation among young There could be more lax enforcement of
adults ages 21 to 24 is substantially MLA 25 in this age group. The tobacco
less than at earlier ages. Developmental industry engages in extensive marketing in
changes in life setting and milestones are bars to which this age group will have legal
likely to reduce motives for smoking. access; many young adults link smoking and
Young adults in this age group have likely drinking behaviors.
developed coping strategies other than
smoking.

under taken by the committee required quantification of these effects;

the next section details the process by which the committee translated the
qualitative terms into specific quantitative estimates of effects on the rate of
initiation of smoking for various age segments. The SimSmoke and CISNET
models used those committee-estimated initiation rates to project the effects
of a change in the MLA on smoking prevalence. Chapter 8 then translates
those quantitative estimates of effects on initiation and prevalence into
quantitative estimates of effects on smoking-related premature mortality,
lung cancer mortality, and maternal and child health outcomes using the
simulation models that track lifetime trajectories of smoking behavior
post-initiation. Chapter 8 also includes a discussion of the likely effect of
changes in tobacco use on the many important adverse health effects not
encompassed by the modeling exercises.
ESTIMATED INITIATION EFFECT SIZES

As described above, the committee used a consensus process to arrive at
estimates for changes in initiation rates. As shown above in Table 7-1, the
committee decided on three qualitative descriptors, labeled small, medium,
and large. The committee attached numeric ranges to each of these magni-
tude descriptors. The ranges increase in increments of 5 percent (to avoid
implying an unrealistic precision in the estimates) from 5 to 30 percent for
potential changes in initiation. Small effects were considered to be 5 and
10 percent; medium effects were 15 and 20 percent; and large effects were
25 and 30 percent.
These estimates can be compared to effect sizes from a variety of tobacco
control policies. The committee provides this brief summary not to make
direct comparisons between other researchers’ findings and the committee’s
estimates but to illustrate that the committee’s informed judgment about the
likely effects of raising the MLA falls within the range of relevant effect sizes
identified by or considered reasonable by other tobacco control researchers.
For example, tobacco control policy modules incorporated into the
SimSmoke model used estimates from an expert judgment process to p roject
the effects of a variety of interventions on adolescent, young adult, and
adult smoking behaviors. As shown in Table D-1, these effects range from
1 percent to 50 percent. Of note, the effect size of the youth access restric-
tion module in SimSmoke range from 2.5 percent to 30 percent. A modeling
exercise assessing the cost-effectiveness of raising the legal smoking age in
California, an effort not dissimilar from that in this report, used a range
of 10 to 50 percent decrease in initiation for the projected effect on those
under age 21 (Ahmad, 2005).
Flay (2007) summarized the effects of school-based prevention pro-
grams using a relative risk reduction calculation and estimated that the

potential medium-term (2–4 years) effects of a national program of well-

implemented, school-based smoking prevention programs of proven effec-
tiveness would be approximately 28 percent, rising to a potential 31 percent
if the programming included a mass media component as well. A recent
review of the effectiveness of increases in tobacco product excise taxes
and fees on initiation of tobacco use in young people reports −0.43 as the
median estimate of the elasticity of adolescent initiation with respect to
price, meaning that a 10 percent increase in price would result in a 4.30 per-
cent decrease in initiation (CPSTF, 2012).
Table 7-2 summarizes the committee’s estimates for percent reductions
in initiation rates and potential rebound effects for the different age group-
ings assuming a change in the MLA to 19, 21, or 25 years of age, referred to
as MLA 19, MLA 21, or MLA 25, respectively. The effect sizes used in the
modeling reflect the committee’s judgment about the effect of an increase in
the MLA on the entire United States. Some regions or subpopulations might
experience larger effects, some smaller. The largest effect size used in other
modeling exercises identified by the committee is 50 percent. The committee
thinks that is overly optimistic and chose to use more conservative estimates
in the modeling, although upper estimates are provided.
In addition, the committee recognized that although there is limited
direct evidence about how raising the MLA might affect tobacco use at
different ages, there is less confidence about some effects than others. The
committee is the most confident about the estimates related to an MLA of
19 and 21 and is much less confident regarding estimates related to an MLA
of 25 because of the greater level of extrapolation needed for estimating
changes. To address this uncertainty, the committee includes ranges (upper
and lower estimates) around each mid-estimate that vary according to the
degree of the committee’s uncertainty. Thus, the estimates for the MLA of
25 used the broader range. The effect ranges do not represent bounds or a
measure of uncertainty in the classical statistical sense. Rather these values
reflect ranges that the committee deemed plausible (see Table 7-3). As will
be discussed, the models simulate national cigarette smoking patterns and,
in Chapter 8, their consequences. However, the committee’s effect sizes are
percentage decreases from the status quo and thus would apply to any ju-
risdiction of any size assuming the jurisdiction roughly mirrors the United
States as a whole. Absolute numbers of people affected would vary with the
size of the population. The implications of this are discussed in Chapter 9.
MODELING
For this report, the committee used the CISNET smoking population
model (hereafter referred to simply as the CISNET model) calibrated to
data from 36 National Health Interview Surveys covering the years 1965–

206
TABLE 7-2 Committee Estimates Regarding Effects on Initiation with Qualitative Descriptors and Numeric Estimates
MLA 19 MLA 21 MLA 25
Qualitative Numeric Qualitative Numeric Qualitative Numeric
Reduction in Initiation by Age Group Descriptor Estimate Descriptor Estimate Descriptor Estimate
Reduction in initiation for adolescents under age 15 small 5% medium 15% medium 15%
Reduction in initiation for adolescents ages 15–17 small 10% large 25% large 30%
Reduction in initiation for young adults age 18 small 10% medium 15% medium 20%
Reduction in initiation for young adults ages 19–20 n/a n/a medium 15% medium 20%
Reduction in initiation for young adults ages 21–24 n/a n/a n/a n/a small 5%

TABLE 7-3 Committee Inputs to the Model for Each MLA Policy Option with Qualitative Descriptors, Numeric
Estimates, and Upper and Lower Range Estimates
MLA 19 MLA 21 MLA 25
Qualitative Qualitative Qualitative
Descriptor Upper and Descriptor Upper and Descriptor Upper and
and Numeric Lower and Numeric Lower and Numeric Lower
Change in Initiation Estimate Estimatesa Estimate Estimatesa Estimate Estimatesa
Reduction in initiation for small (5%) (4.2%, 6%) medium (12.5%, 18%) medium (10.7%, 21%)
adolescents under age 15 (15%) (15%)
Reduction in initiation for small (10%) (8.3%, 12%) large (25%) (20.8%, 30%) large (30%) (21.4%, 42%)
adolescents ages 15–17
Reduction in initiation for small (10%) (8.3%, 12%) medium (12.5%, 18%) medium (14.3%, 28%)
young adults age 18 (15%) (20%)
Reduction in initiation for n/a n/a medium (12.5%, 18%) medium (14.3%, 28%)
young adults ages 19–20 (15%) (20%)
Reduction in initiation for n/a n/a n/a n/a small (5%) (3.6%, 7%)
young adults ages 21–24
Rebound small (5%) (4.2%, 6%) small (5%) (4.2%, 6%) none (0%) n/a
Duration of rebound 2 n/a 2 n/a 0 n/a
(in number of years)
aUpper and lower estimates reflect the uncertainty ranges, with a smaller range (1.2) for MLA 19 and MLA 21, and a larger range (1.4) for MLA

25. The mid-estimate is treated as a geometric mean, not an arithmetic mean, thus, upper estimates are calculated as 1.2(x) and lower estimates as
x/1.2 for MLA 19 and MLA 21 and as 1.4(x) and x/1.4 for MLA 25, resulting in slightly nonsymmetric ranges around the mid-estimates. These
ranges do not represent bounds or a measure of uncertainty in a classical statistical sense. Rather, these values reflect ranges that the committee
deemed plausible as described in the text.
207
2012 (Holford et al., 2014) to simulate age-specific smoking prevalence

and mortality outcomes for birth cohorts projected through 2100. Using
the smoking initiation estimates developed by the committee as inputs,
the model was used to assess the potential effects of raising the MLA on
U.S. smoking patterns by birth cohort and calendar year and the corre-
sponding smoking-attributable mortality, life expectancy, and lung cancer
deaths, using lung cancer projections from the CISNET Yale Lung Cancer
Model (Holford et al., 2012).This report also used a recently updated ver-
sion of the SimSmoke model that assumes the same smoking-rate inputs
used by the CISNET model based on NHIS (Holford et al., 2014) and
beginning in 1965. SimSmoke is able to reproduce the population smoking
patterns by gender and by age in the United States from 1965 through 2012
(Levy, under review; Levy et al., under review), and it predicts the impact
of current and future policies from 2015 through 2100. The initiation rates
from 2015 forward reflect the effect of past policies under the assumption
that policies other than raising the MLA will remain at 2014 levels into the
future. As such, the initiation rates remain constant in future years under
the status quo policy option. More details about the models can be found
in Appendix D.
The models focused on characterizing the potential effects of raising
the MLA on future rates of smoking initiation rates by age (the probability
of becoming an established smoker at a given age, conditional on not hav-
ing started before), considering both the prevention of smoking initiation
for individuals younger than the new MLA (prevented initiation) and the
delay of smoking initiation for some individuals who will start at an older
age because of the policy (delayed initiation). The modeled policy effects
varied by age, with assumed reductions in smoking initiation rates among
individuals younger than the new MLA, while allowing for a potential in-
crease (rebound) in the smoking initiation rates for individuals of or above
the MLA (delayed smoking initiation).
Effects of Raising the MLA on Smoking Initiation

Figure 7-1 shows the initiation rates for the baseline (the current MLA)
for both CISNET and SimSmoke and the corresponding mid-estimate for
the smoking initiation inputs for the MLA 21. The SimSmoke baseline
initiation rates are generally higher than those of the CISNET model. The
specific initiation rates used in both models can be accessed on the CISNET
webpage.1 Applying the CISNET initiation rates directly to a hypothetical
birth cohort of 100,000 individuals would result roughly in 30,000 ever
smokers by age 40. Applying the reductions in the mid MLA 21 scenario
1 See https://resources.cisnet.cancer.gov/projects/#shg/iomr.

CISNET Males CISNET Females

Initiation Probability by Age

Baseline
0.04
0.04
● MLA21 mid−scenario
●
●● ●
0.02
0.02
●
● ● ●
● ●● ●
● ●
● ● ●●
● ● ● ●
●● ●●
● ●● ● ●●
● ●●● ●●●
0.00
0.00
●●●●●●●● ● ●●●●●●●
●●●●●●●●●●● ●●●●●●●●●●●● ●●●●
0 10 20 30 40 0 10 20 30 40
Age Age
SimSmoke Males SimSmoke Females


0.04
0.04
●
●
●●
●
● ●
● ●
0.02
0.02
● ●●
● ●● ●
●
● ●
● ●
● ●● ●
●
● ● ●
0.00
0.00
● ●
●●●●●●●●●●● ●●●●●●●●●●●●● ●●●●●●●●●●●● ●●●●●●●●●●●●●●●
0 10 20 30 40 0 10 20 30 40
Age Age
FIGURE 7-1 Initiation rates by age under baseline and middle scenarios for the
effects of raising the MLA to 21.
would translate into 10 percent fewer smokers, with approximately 3,000

individuals never initiating plus another 600 individuals delaying smoking
initiation until an older age (data not shown).
Smoking Prevalence
The models estimate the impact of reduced and delayed initiation on
future annual U.S. smoking prevalence (described below) and smoking-
related health outcomes (shown in Chapter 8), assuming that the MLA
would change in 2015 and go into full effect immediately (with progres-
sive staggered implementation evaluated in sensitivity analyses). Although
raising the MLA could also affect future rates of cessation and smoking

intensity, the models did not incorporate effects on these rates because of
uncertainty about the potential impacts. Thus, this represents a conserva-
tive assumption (one that underestimates the health benefits of the MLA
policy), considering the substantial evidence linking delayed initiation with
higher smoking cessation and lower smoking consumption rates.
Status Quo Projections

Any projections into the future imply some uncertainty because one
cannot observe future outcomes before they occur. However, models provide
an opportunity to explore the potential outcomes associated with various
policy options and compare with the status quo, assuming all else remains
unchanged. Both of the models employed by the committee begin with the
creation of a baseline projection that assumes no change in the MLA; this
serves as the status quo projection. The models are then run assuming the
altered initiation rates provided by the committee (see Table 7-2) beginning
after 2015. To characterize the incremental impacts of policy changes that
are predicted by the model, the committee subtracted the outcome result for
the baseline or status quo policy from the outcome result for the new MLA
policy and then divided by the baseline outcome result, thus expressing the
change as a percentage.
Both models project the baseline smoking prevalence in the United
States from 2015 to 2100 assuming that smoking initiation and cessation
rates will remain the same in all future years, but they do so in different
ways and at different levels. CISNET projects that the age-specific initiation
and cessation rates by gender estimated for the 1980 birth cohort will apply
to all future birth cohorts. By contrast, SimSmoke assumes that the esti-
mated age-specific initiation and cessation rates by gender observed for the
year 2014 will persist throughout the modeled horizon (effectively assum-
ing that tobacco control policies will remain at current levels). Although
the differences may seem relatively minor, they lead to different projected
smoking rates at the baseline for the two models. To facilitate comparison
of the projected policy consequences associated with raising the MLA, the
focus here will be on the relative effects of the MLA policy (i.e., the per-
centage reductions in smoking, mortality, and other health outcomes) while
noting uncertainty about the absolute magnitude of the status quo.
Effects of Changing the MLA on Smoking Prevalence

Figure 7-2 shows projected smoking prevalence in the United States
from 2014 to 2100 by gender as estimated by the CISNET model for the
status quo and the three MLAs considered. The figure shows that even
under the status quo, the CISNET model predicts a decrease in adult smok-

Adult Smoking Prevalence

Smoking prevalence (%)
Status Quo
8 10 1 14 1 18
MLA9
MLA 21
MLA25
2020 2040 2060 2080 2100

Year
Women Men
10 12 14 16

8 12 16 20
8
6
2020 2060 2100 2020 2060 2100

Year Year
FIGURE 7-2 CISNET model-projected smoking prevalence for the three mid-MLA
scenarios for adults (18+), adult women, and adult men in the United States for
2014–2100.
ing prevalence from 18 percent in 2014 to 9 percent in 2100 (15 percent in

2014 to 7 percent in 2100 for females and 21 percent in 2014 to 11 percent
in 2100 for males).
The figure also suggests that the MLA 21 and MLA 25 options lead
to considerable further reductions in smoking prevalence relative to MLA
19. Switching to a progressive staggered implementation of the policy (i.e.,
for an MLA greater than 19, increasing the MLA by 1 year each calendar
year until reaching the desired MLA) did not significantly change the results
(not shown).
Figure 7-3 shows the corresponding projection from the SimSmoke
model. As can be seen, the SimSmoke model also projects a significant

18 Adult Smoking Prevalence

Status Quo
ML$ 19
16
MLA 21
MLA25
14
12
10
2020 2040 2060 2080 2100

Year
Women 10 12 14 16 18 20 Men
16

14
12
10
8
2020 2060 2100 2020 2060 2100

Year Year
FIGURE 7-3 Projected smoking prevalence predicted by the SimSmoke model for
adults (18+), adult women, and adult men in the United States for 2014–2100.
reduction in adult smoking prevalence—from 17 percent in 2014 to 13 per-

cent in 2100 (15 percent in 2014 to 11 percent in 2100 for females and
19 percent in 2014 to 14 percent in 2100 for males)—to occur in the fol-
lowing decades. As shown in these figures, the SimSmoke model projects
smaller decreases in prevalence than does the CISNET model because of the
underlying lower baseline smoking initiation and higher (not shown) ces-
sation rates in the CISNET model. Both models project that MLA 21 and
MLA 25 would lead to larger reductions in smoking prevalence compared
to MLA 19.
Table 7-4 shows a comparison between the two models of the pro-
jected adult smoking prevalence and the absolute percentage reductions in
prevalence versus the status quo for selected years. The ranges in percentage

TABLE 7-4 Adult (18+) Smoking Prevalence (%) and Percentage (%) Reduction for Selected Years for the Mid-
Estimates of Initiation Inputs from Table 7-3 (lower and upper estimate results shown in parentheses)
MLA/Outcome 2020 2040 2060 2080 2100
SQ (status quo)
SimSmoke prevalence 15.7 13.1 12.7 12.7 12.7
CISNET prevalence 15.2 10.4 9.1 8.8 8.7
MLA 19
reduction versus SQ 0.4% 1.8% 2.7% 3.0% 3.0%
(0.2, 0.6) (1.3, 2.4) (1.9, 3.5) (2.1, 3.9) (2.1, 3.9)
(0.14, 0.21) (1.5, 2.3) (2.4, 3.8) (2.7, 4.3) (2.7, 4.3)
MLA 21
(1.5, 2.4) (5.8, 8.9) (8.3, 12.7) (9.0, 13.7) (9.0, 13.7)
(0.37, 0.53) (5.4, 8.8) (8.8, 12.9) (9.9, 14.5) (10.0, 14.7)
MLA 25

(2.9, 4.8) (9.2, 15.2) (12.6, 20.9) (13.4, 22.3) (13.4, 22.3)
(0.36, 0.71) (5.9, 11.7) (9.8, 19.4) (11.1, 21.9) (11.2, 22.1)
213
reduction represent the results from the lower and upper estimate scenarios
(see Table 7-3) for effects on initiation for each MLA option. The results
in Table 7-4 demonstrate that although the absolute prevalence predictions
differ between the models, the two models predict similar percentage reduc-
tions in smoking for each MLA relative to the status quo. Specifically, both
models estimate a roughly 3 percent decrease in the 2100 prevalence for
the mid-MLA 19, an 11–12 percent decrease for the mid-MLA 21 scenario,
and a 15.7 percent decrease for the mid-MLA 25 scenario.
Summary of Smoking Prevalence Projections

The modeling analysis suggests that raising the MLA for tobacco prod-
ucts could lead to considerable reductions in smoking prevalence. Both
models suggest that it would take about a decade for the reductions in
population-wide smoking prevalence to become meaningful; the delay can
be attributed to the nature of the policy, which primarily affects children,
adolescents, and young adults, so the effects become apparent only after
those individuals affected by the policy have aged. Still, the projections
show that with time the potential reductions and delays in smoking initia-
tion would accumulate and lead to considerable decreases in prevalence.
Both models suggest that there is a considerable difference between the
results of MLA 19 and MLA 21. Increasing the MLA from 21 to 25 leads
to additional reductions, but they are smaller than the changes seen increas-
ing the MLA from 19 to 21. This reflects the uncertainty in the assumed
smoking initiation reductions for each MLA and the overlapping ranges for
MLA 21 and MLA 25 (wider effect ranges for MLA 25).
Finding 7-1: Two policy simulation models project significant reduc-

tions in smoking prevalence from 2015 to 2100 in the United States
in a status quo policy that captures the benefits from prior tobacco
control policies.
Finding 7-2: The models predict that raising the minimum age of legal
access to tobacco products would lead to additional reductions beyond
the status quo in smoking prevalence based on reasonably conservative
assumptions about the potential reductions in smoking initiation rates.
Finding 7-3: Raising the minimum age of legal access to tobacco prod-
ucts to 21 or 25 years would lead to larger reductions in smoking
prevalence than the status quo or an increase of the MLA to 19.


legal access to tobacco products, particularly to age 21 or 25, will likely
lead to substantial reductions in smoking prevalence.
As discussed in Chapter 2, tobacco use is far from uniform among vari-

ous subpopulations and varies, for example, by race and ethnicity, social
and economic status, geography, incarceration status, and the presence of
mental illness (Bachman et al., 2011; Cropsey et al., 2004, 2008; Goodman
and Capitman, 2000; Green et al., 2007; HHS, 2012; Johnson et al., 2000;
Johnston et al., 2014a; Kann et al., 2014; Melnick et al., 2001; Peek et
al., in preparation; SAMHSA, 2012; Welte et al., 2011; Ziedonis et al.,
2008). Tobacco control advocates interested in decreasing tobacco use are
particularly concerned about closing the “equity gap” by reducing tobacco
use among the highest-risk populations. An important consideration for
the committee is whether a change in the MLA would differentially affect
high-risk populations with initiation rates that vary significantly from the
national averages considered in this report, including the rates contained in
the modeling. One possibility is that groups with higher-than-average initia-
tion rates would remain relatively resistant to tobacco control interventions
and the effect would be smaller in those populations, widening the equity
gap. The equity gap could be narrowed if groups with lower-than-average
initiation rates respond less to an increase in the MLA. The third possibility
is that the effects will not vary significantly between groups.
The literature provides little evidence to clarify this issue. Two recent
systematic reviews of the effects of population-level tobacco control inter
ventions on adolescents and young adults found no clear evidence of a
differential impact by social factors. One review found “little evidence of
policies that have the potential to increase inequalities” (Thomas et al.,
2008, p. 235). The second review identified price as the only intervention
with a consistent effect that would decrease the inequalities in smoking
initiation (Brown et al., 2014). Given the extremely limited data available
and the fact that the models are not equipped to analyze according to high-
risk populations, the committee did not produce separate analyses of the
effect of raising the MLA by subpopulation. The committee’s conclusions
also do not anticipate the changing landscape of tobacco products—in
particular, the burgeoning popularity of electronic nicotine delivery systems
(ENDS) (e.g., “e-cigarettes”). This new pattern of tobacco use creates vari-
ous unknowns. The committee has no basis on which to conclude that the
effect of a change in the MLA would have more or less effect on initiation
with ENDS than with other tobacco products. Both of these limitations are
discussed further in Chapter 9.

REFERENCES
Adriani, W., V. Deroche-Gamonet, M. Le Moal, G. Laviola, and P. V. Piazza. 2006. Pre
exposure during or following adolescence differently affects nicotine-rewarding proper-
ties in adult rats. Psychopharmacology 184(3–4):382–390.
Ahmad, S. 2005. Closing the youth access gap: The projected health benefits and cost savings
of a national policy to raise the legal smoking age to 21 in the United States. Health
Policy 75(1):74–84.
Arnett, J. J. 2000. Emerging adulthood: A theory of development from the late teens through
the twenties. American Psychologist 55(5):469–480.
———. 2004. Emerging adulthood: The winding road from the late teens through the twenties.
New York: Oxford University Press.
Bachman, J. G., P. M. O’Malley, J. E. Schulenberg, L. D. Johnston, A. L. Bryant, and A. C.
Merline. 2002. The decline of substance use in young adulthood changes in social activi-
ties, roles, and beliefs. Mahwah, NJ: Lawrence Erlbaum Associates.
Bachman, J. G., P. M. O’Malley, L. D. Johnston, J. E. Schulenberg, and J. M. Wallace, Jr. 2011.
Racial/ethnic differences in the relationship between parental education and substance
use among U.S. 8th-, 10th-, and 12th-grade students: Findings from the Monitoring the
Future project. Journal of Studies on Alcohol and Drugs 72(2):279–285.
Baker, T. B., T. H. Brandon, and L. Chassin. 2004. Motivational influences on cigarette smok-
ing. Annual Review of Psychology 55:463–491.
Brown, T., S. Platt, and A. Amos. 2014. Equity impact of interventions and policies to reduce
smoking in youth: Systematic review. Tobacco Control 23(e2):e98–e105.
CDC (Centers for Disease Control and Prevention). 2014. National Youth Tobacco Survey.
http://www.cdc.gov/tobacco/data_statistics/surveys/nyts (accessed February 8, 2015).
CPSTF (Community Preventive Services Task Force). 2012. Reducing tobacco use and second
hand smoke exposure: Interventions to increase the unit price for tobacco products.
http://www.thecommunityguide.org/tobacco/RRincreasingunitprice.html (accessed Janu-
ary 20, 2015).
Cropsey, K., G. D. Eldridge, and T. Ladner. 2004. Smoking among female prisoners: An ig-
nored public health epidemic. Addictive Behaviors 29(2):425–431.
Cropsey, K. L., J. A. Linker, and D. E. Waite. 2008. An analysis of racial and sex differences
for smoking among adolescents in a juvenile correctional center. Drug and Alcohol
Dependence 92(1–3):156–163.
156(5):818–822.
Ellickson, P. L., J. S. Tucker, and D. J. Klein. 2001. High-risk behaviors associated with early
smoking: Results from a 5-year follow-up. Journal of Adolescent Health 28(6):465–473.
Fagerstrom, K. O., T. F. Heatherton, and L. Kozlowski. 1990. Nicotine addiction and its
assessment. Ear, Nose, and Throat Journal 69(11):763–765.
Flay, B. R. 2007. The long-term promise of effective school-based smoking prevention pro-
grams. Paper commmissioned by the Committee on Reducing Tobacco Use: Strategies,
Barriers, and Consequences, Institute of Medicine. In Ending the tobacco problem (see
Appendix D). Edited by R. J. Bonnie. Washington, DC: The National Academies Press.
Goodman, E., and J. Capitman. 2000. Depressive symptoms and cigarette smoking among
teens. Pediatrics 106(4):748–755.
Green, M. P., K. L. McCausland, H. Xiao, J. C. Duke, D. M. Vallone, and C. G. Healton.
2007. A closer look at smoking among young adults: Where tobacco control should focus
its attention. American Journal of Public Health 97(8):1427–1433.

Health.
Holford, T. R., K. Ebisu, L. McKay, C. Oh, and T. Zheng. 2012. Chapter 12: Yale lung cancer
model. Risk Analysis 32(Suppl 1):S151–S165.
Holford, T. R., D. T. Levy, L. A. McKay, L. Clarke, B. Racine, R. Meza, S. Land, J. Jeon,
and E. J. Feuer. 2014. Patterns of birth cohort-specific smoking histories, 1965–2009.
American Journal of Preventive Medicine 46(2):e31–e37.
IOM and NRC (Institute of Medicine and National Research Council). 2011. The science
of adolescent risk-taking: Workshop report. Washington, DC: The National Academies
Press.
5(Suppl 1):S71–S87.
Johnson, J. G., P. Cohen, D. S. Pine, D. F. Klein, S. Kasen, and J. S. Brook. 2000. Association
between cigarette smoking and anxiety disorders during adolescence and early adulthood.
JAMA 284(18):2348–2351.
Monitoring the Future: National survey results on drug use, 1975–2013. Volume 1:
Secondary school students. Ann Arbor, MI: Institute for Social Research, University of
Michigan.
Monitoring the Future: National survey results on drug use, 1975–2013: Overview, key
findings on adolescent drug use. Ann Arbor, MI: Institute for Social Research, University
of Michigan.
O’Malley Olsen, T. McManus, D. Chyen, L. Whittle, E. Taylor, Z. Demissie, N. Brener, J.
Thornton, J. Moore, and S. Zaza. 2014. Youth risk behavior surveillance—United States,
2013. Morbidity and Mortality Weekly Report 63(4):1–168.
Kassel, J. D., L. R. Stroud, and C. A. Paronis. 2003. Smoking, stress, and negative affect:
Correlation, causation, and context across stages of smoking. Psychological Bulletin
129(2):270–304.
Kobus, K. 2003. Peers and adolescent smoking. Addiction 98(Suppl 1):37–55.
Levy, D. T. Under review. The use of SimSmoke to investigate maternal child health outcomes
with exposure to cigarette smoke in the United States.
Levy, D. T., R. Meza, Y. Zhang, and T. R. Holford. Under review. The role of tobacco control
policies in reducing US smoking prevalence from 1965-2012: Results from the SimSmoke
Tobacco Policy Simulation Model.

92(6):908–916.
Melnick, M. J., K. E. Miller, D. F. Sabo, M. P. Farrell, and G. M. Barnes. 2001. Tobacco use
among high school athletes and nonathletes: Results of the 1997 Youth Risk Behavior
Survey. Adolescence 36(144):727–747.
Peek, K. J., J. Kusel, S. L. Eyre, C. L. Auerswald, and B. L. Halpern-Felsher. In preparation.
Understanding smokeless tobacco initiation and use among rural youth: An unexplored
population.
Prokhorov, A. V., U. E. Pallonen, J. L. Fava, L. Ding, and R. Niaura. 1996. Measuring nicotine
dependence among high-risk adolescent smokers. Addictive Behaviors 21(1):117–127.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2012. National Sur-
vey on Drug Use and Health: Comparison of 2002–2003 and 2010–2011 model-based
prevalence estimates (50 states and the District of Columbia). Rockville, MD: Substance
Abuse and Mental Health Services Administration, Center for Behavioral Statistics and
Quality, Department of Population Surveys.
Sarason, I. G., E. S. Mankowski, A. V. Peterson, Jr., and K. T. Dinh. 1992. Adolescents’
reasons for smoking. Journal of School Health 62(5):185–190.
Settersten, R., and B. E. Ray. 2010. Not quite adults. New York: Bantam Books.
Silk, J. S., L. Steinberg, and A. S. Morris. 2003. Adolescents’ emotion regulation in
daily life: Links to depressive symptoms and problem behavior. Child Development
74(6):1869–1880.
Slotkin, T. A. 2002. Nicotine and the adolescent brain: Insights from an animal model. Neuro
toxicology and Teratology 24(3):369–384.
Spear, L. P. 2000. The adolescent brain and age-related behavioral manifestations. Neuro
science and Biobehavioral Reviews 24(4):417–463.
Steinberg, L. 2007. Risk taking in adolescence: New perspectives from brain and behavioral
science. Current Directions in Psychological Science 16(2):55–59.
Thomas, S., D. Fayter, K. Misso, D. Ogilvie, M. Petticrew, A. Sowden, M. Whitehead, and
G. Worthy. 2008. Population tobacco control interventions and their effects on social
inequalities in smoking: Systematic review. Tobacco Control 17(4):230–237.
Welte, J. W., G. M. Barnes, M.-C. O. Tidwell, and J. H. Hoffman. 2011. Tobacco use, heavy
use, and dependence among adolescents and young adults in the United States. Substance
Use and Misuse 46(9):1090–1098.
Widome, R., J. L. Forster, P. J. Hannan, and C. L. Perry. 2007. Longitudinal patterns of youth
access to cigarettes and smoking progression: Minnesota Adolescent Community Cohort
(MACC) study (2000–2003). Preventive Medicine 45(6):442–446.
Ziedonis, D., B. Hitsman, J. C. Beckham, M. Zvolensky, L. E. Adler, J. Audrain-McGovern,
N. Breslau, R. A. Brown, T. P. George, and J. Williams. 2008. Tobacco use and cessation
in psychiatric disorders: National Institute of Mental Health report. Nicotine & Tobacco
Research 10(12):1691–1715.

Health Benefits of
Raising the Minimum Age of
Legal Access to Tobacco Products
T
he preceding chapter describes the committee’s conclusions regard-
ing the likely effects of raising the minimum age of legal access to
tobacco products (MLA) on initiation of tobacco use by adolescents
and young adults under each of the three policy options: MLA 19, MLA 21,
or MLA 25. The committee uses SimSmoke and Cancer Intervention and
Surveillance Modeling Network (CISNET) simulation modeling to project
numerical estimates of how, through to the year 2100, these effects on ini-
tiation would affect cigarette smoking prevalence, as the cohorts affected
by an MLA increase age into adulthood and, in fact, through middle and
older ages. This chapter uses those changes in initiation and prevalence
to model the likely effects on morbidity and mortality. Projections from
CISNET and SimSmoke include some measures of mortality (premature
deaths, years of life lost [YLL], and lung cancer deaths) and of morbidity
(low birth weight, pre-term birth, and sudden infant death syndrome, or
SIDS). The chapter concludes with the committee’s findings and conclu-
sions on the likely e ffects of raising the MLA on the many other important
health outcomes not included in the modeling exercise. See Appendix D for
a detailed discussion of the models.
PREMATURE DEATHS PREVENTED

The CISNET model provided estimates of the smoking-attributable
mortality by birth cohort (generation) for each policy option for raising
219

the minimum age of legal access to tobacco products.1 The mortality pre-
dictions by birth cohort summarize in a single statistic the cumulative ef-
fects of raising the MLA on the mortality experienced by new generations
throughout their lifetime.
Table 8-1 presents the CISNET model projections of lifetime deaths
prevented by birth cohort (i.e., for the hypothetical population of U.S.
individuals born in 2000–2019, 2020–2039, . . ., and 2080–2099) for the
status quo as well as the premature deaths2 prevented by the mid-scenario
of the three MLA policy options for initiation, along with the percentage
mortality reduction. The projections show that for each MLA the percent-
age reduction in premature deaths appears to be consistent across birth
cohorts; this makes sense because all the cohorts would reach adulthood
after—sometimes substantially after—implementation of the law. Nonethe-
less, the number of deaths prevented for each birth cohort varies because
of differences in the projected size of these different cohorts, with more
lives saved in a larger cohort than in a smaller cohort even with the same
proportionate reductions. The results show that MLA 19 could reduce the
lifetime smoking-attributable deaths versus the status quo by approximately
3 percent, with reductions of 11 percent for MLA 21 and 15 percent for
MLA 25. Hence, the projected reductions in smoking-related deaths track
the long-run projected declines in smoking prevalence. The results show
similar patterns for the upper and lower estimates3 of smoking initiation
(see Appendix D).
Figure 8-1 shows the CISNET model estimates of the cumulative num-
bers of premature deaths prevented from 2014 to 2099 for each MLA;
these cumulative numbers aggregate over all individuals in the birth cohorts
alive during the time period. The lines represent the mid-estimate, and the
shaded regions correspond to the upper and lower (see Table 8-2). The fig-
ure shows the considerable gains achieved by both MLA 21 and MLA 25
1 Modeling results are presented as cohort effects or period effects. Cohort effects are pat-
terns that differentiate individuals born in the same epoch or generation. Period effects are
patterns that characterize individuals who happened to be alive at a certain point in time,
independent of their age or generation.
2 Premature deaths are the difference between the effective mortality rate versus the mor-
tality rate of never smokers multiplied by the corresponding age-specific population (see
Appendix D).
3 As described in Chapter 7, the simulation models include a range of potential values, re-
sulting in upper and lower estimates around the mid-estimate that vary according to the degree
of the committee’s uncertainty, with a broader range for the MLA of 25. The effect ranges do
not represent bounds or a measure of uncertainty in the classical statistical sense. Rather, these
values reflect ranges that the committee deemed plausible. The mid-estimate is treated as a
geometric mean rather than an arithmetic mean; thus, upper estimates are calculated as 1.2(x)
and lower estimates as x/1.2 for MLA 19 and MLA 21 and as 1.4(x) and x/1.4 for MLA 25,
resulting in slightly nonsymmetric ranges around the mid-estimates.

TABLE 8-1 Lifetime Premature Deaths by Birth Cohort—CISNET Model
Deaths Deaths Deaths
Averted Under Averted Under Averted Under
MLA 19 MLA 19 % MLA 21 MLA 21 % MLA 25 MLA 25 %
Status Quo Mid-Scenario Reduction Mid-Scenario Reduction Mid-Scenario Reduction
2000–2019 2,160,000 59,000 2.7% 223,000 10.3% 296,000 13.7%
2020–2039 1,996,000 58,000 2.9% 218,000 10.9% 289,000 14.5%
2040–2059 2,024,000 58,000 2.9% 222,000 10.9% 293,000 14.5%
2060–2079 2,097,000 60,000 2.9% 229,000 10.9% 304,000 14.5%
2080–2099 2,171,000 63,000 2.9% 238,000 10.9% 315,000 14.5%
NOTE: Although the table carries many significant figures to aid in reproducibility, precision is limited to one or two digits.

221
400,000 500,000
MLA19
MLA21
Cumulative premature deaths prevented
MLA25
300,000
200,000
100,000
0
2020 2040 2060 2080 2100

Year
FIGURE 8-1 Predicted number of premature deaths prevented (lives saved) for the
three MLA policies using the CISNET model. Lines correspond to the mid-scenario
for each MLA. Shaded regions represent the area between the upper and lower
scenarios for each MLA.
in comparison with MLA 19, with the mortality benefits beginning many
years after implementation of the policy, because smoking-attributed mor-
tality becomes more significant after age 40 and the policy primarily affects
adolescent and young adult initiation. The figure shows the preservation of
the general patterns across the mid, upper, and lower initiation scenarios.
Table 8-2 shows the predicted number of premature deaths due to
smoking for selected periods as well as the corresponding number of deaths
prevented and the percentage reduction for each of the MLA mid-estimate
scenarios. According to the CISNET model, raising the MLA to 19, 21, or
25 would save approximately 66,000, 250,000, or 330,000 lives, respec-
tively, by 2100. Of those lives saved, 23,000 (MLA 19), 90,000 (MLA 21),
and 120,000 (MLA 25) would be premature deaths avoided among people

HEALTH BENEFITS OF RAISING THE MINIMUM AGE 223
TABLE 8-2 Cumulative Premature Deaths Expected and Prevented by

Period—CISNET
MLA/Outcome 2020–2039 2040–2059 2060–2079 2080–2099 2015–2099
Status Quo
Premature deaths 6,782,000 4,568,000 2,927,000 1,996,000 18,978,000
expected
MLA 19
Deaths prevented — 3,000 20,000 43,000 66,000
Percentage reduction 0.0% 0.1% 0.7% 2.2% 0.0%
(ages <65)
MLA 21
(ages <65)
MLA 25
(ages <65)
NOTE: This assumes the use of mid-scenarios and that the policy is implemented in 2015.
Although the table carries many significant figures to aid in reproducibility, precision is limited
to one or two digits.
younger than 65 years. The table shows that the percentage of p remature
deaths prevented would increase progressively with time, going from ap-
proximately 0.1 percent, 0.2 percent, and 0.3 percent in 2040–2059 to
2.2 percent, 8.2 percent, and 10.8 percent in 2080–2099 for MLA 19,
MLA 21, and MLA 25, respectively, all based on the mid-estimate scenarios.
Figure 8-2 shows the SimSmoke model estimates of the number of
smoking-related deaths that would be prevented from 2014 to 2100 for
each MLA. The model projects more prevented deaths than the CISNET
model primarily because of the higher future smoking prevalence predicted
by the SimSmoke model and the model differences in assumed mortality
rates for current smokers. The CISNET model also allows for differential
age-specific mortality by smoking intensity, which is particularly relevant
due to the significant decreases in smoking intensity levels projected by the
CISNET model under the status quo (see Appendix D).
The relative proportion of deaths prevented between the three MLAs
appears consistent across the two models, with MLA 21 and MLA 25 lead-

1,500,000 MLA19
MLA21
Cumulative premature deaths prevented
MLA25
1,000,000
500,000
0
2020 2040 2060 2080 2100

Year
FIGURE 8-2 Number of premature deaths prevented (lives saved) for the three
MLA policies estimated using the SimSmoke model. Lines correspond to the mid-
input scenario for each MLA. Shaded regions represent the area between the upper
and lower scenarios for each MLA.
ing to significantly greater proportions of lives saved than with MLA 19.
In contrast with the SimSmoke model, the CISNET model’s projections of
premature deaths prevented for the upper MLA 21 scenario and the lower
MLA 25 scenarios overlap, although they still lead to significantly larger
gains compared to MLA 19, just as in SimSmoke. Table 8-3 shows the
SimSmoke model’s projected number of premature deaths due to smoking
for selected periods as well as the corresponding number of deaths pre-
vented and the percentage reduction for each of the MLA mid-scenarios.
The table shows that the SimSmoke model estimates that the percentage
reduction in smoking-attributed mortality increases progressively with time,
from approximately 0.1 percent, 0.8 percent, and 1.5 percent in 2040–2059
to 2.5 percent, 9.9 percent, and 14.5 percent in 2080–2100 for MLA 19,
MLA 21, and MLA 25, respectively. Thus, although the absolute numbers

TABLE 8-3 Cumulative Premature Deaths Expected and Prevented by

Period—SimSmoke
MLA/Outcome 2020–2039 2040–2059 2060–2079 2080–2100 2015–2100
Status Quo
expected
MLA 19
(ages <65)
MLA 21
Deaths prevented 1,000 51,000 229,000 423,000 705,000
Deaths prevented 700 51,000 108,000 89,000 249,000
(ages <65)
MLA 25
Deaths prevented 4,000 99,000 375,000 620,000 1,098,000
(ages <65)
NOTE: Assumes the use of mid-scenarios and that the policy is implemented in 2015. Al-
though the table carries many significant figures to aid in reproducibility, precision is limited
to one or two digits.
of deaths prevented differ considerably between the models, the percent-

age reductions in smoking-attributable deaths appear relatively consistent,
especially for later years.
Tables 8-4 and 8-5 show estimates of the number of YLL in the United
States for each of the MLA scenarios by calendar-year (period) and birth
cohort, respectively. The calendar-year results (see Table 8-4) suggest the
gains in years of life would begin several decades after implementation
of the policy. Nonetheless, the birth cohort results (see Table 8-5) show
large reductions in the lifetime YLL (>10 percent) achieved by MLA 21 or
MLA 25 for new generations, starting with those born in 2000–2019, with
similar patterns observed for the upper and lower smoking initiation input
values (see Appendix D).
Finding 8-1: Model results suggest that reductions in smoking-related

mortality will not be observed for at least 30 years following the in-
crease in the minimum age of legal access to tobacco products.

226
TABLE 8-4 Years of Life Lost (YLL) by Period—CISNET Model

YLL Under YLL Under YLL Under
2000–2019 134,823,000 — 0.0% — 0.0% — 0.0%
2020–2039 106,126,000 — 0.0% — 0.0% — 0.0%
2040–2059 68,217,000 100,000 0.1% 352,000 0.5% 469,000 0.7%
2060–2079 46,490,000 561,000 1.2% 1,979,000 4.3% 2,641,000 5.7%
2080–2099 36,688,000 964,000 2.6% 3,401,000 9.3% 4,542,000 12.4%
TABLE 8-5 Lifetime Years of Life Lost (YLL) by Cohort—CISNET Model

2000–2019 40,116,000 1,180,000 2.9% 4,163,000 10.4% 5,560,000 13.9%
2020–2039 36,447,000 1,134,000 3.1% 4,000,000 11.0% 5,343,000 14.7%
2040–2059 36,084,000 1,123,000 3.1% 3,962,000 11.0% 5,291,000 14.7%

2060–2079 37,412,000 1,164,000 3.1% 4,108,000 11.0% 5,486,000 14.7%

2080–2099 38,874,000 1,210,000 3.1% 4,268,000 11.0% 5,700,000 14.7%
LUNG CANCER DEATHS

The CISNET Yale Lung Cancer Model, in combination with the CISNET
smoking population model, provides estimates of lung cancer deaths pre-
vented (Holford et al., 2012; Moolgavkar et al., 2012). The model uses a
multistage lung carcinogenesis model to translate the population patterns of
smoking projected by the CISNET smoking population model into predic-
tions of lung cancer deaths (Hazelton et al., 2012; Meza et al., 2008). More
details are provided in Appendix D. Figure 8-3 shows the projected number
of annual lung cancer deaths prevented for each of the MLA mid-scenarios.
Figure 8-4 shows the corresponding cumulative number of lung cancer
deaths prevented. The figures show that the reductions in lung cancer mor-
tality would not become observable until the late 2040s because of the time
delay between smoking exposure and lung cancer risk. As in the case with
4,000
MLA19
MLA21
MLA25
Lung cancer deaths prevented per year
3,000
2,000
1,000
0
2020 2040 2060 2080 2100

Year
FIGURE 8-3 CISNET model estimates of the number of lung cancer deaths pre-
vented per year for the three MLAs (mid-scenario).

MLA19
MLA21
80,000
MLA25
Cumulative lung cancer deaths prevented
60,000
40,000
20,000
0
2020 2040 2060 2080 2100
Year
FIGURE 8-4 CISNET model estimates of the number of cumulative lung cancer
deaths prevented per year for the three MLAs (mid-scenario).
overall mortality, the model predicts that raising the MLA to 21 or 25 would
lead to a considerably higher number of lung cancer deaths prevented than
if the MLA was raised only to 19. Table 8-6 shows the projected number of
lung cancer deaths and deaths prevented for selected periods for each MLA
(mid-scenario). The table shows the progressive increase in the percentage
of lung cancer deaths prevented, going from 0.1 percent, 0.3 percent, and
0.4 percent in 2040–2059 to 2.9 percent, 10.5 percent, and 13.6 percent in
2080–2099 for MLA 19, MLA 21, and MLA 25, respectively.
Finding 8-2: Raising the minimum age of legal access to tobacco prod-
ucts to 21 or 25 years would lead to larger reductions in smoking-
attributable mortality than keeping the status quo or raising the MLA
to 19 years.

TABLE 8-6 Lung Cancer Deaths and Prevented Deaths by Period Under
Each MLA (CISNET)
2020–2039 2040–2059 2060–2079 2080–2099
Status Quo 1,388,000 771,000 510,000 431,000
MLA 19 0 1,000 5,000 12,000
averted percentage reduction 0.0% 0.1% 1.0% 2.8%
MLA 21 0 3,000 19,000 45,000
MLA 25 0 3,000 24,000 59,000
NOTE: Although the table carries many significant figures to aid in reproducibility, precision
is limited to one or two digits.
Finding 8-3: Modeling mortality outcomes by birth cohort estimates

that large reductions in lifetime smoking-attributable deaths and years
of life lost would be achieved by raising the minimum age of legal
access to tobacco products to age 21 or 25 for new generations starting
with the cohort born in 2000. It also projects the prevention of a large
number of lung cancer deaths under such scenarios, with most of these
prevented deaths realized after 2050.
MATERNAL AND CHILD HEALTH OUTCOMES

The fetal and early infancy periods in life are particularly critical
eriods for future development, and, as such, adverse exposures are espe-
p
cially harmful during these periods (HHS, 2004). Cigarette smoke exposure
has potent adverse effects that negatively affect the likelihood of concep-
tion, degrade the health of pregnant women and the developing fetus,
increase the risk of pregnancy complications, and reduce the likelihood
of infant survival (HHS, 2004). An increase in the MLA would therefore
have a robust and immediate impact in improving maternal/fetal and infant
outcomes by reducing the likelihood of maternal/paternal smoking. Benefits
would be expected to occur immediately with a change in the MLA, and
they would at first be concentrated within the younger ages of the repro-
ductive years because of the short-term policy impact that would quickly
appear by reducing smoking prevalence in this age range. The impact of a
raise in the MLA would then increase over time as the early birth cohorts
affected by the MLA increase aged into the reproductive ages. The magni-
tude of the benefit would be directly associated with the magnitude of the
decrease in smoking prevalence.

The SimSmoke model projected the effects of raising the MLA on the
incidence of pre-term births (PTBs), low birth weight (LBW), and SIDS.
The focus on maternal health outcomes required modification of the model
to distinguish the number of smoking women who become pregnant and
the number of children born to smoking women. The model calculated
the number of cases of smoking-attributable birth outcomes using stan-
dard a ttribution formulas based on relative risks and projected smoking
prevalence (HHS, 2010; Levin, 1953; Lilienfeld and Lilienfeld, 1980) (see
Appendix D).
Tables 8-7, 8-8, and 8-9 show the predicted cumulative numbers of
LBW, PTB, and SIDS, respectively, for each MLA for the mid-scenario and
the corresponding number of averted cases versus the status quo for se-
lected years. For mothers ages 15 to 49, the SimSmoke model predicts that
about 124,000 LBW cases, 82,000 PTBs, and 1,100 SIDS deaths would be
averted between 2015 and 2100 for MLA 19. These increase to 438,000
LBW cases, 286,000 PTBs, and 4,000 SIDS deaths averted under MLA 21
and to 597,000 LBW cases, 388,000 PTBs, and 5,400 SIDS deaths averted
under MLA 25. Thus, about three times more cases could be avoided under
MLA 21 than under MLA 19, while only about 1.35 times more cases could
be prevented under MLA 25 than under MLA 21.
TABLE 8-7 Smoking Attributable LBW Cases and Averted Cases by

Period Under Each Policy Option (Mothers Ages 15–49) (SimSmoke)
2015–2019 2020–2039 2040–2059 2060–2079 2080–2099
Status Quo 242,000 727,000 854,000 964,000 1,064,000
MLA 19 2,000 22,000 30,000 34,000 37,000
averted percentage 0.8% 3.0% 3.5% 3.5% 3.5%
reduction
MLA 21 10,000 78,000 104,000 117,000 129,000
reduction
MLA 25 16,000 109,000 140,000 158,000 174,000
reduction

TABLE 8-8 Smoking Attributable PTB Cases and Averted Cases by

2015–2019 2020–2049 2040–2059 2060–2079 2080–2099
Status Quo 148,000 442,000 520,000 587,000 648,000
MLA 19 1,000 14,000 20,000 22,000 24,000
reduction
MLA 21 6,000 51,000 68,000 76,000 84,000
reduction
MLA 25 11,000 71,000 91,000 103,000 113,000
reduction
TABLE 8-9 Smoking Attributable SIDS Cases and Averted Cases by

2015–2019 2020–2049 2040–2059 2060–2079 2080–2099
Status Quo 2,280 6,850 8,060 9,090 10,020
MLA 19 20 200 270 300 340
reduction
MLA 21 100 730 950 1,070 1,180
reduction
MLA 25 160 1,010 1,270 1,430 1,580
reduction
Finding 8-4: Modeling estimates that immediate reductions in cases of

low birth weight, pre-term birth, and sudden infant death syndrome
will occur with changes in the minimum age of legal access to tobacco
products.

TABLE 8-10 Reduction (percentage) in Smoking Prevalence for MLA 21

by Year
2020 2040 2060 2080 2100
Smoking prevalence—SimSmoke 2.0% 8.3% 10.3% 11.2% 11.20%
Smoking prevalence—CISNET 0.4% 6.4% 10.6% 11.9% 12.00%
TABLE 8-11 Reduction (percentage) in Health Outcomes for MLA 21 by

Period
2020–2039 2040–2059 2060–2079 2080–2099
Deaths prevented—SimSmoke 0.0% 0.8% 4.6% 9.9%
Deaths prevented—CISNET 0.0% 0.2% 2.6% 8.2%
Years of life lost—CISNET 0.0% 0.5% 4.3% 9.3%
Lung cancer deaths prevented 0.0% 0.3% 3.7% 10.5%
Low birth weight cases 10.8% 12.2% 12.2% 12.2%
Pre-term birth cases 11.6% 13.0% 13.0% 13.0%
Sudden infant death syndrome 16.0% 18.5% 18.5% 18.5%
cases
TIME TO ACCRUE BENEFITS

Tables 8-10 and 8-11 summarize the reductions in smoking prevalence
for selected years and health outcomes by 20-year periods for MLA 21,
showing the relative timing at which different benefits occur. The results
illustrate the longer times required for chronic outcomes compared to short-
term outcomes.
OTHER HEALTH EFFECTS

The previous section laid out the results of the simulation modeling
regarding the likely effects of raising the MLA on cigarette-related mortal-
ity and select health outcomes, limited to the capacity of the commissioned
models. However, such results can only begin to estimate the magnitude of
the effects of reduced tobacco use on individual and population health in
the United States. As the cohorts of adolescents and young adults affected
by a raise in the MLA age, the benefits accrue and grow over time. The
adverse health effects of tobacco use are well documented and described in
Chapter 4. Here, the committee describes qualitatively the wide spectrum
of likely benefits to health throughout the life span from decreased tobacco

initiation in adolescents and young adults and the resulting lowered preva-
lence rates in adulthood. It should be stressed that most of the data about
adverse health effects of tobacco come from studies of cigarette smoking.
Immediate Health Effects

Cigarette smoking causes the immediate adverse health effects of in-
creased oxidative stress; depletion of selected bioavailable antioxidant
micronutrients; increased inflammation; impaired immune status; altered
lipid profiles; poorer self-rated health status; respiratory symptoms such as
coughing, phlegm, wheezing, and dyspnea; and nicotine addiction (HHS,
2004). As summarized above, increasing the MLA would be expected to
reduce the initiation of tobacco use by adolescents and young adults, which
would naturally lead to a decrease in the prevalence of tobacco use. Re-
ducing the prevalence of smoking by any amount will automatically lead
to immediate population health benefits that are directly proportional to
the size of the reduction. Each one of the immediate adverse health effects
caused by cigarette smoking itself compromises the health status of smok-
ers, and when combined, this constellation of immediate adverse health
effects leaves the smoker with a health status that is significantly impaired
and subpar compared to nonsmokers. For example, smokers are less able
to fend off acute infectious diseases and more likely to exhibit respiratory
symptoms (HHS, 2014). The cumulative toll leaves the smoker generally
feeling worse off about his or her health status soon after starting to smoke
(HHS, 2004, 2014). Nicotine addiction makes the smoker more likely to
keep smoking over the long term, which in turn makes the smoker more
and more prone to the immediate and long-term health effects as the life-
time extent of smoking grows.
The immediate adverse health effects of smoking affect people of all
ages, but the immediate impact upon adolescents who initiate smoking
is the most disconcerting from a population health perspective because
these adverse consequences occur during such a critical developmental
period of life. The immediate health effects result in adolescents and young
adults who smoke having compromised educational achievement, dimin-
ished a thletic performance, reduced proficiency in performing occupational
duties, and, for those enlisted in the armed forces, having compromised
military performance (HHS, 2004). In fact, each of these populations of
students, w orkers, and military personnel can be viewed as having a sub-
population of smokers that is physiologically disadvantaged compared to
the nonsmoker portion of the population. Thus, a reduction in smoking
prevalence by any amount is a step toward reducing a population health
disparity that is created by cigarette smoking even in the ostensibly healthy
population of adolescents and young adults. The larger the reduction in

smoking prevalence created by raising the MLA, the larger the commensu-
rate reduction will be in these smoking-caused health disparities. Reducing
the prevalence of these immediate adverse health effects would not only
benefit population health but also have downstream benefits on popula-
tion educational achievement, workforce productivity, and military per-
formance. The higher the MLA, the greater the public health benefit will
be in terms of reducing the size of the population of smokers and hence
decreasing the number who experience the corresponding health deficits.
Further public health benefits will occur from the delays in the age of
starting to smoke that would result from raising the MLA for tobacco.
Within the age range where the delays occur, the delayed age of initiation
would postpone the immediate adverse health effects until the individuals
are older. The child and adolescent population would directly benefit, with
a smaller percentage of the adolescent population smoking and a larger
percentage maintaining a more optimal health status. Delaying smoking in
adolescents until they are older would help protect the tissues and organ
systems that are still in the growth and maturation phase during adoles-
cence and hence are particularly vulnerable to the detrimental effects of
the toxicants in smoke (HHS, 2004). As with the prevention of smoking,
the extent to which smoking initiation will be delayed will be directly re-
lated to how high the MLA is set.
Intermediate Health Effects

Cigarette smoking causes the intermediate adverse health effects of
increased absence from school and work, increased use of medical ser-
vices, subclinical atherosclerosis, impaired lung development and function,
increased risk of lung infections, diabetes, periodontitis, exacerbation of
asthma, subclinical organ injury, and adverse surgical outcomes (HHS,
2004). The reductions in smoking prevalence caused by increasing the
MLA will reduce the entire burden that these intermediate adverse health
effects pose to population health. The estimated amount of reduction in this
burden will be larger with a higher MLA and will grow in magnitude over
time as the policy impact matures.
Reducing the prevalence of smoking will lead to population health
benefits in the near term by reducing the burden of the intermediate adverse
health effects of cigarette smoking. Each of the intermediate adverse health
effects caused by cigarette smoking compromises an individual smoker’s
health status; in total, they combine to exact a severe toll on individuals
and on population health in general. They further widen the health status
differential between smokers and nonsmokers, which commences with the
immediate adverse health effects. The intermediate health effects leave the
smoker not only with subclinical diminished health status but also with

clinically apparent morbidities across multiple organ systems (HHS, 2004).

In turn, the diminished health status and clinical morbidities have a detri-
mental influence on the national economy both by limiting workforce pro-
ductivity via absences and also by increasing health care costs (HHS, 2004).
The morbidities experienced by smokers during this intermediate period are
outward manifestations of the subclinical effects that begin immediately
after smoking initiation, a fact that reinforces the observation that the
health status of smokers is diminished throughout the life span compared
to nonsmokers, even before the impact of clinically apparent morbidities
and then mortality make this difference in health status obvious.
These intermediate adverse health effects affect the entire age con-
tinuum, generating clear smoker–nonsmoker health disparities during early
life (HHS, 2004, 2014). As with the immediate health effects, as smok-
ing persists into adulthood the divergence in markers of health status
between smokers and nonsmokers widens. Cigarette smokers constitute a
sub-population that is physiologically disadvantaged compared with the
nonsmoker population, and a reduction in smoking prevalence resulting
from an increase in the MLA is a step toward achieving a reduction in
smoking-caused population health disparities.
The public health benefits from delayed initiation would not simply
be seen in a decrease in the immediate adverse health effects, but would
continue to have a ripple effect over time, benefiting people at all ages and
stages of life. For example, the fact that delayed initiation reduces the dose
of cigarette toxins ingested by smokers would help to offset the population
burden of intermediate health effects, and because an older age of initiation
is associated with increased likelihood of cessation, this would further ben-
efit population health by leading to further reductions in cigarette smoking
prevalence during those stages of life affected by the intermediate adverse
health effects of cigarette smoking.
Long-Term Health Effects

Cigarette smoking is causally associated with a long list of long-term
health effects that includes 12 different types of cancer, vascular and heart
disease outcomes, respiratory disease, eye disease, rheumatoid arthritis,
and bone health (HHS, 2004, 2014). The immediate, intermediate, and
long-term adverse health effects of cigarette smoking are related as several
of these long-term outcomes are mechanistically linked to the immediate
and intermediate adverse health effects summarized above. Cancer, cardio-
vascular disease, and chronic obstructive pulmonary disease are caused by
smoking and are also the major causes of death in the United States (HHS,
2014); thus, these specific outcomes are also included indirectly in the sta-
tistical modeling of all-cause mortality. On the other hand, although such

long-term adverse health effects as eye disease, rheumatoid arthritis, and

adverse effects on bone health are not direct causes of death, they do pose
a major burden of disability and impaired quality of life in the U.S. popula-
tion (HHS, 2004, 2014). Furthermore, regardless of the ultimate prognosis,
cancer, cardiovascular disease, and chronic obstructive pulmonary disease
all contribute to the burden of disability and impaired quality of life.
Unlike the case with the immediate and intermediate adverse health
effects caused by cigarette smoking, the impact of increasing the MLA on
long-term adverse health effects caused by cigarette smoking would not
become apparent until decades after the policy change occurred because
raising the MLA will primarily affect the initiation and delay of smoking
among children, adolescents, and young adults. Therefore, the impact of
raising the MLA on the long-term adverse health effects would not occur
until the initial birth cohorts affected by an MLA increase were old enough
to be in the older age groups where these chronic diseases typically occur.
The degree of morbidity reduction in these cohorts would be expected to be
directly correlated with the decrease in smoking prevalence and delayed ini-
tiation that the older MLA generated. The population health impact would
be profound even for modest decreases in smoking prevalence because of
the broad spectrum of the long-term health effects caused by smoking and
the population health burden caused by each of these diseases. Delays in
initiating cigarette smoking would result in further reductions in the long-
term adverse health effects caused by smoking because of reductions in
the population-level exposure to tobacco toxins; these reductions would
occur because a later age of initiation leads to individuals smoking fewer
cigarettes per day, on average, and also for fewer years because individuals
who start smoking later are more likely to eventually quit.
The focus here has been specifically on the public health effects of
reducing the prevalence of, and delaying initiation of, cigarette smoking.
Raising the MLA will also reduce the prevalence of smoking combustible
tobacco products other than cigarettes, such as pipes and cigars. As re-
viewed in Chapter 4, although less thoroughly studied than cigarette smok-
ing, smoking other tobacco products also causes significant adverse health
effects that will be prevented with a reduction in prevalence. Furthermore,
raising the MLA will lead to reductions in smokeless tobacco use and hence
a reduction in the adverse health effects caused by smokeless tobacco use.
By reducing the prevalence of smoking of all tobacco products in the
population, raising the MLA will also lead to a reduction in the popula-
tion exposure to secondhand smoke (SHS). A reduction in the prevalence
of exposure to SHS will benefit public health by reducing the spectrum of
adverse health effects, reviewed in Chapter 4, that have been causally as-
sociated with SHS exposure.

IMPLICATIONS OF RAISING THE MINIMUM AGE OF

LEGAL ACCESS TO TOBACCO PRODUCTS ON HEALTH
The modeling analysis suggests that raising the MLA could lead to
considerable reductions in smoking-attributed mortality and morbidity over
time, mirroring the reductions in smoking discussed earlier. Both models
suggest a time delay of a few decades for the overall mortality benefits to
accrue at the population level because of the lag time between smoking
exposure and major health outcomes and because the policy primarily
affects adolescents and young adults. Nonetheless, more immediate effects
would be observed for maternal and child outcomes as well as other acute
outcomes. Moreover, the analysis shows that new generations, starting
with those born between 2000 and 2019, could see significant reductions
in mortality and years of life lost accumulated throughout their lifetimes.
Both models suggest that significant mortality gains occur when going
from MLA 19 to MLA 21. Increasing the MLA from 21 to 25 leads to
additional benefits, but the magnitude of these benefits is less than achieved
when going from MLA 19 to MLA 21, based on conservative assumptions
that reflect uncertainty about extrapolation to MLA 25.
The CISNET model predicts about one-third as many premature deaths
prevented as the SimSmoke model. This occurs largely because of the lower
smoking prevalence projected by the CISNET model for all cases and the
concomitant lower baseline smoking-attributable deaths. Furthermore, the
CISNET model allows for differential mortality by smoking intensity. Thus,
the large reductions in smoking intensity levels projected by this model
translate into fewer estimated smoking-attributable mortality in all cases
than in the SimSmoke model. Nonetheless, the estimated percentage mor-
tality reductions of the different MLAs appear consistent between the two
models, particularly for later years.
Raising the MLA would significantly reduce lung cancer mortality in
the long term, with most of the benefits realized after 2050. Similarly, as
with the overall mortality projections, the models predict considerably
larger reductions when raising the MLA to 21 or 25 versus 19. Raising the
MLA to 19, 21, and 25 will reduce LBW, PTB, and SIDS outcomes, with
these benefits occurring relatively earlier in time.
All models come with limitations because their results depend on the
model structure and assumptions. In this case, uncertainty also arises from
the assumptions about the effects of various MLA policies on smoking ini-
tiation scenarios. The committee used an evidence-driven process to create
the inputs regarding potential ranges for the assumed effects of the MLA
policies. While these inputs are assumptions, they are well reasoned based
on the existing evidence regarding adolescent and young adult smoking
behavior and tobacco control policy responses, as explained in Chapter 7.

The use of two established tobacco control simulation models with

differences in the underlying assumptions related to future baseline initia-
tion and cessation rates led to different estimates of the absolute decrease
in smoking prevalence and different status quo estimates. However, the
two distinct models predict similar results for the percentage reductions
associated with the various MLA options considered. Similarly, although
the models differ in their predicted absolute numbers of deaths prevented,
they agree in their estimated relative reductions and relative effects among
the different MLAs. This provides some confidence about these overall
findings. Sensitivity analyses (see Appendix D) showed that the conclusions
about the relative effects of the different MLAs appear robust to alternative
assumptions on the initiation effects (upper and lower scenarios).
The projections provide somewhat conservative estimates, given that
the models did not account for the possible synergistic effects of reduced
and delayed initiation with increased cessation, and the committee estimates
accounted for greater uncertainty about projection to an MLA 25 policy.
Moreover, the models only considered smoking, ignoring the potential addi-
tional health benefits from reductions in the consumption of other tobacco
products. The models also ignored the potential additional health benefits
from reductions in the consumption of other tobacco products and the
likely synergistic effects of increased cessation on disease risk. The models
also ignore benefits that might accrue because nonsmokers engage in a va-
riety of healthy behaviors compared to smokers. Overall, the results from
both models are consistent with the conclusions from the literature review
and show that raising the MLA would lead to significant health benefits.
Some, such as maternal and child health outcomes, will occur immediately,
while others, such as overall mortality benefits, will take time to accrue.

legal access to tobacco products will likely lead to substantial reduc-
tions in smoking-related mortality.
As described above and in Chapter 4, cigarette smoking causes numer-

ous adverse health effects, and these can be categorized as immediate, in-
termediate, or long term. In assessing the potential public health impact of
raising the MLA, it is worth keeping in mind that this lengthy catalogue of
well-established consequences of cigarette smoking and SHS exposure will
grow as more definitive evidence coalesces for additional health outcomes.
There are many additional adverse health effects currently suspected of
being causally associated with both cigarette smoking and SHS exposure,

but the evidence currently falls short of being definitive; thus, the scope of
adverse health effects will grow over time.
Considering the causes of the health effects of cigarette smoking
throughout the entire life course more accurately characterizes the full ex-
tent of the public health burden imposed by cigarette smoking. It is impor-
tant to emphasize that because the spectrum of adverse health effects caused
by cigarette smoking is so extensive in both the near term and the long
term, even small reductions in smoking prevalence will benefit public health
substantially. The magnitude of the public health impact will be larger for
greater reductions in smoking prevalence; thus, the public health impact
will be greatest for an MLA of 25 years and least for an MLA of 19 years.
Conclusion 8-2: Based on a review of the literature, raising the mini-

mum age of legal access to tobacco products (MLA) will likely imme-
diately improve the health of adolescents and young adults by reducing
the number of those with smoking-caused diminished health status. As
the initial birth cohorts affected by the policy change age into adult-
hood, the benefits of the reductions of the intermediate and long-term
adverse health effects will also begin to manifest. Raising the MLA will
also likely reduce the prevalence of other tobacco products and expo-
sure to secondhand smoke, further reducing tobacco-caused adverse
health effects, both immediately and over time.
Conclusion 8-3: Based on a review of the literature and on the model-

ing, an increase in the minimum age of legal access to tobacco products
will likely improve maternal, fetal, and infant outcomes by reducing the
likelihood of maternal and paternal smoking.
As discussed in Chapter 7 with regard to effects of an increase in the

MLA on tobacco initiation, it is an open question whether raising the MLA
will have a greater or lesser impact on the health of population subgroups
with a higher prevalence of cigarette smoking than on the general popula-
tion. If the reduction in smoking prevalence was proportionally larger in the
subgroups of the population with the highest smoking prevalence, then the
public health impact of raising the MLA might be even greater than antici-
pated. If the converse were true, however, and these population subgroups
were more resistant to the influence of the policy with respect to reducing
smoking prevalence and delayed initiation, then the end result would be to
widen the existing disparities.

REFERENCES
Hazelton, W. D., J. Jeon, R. Meza, and S. H. Moolgavkar. 2012. Chapter 8: The FHCRC lung
cancer model. Risk Analysis 32(Suppl 1):S99–S116.
HHS (Department of Health and Human Services). 2004. The health consequences of smoking:
A report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human
Services, Centers for Disease Control and Prevention, National Center for Chronic Disease
Prevention and Health Promotion, Office on Smoking and Health.
———. 2010. Smoking-attributable mortality, morbidity, and economic costs (SAMMEC).
https://apps.nccd.cdc.gov/sammec/mch_mort_comp.asp?cdc=706 (accessed January 29,
2015).
Holford, T. R., K. Ebisu, L. McKay, C. Oh, and T. Zheng. 2012. Chapter 12: Yale Lung Cancer
Model. Risk Analysis 32(Suppl 1):S151–S165.
Levin, M. 1953. The occurrence of lung cancer in man. Acta Unio Internationalis Contra
Cancrum 9:531–541.
Lilienfeld, A., and D. Lilienfeld. 1980. Foundations of epidemiology. New York: Oxford
University Press.
Meza, R., W. D. Hazelton, G. A. Colditz, and S. H. Moolgavkar. 2008. Analysis of lung cancer
incidence in the Nurses’ Health and the Health Professionals’ Follow-Up Studies using a
multistage carcinogenesis model. Cancer Causes and Control 19(3):317–328.
Moolgavkar, S. H., T. R. Holford, D. T. Levy, C. Y. Kong, M. Foy, L. Clarke, J. Jeon, W. D.
Hazelton, R. Meza, F. Schultz, W. McCarthy, R. Boer, O. Gorlova, G. S. Gazelle, M.
Kimmel, P. M. McMahon, H. J. de Koning, and E. J. Feuer. 2012. Impact of reduced
tobacco smoking on lung cancer mortality in the United States during 1975–2000. Jour-
nal of the National Cancer Institute 104(7):541–548.

Other Considerations for Policy Makers
T
he objective of this report is to predict what the health consequences
would be of raising the minimum age of legal access to tobacco
products (MLA) to 19, 21, or 25. As discussed in Chapter 5, few
jurisdictions, states, or localities in this country have undertaken such
changes, and no other country has done so. None of the state and local
initiatives has been followed by a rigorous evaluation published in the peer-
reviewed literature. Because a review and synthesis of existing empirical
literature cannot answer the question at hand, the committee drew on a
comprehensive review of the relevant scientific literature, on its collective
expertise, and on models of population-level smoking behavior to predict
changes in adolescent and young adult initiation attributable to raising the
MLA and to project the impact of these changes on the prevalence of use
and on health outcomes.
Using conservative assumptions about the enforcement of the MLA in-
creases, the committee concluded that raising the MLA will likely decrease
initiation of tobacco use by adolescents and young adults and thereby,
over time, reduce adult prevalence, leading to longer and healthier lives
for those who would have otherwise used tobacco. More specifically, the
modeling analysis concluded that raising the MLA, particularly to ages 21
and 25, would lead to substantial reductions in smoking prevalence and
thereby prevent considerable numbers of smoking-attributable deaths, in-
cluding lung cancer deaths, and poor maternal and child health outcomes.
However, the committee has greater uncertainty about the magnitude of
the effects of raising the MLA to age 25 rather than to 19 or 21. The re-
sults suggest a range of potential population health benefits that depend on
241

a number of informed assumptions regarding enforcement practices and

behavioral responses to the policy change by retailers and other potential
sources of tobacco products and by underage individuals in different age
and gender groups.
The purpose of this chapter is to help policy makers translate the com-
mittee’s findings and conclusions into the policy context. First, the chapter
highlights a key constraint arising from the committee’s charge: Its quanti-
tative estimates and projections relate to the nation as a whole, while the
traditional responsibility for enacting and enforcing the MLA lies with
states and localities. Second, the chapter revisits several policy assumptions
that were explicitly made by the committee (or that are built into the simu-
lation models) as a basis for its estimates of the effects of raising the MLA
on adolescent and young adult initiation. These assumptions relate to
the scope and enforcement of the MLA policy and to the status of other
tobacco control policies. Reviewing them is important because it will enable
the policy maker to consider the possible effects of different assumptions.
Third, the chapter discusses the possible policy implications of increas-
ing scientific knowledge regarding adolescent development. Finally, the
chapter identifies two factors of possible public health relevance that were
not taken into account in making the estimates and projections described
in the report. The more important of these factors is the possible impact of
the marketing and use of new tobacco products, most notably electronic
nicotine delivery systems (ENDS). The other is the possible impact of rais-
ing the MLA for tobacco use on the use of alcohol or other drugs.
NATIONAL OR STATE ENACTMENT OF MLA

Traditionally, political responsibility for setting the MLA for tobacco
products has rested with the states and, depending on state constitutional
arrangements, with local governments. However, since 1992 the federal
government has played an increasingly significant role. The Family Smoking
Prevention and Tobacco Control Act (hereafter referred to as the Tobacco
Control Act), enacted by Congress in 2009, directed the Food and Drug
Administration (FDA) to revive its 1996 Tobacco Rule, which prescribed a
federal MLA of 18. At the same time, however, Congress precluded FDA
from raising the MLA without congressional action. In effect, the Tobacco
Control Act sets a “floor” of 18 while allowing states and localities to raise
the age if they choose to do so. Hence, unless Congress acts to raise the
age on a national basis or delegates authority to FDA to do so, one might
expect a patchwork of different MLAs in different states and localities, as
existed for alcohol for many decades, rather than a uniform MLA across
all of the 51 jurisdictions.

OTHER CONSIDERATIONS FOR POLICY MAKERS 243
It is important to emphasize that the simulations described in Chap-

ters 7 and 8 model a situation in which increases in the MLA would be
adopted and implemented on a nationwide basis. However, a state-by-state
implementation is more likely. Nationwide implementation would occur
only if Congress raises—or authorizes FDA to raise—the national MLA or
if every state raises the MLA. To the extent that states choose not to raise
the MLA, the effects estimated in Chapters 7 and 8 are not likely to be real-
ized. In addition, to the extent that people who are underage in a high-MLA
state could cross state borders to purchase in a low-MLA state, the effects
estimated in Chapter 8 may be somewhat optimistic, particularly for small
states surrounded by many low-MLA neighbors.
Even if Congress does not choose to set a national MLA higher than
18, there are other mechanisms through which universal or near univer-
sal adoption might be motivated. For example, Congress could provide
incentives for states to do so by making the level of funding under federal
grants contingent on the state raising the MLA. It could do this based on
the approach used in the Synar Amendment (up to 40 percent of a state’s
substance abuse prevention block grant funding is contingent on enforcing
the state’s MLA) by simply defining underage purchasers under the Synar
program as persons under 19, 21, or 25 as the case may be. Alternatively,
Congress can use an approach similar to that taken in the National Mini-
mum Drinking Age Act of 1984,1 which penalized states that did not ban
the purchase and public possession of alcoholic beverages under age 21
by reducing their annual federal highway appropriations by 10 percent.
By 1995 all 50 states and the District of Columbia were in compliance,
thanks to this strong incentive. Although the highway appropriation may
not be seen as the most appropriate type of leverage for tobacco policy,
federal funds related to public health may be viewed as more suitable for
this purpose.
In sum, Congress could decide to raise the MLA at the national level,
to provide federal funding incentives for the states to do so, or to leave
the matter entirely to the states or local jurisdictions. In the absence of a
national MLA, however, the national public health impact of raising the
MLA for tobacco would be dependent, first and foremost, on the degree to
which local and state governments take up this policy.
EFFECTS OF OTHER TOBACCO CONTROL POLICIES

Both simulation models predict the potential effects on future initiation
of increasing the MLA. The SimSmoke model also includes modules for
1 The National Minimum Drinking Age Act of 1984, Public Law 98-363. 98th Cong.
(July 17, 1984). 23 U.S.C. § 158.

modeling the effects of other tobacco control policies: taxation, smoke-free

air, marketing restrictions, health warnings, media campaigns, and cessa-
tion treatment policies. Tobacco control policies are effective; they reduce
tobacco use and hence decrease the adverse health outcomes associated
with use. The effects of these policies are modeled as changes in the initia-
tion and cessation of smoking. The effects of past policies are incorporated
into the initiation and cessation rates in future years. While the models in
general have the capability to project future changes in policies, the models
as used here assume that all current policies other than the MLA will re-
main in effect at their current rates and that no new policies will be imple-
mented. This assumption is useful because it isolates the effects of raising
the MLA from other potential policy changes in the modeling of the effects
of nationwide implementation. However, a significant change, one way or
the other, in the intensity and effectiveness of tobacco control policies in
the country as a whole could alter the figures projected by the models for
prevalence and health outcomes presented in Chapters 7 and 8.
In this connection, it is important to emphasize that there are significant
variations in the strength and efficacy of existing state and local tobacco
control programs. These variations reflect differences in the number and
intensity of tobacco control activities and in the resources allocated to sup-
port them. A comprehensive approach to tobacco control integrates “edu-
cational, clinical, regulatory, economic, and social strategies” (CDC, 2014,
p. 6). Specifically, such an approach includes activities targeted at prevent-
ing initiation of tobacco use, reducing tobacco use and tobacco-related dis-
eases, promoting cessation, and reducing exposure to secondhand smoke,
combined with mass media campaigns and community mobilization efforts
(CDC, 2014; HHS, 2000). Comprehensive, multifaceted strategies have
been shown to effectively reduce tobacco use among adolescents (Farrelly
et al., 2013; Kuiper et al., 2005; Laugesen and Swinburn, 2000; Luke et al.,
2000; Tauras et al., 2005; Wakefield and Chaloupka, 2000), young adults
(Farrelly et al., 2014; Kuiper et al., 2005; Laugesen and S winburn, 2000;
Pierce et al., 2009), and adults (Farrelly et al., 2008; Kuiper et al., 2005;
Laugesen and Swinburn, 2000; Stillman et al., 2003; Zaza et al., 2005),
as well as to reduce tobacco-related death and disease (Jemal et al., 2003;
Kuiper et al., 2005; Laugesen and Swinburn, 2000). Moreover, in a re-
view of comprehensive state-level tobacco control programs, Wakefield
and C haloupka (2000) found that states were able to substantially reduce
teenage smoking despite differences in the specific program components
that the states used. On the other hand, comprehensive statewide tobacco
control programs that lacked optimal funding failed to achieve the full
magnitude of their potential effect, despite achieving substantial reductions
in tobacco use (Farrelly et al., 2008; Tauras et al., 2005). States and locali-
ties that have more comprehensive and intensive tobacco control activities

and that devote more resources to support these activities are likely to have
a lower prevalence of tobacco use than states and localities with weaker
tobacco control programs.
As noted above, the national projections in Chapters 7 and 8 are
grounded in models that essentially aggregate each state’s tobacco con-
trol activities, whether they are strong or weak. To the extent that policy
makers in individual states want to try to derive state-based estimates from
the findings of national modeling exercise, they will have to take into ac-
count whether the existing level of tobacco control activity in their state is
comparable to the investment (and intensity of activity) in the “average”
state. If it is much weaker, the extrapolation from the modeling used in this
report may not be suitable. Similarly, if a state is among the nation’s leaders
in the tobacco control, the reduction in prevalence and in morbidity and
mortality may be greater.
SCOPE AND ENFORCEMENT OF MLA RESTRICTIONS

Before undertaking the task of estimating the effects of raising the MLA
on adolescent and young adult initiation, the committee agreed on certain
key assumptions about the scope and enforcement of the MLA (51 juris-
dictions aggregated nationally). First, the committee assumed that current
levels of enforcement and retailer compliance with the MLA restrictions
will be sustained for all underage purchasers, including those 18 or older
but under the new MLA, if the MLA is raised. Second, the committee as-
sumed that existing bans on noncommercial distribution of tobacco by
friends, proxy purchasers, and other “social sources” will continue to be
weakly enforced whether or not the MLA is raised and that these sources
will continue to provide substantial, though incomplete, substitution for
retail purchases for newly underage buyers. Third, the committee assumed
that the proportion of underage users who purchase tobacco on the illicit
commercial market will remain small. Finally, the committee assumed that
sanctions will continue to be directed primarily toward retailers and will
not be enforced against underage users on a significant scale. The committee
revisits these assumptions here.
Enforcement Against Retailers

Federal support for youth access enforcement, together with funding
incentives, has significantly strengthened state enforcement of youth access
policies and has thereby curtailed retail availability to underage persons.
The committee has assumed that the current levels of enforcement and pen-
alties for violators will continue, creating a credible threat of punishment
sufficient to sustain current levels of compliance. In addition, the committee

assumes that the deterrent threat will be the same for selling to every under-
age purchaser, regardless of where the MLA line is drawn.
It is possible, of course, that the intensity of enforcement could be
significantly increased against all underage users, in which case the com-
mittee’s estimates in Chapter 7 about the impact of raising the MLA on
the adolescent and young adult initiation rates might be too conservative.
On the other hand, it is also possible that increasing the MLA into the
years of “adulthood” could generate a backlash and weaken public sup-
port for enforcing the law. As emphasized in Chapter 6, curtailing retail
access depends on active enforcement and retailer compliance. Those con-
ditions could be undermined if the MLA is set too high. Concerns about
under-enforcement would be particularly pronounced if the MLA were set
at age 25, and for this reason the committee is relatively more confident
about the assumption that current enforcement intensity is more likely to be
maintained if the age is increased to 19 or 21 than if it is set at 25. (This is
one of the reasons why the range between the lower and upper scenarios
is broader in the analysis of the MLA 25 policy option.)
If current levels of enforcement intensity are to be sustained and ex-
tended to the older ages, another key question is whether doing so will
require a significant increase in current funding for enforcement. Recall that
the 1992 Synar Amendment to the Alcohol, Drug Abuse and Mental Health
Administration Reorganization Act2 was designed to incentivize states to
enact, enforce, and continuously evaluate laws that prohibit the sale and
distribution of tobacco products to individuals under age 18. As discussed
in Chapters 1 and 5, states are required to follow specific guidelines for
random compliance inspections, surveillance, and reporting as a condition
of their receipt of federal Substance Abuse Prevention and Treatment block
grant funding. Failure to comply with Synar regulations could result in the
withholding of up to 40 percent of block grant funds.
The language of the Synar Amendment focuses specifically on restrict-
ing access to tobacco products among persons under age 18. Because
the amendment incentivizes states to enforce and track compliance with
tobacco purchase laws only for adolescents under age 18, it is not clear
whether additional resources would be required to extend significant en-
forcement activities to individuals above age 18. Ongoing surveillance and
the associated random inspections/compliance checks are essential, not only
for policy evaluation but also as a strong incentive for retailers and distribu-
tors to comply with the law. Extending the training and surveillance systems
in place for the Synar Amendment to ensure compliance with an MLA of
19 or above might require additional financial and human resource invest-
2ADAMHA Reorganization Act of 1992, Public Law 102-321. 102nd Cong. (July 10,
1992).

ments. Compliance checks must be done with age-appropriate confederates

(e.g., 20-year-olds cannot be used for compliance checks for Synar report-
ing, and under-18 purchasers are not appropriate for surveillance regarding
enforcement among 18- to 20-year-olds).
For an MLA of 21, local enforcement activities might dovetail with
those for alcohol, and assigning responsibility to the same agency, as some
states have already done, might actually reduce the costs of enforcement,
particularly given the overlap of licensees.
The committee understands that the relevant agencies in New York
State and New York City have reached an agreement that facilitates the
enforcement of the city’s new Tobacco 21 law without increasing the cost of
enforcement. The New York City Department of Consumer Affairs (DCA)
enforces the city’s Tobacco 21 law with funding from the state Department
of Health. Their agreement requires DCA to perform compliance checks
with adolescents ages 16 and 17 in compliance with state and federal laws
prohibiting tobacco sales to adolescents under age 18. The state will then
punish violators detected during these inspections. In addition, DCA will
employ a small team of young adults ages 18 to 20 to assess compliance
with the city’s MLA of 21, and will also punish violators.3 The agreement
also requires DCA to verify that tobacco retailers post required minimum
age signage, perform age verification, and comply with other point-of-sale
restrictions. Because New York City’s Tobacco 21 law is more stringent
than both state and federal laws, New York State has agreed that DCA will
inspect for city Tobacco 21 signs (as opposed to state signs for MLA 18)
and to check that retailers ask for proof of age using photo identification
for customers who look under 30 years old (as opposed to state law requir-
ing age verification for customers who look under age 26) (NYCDOHMH,
2014).
In addition to the intensity of enforcement and retailers’ perceived risk
of getting caught, the severity of the penalty for violation would also play
a role in policy effectiveness. For example, in Hawaii County, failure to
post signage regarding the MLA 21 policy results in a $500 fine, and any
person who sells or distributes tobacco products to a person under age 21 is
subject to up to a $2,000 fine. Similarly, the penalties associated with New
York City’s recent Tobacco 21 law include a $500 fine for failure to post
required signage, a $1,000 fine for the first sales violation to someone ages
18 to 20 or any other violation in the same day, and a $2,000 fine for the
second and any subsequent violation within 3 years. In addition, a second
violation may result in the revocation of the retail tobacco license. Although
the committee is not aware of any systematic data regarding the severity of
3Personal communication, K. Munn, New York State Department of Health, October 14,
2014.

penalties imposed on violators, it seems likely that the imposition of penal-

ties at this level, particularly the loss of the tobacco license or, as in some
states, a lottery license, can achieve meaningful deterrence as long as there
is a credible threat of detection for a violation.
Enforcement Against Social Sources

As discussed in Chapter 6, cigarettes obtained from friends, family
members or fellow smokers, or from proxy buyers are very good substi-
tutes for the same product bought directly from a retail outlet. So if direct
sales become unavailable through effective enforcement efforts, underage
users will likely continue to substitute cigarettes obtained from these other
sources. Although existing bans on noncommercial distribution of tobacco
by friends, proxy purchasers, and other “social sources” are weakly en-
forced, the committee has concluded that access to social sources does not
fully substitute for convenient access to retail purchases because, as eco-
nomic theory suggests, the use of social sources is more costly. It requires
additional time and effort, in addition to money, for someone to obtain
cigarettes indirectly instead of purchasing them directly from a store. As
such, forcing underage smokers to find and use indirect sources raises their
costs of obtaining tobacco products, which in turn is likely to reduce their
consumption. It is these additional costs that account for the reduction
in underage use attributable to youth access restrictions, especially when
smoking is reduced among the members of social networks to which the
underage smoker has ready access. The committee has estimated that rais-
ing the MLA to ages 19, 21, or 25 will reduce tobacco use by secondary
school students who lack ready access to social networks of older youth.
That said, the committee expects that social sources, especially proxy
purchases, will remain the primary sources of tobacco for underage per-
sons, and it has been realistic about the high level of continuing availability
to adolescents and young adults who are in the workforce or in college
environments. Our estimates in this respect are predicated on relatively con-
servative assumptions. Although access to social sources could be reduced
significantly if the laws prohibiting transfers to underage persons were ag-
gressively enforced, the committee does not expect such a radical change
in enforcement policy in the foreseeable future, especially under a higher
MLA, because of likely public resistance. However, if a state or locality
decided to ramp up the threat of detection and punishment against social
sources and to sustain this policy, the impact on youth consumption could
be greater than the committee has projected.

Black Market Supply to Adolescents and Young Adults

As noted in Chapter 5, a 2015 National Research Council report
(NRC, 2015) concluded that a sizable illegal market in untaxed tobacco
and cross-border shipments from low-tax states to high-tax states is emerg-
ing. (See also Joossens and Raw, 2012; Shelley et al., 2007.) Cigarettes
are fairly compact and are not highly perishable. A day’s supply weighs
about an ounce, which means that black market operators could smuggle
nontrivial quantities in the trunk of a car or other small spaces. Under a
policy regime that significantly hindered social sources and proxy buyers, it
is theoretically possible that a true black market serving underage s mokers
could emerge (with entrepreneurs organizing their activities to target under-
age consumers). Nonetheless, it seems quite unlikely that enforcement of
the MLA restrictions against social sources and proxy buyers of tobacco
will be intensified so substantially as to create underage demand for black
market tobacco products. Also, it is difficult for a true black market to
emerge when everyone over a certain age is a legitimate purchaser (as has
been the experience with alcohol). As such, the committee thinks it highly
unlikely that raising the MLA will create a black market with “street deal-
ers” and associated violence, the way that prohibiting an entire product
class for all ages (e.g., marijuana) can do and has done. If this supposition
proves to be erroneous, the policy significance of an emerging black market
in tobacco on the streets of our communities goes way beyond the limited
task undertaken here.
Enforcement of PUP Restrictions

As noted in Chapter 5, bans against underage purchase–use–possession
(PUP) restrictions are common. Active enforcement of sanctions for PUP
violations has rarely been attempted and, in the committee’s judgment, is
unlikely to occur on a significant scale in the foreseeable future. However,
this is not to say that the bans have no instrumental effect; indeed, they
empower parents and schools to demand compliance and impose discipline.
If raising the MLA was to be accompanied by greater PUP enforcement
against underage users, then initiation rates could be reduced more than
the committee has estimated. The committee did not attempt to quantify
the effects of increased PUP law enforcement because there is so little basis
in either the deterrence literature or the tobacco youth access literature for
doing so.
Whether laws banning selling tobacco to minors should be accom-
panied by penalties against the underage purchasers themselves has been
debated for a more than a quarter of a century, ever since preventing ado-
lescent and young adult smoking emerged as a key component of tobacco

control in the early 1990s. Tobacco control advocates have typically con-
centrated their attention on the retailers and distributers who provide the
tobacco rather than on the buyers themselves (Jason et al., 2007; Wakefield
and Giovino, 2003). In addition to making enforcement easier, concentrat-
ing policy efforts on the sellers also focuses the moral responsibility for
preventing youth access to tobacco products on the retailers and industry
distributors rather than on the minors themselves (Craig and Boris, 2007;
Forster and Wolfson, 1998).
The case against punishing underage users of tobacco was put force-
fully in Growing Up Tobacco Free:
Imposing penalties on minors for buying, possessing, or using tobacco

products is controversial. At least 21 states currently prohibit smoking and
the use of tobacco products by minors. Proponents of these penalties argue
that they may have some deterrent value, and that the failure to make
possession illegal sends a mixed message, reinforcing the idea that tobacco
use is a trivial infraction. However, the Committee believes that penalizing
minors is an unwise and ineffective strategy. Criminal sanctions or delin-
quency adjudications are grossly disproportionate to the seriousness of the
offense and would not be sought by prosecutors or imposed by judges.
Even if the offense were punishable with a civil fine, like a traffic ticket,
the penalty would rarely be enforced. Because lack of enforcement would
erode whatever deterrent effect the law might otherwise achieve, the only
remaining rationale for such a prohibition is a symbolic one: the failure
to make tobacco use an offense would somehow imply that tobacco use is
not harmful or that it is socially acceptable. In the Committee’s view, such
speculative fears are groundless—social disapprobation is (or should be)
strongly communicated by the laws on distribution, by warning labels, and
by all of the other policies outlined in this report. Young people will not
miss the point simply because their disapproved conduct is not against the
law. Furthermore, purely symbolic prohibitions—laws that are not meant
to be enforced—are harmful because they undermine respect for the law.
Finally, imposing legal penalties on the underage purchaser also impedes
the use of underage buyers to monitor retailer compliance with youth ac-
cess restrictions. The need to obtain waivers unnecessarily increases the
cost of enforcement. (IOM, 1994, pp. 222–223)
Notwithstanding the argument set forth in the 1994 IOM report,

most states have prescribed penalties for underage purchasers, and some
tobacco control advocates have argued that youth access restrictions would
be more effective if sanctions against underage purchasers were prescribed
and enforced. First, their argument goes, PUP laws signal strong social
disapproval by making acquisition and use of tobacco punishable acts (the
declarative effect). Under this view, a law that penalizes retailers who sell

tobacco products but does not penalize the underage individuals for pur-
chasing, possessing, or using the product is sending a “mixed message,”
thereby undermining the social norms against tobacco use that tobacco
policy makers are trying to instill among young adults in work environ-
ments, school settings, and other public and private places. Second, PUP
proponents contend, penalties against the underage purchaser would have
a significant deterrent effect on purchase and would also make it easier to
deter underage proxy sellers. Penalties against underage alcohol users ap-
pear to have been enforced to a greater extent than penalties for underage
tobacco users and may have functioned to some extent as a deterrent to the
purchase and public transport or use of alcohol. For example, some states
have implemented so-called brown jug laws under which businesses that
sell alcohol are allowed to report underage purchase/use and to receive the
fine payments from offenders (IOM and NRC, 2004).
This argument reflects very different views about the effects on PUP
laws on underage smoking than those set forth in Growing Up Tobacco
Free as well as a different perspective on the potential disadvantages and the
costs of punishing young people for this sort of minor transgression. There
are few rigorous studies regarding the effects of PUP laws on underage use,
mainly because the laws are so rarely enforced, and the limited evidence is
mixed. It seems likely, in the committee’s view, that meaningful enforcement
of PUP sanctions against underage persons for purchasing, possessing, or
using tobacco products would deter tobacco use by some underage persons,
most likely those who are at least risk for becoming addicted. However, the
PUP laws on the books in 47 U.S. jurisdictions are essentially unenforced.
Under these circumstances, the operative policy is to capture the declarative
effects of making the behavior illegal and empowering parents and schools
to enforce it without incurring the costs of having to impose legal punish-
ment. The committee assumes that this will be the operative policy in the
foreseeable future, and its estimates reflect this conservative assumption.
ADOLESCENT DEVELOPMENT AND THE MLA FOR TOBACCO

In accordance with the committee’s charge, this report addresses the
“public health implications” of raising the MLA for tobacco products.
However, federal, state, and local lawmakers will likely take into account
factors other than public health benefits, including the economic interests
of tobacco retailers and other businesses that profit from tobacco use. Leg-
islators also will likely give some weight to arguments by and on behalf of
young adults that they should be entitled to make their own decisions about
whether to use tobacco products, especially in light of the fact that the “age
of majority” for many legal purposes is 18 in all but four states (JRank,
2014). This argument may be grounded in a deeper concern about the role

of government, especially in the realm of public health, and these ethical

concerns about the “nanny state” may affect not only the level of political
support for proposals to raise the MLA4 but also the level of community
willingness to enforce a higher MLA if it is enacted. Naturally the strength
of these concerns is likely to increase as the proposed MLA is raised from
19 to 21 to 25, all the more so when the policy lever is an outright prohibi-
tion rather than an excise tax or a public smoking restriction, which would
limit use without banning it completely. A lack of public support could
erode the potential public health benefits of raising the MLA.
The policy judgment regarding where to draw the line for the MLA
involves a burgeoning scientific literature on adolescent development on
which the Supreme Court has recently relied to explain why the Constitu-
tion mandates differential treatment of adolescents in the context of crimi-
nal punishment (Bonnie and Scott, 2013). That body of research, reviewed
in the National Research Council’s report Reforming Juvenile Justice: A
Developmental Approach (2013) and summarized in Chapter 3, docu-
ments various distinctive features of adolescent judgment as compared with
adults, including a heightened sensitivity to rewards, lower impulse control,
and tendencies to take risks—especially when in the presence of peers—and
to discount the long-term consequences of actions. These behavioral ten-
dencies are rooted in the different pace of maturation between the brain’s
motivational and reward systems and the systems in the brain that are
responsible for self-regulation and cognitive control. These developmental
factors, along with adolescents’ vulnerability to the rewarding effects of nic-
otine and their risk of addiction, are widely thought to justify policies that
curtail access to tobacco products by teenagers (IOM, 1994, 2007; IOM
and NRC, 2011). It is noteworthy that John Stuart Mill’s justly famous
defense of the anti-paternalism principle in his essay On Liberty (1859)
acknowledged that the individual’s sovereign control over self-regarding
choices applies only to persons “in the maturity of their faculties.” Indeed,
these same concerns about adolescent vulnerability and immature judgment
have been invoked to justify non-prohibitory efforts to curtail smoking by
addicted adults. As explained in Ending the Tobacco Problem in 2007:
It can also be argued that paternalism in this context is a justified response

to irremediable deficiencies in smokers’ capacity to successfully exercise
self-interested decision making about whether they should continue to
smoke. Although the committee’s blueprint need not rest on this argument,
many committee members do find elements of it convincing, and that is
4 A recent publication indicates that more than 70 percent of adults surveyed support raising
the age of sale of tobacco products to 21 years of age; majority support is seen across smoking
status, geographic region, race, sex, education, and age (Winickoff et al., 2015).

why we summarize it here. The argument runs as follows: (1) Virtually all
addicted adults begin smoking (and probably become addicted) while they
are adolescents, before they have developed the capacity to exercise mature
judgment about whether or not to become a smoker; (2) the preferences
expressed when people begin to smoke, which tend to ignore long-term
health risks, are inconsistent with the health-oriented preferences they later
come to have, and they soon regret the decision to have become a smoker;
and (3) once smokers begin to be concerned about the health dangers of
smoking, their judgment is often distorted by optimism bias (“the harms
will happen to other people, not to me”), thereby weakening their motiva-
tion to quit. (IOM, 2007, p. 150)
Although adolescents’ vulnerability to addiction and immaturity of

judgment support an underage access restriction, these developmental con-
cerns do not resolve the policy question about the specific age at which
the line should be drawn. The argument against raising the MLA above
18 is predicated on the assumption that adolescents older than 17 are
mature enough to make their own decisions about what is in their best
interests. However, experts on developmental psychology and neuroscience
(e.g., IOM and NRC, 2014; Steinberg, 2012) and also specialists in family
and adolescent and young adult policy (Goldfarb, 2014; Hamilton, 2012;
Scott, 2013) have called attention to the evidence that capacities related to
mature judgment, especially judgment in emotionally charged situations or
in situations in which peer influence plays are role, are still developing into
the early 20s. (See also Chapter 3.) Authorities on adolescent development
generally agree that the period of development that is typically labeled ado-
lescence stretches from the onset of puberty into the early 20s (Steinberg,
2012). Many young people in their late teens and early 20s may also still
be at elevated risk, developmentally speaking, to becoming addicted to
nicotine.
A review of age-specific public policies demonstrates that policy judg-
ments about where to draw age lines relating to adulthood are highly
contextual, ranging from ages 14 to 16 (medical decision making) to
age 21 (the purchase, use, and possession of alcohol and firearms, fiduciary
appointments, and most professional occupational licenses).5 In short, a
balance needs to be struck between the personal interest of young adults in
making their own choices and society’s legitimate concerns about protect-
ing the public health and protecting young people from decisions they may
later regret (IOM, 2007; IOM and NRC, 2004). None of this is to say that
the line should be drawn based solely on developmental science; it is only
5 Although not directly relevant in the present context, it is worth noting that the legally
relevant age of eligibility for various types of parental and social support in young adulthood
is often around 25 or 26 (IOM and NRC, 2014).

to say that 18 is not the only developmentally plausible place to draw the
line. The so-called age of majority functions as a default, and every state
sets the legal age for certain activities higher or lower for different policy
purposes. In short, state legislators will likely continue to draw the line in
different places in different policy contexts, and tobacco will be no excep-
tion (Bonnie and Scott, 2013; Hamilton, 2010; Steinberg, 2012).
One inevitable comparison in any discussion of the MLA for tobacco
is the 21-year-old MLA for alcohol in all states. The developmental jus-
tification for such a comparison is fairly strong in light of the addictive
properties of these drugs and the long-term consequences of initiating use
during adolescence. However, the intoxicating properties of alcohol are also
associated with harm to other persons, especially in relation to driving and
aggression, and not only with harm to oneself. The likely counterargument
is that the public health burden of tobacco use exceeds the toll associated
with any other self-regarding behavior or with the use of any other legal
product, making a case for “tobacco exceptionalism” in public health
policy (Collin, 2012; Malone and Warner, 2012). Whether this argument
is sufficient to trump otherwise strong commitments to individual choice is
being played out in the policy arena.
POSSIBLE PUBLIC HEALTH EFFECTS OF

NEW TOBACCO PRODUCTS
The prevalence of use of electronic nicotine delivery systems among
adolescents and young adults appears to be increasing substantially (see
Chapter 2; Arrazola et al., 2013; Wadley and Bronson, 2014). ENDS in-
clude electronic cigarettes (e-cigarettes), e-hookahs, and other vapor emit-
ting devices. FDA has begun the process of deeming these products to be
“tobacco products” under the Tobacco Control Act and thereby bringing
them within the agency’s regulatory jurisdiction. States have also been
gradually including these products in youth access statutes. The committee
assumes that FDA will eventually regulate these products, that they will be
subject to the MLA in all states, and that the committee’s findings regarding
enforcement of the MLA will apply to ENDS and other novel products. It is
also important to emphasize that the simulation models used in Chapters 7
and 8 are calibrated to project cigarette use and related outcomes and do
not include the public health effects of use of other tobacco products.
The question of greatest relevance to the committee’s task is how the
use of ENDS or other novel tobacco products is likely to affect the public
health impact of increasing the MLA. Assessing this impact is difficult,
given the relatively recent introduction of these products and the lack of
detailed data on the patterns of ENDS use over time, its relation to ciga-
rette use, and its health effects. Nevertheless, it is possible to speculate in

broad terms about several ways in which ENDS use might affect initiation
and prevalence of cigarette use and the public health and possibly alter the
projections described in Chapters 7 and 8.
Preliminarily, it should be emphasized that even if increasing ENDS use
has no effect on current patterns of initiation of cigarette use, it is likely
to affect the prevalence of cigarette use over the long term. The challenge
in evaluating its impact is that the net effect on conventional cigarette use
could be in either direction. For example, it is plausible that some persons
already using conventional cigarettes may quit using cigarettes and instead
switch to ENDS. In this scenario, there is likely a public health benefit in
that early data suggest that, while not harm free, ENDS are probably less
harmful than conventional cigarettes (Bhatnagar et al., 2014; Farsalinos
and Polosa, 2014; Grana et al., 2014). However, it is also plausible that
some persons already using conventional tobacco cigarettes may become
dual users of conventional cigarettes and ENDS (Bhatnagar et al., 2014;
Dutra and Glantz, 2014; Grana et al., 2014; Pearson et al., 2012; Regan et
al., 2013) because it costs less, helps the user reduce conventional cigarette
consumption, or serves as a “bridge” for nicotine use during times when
smoking conventional cigarettes is prohibited or inconvenient. Emergence
of this “dual use” scenario may increase the public health harm attribut-
able to tobacco use if it increases nicotine dependence (due to increased
consumption of nicotine), making smoking cessation more difficult, or
otherwise prolongs conventional cigarette smoking. While these scenarios
are postulated to have no effect on the initiation of cigarette use and are
therefore unaffected by raising the MLA, they would affect the quantita-
tive estimates of health benefits attributable to raising the purchase age by
reducing the estimated benefits in the first scenario (of increased conven-
tional cigarette cessation) and increasing them under the latter scenario (of
increased nicotine dependence and prolonged smoking).
The question of greatest relevance to this report is whether and how use
of ENDS will affect initiation of cigarette use. Broadly speaking, there are
three possibilities. One scenario is that initiation of ENDS use will reduce
initiation of cigarette use; that is, some portion of adolescents and young
adults who otherwise would have initiated cigarette use will not do so, be-
coming ENDS users instead. Under this scenario, there may be net public
health benefits over the long term, but some portion of those benefits would
be attributable to the initiation of ENDS, not to the raising of the MLA.
A second possibility is that initiation of ENDS would delay conventional
tobacco use, as adolescents and young adults who begin with ENDS switch
to conventional cigarettes at a later time, due in part to nicotine dependence
and to the relatively lower levels of nicotine delivery from ENDS compared
to conventional cigarettes. This scenario, involving the possibility of ENDS
serving as a gateway to conventional cigarettes, would be particularly wor-

risome because it would increase the prevalence of cigarette use, possibly

offsetting some portion (if not all) of the public health gains of raising the
MLA. Finally, as more recent data suggest (Wills et al., 2015), it is possible
that some of those who have never considered using conventional tobacco
products will initiate with ENDS and only use ENDS. In that case, the
net public health effect would be entirely attributable to the yet unknown
health effects of ENDS use.
All three of these patterns and trajectories of tobacco use—as well as
other variations—are likely to emerge, and the committee has no b asis for
estimating the proportions of adolescents and young adults that will take
each path, much less the net effect of ENDS use on initiation of cigarette
use. What can be said, then, about the possible effects of raising the MLA
for ENDS use on the likelihood of these scenarios? For this purpose, the
committee assumes that the MLA will be increased for all tobacco products,
including ENDS, and that the intensity of enforcement will be the same for
all products. The committee sees no reason to believe that the effects of the
legal norm and its enforcement on retailer compliance, retail availability, or
access to social sources would differ materially for ENDS compared with
other tobacco products. Given the evidence that adolescents who currently
initiate tobacco use with ENDS rather than with conventional tobacco
products are younger (Wills et al., 2015), the main effect of raising the
MLA for ENDS will likely be to reduce the number of adolescents who
initiate tobacco use with ENDS. That may translate into reduced initiation
of cigarette use for some, but it also may translate into delayed initiation
of cigarette use for others, including some proportion who would not have
otherwise used conventional cigarettes. Presumably FDA and state policy
makers will take these possibilities into account in setting the MLA and
will carefully monitor the promotion and use of ENDS, especially by ado-
POSSIBLE EFFECTS OF RAISING THE TOBACCO MLA

ON USE OF ALCOHOL AND OTHER DRUGS
In summarizing the estimated health effects of raising the tobacco MLA
presented in Chapter 8, the committee has not taken into account the pos-
sibility that reducing adolescent and young adult tobacco use could affect
the use of alcohol, marijuana, or other illegal drugs and thus has ignored
the substantial mortality and morbidity associated with use of those sub-
stances. However, it is possible that raising the MLA for tobacco could
have indirect effects on the use and abuse of other substances, either by
increasing their use (and thereby having a negative effect on public health
that might offset some of the effects of reduced tobacco use) or by decreas-
ing their use (and thereby augmenting the public health benefit of reducing

tobacco use). Because tobacco use is correlated with the use of many other
substances, it could be important to consider the indirect effects of reducing
tobacco use on the use of other substances, but definitive statements are
difficult to make since the associations need not be causal. The mere fact
that people who smoke today have greater rates of abuse or dependence
on other substances is not sufficient to infer that an intervention that re-
duces smoking—such as raising the MLA—would necessarily reduce rates
of abuse or dependence on that second substance. There is, however, some
literature examining the effects of tobacco control interventions on the use
of other substances.
The empirical literature on spillover effects of tobacco policies on alco-
hol use and abuse is mixed. Picone et al. (2004) found that smoking bans
reduce alcohol consumption in older adult females. Gallet and Eastman
(2007) obtained a similar but more general result, but Hahn et al. (2010)
found no such effect. Young-Wolff et al. (2014) reported that increasing
tobacco taxes was associated with modest to moderate reductions in al-
cohol use in vulnerable groups. McKee and colleagues, in a series of three
studies (Kasza et al., 2012; McKee et al., 2009; Young-Wolff et al., 2013),
found evidence for the proposition that smoking bans reduce alcohol use
and related problems. However, Bernat et al. (2013) did not observe a
decline in alcohol-related vehicle accidents when analyzing California and
New York’s statewide smoke-free policies.
There is also a modest literature investigating whether tobacco and
alcohol are “substitutes” or “complements” in the economic sense of these
terms. Although some studies find that cigarettes are substitutes for either
alcohol in general (Decker and Schwartz, 2000) or liquor in particular
(Goel and Morey, 1995), a more common finding is that they are instead
complements (e.g., Bask and Melkersson, 2004; Cameron and Williams,
2001; Jones, 1989; Pierani and Tiezzi, 2009; Tauchmann et al., 2013; Zhao
and Harris, 2004); that is, they enhance each other’s value to a user, and a
decrease in the use of one is likely to be associated with a decrease in the
use of the other. Thus, the research would suggest that interventions that
reduce tobacco use will not increase alcohol use. A study by Hughes (1993)
found that smoking cessation treatment among adults does not increase
alcohol intake, even among former alcohol abusers.
In theory, tobacco control policies could have indirect effects on the
consumption of illicit drugs. However, the literature on this subject is quite
sparse and mostly limited to effects on marijuana use. A few studies, such
as Cameron and Williams (2001), Chaloupka et al. (1999), and Zhao and
Harris (2004), find “complementarity” between tobacco and marijuana;
that is, when cigarette prices go up, marijuana use declines. However,
Cameron and Williams (2001) found that increases in tobacco prices did
not affect cannabis use.

A conceptually separate issue concerns how changes in marijuana policy

might affect tobacco use and, hence, the effects of raising the MLA on
tobacco use. Marijuana policy is in a state of flux, and there is consider-
able overlap between the populations who use marijuana and those who
use tobacco. Thus, changes in marijuana policy might spill over to affect
tobacco use, and vice versa. It is extremely challenging to estimate legaliza-
tion’s effects on marijuana use (Kilmer et al., 2010), let alone its spillover
effects on the use of other substances, including whether any spillover effects
would enhance or undermine the value of raising the MLA for tobacco. This,
however, does not imply that any such effects would be small. If marijuana
and tobacco were substitutes, increased marijuana use might lead to lower
tobacco use. As noted above, however, what little literature exists on the
subject suggests that marijuana and tobacco are more likely to be comple-
ments, not substitutes.
Furthermore, although the overlap in North America has tended to be
by user (with marijuana smokers more likely than others to smoke tobacco
and vice versa), in Europe it is quite common to mix tobacco and marijuana
in the same cigarette (UNODC, 2006), as also occurs already in the United
States with “blunts.” Hence, it is plausible that what adolescents and young
adults primarily construe as “marijuana use” might become the vehicle for
first exposure to nicotine. Also, the relaxation of marijuana laws has been
accompanied by a proliferation of modalities of use, including vaporization
as opposed to combustion. It is conceivable that a proliferation of vaporizer
pens or other devices acquired initially for marijuana use might facilitate
the uptake of consumption of nicotine via ENDS or increase the social ac-
ceptability of “vaping.”
In sum, it seems plausible that to the extent that raising the MLA
reduces tobacco use, it might have some beneficial spillover in the form
of indirect effects on the use of and harm from alcohol and, potentially,
marijuana. And it seems plausible that changes in marijuana policy and
patterns of use could modulate the effects of raising the MLA on tobacco
use. However, the existing empirical literature does not allow estimating a
specific magnitude or even a potential range of estimates of those effects
in the population overall, let alone among adolescents and young adults
specifically.
CONCLUDING REMARKS
The committee was charged with assessing the potential public health
implications of raising the minimum age of legal access to tobacco products.
Studies investigating the effects of setting or raising the MLA for tobacco
are sparse. In order to carry out its charge, the committee undertook a
thorough review of the available evidence related to tobacco use by ado-

lescents and young adults, the effects of raising the MLA for alcohol, and
enforcement of the existing MLA restrictions for tobacco products. This
evidence provided a solid foundation for the critical phase of the commit-
tee’s work—using its collective expert judgment to estimate the effects of
raising the MLA on initiation rates at various ages. Using these estimates
as inputs, the committee commissioned new modeling studies of aggregate
smoking behavior with which to project likely population-level outcomes
of changes in the MLA. The most important assumptions required for
these estimates have been discussed in this chapter, as have been additional
policy-relevant considerations.
Among the key assumptions are relative stability in the intensity of
tobacco control activities and the continuation of the MLA enforcement at
existing levels. These are relatively conservative assumptions, and the public
health benefits could be greater if tobacco control policies and the MLA
enforcement were substantially strengthened. It is important to recognize,
however, that public health gains also have to be weighed against the costs
and other social consequences of enforcing more restrictive MLA policies.
It is also important to emphasize that the committee’s modeling esti-
mates are based on nationwide adoption of the increased MLA, although
public health benefits of that magnitude will occur only if Congress facili-
tates federal action or if states with a substantial portion of the nation’s
population raise the MLA. Over the short term, at least, the projected
public health benefits will need to be translated into state-by-state estimates.
Although the full benefits of preventing initiation of tobacco use will
take decades to accrue, some direct health benefits, including those from
reduced secondhand smoke exposure, will be immediate. Perhaps the great-
est uncertainty in the committee’s assessment is the currently unpredictable
effects of the marketing and use of electronic nicotine delivery systems and
other novel tobacco products. However, in the absence of transformative
changes in the tobacco market, social norms and attitudes, or the epidemi-
ology of tobacco use, the committee is reasonably confident that raising the
MLA will reduce tobacco initiation, particularly among adolescents 15 to
17 years of age, will improve health across the life span, and will save lives.
REFERENCES
62(45):893–897.
Bask, M., and M. Melkersson. 2004. Rationally addicted to drinking and smoking? Applied
Economics 36(4):373–381.
Bernat, D. H., M. Maldonado-Molina, A. Hyland, and A. C. Wagenaar. 2013. Effects of
smoke-free laws on alcohol-related car crashes in California and New York: Time series
analyses from 1982 to 2008. American Journal of Public Health 103(2):214–220.

Bhatnagar, A., L. P. Whitsel, K. M. Ribisl, C. Bullen, F. Chaloupka, M. R. Piano, R. M.

Robertson, T. McAuley, D. Goff, and N. Benowitz. 2014. Electronic cigarettes: A policy
statement from the American Heart Association. Circulation 130(16):1418–1436.
Bonnie, R. J., and E. S. Scott. 2013. The teenage brain: Adolescent brain research and the law.
Cameron, L., and J. Williams. 2001. Cannabis, alcohol and cigarettes: Substitutes or comple-
ments? Economic Record 77(236):19–34.
CDC (Centers for Disease Control and Prevention). 2014. Best practices for comprehensive
tobacco control programs—2014. Atlanta, GA: U.S. Department of Health and Human
Services, Centers for Disease Control and Prevention, National Center for Chronic Dis-
ease Prevention and Health Promotion, Office on Smoking and Health.
Chaloupka, F. J., R. L. Pacula, M. C. Farrelly, L. D. Johnston, and P. M. O’Malley. 1999. Do
higher cigarette prices encourage youth to use marijuana? National Bureau of Economic
Research Working Paper Series No. 6939.
Collin, J. 2012. Tobacco control, global health policy and development: Towards policy coher-
ence in global governance. Tobacco Control 21(2):274–280.
Decker, S. L., and A. E. Schwartz. 2000. Cigarettes and alcohol: Substitutes or complements?
National Bureau of Economic Research Working Paper Series No. 7535.
Dutra, L. M., and S. A. Glantz. 2014. Electronic cigarettes and conventional cigarette use
among U.S. adolescents: A cross-sectional study. JAMA Pediatrics 168(7):610–617.
Farrelly, M. C., T. F. Pechacek, K. Y. Thomas, and D. Nelson. 2008. The impact of tobacco
control programs on adult smoking. American Journal of Public Health 98(2):304–309.
103(3):549–555.
Farsalinos, K. E., and R. Polosa. 2014. Safety evaluation and risk assessment of electronic
cigarettes as tobacco cigarette substitutes: A systematic review. Therapeutic Advances in
Drug Safety 5(2):67–86.
Forster, J. L., and M. Wolfson. 1998. Youth access to tobacco: Policies and politics. Annual
Review of Public Health 19(1):203–235.
Gallet, C. A., and H. S. Eastman. 2007. The impact of smoking bans on alcohol demand. The
Social Science Journal 44(4):664–676.
Goel, R. K., and M. J. Morey. 1995. The interdependence of cigarette and liquor demand.
Southern Economic Journal 62(2):451–459.
Goldfarb, S. F. 2014. Who pays for the “boomerang generation”? A legal perspective on finan-
cial support for young adults. Harvard Journal of Law & Gender 37:45–106.
Grana, R., N. Benowitz, and S. A. Glantz. 2014. E-cigarettes: A scientific review. Circulation
129(19):1972–1986.
Hahn, E. J., M. K. Rayens, S. L. Ridner, K. M. Butler, M. Zhang, and R. R. Staten. 2010.
Smoke-free laws and smoking and drinking among college students. Journal of Com-
munity Health 35(5):503–511.
Hamilton, V. E. 2010. Immature citizens and the state. Brigham Young University Law Review
1055(4):1055–1148.
———. 2012. The age of marital capacity: Reconsidering civil recognition of adolescent
marriage. Boston University Law Review 92:1817–1863.

HHS (Department of Health and Human Services). 2000. Reducing tobacco use: A report
of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services,
Centers for Disease Control and Prevention, National Center for Chronic Disease Preven-
tion and Health Promotion, Office on Smoking and Health.
Hughes, J. R. 1993. Treatment of smoking cessation in smokers with past alcohol/drug prob-
lems. Journal of Substance Abuse Treatment 10(2):181–187.
Academy Press.
———. 2011. The science of adolescent risk-taking: Workshop report. Washington, DC: The
Jason, L. A., S. B. Pokorny, M. Adams, Y. Hunt, P. Gadiraju, T. Morello, M. Schoeny, and
C. Dinwiddie. 2007. Youth caught in violation of tobacco purchase, use, and possession
laws: Education versus fines. Behavior Modification 31(6):713–731.
Jemal, A., V. E. Cokkinides, O. Shafey, and M. J. Thun. 2003. Lung cancer trends in young
adults: An early indicator of progress in tobacco control (United States). Cancer Causes
and Control 14(6):579–585.
Jones, A. M. 1989. A systems approach to the demand for alcohol and tobacco. Bulletin of
Economic Research 41(2):85–106.
Joossens, L., and M. Raw. 2012. From cigarette smuggling to illicit tobacco trade. Tobacco
Control 21(2):230–234.
JRank. 2014. Legal ages laws—Information on the law about legal ages. http://law.jrank.org/
pages/11848/Legal-Ages.html (accessed October 30, 2014).
Kasza, K. A., S. A. McKee, C. Rivard, and A. J. Hyland. 2012. Smoke-free bar policies and
smokers’ alcohol consumption: Findings from the International Tobacco Control Four
Country Survey. Drug and Alcohol Dependence 126(1–2):240–245.
Kilmer, B., J. P. Caulkins, R. L. Pacula, R. J. MacCoun, and P. H. Reuter. 2010. Altered state?:
Assessing how marijuana legalization in California could influence marijuana consump-
tion and public budgets. RAND Corporation Occassional Paper Series. http://www.
rand.org/content/dam/rand/pubs/occasional_papers/2010/RAND_OP315.pdf (accessed
February 22, 2015).
Kuiper, N. M., D. E. Nelson, and M. Schooley. 2005. Evidence of effectiveness: A summary
of state tobacco control program evaluation literature. Washington, DC: U.S. Depart-
ment of Health and Human Services, Public Health Service, Centers for Disease Control
and Prevention.
Laugesen, M., and B. Swinburn. 2000. New Zealand’s tobacco control programme, 1985–
1998. Tobacco Control 9(2):155–162.
Luke, D. A., K. A. Stamatakis, and R. C. Brownson. 2000. State youth-access tobacco control
policies and youth smoking behavior in the United States. American Journal of Preventive
Medicine 19(3):180–187.
Malone, R. E., and K. E. Warner. 2012. Tobacco control at twenty: Reflecting on the past,
considering the present and developing the new conversations for the future. Tobacco
Control 21(2):74–76.

McKee, S. A., C. Higbee, S. O’Malley, L. Hassan, R. Borland, K. M. Cummings, G. Hastings,

G. T. Fong, and A. Hyland. 2009. Longitudinal evaluation of smoke-free Scotland on pub
and home drinking behavior: Findings from the International Tobacco Control Policy
Evaluation Project. Nicotine & Tobacco Research 11(6):619–626.
Mill, J. S. 1859. On liberty. London: Longman, Roberts & Green.
NRC (National Research Council). 2013. Reforming juvenile justice: A developmental ap-
proach. Edited by R. J. Bonnie, R. L. Johnson, B. M. Chemers, and J. Schuck. Washing-
ton, DC: The National Academies Press.
———. 2015. Understanding the U.S. illicit tobacco market: Characteristics, policy context,
and lessons from international experiences. Washington, DC: The National Academies
Press.
NYCDOHMH (New York City Department of Health and Mental Hygiene). 2014. New
law prohibiting sale of cigarettes, tobacco products, and electronic cigarettes to people
under age 21 in New York City: What you need to know. http://www.nyc.gov/html/doh/
downloads/pdf/smoke/tobacco-21-faq.pdf (accessed January 29, 2015).
Pearson, J. L., A. Richardson, R. S. Niaura, D. M. Vallone, and D. B. Abrams. 2012.
E-cigarette awareness, use, and harm perceptions in U.S. adults. American Journal of
Public Health 102(9):1758–1766.
Picone, G. A., F. Sloan, and J. G. Trogdon. 2004. The effect of the tobacco settlement and
smoking bans on alcohol consumption. Health Economics 13(10):1063–1080.
Pierani, P., and S. Tiezzi. 2009. Addiction and interaction between alcohol and tobacco con-
sumption. Empirical Economics 37(1):1–23.
Regan, A. K., G. Promoff, S. R. Dube, and R. Arrazola. 2013. Electronic nicotine delivery
systems: Adult use and awareness of the “e-cigarette” in the USA. Tobacco Control
22(1):19–23.
Scott, E. S. 2013. “Children are different”: Constitutional values and justice policy. Ohio State
Journal of Criminal Law 11(1):71-105.
Shelley, D., M. J. Cantrell, J. Moon-Howard, D. Q. Ramjohn, and N. VanDevanter. 2007. The
$5 man: The underground economic response to a large cigarette tax increase in New
York City. American Journal of Public Health 97(8):1483–1488.
Stillman, F. A., A. M. Hartman, B. I. Graubard, E. A. Gilpin, D. M. Murray, and J. T. Gibson.
2003. Evaluation of the American Stop Smoking Intervention Study (ASSIST): A report
of outcomes. Journal of the National Cancer Institute 95(22):1681–1691.
Tauchmann, H., S. Lenz, T. Requate, and C. M. Schmidt. 2013. Tobacco and alcohol: Comple-
ments or substitutes? Empirical Economics 45(1):539–566.
Tauras, J. A., S. Markowitz, and J. Cawley. 2005. Tobacco control policies and youth smoking:
Evidence from a new era. Advances in Health Economics and Health Services Research
16:277–291.
UNODC (United Nations Office on Drugs and Crime). 2006. Bulletin on narcotics: Review
of the world cannabis situation, Volume LVIII, nos. 1 and 2. Vienna: United Nations
Office on Drugs and Crime.
Wadley, J., and A. Bronson. 2014. E-cigarettes surpass tobacco cigarettes among teens. Ann
Arbor, MI: University of Michigan News Service. http://ns.umich.edu/new/multimedia/
videos/22575-e-cigarettes-surpass-tobacco-cigarettes-among-teens (accessed February 6,
2015).

Wakefield, M., and F. Chaloupka. 2000. Effectiveness of comprehensive tobacco control

programmes in reducing teenage smoking in the USA. Tobacco Control 9(2):177–186.
Wakefield, M., and G. Giovino. 2003. Teen penalties for tobacco possession, use, and pur-
chase: Evidence and issues. Tobacco Control 12(Suppl 1):i6–i13.
Wills, T. A., R. Knight, R. J. Williams, I. Pagano, and J. D. Sargent. 2015. Risk factors for
exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pedi-
atrics 135(1):e43–e51.
Winickoff, J. P., R. McMillen, S. Tanski, K. Wilson, M. Gottlieb, and R. Crane. 2015. Public
support for raising the age of sale for tobacco to 21 in the United States. Tobacco Con-
trol. doi: 10.1136/tobaccocontrol-2014-052126. [Epub ahead of print]
Young-Wolff, K. C., A. J. Hyland, R. Desai, J. Sindelar, C. E. Pilver, and S. A. McKee. 2013.
Smoke-free policies in drinking venues predict transitions in alcohol use disorders in a
longitudinal U.S. sample. Drug and Alcohol Dependence 128(3):214–221.
Young-Wolff, K. C., K. A. Kasza, A. J. Hyland, and S. A. McKee. 2014. Increased cigarette
tax is associated with reductions in alcohol consumption in a longitudinal U.S. sample.
Alcoholism, Clinical and Experimental Research 38(1):241–248.
Zaza, S., P. A. Briss, and K. W. Harris. 2005. The guide to community preventive services:
What works to promote health. New York: Task Force on Community Preventive
Services.
Zhao, X., and M. N. Harris. 2004. Demand for marijuana, alcohol and tobacco: Participa-
tion, levels of consumption and cross-equation correlations. The Economic Record
80(251):394–410.


Appendix A
State and Local Laws on the

Minimum Age of Legal Access
to Tobacco Products
T
he 2009 Family Smoking Prevention and Tobacco Control Act estab
lished 18 as the minimum age of legal access to tobacco products
(MLA) while also allowing states and localities to impose a higher
MLA. In the last decade, this has been an active area of tobacco control
policy. In Massachusetts, in particular, where state law grants local Boards
of Health authority to make “reasonable health regulations,”1 nearly two
dozen communities have raised the MLA above age 18, and numerous
others are currently considering proposals. A smaller number of jurisdic-
tions outside of Massachusetts, most notably New York City, have also
taken similar steps. This appendix provides a set of tables detailing activi-
ties at the state and local levels to raise the MLA. These tables provide a
selection of the recent activity in this area of the law and aim to provide
examples of the type, range, and scale of such activity as of September
2014. As such, this appendix should be considered a list of illustrative
examples rather than a comprehensive and exhaustive list of jurisdictions
that have raised or are considering raising the MLA. Tables A-1 and A-2
list states and localities that have raised the MLA to 19 and 21, respectively.
Tables A-3 and A-4 list select states and localities that are considering pro-
posals to raise their MLA to 19 and 21, respectively. Tables A-5 and A-6 list
select states and localities that have considered but not enacted proposals
to raise the MLA.
1 MASS. GEN. LAWS ch. 111 § 31.
265

TABLE A-1 Select States and Localities That Have Established a

Minimum Age of Legal Access to Tobacco Products of 19
Jurisdiction
(state/county/ Year Regulation/
town) Enacted Legislation Status Details of Law
Alabamaa 1997 Legislation Enacted. Defines a “minor” for the
purposes of this tobacco
law as any individual
under 19 years of age.
Alaskab 1988 Legislation Enacted. It is illegal to sell or give
tobacco or any product
containing nicotine to a
minor, defined as someone
Massachusetts—Select Towns
Brookline 2013 Special town Approved. Voted to raise the MLA
(Parker, 2013) meeting from 18 to 19 years.
action to
amend town
law
Newburyport 2014 Town board Approved. The board of health
(Hendrickson, of health initially proposed raising
2013; Quinn, regulation the MLA from 18 to 21
2014; Wade, years. Mayor Donna
2013) Holaday vowed to fight
the measure and litigate
if necessary. The board
subsequently considered
and approved a measure
to raise the age to 19.
Sudburyc 2013 Town board Approved. The board of health held a
of health public hearing and, at the
regulation subsequent board meeting,
approved raising the MLA
for tobacco products from
18 to 19 years.
Walpoled 2013 Town board Approved. The board of health
of health Further action approved raising the MLA
regulation to raise the for tobacco products from
MLA to 21 will 18 to 19 years. The board
be discussed in also discussed phasing
mid-2015. in an MLA of 21 years,
and will revisit the issue
in mid-2015 to consider
further action.

APPENDIX A 267
TABLE A-1 Continued

Jurisdiction
Watertowne 2012 Town board Approved. The board of health
of health adopted a new tobacco
regulation regulation that defines
a “minor” (relative to
tobacco) as someone
Westwoodf 2013 Town board Approved. The board of health
of health adopted a new tobacco
regulation regulation that defines
a “minor” (relative to
tobacco) as someone
New Jerseyg 2006 Legislation Enacted. Amended earlier tobacco
laws, raising MLA for
purchase and sale of
tobacco products from 18
to 19 years.
New York—Select Counties
Nassauh 2006 Legislation Enacted. Local laws amended to
add provisions to Nassau
Administrative Code
to raise the MLA for
tobacco products from 18
to 19 years.
Onondagai 2009 Legislation Enacted. Raised the MLA for
tobacco products from
18 to 19 years, with the
exception of individuals
who are 18 years of age
and have a valid military
ID, who are exempt and
may purchase tobacco
products.
continued

TABLE A-1 Continued

Jurisdiction
Suffolkj 2005 Legislation Enacted, but New law filed, repealing
amended by earlier legislation raising
2014 law minimum purchasing age
raising the age from 18 to 19 years.
to 21. (See
table of U.S.
Jurisdictions
with an
MLA of 21
to Purchase
Tobacco
Products—
Enacted.)
Utahk 1973 Legislation Enacted. New Prohibits sales tobacco
legislation and tobacco products to
to raise the individuals under 19 years
MLA from 19 of age. Prohibits business
to 21 years owners from allowing
introduced individuals under age 19
in 2013. (See from entering businesses
table of U.S. while the underage
Jurisdictions person is using tobacco.
with an “Smoking paraphernalia”
MLA of 21 was added in 2010.
to Purchase
Tobacco
Products.)
a The Code of Alabama 1975 § 13A-12-3, 28-11-9, 28-11-13.
b Alaska Statutes § 11.76.100, 11.76.105-109.
c Sudbury, Massachusetts, Board of Health. Minutes of Meeting of September 10, 2013. Avail-
able: https://sudbury.ma.us/boardofhealth/?attachment_id=283 (accessed February 25, 2015).

d Walpole, Massachusetts, Board of Health. Minutes of Meeting of March 12, 2013.
Available: http://www.walpole-ma.gov/sites/walpolema/files/minutes/minutes-file/minutes_
march_12_2013.pdf (accessed February 25, 2015).
e Watertown, Massachusetts, Board of Health. Minutes of Meeting of August 15, 2012. Avail-
able: http://www.ci.watertown.ma.us/Archive/ViewFile/Item/1824 (accessed February 25, 2015).

f Town of Westwood, Massachusetts, Board of Health Regulations. Regulation restricting the
sale of tobacco products and nicotine delivery product (February 12, 2013).
g New Jersey State Legislation. P.L. 2005 c. 384; S2783 (January 15, 2006).
h The Nassau County Administrative Code. Title H § 9-25.1–9-25.7. Local Law 5-2006
(April 26, 2006).

i Onondaga County, New York, Local Law No. 2-2009 (January 12, 2009).
j Suffolk County, New York, Local Law No. 5-2005 (January 3, 2005).
k Utah State Legislature. Ban on sale of smoking paraphernalia to minors section 76-10-104.1
(March 29, 2010).

APPENDIX A 269
TABLE A-2 States and Localities That Have Established a Minimum Age
of Legal Access to Tobacco Products of 21
Jurisdiction
town) Enacted Legislation Status Details of the Law
California—Select City
Healdsburg 2014 Regulation Passed by city Prohibits the sale of
(Alexander council. cigarettes, chew, and
and Williams, other tobacco products
2014; Mason, to anyone under age 21
2013) and institutes new annual
license to sell tobacco,
with revenues earmarked
for enforcing tobacco
laws.
Hawaii—Select County
Hawaii 2013 Legislation Enacted. In The bill raised minimum
Countya effect June 30, purchasing age from 18
2014. years of age to 21.
Illinois—Select City
Evanston 2014 Regulation Passed by city The new ordinance raises
(Blakley, council. the MLA from 18 to
2014) 21. An earlier version of
the ordinance proposed
making underage
possession a crime, but
the final ordinance holds
only retailers responsible
for violations.
Arlingtonb,c 2013 Town board Approved May From July 1, 2013, to
of health 2013, in effect July 1, 2014, the MLA is
regulation July 2013, and 19. From July 1, 2014, to
subject to a July 1, 2015, the MLA is
3-year phase-in, raised to 20. As of July 1,
raised to 20 on 2015, the MLA is raised
July 1, 2014. to 21.
Ashlandd 2013 Town board Approved. Raised the MLA from 18
of health to 21. Discussed but did
regulation not vote on e-cigarette
regulations.
continued

TABLE A-2 Continued

Jurisdiction
Belmonte 2012 Town board Approved. Amendment of existing
(Eisenstadter, of health regulations of the sale
2014) regulation of tobacco to minors to
raise the age of a “minor”
(relative to tobacco
purchasing) from 18 to
19 years of age. The MLA
raised to 21 in September
2014, to take effect on
January 1, 2015.
Braintree 2014 Town board Approved. Regulation raises the
(Aicardi, of health MLA to 21 products and
2014) regulation establishes an MLA of 21
for electronic cigarettes
and other “nicotine-
delivery” products.
Canton 2013 Town board Approved. Raises minimum
(Donga, 2013; of health purchasing age from 18
Turner, 2013) regulation to 21 years. Contains
a “sunset” clause,
whereby, in 5 years,
the MLA will revert to
18 years if there is no
major reduction in teen
smoking rates. If there is
a demonstrable reduction,
the MLA of 21 years will
automatically renew. With
approval from the school
superintendent, the board
of health will conduct an
annual survey of Canton
middle and high school
students to determine
the efficacy of the new
regulation.
Dedhamf 2013 Town board Approved. Amended existing
of health regulations, raising the
regulation MLA from 18 to 21.
Doverg 2013 Town board Approved. Amended existing
of health regulations, raising the
regulation MLA from 18 to 21.

APPENDIX A 271
TABLE A-2 Continued

Jurisdiction
Foxboroughh 2013 Town board Approved. Held a public hearing
of health and adopted regulations
regulation restricting the sale of
tobacco, including raising
the MLA from 18 to 21.
Hudson 2014 Town board Approved. Raised the MLA from 18
(Bartlett, of health to 21.
2014b; regulation.
Malachowski,
2014)
Hull (Hanson, 2014 Town board Approved. Raised the MLA from
2014) of health 18 to 21 and bans
regulation smoking in town parks,
playgrounds, and adjacent
parking lots.
Malden (City 2014 Town board Approved. Amends board of health
of Malden, of health rules and regulations,
2014) regulation raising the MLA from 18
to 21.
Medwayi 2013 Non- Approved by Board of selectmen voted
(Comeau, binding the board of unanimously in March
2014) referendum selectmen for 2014 to include the
question. a public vote proposed raising of the
at the annual MLA to purchase tobacco
town election, products from 18 to 21 as
May 20, 2014. a non-binding referendum
question on the May 20
town election ballot, one
of three legal options to
put the question up for
public vote. In 2013 the
board of health discussed
raising the MLA from 18
to 21 in May, June, and
July 2013, but it could
not agree whether to
make changes to tobacco
regulations as a board,
wait for quorum, or to go
to a public vote.
continued

TABLE A-2 Continued

Jurisdiction
Melrose 2014 Town board Approved. Prohibits the sale
(Sacco, 2014) of health of cigarettes, cigars,
regulation electronic “vaping
devices,” snuff, and
related products to
persons under age 21.
Needhamj 2003 Town board Approved. From April 1, 2003, to
of health In effect April 1, 2004, the MLA
regulation April 1, 2003, was raised to 19. From
and subject to a April 1, 2004, to April
3-year phase-in. 1, 2005, the MLA was
raised to 20. As of April
1, 2005, the MLA was
raised to 21.
Newtonk 2014 Town board Passed Revised town ordinances
of aldermen ordinance. to raise the MLA to
21 and also cover all
nicotine delivery products
(including e-cigarettes).
Norwoodl 2014 Town board Approved. The board of health
of health issued new regulations
regulation defining “minor” as “any
individual who is under
the age of twenty-one
(21).”
Scituate 2014 Town board Approved. Raised the MLA from 18
(Bartlett, of health to 21.
2014a,b) regulation
Sharonm 2013 Town board Approved. Raised the MLA from 18
of health to 21.
regulation
Wakefieldn 2014 Town board Approved. Raised the MLA from 18
of health to 21.
regulation
Wayland 2014 Town board Approved. In Raised the MLA from 18
(Wagner, of health effect beginning to 21 and bans electronic
2014) regulation January 1, smoking devices anywhere
2015. smoking is prohibited in
workplaces.
Wellesleyo 2014 Town board Approved. Raised the MLA from 18
of health to 21.
regulation

APPENDIX A 273
TABLE A-2 Continued

Jurisdiction
Westford 2014 Town board Approved. Prohibits the sale of
(Allen, 2014) of health cigarettes and electronic
regulation cigarettes to persons
under age 21.
Winchester 2014 Town board Approved. Raises the MLA and
(McLean, of health nicotine delivery products
2014) regulation (including e-cigarettes)
from 18 to 21.
New Jersey—Select Towns
Englewood 2014 City board Approved. Raises the MLA from 19
(Noda, 2014; of health to 21 for the purchase
Perez, 2014) resolution of tobacco or tobacco-
related products from any
vendor in the city.
Sayreville 2014 Borough Approved. Raises the MLA from 19
(Loyer, 2014) council to 21.
ordinance
New York— Select Counties
New York 2013 Legislation Passed by city Local law amends the
Cityp council and administrative code of
signed into the city of New York
law by mayor. (including New York,
In effect as of Bronx, Kings, Queens,
May 2014. and Richmond Counties),
raising the MLA from
18 to 21 years of age.
Establishes an MLA of 21
for electronic cigarettes.
Suffolk 2014 Legislation Signed into law, Raises the MLA from 19
Countyq to go into effect to 21.
(Schwartz, January 1,
2014) 2015.
a The Hawaii County Code 1983 (2005). Bill no. 135. Ordinance no. 13 124 (December 13,
2013).
b Town of Arlington, Massachusetts, Department of Health and Human Services; Office of
the Board on Health. Regulation restricting the sale of tobacco products and nicotine delivery
products (May 15, 2013).
c Arlington, Massachusetts, Board of Health. Minutes of Meeting of May 15, 2013. Available:
http://www.arlingtonma.gov/Home/ShowDocument?id=1084 (accessed February 25, 2015).

d Ashland, Massachusetts, Board of Health. Minutes of Meeting of September 10, 2013. Avail-
able: http://www.ashlandmass.com/document-center/minutes (accessed February 25, 2015).
continued

TABLE A-2 Continued

e Belmont, Massachusetts, Board of Health. Minutes of Meeting of April 30, 2012. Available:
http://www.belmont-ma.gov/sites/belmontma/files/minutes/minutes-file/4-30-12.pdf (accessed
February 25, 2015).
f Board of Health Regulations, Dedham, Massachusetts, Part IV: Regulation affecting smoking
and the sale and distribution of tobacco and nicotine delivery products in Dedham (November
25, 2013).
g Dover, Massachusetts, Board of Health. Minutes of Meeting of May 13, 2013. Available:
http://www.doverma.org/wp-content/uploads/2013/02/boh-5-13.pdf (accessed February 25,

2015).
h Foxborough, Massachusetts, Board of Health. Minutes of Meeting of June 24, 2013.
Available: http://www.foxboroughma.gov/Pages/FoxboroughMA_HealthMin (accessed Feb-

ruary 25, 2015).
i Medway, Massachusetts, Board of Health. Minutes of Meeting of July 22, 2013, and Novem-
ber 20, 2013. Available: http://www.townofmedway.org/Pages/MedwayMA_Bcomm/BOH/

Minutes/2013 (accessed February 25, 2015).
j Town of Needham Board of Health Regulations. Section 1.6. Retail sale of tobacco products
(updated February 4, 2014).

k City of Newton Board of Alderman. Ordinance No. A-42 (June 16, 2014).
l Norwood, Massachusetts, Board of Health Regulations. Restricting the sale of tobacco
products and nicotine delivery products (February 27, 2014).

m Sharon, Massachusetts, Board of Health. Minutes of Meeting of April 8, 2013. Available:
http://www.townofsharon.net/node/2013/minutes (accessed February 25, 2015).

n Wakefield, Massachusetts, Board of Health Regulations. Regulation of the Wakefield Board
of Health restricting the sale of tobacco and nicotine delivery devices (March 19, 2014).
o Wellesley, Massachusetts, Board of Health. Minutes of Meeting of September 12, 2013.
Available: http://www.wellesleyma.gov/Pages/WellesleyMA_HealthMin/2013 (accessed Febru-

ary 25, 2015).
p New York City Administrative Code. Section 17-706a-c, as amended by Local Law 69-2009
(November 19, 2013).

q A local law to raise the legal age for the sale of tobacco products in Suffolk County. Resolu-
tion No. 1039-2014. Suffolk County Legislation (January 2, 2014).

APPENDIX A 275
TABLE A-3 States and Localities Currently Considering Proposals to

Raise the Minimum Age of Legal Access to Tobacco Products to 19
Jurisdiction
(state/county/ Year of Regulation/
town) Proposal Legislation Status Details of Law
Massachusetts—Select Town
Franklina 2013 Town board Proposal The board of health
(Tota, 2014) of health discussed and held a public hearing
regulation currently open on new proposed
for public tobacco regulations,
comment. including raising the
MLA from 18 to 19
years.
Missouri—Select City
Columbia 2014 Legislation Under review Proposal raises the
(Denney, by the city’s MLA to 21 and also
2014) board of health bans e-cigarette use
and substance indoors.
abuse advisory
commission.
New York—Select County
Westchester 2014 Legislation Proposed New law would
(Ganga, 2014) and under raise the MLA for
consideration “cigarettes and
by the county tobacco-related
board of products” from 18
legislators. to 19.
a Town of Franklin, Massachusetts, Code. Chapter 262: Regulation affecting smoking and
the sale and distribution of tobacco and nicotine delivery products in the town of Franklin
(March 19, 2014).

TABLE A-4 States and Localities Currently Considering Proposals to

Raise the Minimum Age of Legal Access to Tobacco Products to 21
Jurisdiction
town) Proposal Legislation Status Details of the Law
Hawaii—Select City
Honolulu 2014 Legislation Bill advanced Bill would make it illegal
(Sadoy, 2014) by the for people under age
Honolulu 21 to purchase tobacco
city council’s or electronic smoking
Public Safety devices.
and Economic
Development
Committee.
Andover 2014 Town board Board of Proposed regulations
(Lima, 2014b) of health health expected include raising the MLA
regulation to vote on to 21 and requiring
proposed tobacco retailers to
regulations on include at least two
November 17, smoking cessation
2014 (Lima, products on half of a sales
2014a). display featuring tobacco
and nicotine products.
Lawrence 2014 Town board Public hearing New antismoking
(Tennant, of health scheduled for regulations include several
2014) regulation November 18, sales restrictions (e.g.,
2014. permitting restrictions,
minimum pack size and
bans on other tobacco
products, etc.) as well
as raising the MLA to
purchase tobacco to
age 21.
Saugus 2014 Town board Public hearing Proposed amendment to
(Gaffney, of health scheduled for town tobacco regulations
2014) regulation November 3, includes raising the
2014. MLA to purchase
tobacco to age 21 and
banning flavored tobacco
products.

APPENDIX A 277
TABLE A-4 Continued

Jurisdiction
town) Proposal Legislation Status Details of the Law
New Jerseya 2014 Legislation Passed by Amendment of existing
(Wilson, 2014) State Senate, laws, raising the MLA
awaiting House for purchase and sale of
vote in 2014 tobacco and electronic
(Mickle, 2014). smoking devices from 19
to 21.
Washington 2014 Legislation King County Resolution calls for
(Rhodan, 2014) (covers Seattle) Washington State
Alcoholism legislators to change
and Substance the legal age of tobacco
Abuse purchase to 21 across the
Administrative state.
Board approved
resolution, for
consideration
by Washington
State legislators.
a Raises minimum age for purchase and sale of tobacco products and electronic smoking
devices from 19 to 21. New Jersey Senate Bill 602. State of New Jersey 216th Legislature,
passed Senate June 30, 2014. Received in the Assembly July 11, 2014; referred to Health and
Senior Services Committee; New Jersey Assembly Bill 3254.

TABLE A-5 States and Localities That Have Considered But Not Enacted
Proposals to Raise the Minimum Age of Legal Access to Tobacco Products
to 19
Jurisdiction
town) Proposal Legislation Status Details of Law
Illinoisa 2001 Legislation Proposed. Not Proposed to amend
enacted. existing law to raise
the MLA for tobacco,
tobacco products, and
other smoking herbs from
18 to 19 years.
Maineb 2003 Legislation Proposed. Not Proposed to amend
enacted. existing law to raise the
MLA from 18 to 19 years.
Massachusettsc 2005 Legislation Proposed. Not Proposed to prohibit
enacted. the sale and possession
of tobacco products to
persons under 19 years
of age.
New Yorkd,e 2005 Legislation Proposed. Not Proposed raising the MLA
enacted. from 18 to 19. Proposed
to prohibit the sales of
tobacco products and
herbal cigarettes to any
individual under the age
of 19.
North Dakotaf,g 2005 Legislation Failed to pass. Proposed to amend the
North Dakota Century
Code to raise the MLA
for sales, purchase,
possession, and use of
tobacco to 18, and to
provide a penalty.
a An act concerning tobacco. HB1034, 92nd Illinois General Assembly Legislation, 2001-2002
regular sess. (2001).
b An act to increase the legal age for the purchase of tobacco products, Title 22, Chapter 262-A
§ 1555-B, § 1557, § 1558, and § 1559, Maine Revised Statutes 121st Legislature, 1st reg. sess.
c HB 1824, Part IV, Title I, Chapter 270, § 6, Commonwealth of Massachusetts, presented
by Rep Jones (2005).

d Regulation of Tobacco Products and Herbal Cigarettes; Distribution to minors (also known
as “Adolescent Tobacco Use Prevention Act”), Article 13-F, § 1399-aa–§ 1399-ee, New York
State Public Health Laws.
e An act to amend the public health law and the penal law, in relation to increasing the pur-
chasing age for tobacco products from eighteen to nineteen. Bill No. A5883-A, New York
State Public Health Laws, reg. sess. (March 2, 2005).

APPENDIX A 279
TABLE A-5 Continued

f Miscellaneous Offenses, Title 12.1-31-03, Criminal Code, North Dakota Century Code,
North Dakota Legislative Branch.
g A bill for an Act to amend and reenact section 12.1-31-03 of the North Dakota Century
Code, relating to the sale of tobacco to individuals under the age of nineteen and the use
of tobacco by minors; to provide a penalty; and to provide for application. HB 1183, 59th
Legislative Assembly of North Dakota, introduced by Rep. DeKrey (January 7, 2005, date of
last action on bill).

TABLE A-6 States and Localities That Have Considered But Not
Enacted Proposals to Raise the Minimum Age of Legal Access to Tobacco
Products to 21
Jurisdiction
town) Law Legislation Status of Law Details of the Law
Californiaa 2003 Legislation Proposed. Not Proposed to amend
enacted. existing Code to raise the
MLA from 18 to 21.
Coloradob 2014 Legislation House Finance Amends existing law to
(Lee, 2014) Committee raise the MLA from 18
voted 7 to 6 to 21 for sales and other
to reject the forms of distribution.
bill, March 19, Would grandfather in
2014. persons born on or before
June 30, 1996 (currently
18).
Connecticutc,d,e 2003 Legislation Proposed. Not Proposed that the general
enacted. statutes be amended to
raise the legal age for
use of tobacco products
to 21.
District of 2013 Legislation Introduced. The bill would raise
Columbiaf Failed to prohibition of sales of
make it out of tobacco products to
committee. minors by redefining
minors from age 18 to
21. Prohibits licenses to
operate vending machines
selling tobacco products
for establishments that
admit individuals under
age 21.
Marylandg 2014 Legislation Introduced. The bill raises prohibition
Unfavorable of sales of tobacco
report by products to minors and
judiciary restricting minors from
committee. purchasing or possessing
tobacco by redefining
minors from age 18 to 21.

APPENDIX A 281
TABLE A-6 Continued

Jurisdiction
Massachusetts—
Select Towns
Cohasset 2014 Town board Proposed. Not Board approved a ban
(Dale, of health enacted. on selling tobacco
2014a,b) regulation products at pharmacies
and restricting the sale of
e-cigarettes, but kept the
MLA at 18. The proposal
would have raised the
MLA from the state
minimum of 18 to 21.
Hopkinton 2013 Town board Rejected. The Proposal would amend
(Krantz, 2013) of health board of health regulations to raise the
regulation discussed MLA from 18 to 21.
holding a town
meeting for
public vote.
New York Stateh,i 2005 Legislation Proposed. Not Proposed raising the MLA
enacted. from 18 to 19 to 20 to 21
over time.
Amherst 2014 Legislation Proposed. Not City council dropped
(Habuda, enacted. provision to raise the
2014; Rey, MLA from 18 to 21 in
2014) final approved resolution
to prevent and reduce
underage smoking.
Nassau 2014 Legislation Introduced into Proposed raising the MLA
(Brodsky, Nassau County from 19 to 21, in line
2014) legislature, with neighboring Suffolk
and blocked County and New York
from a vote City.
by Republican
county
legislators.
Oregonj 2013 Legislation Introduced. Proposed new laws and
Failed to amendments to make
make it out of it a crime to distribute,
committee. sell, or cause to be sold
tobacco in any form to an
individual under the age
of 21.
continued

TABLE A-6 Continued

Jurisdiction
South 2003 Legislation Introduced. Proposed amending the
Carolinak,l,m Failed to Code of Laws of South
make it out of Carolina raising the MLA
committee. from 18 to 21.
Texasn 2013 Legislation Proposed. Proposed raising the MLA
Not enacted. from 18 to 21.
Similar bill also
introduced in
2003.
Utaho 2014 Legislation State senate Proposes raising the
voted against MLA from 19 to 21.
the measure on Prohibits sales of tobacco
March 3, 2014. and tobacco products to
individuals under 21 years
of age. Prohibits business
owners from allowing
individuals under age 21
from entering businesses
while the underage person
is using tobacco.
Vermontp 2005 Legislation Introduced. Proposed raising the MLA
Failed to from 18 to 21.
make it out of
committee.
a An act to amend Sections 17537.3, 22952, 22956, 22958, and 22963 of, and to add Section
22964 to, the Business and Professions Code, and to amend Section 308 of the Penal Code,
relating to tobacco, SB1821, California General Assembly, 2003-04 reg. sess., introduced by
Sen. Dunn (February 20, 2004).
b A bill for an act concerning the prohibition of tobacco transactions for persons under
twenty-one years of age, HB14-1263, 69th General Assembly, State of Colorado, 2nd reg.
sess. (March 19, 2014).
c General Statutes of Connecticut. Vol. 4, Title 12: Taxation, Chapter 214 § 12-295 (a, b, c, d
& e), Vol. 13, Title 53: Crimes, Chapter 946 § 53-344 (revised to January 13, 2013).
d An Act Raising the Legal Age for Use of Tobacco Products. Proposed SB 769, LCO No.
2394, Connecticut General Assembly, January sess. (January 27, 2003).

e An Act Prohibiting the Possession of Tobacco by Minors. Proposed HB 5035, LCO No.
2850, Connecticut General Assembly, January sess. (February 6, 2003).

f Prohibition Against Selling Tobacco Products to Individuals Under 21 Amendment Act of
2013. B20-0567, 20th Council of the District of Columbia, 23rd sess. (November 5, 2013).
g An Act Concerning Criminal Law—Tobacco Products—Minimum Age. HB 278, Maryland
General Assembly, Department of Legislative Services, Regular sess. (February 17, 2014).
h New York State Regulation of Tobacco Products, Herbal Cigarettes and Smoking Parapher-
nalia; Distribution to Minors. Article 13-F, § 1399-aa–§ 1399-ee. 2012.

APPENDIX A 283
TABLE A-6 Continued

i Raises the minimum age for the sale of tobacco and tobacco products from 18 years of age
to 19 years of age to 20 years of age to 21 years of age over time. Bill no. S05301, New York
State Assembly (2005).
j A bill for an act relating to tobacco; creating new provisions; and amending ORS 163.575,
165.800, 165.813, 167.400, 167.401, 167.402, 167.404, 167.407, 323.718, 339.883,

431.840, 431.853, 433.835 and 807.500. HB 2974, 75th Oregon Legislative Assembly, 2009
regular sess. (June 29, 2009).
k South Carolina Code of Laws, Title 16 Crimes and Offenses, Chapter 17 Offenses Against
Public Policy § 16-17-500-504 (eff. June 7, 2013).

l HB A35, R67, H3538, 120th sess., South Carolina General Assembly (signed June 7, 2013).
m HB 3084, general bill, South Carolina General Assembly, 115th sess., sponsored by Rep.
Talley and others (2003–2004).

n A bill to be entitled an Act relating to the distribution, possession, purchase, consumption,
and receipt of tobacco products; providing penalties. SB No. 313, 83rd Texas State Senate,
83rd sess. (2013).
o Ban on Smoking Paraphernalia to Minors. HB 206, Section 1 § 76-10-104.1, 56th Utah
House of Representatives, 2010 general sess. (2010).

p
An Act Relating to Increasing the Legal Smoking Age. H.0105, General Assembly of the
State of Vermont, 2005–2006 legislative sess. (2005).

REFERENCES
Aicardi, R. 2014. Braintree tobacco purchase age increasing to 18. The Braintree Forum
(June 25, 2014). http://braintree.wickedlocal.com/article/20140625/News/140627895
Alexander, K., and K. Williams. 2014. No tobacco purchases until age 21? Healdsburg
passes ban. SFGate (October 22, 2014). http://www.sfgate.com/bayarea/article/In-a-first-
Healdsburg-bans-under-21-tobacco-sales-5837820.php (accessed January 13, 2015).
Allen, S. 2014. Westford board votes to ban cigarette sales to those younger than 21. Low-
ell Sun (April 15, 2014). http://www.lowellsun.com/local/ci_25569105/westford-board-
votes-ban-cigarette-sales-those-younger (accessed January 13, 2015).
Bartlett, J. 2014a. Scituate increases tobacco-purchasing age to 21. Boston.com (February
25, 2014). http://www.boston.com/yourtown/news/scituate/2014/02/scituate_increases_
tobacco_purchasing_age_to_21.html (accessed January 13, 2015).
———. 2014b. Two more towns hike cigarette age. The Boston Globe (March 2, 2014). http://
www.bostonglobe.com/metro/regionals/south/2014/03/02/scituate-and-hudson-raise-
tobacco-buying-age/nwLUD2x0rv7w1Tp2Iic7PJ/story.html (accessed January 13, 2015).
Blakley, D. 2014. Evanston becomes first Illinois city to raise cigarette-buying age to 21. CBS
Chicago-Local. http://chicago.cbslocal.com/2014/11/07/evanston-becomes-first-illinois-
city-to-raise-cigarette-buying-age-to-21 (accessed February 24, 2015).
Brodsky, R. 2014. Nassau GOP to block vote on bill raising age limit to buy tobacco products.
Newsday (March 30, 2014). http://www.newsday.com/long-island/nassau/nassau-gop-to-
block-vote-on-bill-raising-age-limit-to-buy-tobacco-products-1.7551518 (accessed Janu-
ary 14, 2015).
City of Malden. 2014. City of Malden board of health raises minimum age to buy cigarettes to
age 21. http://www.cityofmalden.org/content/city-malden-board-health-raises-minimum-
age-buy-cigarettes-age-21 (accessed February 24, 2015).
Comeau, Z. 2014. Medway selectmen approve tobacco age ballot question, talk region-
alization. Milford Daily News (March 18, 2014). http://www.milforddailynews.com/
article/20140318/News/140316629 (accessed January 13, 2015).
Dale, E. 2014a. Cohasset health board keeps age for tobacco purchase at 18. The
Cohasset Mariner (June 12, 2014). http://cohasset.wickedlocal.com/article/20140612/
NEWS/140618682/0/SEARCH (accessed January 14, 2015).
———. 2014b. Should tobacco age be 21? Patriot Ledger (March 14, 2014). http://www.
patriotledger.com/article/20140314/News/140317981 (accessed January 14, 2015).
Denney, A. 2014. Chadwick proposes increasing legal age to buy tobacco products. C olumbia
Daily Tribune (September 16, 2014). http://www.columbiatribune.com/news/local/
chadwick-proposes-increasing-legal-age-to-buy-tobacco-products/article_e77b76c4-8c57-
5eef-ab9d-f2d531dcba0f.html (accessed January 14, 2015).
Donga, E. 2013. Canton raises age to buy tobacco products, doesn’t ban e-cigarettes. The
Patriot Ledger (August 13, 2014). http://www.patriotledger.com/news/x273448480/
Canton-raises-legal-age-to-buy-cigarettes (accessed January 13, 2015).
Eisenstadter, D. 2014. Belmont raises age for purchasing tobacco to 21. The Belmont Citizen-
Herald. http://belmont.wickedlocal.com/article/20140918/News/140916159 (accessed
February 4, 2015).
Gaffney, M. 2014. Board considers raising tobacco purchase age to 21 in Saugus. The
Saugus Advertiser (October 16, 2014). http://saugus.wickedlocal.com/article/20141016/
News/141018204 (accessed January 14, 2015).
Ganga, E. 2014. Westchester legislators tout bill to increase age to buy tobacco. LoHud
(April 2, 2014). http://polhudson.lohudblogs.com/2014/04/02/westchester-legislators-
tout-bill-increase-age-buy-tobacco (accessed January 14, 2015).

APPENDIX A 285
Habuda, J. 2014. Amherst OKs measure to limit youth smoking. The Buffalo News (August 4,
2014). http://www.buffalonews.com/city-region/amherst/amherst-oks-measure-to-limit-
youth-smoking-20140804 (accessed January 14, 2015).
Hanson, F. 2014. Hull raising age to buy tobacco products to 21. The Patriot Ledger (October
10, 2014). http://www.patriotledger.com/article/20141001/News/141009750 (accessed
January 13, 2015).
Hendrickson, D. 2013. Newburyport lightens up on new tobacco regs. Newburyport News
(December 21, 2013). http://www.newburyportnews.com/news/local_news/newburyport-
lightens-up-on-new-tobacco-regs/article_f63ce3ec-93f6-5a8a-b3d4-166ada84e8e3.html
Krantz, L. 2013. Hopkinton board rejects tobacco age increase. MetroWest Daily (December
7, 2013). http://www.metrowestdailynews.com/x1275649067/Hopkinton-board-rejects-
tobacco-age-increase (accessed January 14, 2015).
Lee, K. 2014. Raising age to 21 for purchase of cigarettes in Colorado rejected. Denver
Post (March 19, 2014). http://www.denverpost.com/news/ci_25377056/raising-age-21-
purchase-cigarettes-colorado-tough-climb#axzz2wRfaI6yy (accessed January 14, 2015).
Lima, T. 2014a. Retailers challenge tobacco changes. Eagle-Tribune (October 21, 2014).
http://www.eagletribune.com/news/retailers-challenge-tobacco-changes/article_52fd512c-
5038-591a-8002-b15975f7e3cf.html (accessed January 14, 2015).
———. 2014b. Retailers challenge tobacco changes; argue higher age will hurt business,
spark bootlegging. The Andover Townsman Online (October 23, 2014). http://www.
andovertownsman.com/news/local_news/retailers-challenge-tobacco-changes-argue-
higher-age-will-hurt-business/article_05b7e327-9743-578b-a26e-47faa7a68e6e.html
Loyer, S. 2014. Sayreville bans the sale of tobacco to anyone under 21. http://www.
mycentraljersey.com/story/news/local/middlesex-county/2014/09/12/sayreville-
smoking-ban-teens/15535341/;%20http:/halfwheel.com/sayreville-n-j-raises-tobacco-
purchase-age-21/69995 (accessed January 13, 2015).
Malachowski, J. 2014. Hudson raises age to buy cigarettes to 21. MetroWest Daily
News (February 25, 2014). http://www.metrowestdailynews.com/article/20140226/
News/140227597 (accessed January 13, 2015).
Mason, C. 2013. Healdsburg: Ban tobacco sales to those under 21. The Press Demo-
crat (December 5, 2013). http://www.pressdemocrat.com/article/20131205/
articles/131209729#page=0 (accessed January 13, 2015).
McLean, D. 2014. Winchester BoH raises tobacco sales age to 21. Winchester Star (March 19,
2014). http://winchester.wickedlocal.com/article/20140319/News/140315825 (accessed
January 13, 2015).
Mickle, T. 2014. More cities raise tobacco age to 21. The Wall Street Journal (October 28,
2014). http://www.wsj.com/articles/more-cities-raise-tobacco-age-to-21-1414526579 (ac-
cessed January 14, 2015).
Noda, S. 2014. Englewood raises age limit to 21 to buy tobacco products. Northern Valley
Suburbanite (August 7, 2014). http://www.northjersey.com/news/city-raises-age-limit-for-
cigarettes-1.1063714 (accessed January 13, 2015).
Parker, B. 2013. Law raising age to buy cigarettes to 19 takes effect in Brookline. Boston
Globe (June 3, 2013). http://www.boston.com/yourtown/news/brookline/2013/06/law_
raising_age_to_buy_cigaret.html (accessed January 13, 2015).
Perez, C. 2014. Englewood board of health backs smoke-free policy. Northern Valley Sub-
urbanite (April 11, 2014). http://www.northjersey.com/news/health-board-backs-smoke-
free-policy-1.917641?page=all (accessed January 13, 2015).

Quinn, C. 2014. Newburyport latest Massachusetts community to raise tobacco-buying age.

Mass Live (April 3, 2013). http://www.masslive.com/news/boston/index.ssf/2014/04/
newburyport_latest_massachuset.html (accessed January 13, 2015).
Rey, J. 2014. Amherst to take up stricter tobacco laws. The Buffalo News (August 1, 2014).
http://www.buffalonews.com/city-region/amherst/amherst-to-take-up-stricter-tobacco-
laws-20140801 (accessed January 14, 2015).
Rhodan, M. 2014. New push to raise tobacco age in Washington state. Time. http://time.
com/74400/tobacco-age-21-washington-state-proposal (accessed February 23, 2015).
Sacco, J. 2014. Melrose raises purchasing age to 21 for tobacco, nicotine products.
Melrose Free Press (April 24, 2014). http://melrose.wickedlocal.com/article/20140424/
NEWS/140428078/12448/LIFESTYLE (accessed January 13, 2015).
Sadoy, A. 2014. Should the legal age to buy tobacco products on O’ahu be raised to 21. Ka
Leo (September 23, 2014). http://www.kaleo.org/opinion/do-you-think-it-s-fair-that-the-
legal-age/article_5e86778e-435a-11e4-9f0f-0017a43b2370.html (accessed January 14,
2015).
Schwartz, D. 2014. Suffolk bill raising age to purchase tobacco to 21 becomes law. Newsday
(April 14, 2014). http://www.newsday.com/long-island/suffolk/suffolk-bill-raising-age-to-
purchase-tobacco-to-21-becomes-law-1.7703911 (accessed January 13, 2015).
Tennant, P. 2014. Beginning March 1, tobacco buyers must be 21 in Lawrence. The Eagle-
Tribune (December 17, 2014). http://www.eagletribune.com/news/beginning-march-
tobacco-buyers-must-be-in-lawrence/article_203d53e7-cf32-5c17-a175-6f019781e8cd.
html (accessed February 23, 2015).
Tota, M. 2014. Tobacco purchase age still 18. Milford Daily News (January 8, 2014). http://
www.milforddailynews.com/article/20140107/NEWS/140109804/0/SEARCH (accessed
January 14, 2015).
Turner, J. 2013. BOH approves landmark tobacco regulations. The Canton Citizen
(August 15, 2013). http://www.thecantoncitizen.com/2013/08/15/board-of-health-
approves-landmark-tobacco-regulations (accessed January 13, 2015).
Wade, C. M. 2013. Proposal on teen smoking decried. Boston Globe (December 1, 2013).
http://www.bostonglobe.com/metro/regionals/north/2013/12/01/board-plan-increase-
tobacco-buying-age-newburyport-riles-store-owners-mayor/9IcbV2ra3L5nCi06jILJJK/
story.html (accessed January 13, 2015).
Wagner, S. L. 2014. Raising legal age in Wayland for tobacco sales. Wayland Town Crier
(October 1, 2014). http://wayland.wickedlocal.com/article/20141001/News/141009775
Wilson, R. 2014. New Jersey smoking age could be raised to 21. The Washington Post (July
1, 2014). http://www.washingtonpost.com/blogs/govbeat/wp/2014/07/01/new-jersey-
smoking-age-could-be-raised-to-21 (accessed February 23, 2015).

Appendix B
State Laws—
Tobacco Transfers to Minors
T
he following table (see Table B-1) summarizes state laws for all
50 states and the District of Columbia specifically in reference to the
transfer of a tobacco product to a minor by both commercial and
noncommercial sources.
The information for this table is adapted from the State Legislated
Actions on Tobacco Issues (SLATI) database,1 which is maintained by the
American Lung Association. It should not be considered a comprehensive
analysis of state law but rather an illustration of state-level variance in
tobacco control legislation.
1 American Lung Association. State Legislated Legal Actions on Tobacco Issues (SLATI)
State Pages. http://www.lungusa2.org/slati/about.php (accessed October 8, 2014).
287

TABLE B-1 State Laws—Tobacco Transfers to Minors
Applies to Explicitly Classification

State Whom Illegal to… Product in Question of the Transfer
Alabama Any person Sell, barter, Cigarettes, cigarette Not specified*
exchange, or tobacco, cigarette
give away* paper or substitute
for either of them*
Permit holder,
member,
employee,
officer
Alaska Any person Sell, exchange, Cigarette, cigar, or Violation
(non licensee) or give* tobacco product; a
product containing
nicotine*
Licensee Not specified
Arizona Any person Knowingly sell, Tobacco product, Petty offense

give, or furnish vapor product
including
e-cigarettes, or any
instrument designed
for smoking/
ingestion of tobacco
Arkansas Any person Sell, give, or Tobacco products Not specified
barter* or cigarette papers;
alternative nicotine
product, cartridge or
component of such;
e-cigarette*
Licensee Violation
* Applies to all tobacco transfers in the given state.

APPENDIX B 289
Increase for
Penalty for Subsequent Suspend Employee or Affirmative
First Offense Offenses? License? Licensee Punished? Defense?
Fine, 10<x<50 Not specified Not Not specified* Not specified*
and may also be specified
imprisoned for
<30 days
Fine Yes Yes
Fine >300 Not specified Not Not specified* Not specified*

specified
Suspend license + Yes Yes

civil penalty
Not specified Not specified Not Not specified Not specified
specified
Fine* Not specified Not Employee of Not specified

specified retail permittee in
violation is subject
to fine <100
in addition to
business owner’s
fine*
Yes Yes Yes

continued

TABLE B-1 Continued

California Person, firm, Knowingly sell, Tobacco products Misdemeanor
corporation give, or furnish or paraphernalia or civil action
including blunt
wraps; e-cigarettes
Colorado Any person Sell, distribute, Cigarettes or Class 2 petty

or offer for sale tobacco products offense
including
e-cigarettes, cigars,
Retailer Sell or permit Not specified
cigarillos and pipes*
the sale of
Connecticut Any person Sell, give, or Tobacco Not specified*
deliver to*
Dealer or Cigarettes or
distributor tobacco products
Delaware Any person; Sell or Tobacco product Not specified

excluding distribute,
parent/ purchase on
guardian behalf of

APPENDIX B 291
Increase for
Fine Yes Not California Yes
specified Department of
Health Services
may assess civil
penalties against a
the owner/licensee
in addition to
the criminal/civil
penalties against
an individual.
An employee
against whom
civil penalties are
sought cannot
additionally have
criminal penalties
Fine Not specified Not Not specified* Yes*
specified*
Written warning Yes
Fine <200* Yes Not The dealer/ Yes*

specified distributor is
assessed a penalty.
Yes— Yes
The employee
retailer and
who perfomed the
employee
transaction may
also be fined*
Fine Yes Yes Licensee/owner Yes: affirmative
is responsible for defense for
fine, employee licensee if can
may ALSO be prove that
charged purchaser showed
valid or seemingly
valid proof of
age, affirmative
defense for
retailer/employer
if can prove that
policies were in
place to prevent
illegal sales

continued

TABLE B-1 Continued

District of Any person Sell, give or Tobacco product Misdemeanor
Columbia furnish
Florida Any person Sell, deliver, Tobacco product Misdemeanor

barter, or (2nd degree)
furnish,
directly or
indirectly
Georgia Any person Knowingly Cigarettes or Misdemeanor

sell or barter, tobacco related
directly or object, including
indirectly, to cigar wraps
advise, counsel,
or compel
any minor to
smoke, inhale
chew, or use
cigarettes or
tobacco-related
objects
Hawaii Any person Sell or furnish Tobacco, including Not specified
chewing tobacco
and snuff, and
electronic smoking
device
Idaho Non-permittee Sell, distribute, Tobacco products or Not specified*
or offer e-cigarettes
Permitee Sell or Tobacco products

distribute

APPENDIX B 293
Increase for
Fine 100<x<500 Yes Yes Not specified Not specified
and/or
imprisonment
<30 days
Fine Yes Yes May mitigate Yes
penalties against a
dealer if employee
performed illegal
sale and dealer
had provided
adequate training
beforehand
specified
Fine Yes Not Retail clerks, and Not specified

specified not the owners or
licensees are cited
for violations
Fine Not specified Not Penalty appears to Yes*

specified be for permittee
ONLY
Warning Yes Yes

continued

TABLE B-1 Continued

Illinoisa Any person Sell, buy for, Tobacco products Petty offense
distribute
samples of, or
furnish
Knowingly sell, Cigarette papers Class C

deliver, or give or other tobacco misdemeanor
away accessories
Wrapping paper or Petty offense
leaf for rolling
Indiana Any person Knowingly sell Tobacco products, Class C
or distribute, including dissolvable infraction
purchase for tobacco products
delivery to a and e-cigarettes*
minor
Retailer Sell or Not specified

distribute
Iowa Any person Sell, give, or Tobacco products* Simple
otherwise misdemeanor
supply*
Retailer/ Not specified

employee of
retailer
a In Illinois, “classification of transfer” and “penalty for first offense” vary based on the
product and the illegal action (e.g., knowingly selling a pack of cigarette papers is a Class C
misdemeanor while knowingly selling a leaf for rolling is a petty offense).

APPENDIX B 295
Increase for
Fine 200 Yes Not Not specified* Not specified*
specified*
Not specified Not specified
Fine 100<x<1,000 Not specified
Fine <500 Not specified Yes* Not specified* Yes*
Fine 200<x<1,000 Yes
Not specified Not specified Yes* If an employee Not specified*

of the retailer
commits the
violation, the
retailer is not
charged if the
employee took
the proper
state tobacco
compliance
training.
Otherwise, penalty
to the retailer
Fine + potential Yes
additional civil
penalty

continued

TABLE B-1 Continued

Kansas Any person Sell, furnish, or Cigarettes, Class B
distribute e-cigarettes, or misdemeanor
tobacco products
Kentucky Any person Sell or cause to Tobacco products Not specified

be sold, solicit
a minor to
purchase
Louisiana Any person Sell or Tobacco products* Not specified*
distribute*
Permittee
(manufacturer,
distributor,
dealer, retailer)
Maine Any person Sell, furnish, Tobacco products Civil violation

give away, or
offer to sell,
furnish or give
away

APPENDIX B 297
Increase for
Fine >200 Not specified Yes The person who Yes
violates the law
is the individual
directly selling the
tobacco product.
A licensee can be
assessed additional
civil penalties for
selling to minors
Fine 100<x<500 Not specified Not Not specified Not specified
specified
Fine <50 Yes* Yes* Sale of tobacco Not specified*

products to a
minor by a retail
dealer’s employee
is considered an
Subject to
act of the retailer
suspension or
for the purpose
revocation of
of suspension
permit and/or
or revocation of
civil penalties
a permit or the
assessment of civil
penalties, unless
employee attends
state approved
training program*
Fine 50<x<1,500 Not specified Yes The employer of Yes
+ court costs the person who
violates the law
also commits a
civil violation
which can be fined
50<x<1,500 +
court costs. The
district court can
also impose fines
listed above or
suspend/revoke
licenses for
violation of sales
laws
continued

TABLE B-1 Continued

Marylandb Retailer Distribute Tobacco products or Not specified
paraphernalia
Any person Purchase for, Tobacco products or Not specified

deliver to, sell paraphernalia
Sell, distribute, Electronic device Misdemeanor

or offer for sale that can be
used to deliver
nicotine, including
e-cigarettes, cigars,
cigarillos, and pipes
Massachusettsc Any person Sell or give Tobacco in any form Not specified*
excluding
parent/
guardian
Sell Rolling papers
Michigan Any person Sell or furnish Tobacco products Misdemeanor
excluding
parent/
guardian
Minnesota Any person Sell or furnish Tobacco or tobacco- Misdemeanor
related devices;
product containing
or delivering
nicotine or lobelia
intended for
consumption that is
not tobacco
b In Maryland, the law applying to “any person” applies different penalties according to the
type of item and transfer. For example, the penalty for purchasing tobacco products on behalf
of a minor is a fine not to exceed $300, while the misdemeanor offense of selling an e-cigarette
to a minor is punishable by a fine of up to $1,000 per violation.
c In Massachusetts, the penalty for first offense depends on the product in question. For ex-
ample, the sale of tobacco to a minor is punishable by a fine no less than $100, while the sale
of rolling papers is punishable by a fine of no less than $25.

APPENDIX B 299
Increase for
Not specified Not specified Not Not specified* Yes*
specified*
Fine <300 Yes
Fine <1,000 per Not specified

violation
Fine >100 Yes* Not Not specified* Not specified*

specified*
Fine >25
Fine <50 Not specified Not Not specified Yes
specified
Not specified Escalation in Yes If a licensee or Yes

criminality an employee of a
with licensee violates
subsequent the law, the
offenses licensee is charged
an administrative
penalty. The
individual must
also be charged
an administrative
penalty. Penalty
can escalate with
subesequent
offenses

continued

TABLE B-1 Continued

Mississippi Any person or Sell, barter, Tobacco products or Not specified*
retailer deliver, or give rolling papers
Directly or Alternative
indirectly nicotine product,
(by agent, any cartridge or
employee, component of an
or vending alternative nicotine
machine) sell, product
offer for sale,
give, or furnish
Missouri Any person Sell, provide, Tobacco products or Not specified

(excluding or distribute rolling papers
family
members
on private
property)
Montana Any person Sell or Tobacco products Not specified

distribute

APPENDIX B 301
Increase for
Fine 50* Yes* Yes* The permittee will Yes
be sent a warning
letter for the first
violation, and
required to enroll
in and complete
a state tobacco
retailer education
Not specified
program. If
the retailer
has directed
employees to sign
an agreement
stating that
they understand
the state laws
regarding youth
tobacco sales*
Fine 100 Yes Yes Owner of Not specified
establishment
issued a reprimand
in addition to
penalties listed.
Exempt from
above penalties
if they have
an employee
compliance
program in place
Verbal Yes Yes Employee pays Not specified
notification $25 per violation
if not licensee

continued

TABLE B-1 Continued

Nebraska Any person Sell or furnish Tobacco products* Class III
misdemeanor*
Licensee Sell, give, or

furnish in
any way, or
allow to be
taken from
their place of
business
Nevada Any person Sell, distribute, Tobacco in any form Not specified
or offer to sell or cigarette papers
New Any person Sell, give, Tobacco products, Violation

Hampshire furnish, or e-cigarettes, or (2nd and on,
cause or allow liquid nicotine misdemeanor)
or procure to
be sold, given,
furnished
Sell, give, or Rolling papers Violation

furnish

APPENDIX B 303
Increase for
Not specified* Not Yes* Licensee is subject Not specified*
specified* to forfeiture. Any
officer, director,
or manager of
the business of
any corporation
that violates the
provision, if they
are aware, is
subject to the same
penalty*
Fine <500 + civil Not specified Not The owner of an Yes

penalty <500 specified establishment
is not held
responsible for
an employee
violation of the
law if they had
no knowledge of
the violation and
establish employee
education to
prevent future
violations
Civil penalty Yes* Yes* Not specified* Not specified
<250
<250

continued

TABLE B-1 Continued

New Jersey Any person Directly or Cigarettes, cigarette Petty
indirectly paper, tobacco in disorderly
(by agent, any form including persons
employee, smokeless, or any offense
or vending electronic smoking
machine) sell, device including
offer for sale, e-cigarettes, cigars,
distribute for cigarillos, pipes,
commercial or any cartridge
purpose at or component or
no cost or related product
minimal cost
with coupons
or rebate
offers, give or
furnish
New Mexico Any person Knowingly sell Tobacco products Misdemeanor
or offer to sell
New York Retailer Sell Tobacco products, Not specified

herbal cigarettes,
shisha, e-cigarettes,
or smoking
paraphernalia
North Any person Distribute, aid, Tobacco products, Class 2
Carolina assist, or abet including tobacco- misdemeanor
in distribution, derived products
purchase on or vapor products,
behalf of cigarette wrapping
papers
North Dakota Any person Sell or furnish, Tobacco products Infraction,
procure on criminal
behalf of misdemeanor
Ohio Manufacturer, Sell or Tobacco products 4th degree
distributor, otherwise misdemeanor
wholesaler distribute
or retailer,
or employee
thereof

APPENDIX B 305
Increase for
Civil penalty 250 Yes Yes The licensee Yes
is subject to
administrative
charges, although
the individual
responsible for the
sale is liable for
the penalty
Imprisonment Not specified Not Not specified Not specified

<year +/or fine specified
<1,000
Fine 300<x<1,000 Yes Yes Not specified Yes
Not specified Not specified Not Not specified Yes

specified

specified
Not specified Escalation in Not Not specified Yes

criminality specified
with
subsequent
offenses

continued

TABLE B-1 Continued

Oklahoma Any person Sell, give, or Tobacco products Not specified
furnish in any
manner
Oregon Any person Knowingly Tobacco product or “Endangering

distributes, device for the use of the welfare of
sells, or causes tobacco a child”/Class
to be sold A violation
Pennsylvania Any person Sell or furnish* Tobacco products* Summary
offense*
Retailer
Rhode Island Any person Sell, give, Tobacco in the form Not specified*
deliver of cigarettes, bidi
cigarettes, cigars,
little cigars (flavored
and unflavored),
blunt wraps,
cigarette rolling
papers, cigarillos,
tiparillos, pipe
tobacco, chewing
tobacco, or snuff
Licensee or Sell, distribute, Tobacco product

employee or deliver
thereof
South Carolina Any person Sell, furnish, Tobacco product or Misdemeanor
give, distribute, alternative nicotine
purchase for or product
provide

APPENDIX B 307
Increase for
Fine <100 Yes Yes If the sale is made Yes
by the employee
of the licensee,
the employee
shall be guilty of
the violation and
subject to the fine.
Each violaton by
an employee shall
be a violation
against the owner
for purposes of a
license suspension
Fine >100 Not specified Not Not specified Not specified
specified
Fine 100<x<250 Yes* Yes* Not specified* Not specified
Fine 100<x<500 Yes

Not specified Not specified Yes* The licensee is Not specified*
responsible for
all violations
that occur at
the location for
which the license
is issued. If courts
find the licensee
has taken adequate
measures to ensure
employees are not
performing illegal
transactions, courts
may chose to not
suspend license*
Fine 250 Yes
Fine 100<x<200 Yes Not Not specified Not specified

specified

continued

TABLE B-1 Continued

South Dakota Any person Knowingly Tobacco products Class Two
distribute, misdemeanor
purchase on
behalf of, give
Tennessee Any person Sell, distribute, Tobacco product or Not specified
or purchase e-cigarette
on behalf of.
Persuade,
entice, send, or
assist a minor
to purchase,
acquire, receive
or attempt
to purchase,
acquire or
receive
Texas Any person With criminal Cigarette or tobacco Class C

negligence, sell, product misdemeanor
give, or causes
to be sold or
given

APPENDIX B 309
Increase for
Not specified Not specified Not Not specified Yes
specified
Warning letter Yes Not Commissioner Yes

specified of Agriculture
is authorized to
assess the penalty
against any
person or persons
determined to be
responsible, in
whole or part,
for contributing
to or causing the
violation to occur,
including but not
limited to the
owner, manager,
or employee of the
store at which the
violation occurred
Not specified Not specified Not If the offense Yes
specified occurs in
connection
with a sale by
an employee
of a licensee,
the employee
is criminally
responsible
and subject to
prosecution

continued

TABLE B-1 Continued

Utah Any person Knowingly, Cigar, cigarette, Class C
intentionally, e-cigarette or misdemeanor*
recklessly, or tobacco in any
with criminal form, or tobacco
negligence paraphernalia
provides
Sell, offer for Clove cigarette

sale, give, or
furnish
Vermont Any person Sell or provide Tobacco products, Not specified
tobacco substitutes
(e-cigarette and
related products),
or related
paraphernalia
Virginia Any person Sell, distribute, Tobacco product Not specified
purchase for, including but not
or knowingly limited to cigarettes,
permit the cigars, bidis and
purchase by wrappings
Washington Any person Sell, give, Cigar, cigarette, Gross
(non licensee) permit to be cigarette paper or misdemeanor*
sold or given* wrapper, tobacco in
any form or a vapor
product*
Licensee
West Any person Sell, give or Tobacco product or Misdemeanor*
Virginia furnish, cause cigarette paper*
to be sold or
furnished*
Firm or
corporation

APPENDIX B 311
Increase for
Not specified* Escalation in Yes* If the licensee Not specified*
criminality or employee
with thereof has
subsequent sold tobacco to
offenses* someone <19 years
old, the agency
may impose
adminstrative
penalties on the
licensee*
Fine <100 Yes Yes Not specified Not specified
Fine <100 Yes Not Not specified Yes

specified
Fine 50 Yes* Yes* Not specified* Yes*
Fine 100
Fine <100 Yes* Not The employer Yes*
specified* may dismiss
an employee
for selling or
furnishing tobacco
Fine 50
products to
minors*

continued

TABLE B-1 Continued

Wisconsin Any person Purchase on Cigarettes, tobacco Not specified*
behalf of, or products or nicotine
provide to products*
Retailer, direct Sell or

marketer, provide for
manufacturer, nominal or no
distributor or consideration
agent/employee
thereof
Wyoming Any person Sell, offer for Tobacco products* Misdemeanor*
sale, give, or
deliver
Retailer Sell, permit the

sale of, offer
for sale, give,
or deliver

APPENDIX B 313
Increase for
Fine <500 if Yes* Yes* Not specified* Not specified
no violation
in previous 30
months or
Fine <500
+/or 1 month
imprisonment
if violation in
previous 30
months
Fine <500 Yes
Fine <50* Yes* Yes* No penalty to Yes*

the permittee for
first violation if
permittee can
prove plan in place
to prevent illegal
sale to minors*


Appendix C
State Laws—
Tobacco Purchase–Use–Possession
by Minors
T
he following table (see Table C-1) summarizes state law for all
50 states and the District of Columbia specifically in reference to
purchase–use–possession laws (PUP) of a tobacco product by a mi-
nor. Throughout this report, the minimum age of legal access to tobacco
products (MLA) focuses on youth access laws and enforcement policies
that curtail retail access to tobacco products by underage persons, with
little emphasis on punishing underage users of tobacco products. Despite
that focus, MLA in the following table, as in the report, therefore covers
youth access restrictions that both punish distributors of tobacco products
to underage users and the underage users themselves.
The information for this table is adapted from the State Legislated
Actions on Tobacco Issues (SLATI) database,1 which is maintained by the
American Lung Association. It should not be considered a comprehensive
analysis of state law but rather an illustration of state-level variance in
tobacco control legislation.
1 American Lung Association. State Legislated Legal Actions on Tobacco Issues (SLATI)
State Pages. http://www.lungusa2.org/slati/about.php (accessed October 8, 2014).
315

TABLE C-1 State Laws—Tobacco Purchase–Use–Possession Laws for

Minors
Penalties for PUP
Explicitly Illegal for Violations Among
State MLA Minors to… Classification Minors
Alabama 19 Purchase, use, Not listed Fine and notify parent/
possess, transport guardian
Alaska 19 Knowingly possess Violation Not listed
Arizona 18 Purchase, possess, Petty offense Fine or community
knowingly accept service
Arkansas 18 Use, possess, Not listed Confiscate the tobacco

purchase, or attempt product, may require
to purchase community service and
enrollment in tobacco
education program
California 18 Purchase, receive, Not listed Fine and community
possess service
Colorado 18 Purchase or attempt Class two Fine or participate
to purchase petty offense in tobacco education
program; may perform
community service
instead of fine
Possession Noncriminal Not listed
offense
Connecticut 18 Purchase Not listed Fine
Delaware 18 Purchase, accept Not listed Fine and community

receipt of, use a service
coupon for
District of 18 Purchase, attempt to Not listed Fine

Columbia purchase, possess,
attempt to possess
Florida 18 Knowingly possess, Not listed Fine or community
purchase, attempt to service; minor must
purchase attend anti-tobacco
education
Georgia 18 Purchase, attempt to Not listed Community service,
purchase, possess attendance at a lecture
on the hazards of
smoking or both

APPENDIX C 317
Illegal to Suspend
Present Fake or Driver’s
Graduated Penalty? Applies to… Borrowed ID License?
Not listed Tobacco, tobacco products, ü
alternative nicotine products
Not listed Tobacco products
Not listed Tobacco products, vapor product ü
including e-cigarettes, hookah,
waterpipe
Not listed Tobacco in any form, cigarette ü
papers
Not listed Tobacco product or

paraphernalia
Yes; increasing fines Tobacco products ü

only
Yes; more community Tobacco products ü
service hours for
second and subsequent
offenses
Yes; increasing fine for Cigarette or other tobacco ü
fake ID product
Yes Tobacco products ü ü
Not listed Cigarettes or tobacco related

objects
continued

TABLE C-1 Continued

Penalties for PUP
Hawaii 18 Purchase Not listed Fine
Idaho 18 Purchase, receive, Misdemeanor Imprisonment and/or a

sell, possess, use, fine; may also require
distribute, consume attendance at tobacco
awareness programs,
community service
Illinois 18 Purchase, possess Petty offense Fine and community
service
Indiana 18 Purchase, accepts Class C Fine
for personal use or Infraction
possess
Iowa 18 Smoke, use, possess, Violation Fine and community
purchase, or attempt service
to purchase
Kansas 18 Purchase, attempt to “Tobacco Fine
purchase; possess or infraction”
attempt to possess
Kentucky 18 Possess or use Not listed Product is confiscated
Purchase or accept Not listed Fine and community
receipt of service
Louisiana 18 Purchase or possess Not listed Fine

Maine 18 Present fraudulent Civil violation Fine and/or community
ID for the purposes service
of purchasing/
possessing/using
Maryland 18
Massachusetts 18
Michigan 18 Purchase, attempt to Misdemeanor Fine and community
purchase, possess, service, health
attempt to possess, promotion program
use participation

APPENDIX C 319
Illegal to Suspend
Yes, fine increases and Tobacco product or electronic
community service for smoking device
second and subsequent
offenses
Not listed Tobacco products or electronic ü
cigarettes
Yes Tobacco products ü
Not listed Tobacco products or electronic

cigarettes
Yes; higher fines and Tobacco products ü

more community
service
Not listed Cigarettes, e-cigarettes, or
tobacco products
Not listed Tobacco products ü

Yes; higher fines and Tobacco products
more community
service
Yes; increase in fine Tobacco products
Yes; increase in fine Tobacco products ü
and community service
Yes; increase in Tobacco product ü

number of hours; fine
stays the same
continued

TABLE C-1 Continued

Penalties for PUP
Minnesota 18 Use fraudulent ID to Misdemeanor If fake ID is used—
purchase, attempt to driver’s license is
purchase suspended
Possesses, smokes, Petty If fake ID is used—
chews, ingests, misdemeanor driver’s license is
purchases, attempts suspended
to purchase
Mississippi 18 Purchase, possess Not listed Fine and/or community
service
Missouri 18 Purchase, attempt to Infraction Confiscation of
purchase, possess product
Montana 18 Knowingly possess or Not listed Fine and/or community

use; purchase service/tobacco
cessation education
Nebraska 18 Use Class V Not listed
misdemeanor
Nevada 18
New 18 Purchase, attempt Violation Fine and/or community
Hampshire to purchase, use, service, maybe also
possess tobacco education
New Jersey 19
New Mexico 18 Procure or attempt to Not listed Fine or community
procure service
New York 18
North 18 Purchase, accept Class two Not listed
Carolina receipt, attempt to misdemeanor
purchase or accept
receipt
North Dakota 18 Purchase, possess, Noncriminal Fine
smoke, use offense
Ohio 18 Consume, possess, Not listed Fine and/or tobacco
purchase, attempt to education
purchase

APPENDIX C 321
Illegal to Suspend
Not listed Tobacco or tobacco related ü ü
devices
Not listed Tobacco or tobacco related

devices
Not listed Tobacco product ü
Second offense: Cigarettes or tobacco products ü

tobacco education/
smoking cessation
program; none else
listed
Yes; increased fines Tobacco products
listed
Not listed Tobacco product, e-cigarette, ü

liquid nicotine; rolling paper
Not listed Tobacco products including ü

tobacco derived products or
vapor products
Yes; increase fine, Tobacco products ü ü

require community
service, suspend
license/learner permit
continued

TABLE C-1 Continued

Penalties for PUP
Oklahoma 18 Purchase, receive, Not listed; Fine, imprisonment
possess if does not (only for the
provide seller misdemeanor)
when asked,
misdemeanor
Oregon 18 Purchase, attempt to Violation Tobacco education or
purchase or acquire community service
Possess Class D Not listed

violation
Pennsylvania 18 Purchase, attempt to Summary Fine and/or tobacco
purchase offense education/community
service/suspend license
Rhode Island 18 Purchase Not listed Not listed
Smoke, chew, possess Not listed Community service or

tobacco education
South 18 Purchase, attempt to Noncriminal Fine or tobacco
Carolina purchase, possess, offense education or
attempt to possess community service
South Dakota 18 Purchase, attempt Class Two Not listed
to purchase, receive, misdemeanor
attempt to receive,
possess, consume
Tennessee 18 Purchase, possess, Civil offense Fine
accept
Texas 18 Possess, purchase, Not listed Fine and tobacco

consume, accepts education or
community service
Utah 19 Purchase, attempt to Class C Fine and tobacco
purchase, possess misdemeanor education
Vermont 18 Possess, purchase, Not listed Confiscate product
attempt to purchase and fine
Virginia 18 Purchase, attempt to Not listed Fine or community
purchase, possess service

APPENDIX C 323
Illegal to Suspend
Increase in fine; may Tobacco products ü ü
suspend driver’s license
Yes; subsequent Tobacco products ü ü

offense = fine, suspend
license
Not listed Tobacco product ü ü
Not listed Tobacco in the form of cigarettes,

bidi cigarettes, cigars, pipe
tobacco, chewing tobacco, snuff
Not listed Tobacco in any form, cigarette
papers
Not listed Tobacco product, alternative ü ü
nicotine product
Yes; add tobacco Tobacco products ü

education and
community service for
subsequent offenses
Not listed Cigarette or other tobacco ü ü
product
Not listed Cigar, cigarette, e-cigarette, or ü

tobacco in any form
Not listed Tobacco products, tobacco ü ü
substitutes or paraphernalia
Yes, higher fine or Tobacco product
more community
service hours
continued

TABLE C-1 Continued

Penalties for PUP
Washington 18 Purchase, possess, Class Three Fine and/or community
attempt to purchase, civil infraction service; maybe tobacco
or obtain education
West Virginia 18 Possess Not listed Fine and community
service
Wisconsin 18 Possess, purchase Not listed Not listed
Wyoming 18 Possess, use, Misdemeanor Fine or community

purchase, attempt to service or tobacco
purchase cessation

APPENDIX C 325
Illegal to Suspend
Yes, higher fine and Tobacco products

more community
service hours
Not listed Tobacco products, nicotine ü
products
Yes, increase in fine Tobacco products ü


Appendix D
Supplemental Information
About the Models
By Theodore R. Holford and David T. Levy
T
he CISNET smoking population model tracks individuals by age
and U.S. birth cohort beginning in 1864 as the individuals progress
through various smoking stages (i.e., reconstructed smoking preva-
lence, initiation, and cessation rates) to estimate the smoking prevalence
and the rates of smoking initiation, cessation, and intensity in the United
States by age and gender from 1964 through 2012 (Anderson et al., 2012;
HHS, 2014; Holford et al., 2014a). The model reports population levels of
smoking, non-lung cancer mortality, and overall mortality (Anderson et al.,
2012; Holford and Clark, 2012; Holford et al., 2014b; Jeon et al., 2012).
The CISNET smoking population model can also simulate individual smok-
ing trajectories using a “smoking history generator” (Jeon et al., 2012). The
CISNET smoking population model was recently used to assess smoking
patterns and estimate the smoking rates of initiation and cessation in the
United States from 1964 to 2012 (Holford et al., 2014a) and the number
of premature deaths prevented in the United States by tobacco control
from 1964 to 2014 (Holford et al., 2014b). The model can also simulate
lung cancer incidence and mortality when coupled with lung cancer natural
history models (de Koning et al., 2014; McMahon et al., 2014; Meza et
al., 2014; Moolgavkar et al., 2012). The CISNET smoking and lung can-
cer models were used to estimate the number of lung cancer deaths in the
United States prevented by historical tobacco control efforts from 1975 to
2000 (Moolgavkar et al., 2012). More recently, the CISNET smoking and
lung cancer models were used to provide estimates of the potential ben-
efits and harms of computerized tomography lung cancer screening in the
United States (de Koning et al., 2014; McMahon et al., 2014; Meza et al.,
327

2014). The CISNET smoking population model accounts for differences in

mortality rates by gender, age, birth cohort and smoking status (Holford et
al., 2014b; Rosenberg et al., 2012), and it breaks the population into never
smoked, former smoker, and six categories of current smokers varying by
intensity. The same approach is applied to projections of U.S. mortality
rates based on the Lee-Carter model (Sprague, 2009). The CISNET smok-
ing model does not account explicitly for the effects of tobacco control
policies. Instead, the model uses historical U.S. rates of smoking prevalence,
initiation, cessation and intensity by age, gender, and birth cohort estimated
from National Health Interview Survey (NHIS) data. These rates capture
the temporal variations in U.S. smoking patterns and the indirect effects of
tobacco control policies as implemented historically.
The SimSmoke model tracks the number of never, current, and former
smokers by age and gender in the modeled population by year and evalu-
ates the impacts of tobacco control policies through their effects on smoking
prevalence as a function of the assumed associated changes in smoking initia-
tion and cessation rates based on literature review and expert judgment (Levy
et al., 2005, 2010b, 2012a). The SimSmoke model estimates the number of
annual smoking-attributable deaths and the effects of tobacco control poli-
cies on smoking prevalence and attributable mortality, with applications to
the entire United States (Levy and Friend, 2000, 2001; Levy et al., 2000a,b,
2001a, 2004, 2005, 2010b) as well as at the state level (Levy et al., 2007,
2008, 2012b), and other countries (Levy et al., 2010a, 2012a, 2013a,b,
2014, in press).
The tobacco control policies modeled in SimSmoke include tax changes,
smoke-free air laws, health warnings, the Fairness Doctrine, advertising
restrictions, mass media interventions, availability of cessation treatments,
and youth access policies considered individually and in combination (Levy
et al., 2005, 2010b, 2012a). The initial development and validation of
the SimSmoke model used data from the Tobacco Use Supplement of the
Current Population Survey (TUS-CPS) (Levy et al., 2005). The SimSmoke
model does not consider smoking intensity and does not account for vary-
ing patterns by birth cohort.
As is the case with CISNET, the SimSmoke model considers differential
mortality rates by gender, age, and smoking status (Holford et al., 2014b;
Rosenberg et al., 2012) and breaks the population into never smoked, current
smoker, and 16 categories of former smokers differentiated by the number
of years since they quit. The SimSmoke model explores the potential effects
of raising the minimum age of legal access to tobacco (MLA) on smoking
initiation rates in order to make predictions of the policy effects on future
smoking prevalence and smoking-attributable deaths and maternal and child
health outcomes (i.e., low birth weight, pre-term births and sudden infant
deaths) while simultaneously accounting for ongoing tobacco control efforts.

APPENDIX D 329
CISNET MODEL (BY HOLFORD)
Smoking History Summary for the United States, 1965–2012

The data from 36 NHISs conducted from 1965 to 2012 were analyzed
using the method employed by CISNET (Holford et al., 2014b). Results
from this analysis provided summary estimates for birth cohorts starting
in 1864 and ending in the calendar year 2012. These estimates included
a. Current smoker prevalence;

b. Former smoker prevalence;
c. Never smoker prevalence;
d. Yearly smoking initiation probabilities for never smokers;
e. Yearly smoking cessation probabilities for current smokers; and
f. Distribution of categories for reported daily cigarettes per day (CPD):
CPD≤5, 5<CPD≤15, 15<CPD≤25, 25<CPD≤35, 35<CPD≤45,
45<CPD.
Smoking Prevalence Model

A compartment (macro) model that characterizes a typical smoking
history in which a subject begins to smoke at some point (never → current
smoker) after which they may quit (current → former smoker) was used.
While this over simplifies what can be much more complex in reality, it does
provide a useful characterization of the experience for most of the popula-
tion. Smoking cessation can be especially difficult to characterize because
it is often not successful on the first attempt. Hence, we adopted the rule
that subjects who report quitting must have done so at least 2 years before
the interview, otherwise their period of observation is regarded as being
truncated at the given age at cessation.
We defined the basic quantities of interest conditional on a hypothetical
case with no transitions to death. Let a represent age, t period or calendar
year, and c cohort or year of birth, and all three temporal components may
play a role when constructing the basic parameters affecting smoking his-
tory. These temporal indicators are related by c = t − a, therefore, when
presenting the relationships among the basic model parameters, we can
without loss of generality represent them as functions of age and cohort.
The smoking initiation probability, p(a,c), is the conditional probability of
smoking initiation at age a for cohort c, if not a smoker at a − 1, i.e.,
p(a,c) = Pr{Smoker at a | Not smoker at (a – 1),c}.

It is related to the cumulative proportion of ever smokers at a conditional

on remaining alive,
a
PE ( a, c ) = 1 − ∏ 1 − p ( i, c ) 
i −1
= 1 − 1 − PE ( a − 1, c )  1 − p ( a, c ) 
(1)
where PE(0,c) = 0, which is equivalent to the actuarial approach for estimat-

ing the survival curve. If smoking did not affect mortality then one would
expect equation (1), which is conditional on remaining alive, to also hold
in a population followed over time. But, of course, mortality is affected
by smoking so that the observed proportion of the population who have
ever smoked at a particular age is given by PE*(a,c) ≤ PE(a,c). Initiation
probabilities estimated at a particular survey would be similarly affected
by differential mortality; and we represented these by p*(a,c) = p(a,c)/Cp
where Cp ≥ 1 is a constant correction factor introduced to adjust for this
effect. We assumed that differential mortality among smoking categories
had little effect early in life and the impact intensified with age. Cohorts
born before 1935 would only have survey data for ages over 30 when one
might expect differential mortality to begin to introduce substantial bias
in the unadjusted estimate, pˆ * ( a, c ). In recent cohorts, almost all smoking
initiation occurred before age 30, but for those born early in the twentieth
century it was not so uncommon for initiation to occur later in life, espe-
cially in women. Later smoking initiation would also tend to postpone the
effect of differential mortality in the cohort. We assumed that the differ-
ential mortality resulting from cigarette smoking occurred at ages, a ≥ a0,
and PE*(a,c) = PE(a,c) for a < a0. Initiation probabilities corrected for dif-
ferential mortality were found by solving
a0
PE ( a0 , c ) = 1 − ∏ 1 − Cp p * ( i, c ) 
i −1
for Cp, i.e., by matching the cumulative initiation rates to the estimated
prevalence at age a0. We assumed that a0 was the age at first survey in 1965
or 30, whichever was older.
Smoking cessation was assumed to be a function of age for each cohort.
The smoking cessation probability conditional on the subject being alive
and currently smoking is
q(a,c) = Pr{Former smoker at a | Smoker at (a – 1),c}.

APPENDIX D 331
We assumed that q(a,c) = 0 for a < 15 and we estimated it for 15 ≤ a ≤ 99.

The cumulative proportion of smokers in cohort c who had not ceased
smoking by age a is given by
a
Q ( a, c ) = ∏ 1 − q (i, c ) . (2)
i −15
For simplicity, we assumed that this quantity does not depend on the
age an individual started smoking, number of cigarettes per day or other
factors that may be related to an individual’s success in quitting. Because
initiation tends to occur in a fairly narrow age range, variation in age of
initiation becomes less of a factor affecting mortality as a cohort gets older.
Introducing intensity of smoking into a model for cessation would require
detailed lifetime histories of smoking which were not commonly obtained
by NHIS, a limitation in the available data.
Current smokers represent ever smokers who have not quit, and given
our assumption that this only depends on age for a given cohort, the preva-
lence is
PC(a,c) = PE(a,c)Q(a,c).
Former smokers are those who have smoked at some point in their
lives, but quit before age a, and the proportion of these individuals is
PF(a,c) = PE(a,c) – PC(a,c)

= PE(a,c)[1 – Q(a,c)].
Finally, the proportion of cohort c who have never smoked is the

complement of those who ever smoked,
PN(a,c) = 1 – PE(a,c).
For a given age and cohort, the sets of current, former, and never smokers
are exhaustive, i.e.,
PC(a,c) + PF(a,c) + PN(a,c) = 1.
Estimation of smoking parameters Data were only obtained for a re-

stricted range of ages, a in [amin,amax], and periods, t in [tmin,tmax]
so that the earliest cohort would be cmin − tmin − amax and the latest
cmax = tmax − amin. Available data for a given cohort c, would cover an
age range that would vary by cohort, i.e., a in [tmin − c, tmax − c]. To fill
in smoking history that was not represented in the survey, we represented

each temporal effect as a nonparametric function that we applied outside

the range of observed data.
Cross-sectional estimates of ever smokers For years covered by sur-
veys, i.e., 1965–2012, participants provided information that could be used
to estimate the prevalence of ever smokers by age, a, for the corresponding
cohort, c = t − a. Let Yi be 1 if the i-th individual ever smoked and 0 other-
wise, where the probability of the response is a function of age and cohort,
PE(a,c). We assume an additive logistic model for Yi, so that
logit {PE(a,c)} = b0 + ba(a) + bc(c)
where β0 is an intercept and β.(•) is a function given by a constrained

natural spline. The model was fitted using PROC GENMOD in SAS® with
knots specified as
Age: 40, 50, 60, 70

Cohort: 1910, 1920, 1930, 1940, 1945, 1950, 1955, 1960, 1965, 1970, 1980
We assumed that the cohort effect remained constant for those born after
1979, the most recent cohort that would provide data to a survey regarding
smoking history after age 30 in 2012 which was the age used to identify
Cp. Values used for subsequent cohorts were set to be identical to those for
the 1982 birth cohort.
Smoking initiation probability Unadjusted estimates of annual age-
specific smoking initiation probabilities for a given cohort, pˆ * ( a, c ) , were
directly derived from NHIS data. For each cohort represented in a survey,
we determined the number of subjects who started to smoke, d(a,c), and
who had never smoked to that point, n(a,c). These comprised the response
data introduced into a linear logistic model in which the temporal factors
were nonparametric functions to be estimated. Each NHIS survey repre-
sented participants who survived until that time, and because this group
would overrepresent individuals in a cohort who started smoking late or
not at all, these cohort-specific initiation probabilities would be biased
downward. The correction factor was found by specifying the target value
for the estimated cumulative initiation at a specific age, a0, to be equal to
the value estimated from the cross-sectional analysis, i.e.,
a*
PˆE ( a*, c ) = 1 − ∏ 1 − Cˆ p pˆ * ( a, c ) 
i =1
and finding Ĉp which satisfies this condition.

To determine the crude initiation probability estimates, an age-period-
cohort model was fitted to the tabulated data given number of subjects who
start smoking and are at risk of starting at a given age,

APPENDIX D 333
logit {p*(a,c)} = b0 + ba(a) + bt(t) + bc(c)
where β0 is an intercept and β.(•) is given by a constrained natural spline.

We were only interested in the fitted values for the initiation probabilities,
which were not affected by the well-known identifiability problem in age-
period-cohort models. Knots were specified as:
age: 10, 15, 20, 50, 60

period: 1910, 1920, 1930, 1940, 1950, 1960, 1970, 1980
cohort: 1910, 1920, 1930, 1940, 1945, 1950, 1955, 1960, 1965, 1970, 1980
Age for the target used to determine the correction factor was age in
1965 (year of the first NHIS survey) or 30, whichever was older,
a* = max{1965 – c,30}. The target value for the cumulative probability of
being a smoker was the estimate derived in the analysis of the prevalence
curve, Πˆ ( a*, c ) .
Smoking cessation probability An individual was identified as having
quit smoking if they had not smoked for 2 years. Because of the 2-year lag
used in the definition of quitting, an individual who reports cessation at age
a − 2 or later could not be classified and they would be truncated at that
age. Hence, current smokers were similarly truncated at age a − 2. Data
used for this analysis were from surveys conducted from 1970–2012, in-
cluding subjects reporting ages from 17–98. If the reported age of cessation
was younger than 8, it was set to 8. For each year of age following smoking,
a binary response was created based on our definition of quitting. Yearly
estimates of the linear logistic age-period-cohort model were fitted in which
logit {q(a,t,c)} = b0 + ba(a) + bt(t) + bc(c)
where β0 is an intercept and β.(•) are given by a constrained natural splines.

We were only interested in the fitted values for the cessation probabilities,
which are not affected by the well-known identifiability problem in age-
period-cohort models. Knots were specified as follows:
age: 30, 40, 50, 60

period: 1920, 1930, 1940, 1950, 1960, 1970, 1980
cohort: 1910, 1920, 1930, 1940, 1950, 1960, 1970
Estimates of the yearly cessation probability for age a and cohort c

were the fitted values for ages 15–99, q̂ (a,a + c,c). The conditional cessa-
tion probabilities were used to generate the cumulative probabilities of not
quitting, Q̂ (a,a + c,c), using equation (2).

Cigarettes smoked per day Reports of the number of cigarettes

smoked per day showed an extremely high degree of digit preference, espe-
cially concentrated at half or whole U.S. packs. Therefore, dose was ana-
lyzed as an ordered categorical response with half pack being at the center
of the category, which was also usually the mode and close to the mean.
The intervals (approximate interval center) employed were: CPD≤5 (3);
5<CPD≤15 (10); 15<CPD≤25 (20); 25<CPD≤35 (30); 35<CPD≤45 (40);
and 45<CPD (60). A cumulative logistic model was fitted to the data us-
ing PROC LOGISTIC in SAS® with age, period and cohort represented by
additive nonparametric factors function of time using constrained natural
splines. Knots were specified as:
age: 25, 30, 35, 40, 45, 50, 55, 60, 65, 70
period: 1970, 1975, 1980, 1985, 2000, 2005
cohort: 1910, 1920, 1930, 1940, 1950, 1960, 1970, 1980
The fitted estimates of the probabilities for each category of smoking dose
for each cohort for ages 0 to 99 were used as parameters for the smoking
history generator. Estimates for cohorts born before 1920 were constrained
to be the same as for the 1920 birth cohort. Similarly, estimates for cohorts
born after 2002 were constrained to be identical to those of the 2002
cohort, who would be 7 in 2012, i.e., the year before the earliest age at
initiation considered in this analysis.
Estimation of current, former, and never smokers for 1-year cohorts
Estimates of smoking prevalence were derived from the estimated curves
for ever smokers, P̂E(a,c), and the corresponding survival function for not
quitting, Q̂ (a,c). The estimated prevalence of current smokers by age and
cohort is
P̂C (a,c) = P̂E(a,c) Q̂ (a,c).
Prevalence of former smokers is
P̂F (a,c) = P̂E(a,c) – P̂C (a,c)

= P̂E(a,c)[1 – Q̂ (a,c)].
Finally, prevalence of never smokers is
P̂N (a,c) = 1 – P̂E(a,c).

APPENDIX D 335
Estimated Smoking History Measures, 2012–2100

In order to assess the impact of a change in the minimum legal age
of purchase of cigarettes we assume that the impact will primarily affect
smoking initiation probabilities and not cessation probabilities or smoking
intensity distribution, which were assumed to remain unchanged for future
birth cohorts who would experience the policy change. The model status
quo initiation and cessation rates are available on the CISNET resources
website.1 We assumed that a change to the MLA in 2015 would primarily
affect those who were 15 or older, i.e., the 2000 or later birth cohorts.
Using the methods described elsewhere (Holford et al., 2014b), the pos-
tulated changes in initiation probabilities yielded ever-smoker prevalence
estimates in the subsequent birth cohorts. In addition, using the methods
described, we obtained estimates of current and former smoker prevalence
estimates. The distribution of smoking intensity categories was assumed to
remain the same in future birth cohorts.
Smoking status–specific death rates (current smoker by intensity, for-
mer smoker, and never smoker), µi(t), were obtained using the Human
Mortality Database (HMD) mortality rates for the United States by age, cal-
endar year and gender combined with the method described by Rosenberg
et al. (2012). HMD rates were then projected to 2100 using the Lee-Carter
model (Sprague, 2009) and then further broken down by smoking status
using the Rosenberg et al. method. The death rate at age t for a particular
scenario was determined by the given distribution of smoking status, pi(t),
µ ( t ) = ∑ pi ( t ) µi ( t ) ,
i
where i represents the various smoking status/intensity combinations. These

death rates were used to modify the population distribution in order to re-
flect the effect of the change in mortality rate (Holford et al., 2014a). The
resulting death rates and estimated populations for 2012–2100 were used
to obtain summary measures of the effect of a given scenario.
Premature Deaths Due to Tobacco Use

The excess death rate resulting from tobacco use was estimated by the
difference between the death rate under the given scenario and the death
rate for never smokers,  µ ( t ) − µ0 ( t )  , where µ0(t) is the death rate at age
t for never smokers. If P(t) is the population size at age t, then the number
of premature deaths due to tobacco use is given by

∑ P (t )  µ (t ) − µ0 (t ) .
t
Population estimates by single years of age (0–84 and 85+) were ob-
tained from the U.S. Census for years 1964 through 2060 (USCB, 2013a,b).
Population estimates going to 2100 were obtained by following the popu-
lation from the previous year and assuming that the proportional change
would remain the same. For age 0, we assume that after 2060 the propor-
tional increase will be the same as the change estimated by the U.S. Census
for 2059 to 2060.
Years of Life Lost

Using the age-specific death rates for a cohort, the expected years of
life remaining at age t, e(t), was calculated. For a death that occurs at age
t, life expectancy for never smokers, e0(t), would estimate the mean number
of years of life lost. The total years of life lost by smokers who died early
is given by
∑ P (t )  µ (t ) − µ0 (t ) e0 (t ) .
t
Lung Cancer Deaths Avoided

The two-stage clonal expansion (TSCE) model (Hazelton et al., 2012;
Meza et al., 2008) was used to estimate the lung cancer mortality rate.
For current smokers, the model gives the rate at age t as a function of
age at initiation and smoking intensity. We assume that the age at initia-
tion is independent of intensity, so the joint distribution is obtained as the
product of the probability of initiation at a given age and the probability
of a particular smoking intensity level. These were then used to obtain the
mean lung cancer mortality rate for current smokers. Similarly, for former
smokers we obtain the mean lung cancer mortality rate for a particular age
of interest by first determining the joint distribution of age at initiation,
age at cessation, and smoking intensity. Multiplying this joint distribution
by the lung cancer mortality rate obtained from the TSCE model and then
summing over all combination of initiation and cessation times provides the
probability of lung cancer death at age t. Finally, we obtain the overall rate
for a particular scenario by taking a weighted average of the lung cancer
mortality rates for never, current, and former smokers. The excess rate is
obtained by taking the difference between the rate for the scenario and the
rate for never smokers. Multiplying this by the population gives the number

APPENDIX D 337
of lung cancer deaths avoided for a given age, and taking the sum over all
ages provides the overall number of lung cancer deaths avoided.
Birth Cohort and Period Temporal Perspectives

The models used to estimate the fundamental parameters of smoking
initiation, cessation, intensity, and ever-smoker prevalence were derived
from the birth cohort perspective. This captures the life course of different
generations, and it provides useful summaries of the groups that would
have experienced a change in the MLA at a point in life in which they are
most likely to initiate cigarette smoking. Viewed from this perspective, we
determined life expectancy, premature deaths from smoking, and excess lung
cancer deaths caused by smoking from the temporal perspective of following
these individuals through life. Because these individuals would be classified
as belonging to a group upon enactment of a change in the MLA, the model
assumptions would result in no further changes in the birth cohort smoking
history, which would result in constant age-specific death rates for all causes
and specific causes. Any changes in the number of subjects affected reflect
trends in the size and the age structure of the population.
Results are also presented from the period or calendar year perspective,
which describe the view experienced by the health community. Mortality
rates differ little in absolute magnitude until after age 40, and diseases like
lung cancer have long latency. Hence, the effect of changing MLA would
not be discernable until decades after enactment of a change. However, once
the effect becomes observable, it will continue to increase until it reaches
steady state. The summaries from the period perspective include estimates
of the number of premature deaths from all causes or from a specific cause
in a given calendar year. Life expectancy estimates given for a period rep-
resent a summary of the age-specific death rates in a given year, which is
identical to the traditional demographic summary that is commonly used
as one summary measure of the health of the country. In contrast to the
estimates derived from the cohort perspective, this summary does not cor-
respond to the life course of a population.
SIMSMOKE (BY LEVY)
The U.S. SimSmoke Model

SimSmoke divides the population in 1965 into (1) never smokers
(Neversmokers, indicated in subscripts by “ns”), (2) smokers (Smoker,
indicated in subscripts by “s”), and (3) 15 categories of former smokers
(Formersmokersk, where k = 1, . . . ,14, 15+, corresponding to the year
quit). Individuals are classified as never smokers from birth until they initi-

ate smoking or die, as shown below, where t is the year and a is the age of
the individual:
Neversmokerst,a =
Neversmokerst–1,a–1 * (1 – MortRatet–1,a–1,ns)*(1 – Initiation ratet–1,a–1).
Never smokers can become smokers through initiation. Once they have
become smokers, individuals continue in that category until they quit or
die. Former smokers continue in that category until they die or re-enter the
group of smokers through relapse. The number of smokers is tracked as:
Smokerst,a =
Smokerst–1,a–1*(1 – MortRatet–1,a–1,s)*(1 – Cessation ratet–1,a–1)
15+
+ ∑ k = 2 Former smokers t −1, a −1,k*(1 – MortRatet–1,a–1,k)*(Relapse ratea–1,k)
+ Neversmokerst–1,a–1*(1 – MortRate t–1,a–1,ns)*Initiation ratet–1,a–1.
First-year former smokers are determined by the first-year cessation

rate applied to surviving smokers in the previous year. After the first year
quit, individuals who have been former smokers for k = 2, . . . , 14 are
defined as:
Former smokerst,a,k =
Former smokerst–1,a–1,k–1*(1 – MortRatea,k)*(1 – Relapse ratea,k–1).
For those who have quit smoking for 15 or more years, 15+, the equa-
tion above includes all individuals who have quit more than 15 years from
the previous year.
Data on smoking rates are from Holford et al. (2014a) and are based
on NHIS. Smoking prevalence is defined as the percentage of people in the
population who have smoked 100 cigarettes during their lifetime and cur-
rently smoke. Initiation rates through age 30 are based on responses for age
of initiation and having smoked 100 cigarettes. Cessation is tracked from
age 16, because data from NHIS on cessation begins at that age. Cessa
tion rates are defined in terms of having quit for 2 years, which reflects a
trade-off between higher cessation rates in the first year and relapse in later
years. After 1965, relapse rates are also distinguished by age and gender,
and the number of years since quitting (HHS, 1990; Hughes et al., 2008).
The SimSmoke model status quo initiation and cessation rates are available
on the CISNET resources website.2

APPENDIX D 339
Smoking-Attributable Deaths
To estimate smoking-attributable deaths (SADs), we used age- and
gender-specific current and never-smoker mortality rates used by Holford et
al. (2014b) based on the Cancer Prevention Studies (CPS-I and CPS-II) and
the Nutrition Follow-up to CPS-II. For smokers, SADs are defined in terms
of the excess death rate of current smokers (smoker mortality rate minus
never-smoker mortality rate). Mortality rates for former smokers decrease
progressively from the current smoker toward the never-smoker level as
years since quitting increase, according to CPS-II data (Burns et al., 1997).
SADs are estimated for current and former smokers by age (a), gender (g),
and year (t) by summing over age (a) as:
∑ a= 40 excess death riska,g,t * prevalencea,g,t * projected populationa,g,t .

85
Policy Analysis
Separate policy modules estimate the effects of past tax changes, smoke-
free air laws, health warnings, advertising restrictions, mass media, cessa-
tion treatment, and youth access policies. The original policy parameters
used to generate the predicted effects are based on thorough reviews of the
literature and the advice of an expert panel (Friend and Levy, 2001, 2002;
Levy and Friend, 2000, 2001, 2002a,b; Levy et al., 2000a,b, 2001a,b,
2004). The policy effects (PEs) are calculated in percentage terms, i.e., PE =
(post-policy rate – initial rate)/initial rate, with PE < 0, assuming an effective
policy where the post-policy rate is less than the initial rate. For most poli-
cies, the greatest effect will occur in the first few years in which the policy
is in effect, modeled as a permanent additive effect on smoking prevalence,
i.e., Smokerst,a * (1 + PPEi,t,a) for policy i at time t, age a, with PPE defined
as the prevalence of PE. If the policy is maintained, the effects of the policy
are sustained throughout future years as: Initiation ratea * (1 + IPEi,a), with
IPE defined as the initiation PE. The effect of a maintained policy increase in
the cessation rate over time is given as: Cessation ratea * (1 – CPEi,t,a), with
CPE defined as the cessation PE. SimSmoke projects smoking rates through
2014, based on policies that were implemented over the period 1964–2014.
The effect sizes are shown in Table D-1.
Data on the levels of policies were input into the SimSmoke model for
the years 1965 through 2012. We calibrated model cessation rates against
data on smoking prevalence through 1985, leading to a reduction in those
rates of 9 percent for females, 10 percent for males ages 55–64, and 20 per-
cent for males of ages 65 and above. Table D-2 shows the sources and
specifications for the data used in SimSmoke in this report.

TABLE D-1 Policy Inputs and Effect Sizes in SimSmoke

Policy Description Potential Percentage Effecta
Cigarette Taxes (Levy et al., 2000a)
Cigarette price The state level average price for For each 10% price
a pack of cigarettes (including increase:
branded and generic), including 6% reduction ages 15–17,
state and federal excise taxes. 4% reduction ages 18–24,
Tobacco Institute (Orzechowski 2% reduction ages 25–34,
and Walker, 2012), adjusted for and
inflation using the consumer price 1% reduction ages 35 and
index (www.bls.gov/cpi/home.htm) above
Smoke-free Air Laws (Levy et al., 2001b)
Worksite ban, Smoking banned in all indoor 6% reduction
well-enforced worksites in all areas
Worksite restrictions, Smoking in restricted areas only 2% reduction
weak
Restaurant and bar ban, Ban in all indoor restaurants in 2% reduction
well-enforced all areas
Restaurant ban, weak Smoking in restricted areas only 1% reduction
Other places bans Ban in three of four (retail stores, 1% reduction
arenas, public transportation, and
elevators)
Enforcement and Compliance reflecting norms Effects reduced by as much
publicity and publicity as tobacco control as 50% if no compliance
campaign variable or publicity
Fairness Doctrine and Advertising Restrictions (Lewit et al., 1981; Warner, 1989; Warner
and Murt, 1983)
Existence of fairness Airing of antismoking messages 39% reduction in
doctrine on radio and television from July initiation rates, 8%
1, 1967, to January 1, 1971, and increase in cessation rates
banning of cigarette advertising
on radio in 1970 and television
in 1971
Tobacco Control Campaigns (Levy and Friend, 2001)
Well-funded campaign Campaign expenditures meeting 6.5% reduction
the pre-2009 CDC minimum
recommended
Moderately funded Campaign expenditures meeting 3.6% reduction
campaign 50% of the pre-2009 minimum
recommended
Low funded campaign Campaign expenditures meeting 1.2% reduction
<25% of the pre-2009 minimum
recommended

APPENDIX D 341
TABLE D-1 Continued

Policy Description Potential Percentage Effecta
Health Warnings (Azagba and Sharaf, 2013; Hammond et al., 2007; Huang et al., 2014)
Weak health warnings Non-graphic warning covers less 1% reduction in
than one-third of the package. prevalence and 2%
Reports, score = 2 increase in cessation only
Cessation Treatment Programs (Levy and Friend, 2002a; Levy et al., 2010b)
Availability of NRT and If NRT is provided by either 1% reduction if score of 3b
Bupropion general store or pharmacy with
Rx = 1 and = 2 If NRT is provided
by general store or pharmacy
(no Rx required). If Bupropion is
provided by either general store or
pharmacy with Rx = 1.
Provision of treatments Types of facilities distinguished, 2.25% reduction if
specified as primary care facilities, indicator = 2 for all
hospitals, offices of health facilities and program is
professionals. Community and well publicizedb
other. MPOWER: 0 = None, Yes
in some = 1, Yes in most = 2.
Quit line Operating active quit line 0.5% reductionb
Comprehensive A proactive quit line with NRT, ~3% reduction in
cessation treatment complete treatment coverage prevalence, and 20%
through insurance increase in cessationb
Youth Access Restrictions (Levy et al., 2001a)
Strongly enforced and Compliance checks are conducted 20% reduction for those
publicized 4 times per year per outlet, ages 16–17 and 30%
penalties are potent and enforced, reduction for those
and with heavy publicity and age <16c
community involvement
Moderate enforcement Compliance checks are conducted 10% reduction for those
at least once per year per outlet, ages 16–17 and 15%
penalties are moderate, and with reduction for those
some publicity age <16c
Low enforcement Compliance checks are conducted 2.5% reduction for
sporadically, penalties are weak, those ages 16–17 and
there is little merchant awareness 4% reduction for those
and minimal community age <16c
participation
a The effect sizes are shown relative to the absence of any policy. Unless otherwise specified,
the same percentage effect is applied as a percentage reduction in the prevalence in the initial
year and as a percentage reduction in initiation rate and a percentage increase in the cessation
rate in future years, and is applied to all ages and both genders.
b Applied to prevalence and first year quit rates only.
c Applied to initiation and prevalence only.

TABLE D-2 Data Used in SimSmoke

Variable Current Source Current Specifications
I. P
opulation model
A. Population 1965–2065 Census and Census Breakdowns by age and
Projections and Projections by Ted gender
Holford from 2066 through 2100
B. M
ortality rates 1965–2065 Multiple Cause-of-Death File Breakdowns by age,
and Cancer Prevention Study I and II and gender, and smoking
the Nutrition Follow-up to CPS-II status (current, former,
Projected U.S. mortality rates based on never)
the Lee-Carter model (Sprague, 2009)
II. S moking model—initialized in 1965, with future changes in initiation and cessation
rates due to policies through policy modules
A. B
aseline smoking 1965 National Health Interview Survey 100+ cigarettes
rates for current (NHIS) for age 10+ lifetime, distinguished
and ex-smokers by current and former
smokers. (<1, 1–2,
3–5, 6–10, 11–14,
15+ years) by age and
gender
B. Initiation rates 1965 National Health Interview Survey Breakdowns by age and
(NHIS) for age 10+ gender
C. F
irst year 1965 National Health Interview Survey Breakdowns by age and
cessation rates (NHIS) for age 16+ gender
D. Relapse rates Previous studies (Gilpin et al., 1997; Breakdowns by age and
HHS, 1990; Hughes et al., 2008) gender
E. E
xcess death risks 1965–2100 death rates by current, Breakdowns by age,
of smokers and former, and never smokers as developed gender, and smoking
ex-smokers by CISNET (Holford et al., 2014b) status
III. Policy modules—levels from 1965–2014
A. P
rice and taxes Orzechowski and Walker, (2013) www. Prices and CPI for
bls.gov/cpi/home.htm 1965–2014
B. S moke-free air www2.cdc.gov/nccdphp/osh/state/report_ Different types of laws
laws index.asp and www.impacteen.org and their stringency
and compliance rates
C. M
edia and other CDC and tobaccofreekids.org Expenditures per capita
educational by state
campaigns
D. Cessation MPOWER Reports (Levy et al., 2010b; Indicators of when
treatment WHO, 2008, 2013) pharmacotherapies
programs became available,
cessation treatment
locations and quitlines

APPENDIX D 343
TABLE D-2 Continued

Variable Current Source Current Specifications
E. H
ealth warnings HHS (2014) Indicator of strength
F. F
airness doctrine Warner (Warner, 1989; Warner and Indicator of extent of
Murt, 1983) implementation
G. Y
outh access CDC, SAMHSA (Levy et al., 2001a) Enforcement checks,
enforcement penalties, community
campaigns, self-service
and vending machine
bans
The percent changes in smoking prevalence from SimSmoke were vali-

dated against the percent change in NHIS rates for four age groups (18–24,
25–44, 45–64, 65+) over the period 1965–2012. By 2012 male adult smok-
ing prevalence (18 and above) from both NHIS and SimSmoke showed
a decline of 61 percent relative to the initial 1965 level. Female smoking
prevalence from the NHIS declined 54 percent, compared with 53 per-
cent from SimSmoke. Generally, SimSmoke predicted prevalence rates for
females and males that were similar to the NHIS rates, except for under-
estimating the reduction for males between the late 1970s and late 1990s.
By 2012 SimSmoke obtains estimates for male smoking prevalence by age
group that are very similar to the NHIS estimates, but it underestimated
rates during the 1980s and early 1990s. For females, SimSmoke predicted
the relative decline in smoking prevalence by 2012 well for all age groups,
except for the 65-and-above age group.
The effects of a change in the MLA are modeled through initiation
rates beginning in 2015. The 2015 initiation rates used to predict the ef-
fects of the change in the MLA are those derived from SimSmoke based on
the policy effects applied to changes in policy levels between 1965 through
2014. The initiation rates are constant from 2015 through 2100.
Estimating Smoking-Attributable Birth Outcomes3

SimSmoke considers three smoking-attributable adverse birth outcomes
(SAABOs): sudden infant death syndrome (SIDS), low birth weight (LBW),
and pre-term birth (PTB). To calculate the number of cases of modeled
SAABO, we use the method employed in the Smoking-Attributable Mortal-
ity, Morbidity, and Economic Costs (SAMMEC) software (Melvin et al.,
3The methods and data regarding maternal and child health outcomes are presented in
more detail than other outcomes for transparency, because they have not yet been published.

2000). SAFs, based on the attributable-fraction formula originally described

by Levin (1953) and expanded upon by Lilienfeld (1980), are multiplied by
the total number of events of each modeled outcome. Separate estimates are
calculated for each year (t) by age group (a), i.e, for each outcome.
SAABOa,t = Number of Observed Events a,t × SAFa,t
Number of Observed Events represents the total observed adverse

birth cases for a given outcome in the population, including those for both
smoking and nonsmoking mothers. Measures of outcome prevalence are
multiplied by the size of the corresponding population. In SimSmoke, the
number of outcomes is available for each smoking outcome by age of the
mother for SIDS, LBW, and PTB as well as for others aggregated over all
maternal ages. SAFs are calculated using the smoking prevalence and rela-
tive risk of current maternal smokers aged 15–49, or some subset thereof.
SAFs for each outcome by year (t) and age group (a) are derived using the
following formula:
SAFa,t = [(1 – pa,t) + pa,t × RR - 1] / [(1 – pa,t) + pa,t × RR], where

p = percentage of pregnant women who are nonsmokers during pregnancy, and
RR = relative risk of outcome where maternal smokers relative to nonsmokers.
While the maternal smoking prevalence can vary by age and year in
the above formulation, we assume that relative risks are constant over time
and by age since past studies do not adequately distinguish by age. Sum-
ming across age categories for a particular year (t) provides the estimate of
SAABO for each health outcome for that year.
Figure D-1 provides a flowchart of the estimation process for maternal
and child health (MCH) outcomes.
Data
Adverse MCH outcomes For LBW and PTB, National Center for Health
Statistics (NCHS) data on adverse MCH outcomes by age and gender for
2012 were obtained from CDC Wonder, the epidemiological database
operated by the Centers for Disease Control and Prevention (CDC) (HHS
et al., 2014b). For SIDS, data were obtained for all ages from CDC W onder
for 2011 (the most recent year) (HHS et al., 2014a), and the propor-
tions by age group were based on overall infant mortality (Matthews and
MacDorman, 2013). Because overall rates for each MCH outcome have
been relatively constant in recent years, the percent of maternal outcomes
in 2012 is maintained for all future years. The data by age are presented in
Table D-3. LBW, PTB, and SIDS are highest at younger ages and for above

APPENDIX D 345
and
and
FIGURE D-1 Relationship of the components for each maternal and child health
(MCH) outcome.
TABLE D-3 Gestational Age and Birth Weight, 2012

PTB, PTB FTB, All All FTB,
Age group LBW NW LBW PTB LBW NW Total
<15 8.8% 12.1% 3.7% 20.9% 12.5% 75.4% 3,657
15–19 7.3% 7.3% 3.5% 14.6% 10.8% 81.9% 309,849
20–24 5.1% 6.4% 3.2% 11.6% 8.3% 85.2% 904,623
25–29 5.4% 5.7% 2.5% 11.2% 7.9% 86.3% 1,130,250
30–34 5.1% 5.9% 2.4% 10.9% 7.5% 86.7% 1,011,765
35–39 6.0% 6.8% 2.6% 12.8% 8.6% 84.6% 471,499
40–44 8.3% 8.7% 3.4% 17.0% 11.7% 79.6% 103,127
45–49 14.0% 10.7% 5.1% 24.7% 19.1% 70.2% 7,122
NOTE: FTB = non-preterm; LBW = low birth weight; NW = normal weight; PTB = pre-term
birth.

age 40. In addition, it should be noted that there is overlap between PTB
and LBW, especially at younger ages. We assume that the rates are constant
from 2012 onward, but the model is flexible enough to allow for trends in
the prevalence of MCH outcomes over time.
The total number of adverse MCH outcomes over time depends on
fertility rates. The fertility rates were obtained from the NCHS through
CDC WONDER (U.S. Department of Health and Human Services for the
year 2012 categorized by age of the mother) (HHS et al., 2014b). We use
projected fertility rates by the United States (CDC, 2014; USCB, 2014)
through 2060 to extrapolate for future years through 2100. The projec-
tions show fertility rates that are slowly decreasing over time, and it is as-
sumed that rates after 2060 stay constant. Since the projected rates are not
distinguished by age, the age distribution from 2012 is assumed to remain
constant. The fertility rates are multiplied by the projected population used
in the model (see above) to obtain the total number of births by age.
Relative risks to MCH The estimates of relative risks to MCH are based
on reviews (Cnattingius, 2004; HHS, 2004) and recent studies (Aliyu et al.,
2010, 2011; Anderka et al., 2010; Dietz et al., 2010; Steyn et al., 2006; van
den Berg et al., 2013; Zhang and Wang, 2013) for each of the outcomes:
PTB, LBW, and SIDS. We estimate a relative risk of 1.4 for PTBs, 2.0 for
LBW and 2.5 for SIDS. We assumed the same relative risks for all women
(ages 15 to 49) and for all years. In addition, we do not distinguish risks
by smoking intensity or by the month of quitting if the women stopped
smoking at some point during pregnancy. The relative risks used to estimate
MCH outcomes are shown in Table D-4.
Maternal smoking prevalence Data were developed based on women

reporting smoking while pregnant. Data on the prevalence of pregnant
women for the United States were obtained from NCHS for 2012 dis
aggregated by age-group (15–19, 20–24, . . . , 40–44, 45–49) (HHS et al.,
2014b). These data are based on birth certificates and now cover most
of the states. Many states were excluded after the 2003 revision in how
TABLE D-4 Relative Risks Used in Estimating Maternal and Child

Health Outcomes in SimSmoke
MCH Outcome Best Estimate Lower Bound Upper Bound
Pre-term Birth (PTB) 1.4 1.1 1.7
Low Birth Weight (LBW) 2.0 1.5 2.5
Sudden Infant Death Syndrome (SIDS) 2.5 1.4 4.0

APPENDIX D 347
t obacco was reported because they did not adopt the 2003 revision, but by
2012 only 13 states (Alabama, Alaska, Arizona, Arkansas, Connecticut,
Hawaii, Maine, Michigan, Mississippi, New Jersey, Rhode Island, Virginia,
and West Virginia) were excluded. We confined the analysis to the year
2012, when most states’ reports had adopted the 2003 revisions, and proj-
ect forward from that year. The data are shown in Table D-5.
After comparing the prenatal smoking rates from NCHS to estimates
from the Pregnancy Risk Assessment Monitoring System (PRAMS) and the
National Survey on Drug Use and Health (NSDUH) for 2011, the most re-
cent year for which data from all three datasets are available, we found the
rates from NCHS (9.0 percent) were lower than from PRAMS (10.5 percent)
and NSDUH (15.9 percent). The NCHS data and also the PRAMS data are
known to consistently underestimate smoking rates because of underreport-
ing. For example, Tong et al. (2013) analyzed the PRAMS 2008 question-
naire and the eight states that also used the 2003 BC revision. Using the same
age stratification, the 20–24 age group again had the highest prevalence in
both the BC and PRAMS prevalence, but the combined prevalence for the
<20 age group had a slightly higher prevalence (22.6 percent compared to
22.5 percent). Tong and colleagues also found that the NCHS data under-
stated smoking prevalence compared to the combined estimates by 65 per-
TABLE D-5 Tobacco Use by Pregnant Women by Age of Mother, Data

from NCHS
2012
Age
Group <15 15–19 20–24 25–29 30–34 35–39 40–44 45–49 Total
Total 3.0 254.4 761.8 939.5 846.6 395.7 91.8 6.1 3,298.8
births*
% 2.8% 10.9% 13.5% 8.9% 5.6% 4.3% 4.2% 2.6% 8.6%
tobacco
use
% 2.6% 2.0% 2.0% 1.7% 1.6% 1.7% 1.8% 2.2% 1.7%
not
stated
% 17.1% 16.7% 16.9% 16.4% 16.5% 16.2% 16.2% 15.4% 16.5%
births,
tobacco
use not
reported
* In thousands; “total births” limited to births in those states for which tobacco use is re-
ported. Births in states that used incompatible birth certificates version are omitted from the
“total births.” “% births, tobacco use not reported” is the percentage of all births from all
states for that year.

cent for the <20 age group (13.7 percent in NCHS versus. 22.6 percent
combined), by 35 percent for the 20–24 age group (16.7 percent in NCHS
versus. 22.5 percent combined), by 27 percent for the 25–29 age group
(13.2 percent NCHS versus. 16.7 percent combined) and by 30 percent for
the 30 and above group (6 percent NCHS versus. 7.8 percent combined).
We applied these correction factors to the NCHS data.
We calibrated the 2012 smoking prevalence from SimSmoke to the
adjusted NCHS maternal smoking prevalence by determining adjustment
factors that equilibrated the smoking prevalence to the adjusted maternal
prevalence by 5-year age groups (15–19, 20–24, . . . , 45–49). In SimSmoke,
estimates of prenatal smoking prevalence may change as a result of policies
through changes in prevalence, initiation or cessation rates (see above). For
changes in the MLA, the changes only take place through the initiation rate.
Detailed Results for the MLA from SimSmoke

The status quo policy level for smoking rates among female adults (age
18 and above) is predicted to decline from 15 percent in 2015 to 10.8 per-
cent in 2065 and then to remain steady at that rate. The prevalence shows
little fluctuation throughout the years for females in age group 15–17, with
only small reductions for those ages 18–20 and 21–24.
The results for changes in the MLA are based on best estimates of
their effects with the lower and upper plausible ranges in parentheses. The
relative percentage reductions in smoking rates for each age group increase
with the MLA. For example, by implementing the new MLA in 2015, the
smoking prevalence of adult (ages 18 and above) females in the year 2010
is projected to fall relative to the status quo by 3.1 percent (range 2.2 to
4.1 percent) under an MLA of 19, 11.6 percent (range 9.4 to 14.2 percent)
under MLA 21, and 16.5 percent (range 11.7 to 23.2 percent) under MLA
25. Due to the assumption of a 2-year initiation rebound for MLA 19 and
MLA 21, slight increases in smoking prevalence and MCH outcomes for the
age of the MLA and the next age are predicted in the early years.
LBW Under the status quo, in 2015 the incidence rate of smoking-at-
tributable LBW babies is about 0.8 percent among the total births for all
the women of childbearing age (ages 14–49), but 1.3 percent for the ages
20–24 years. The rates decrease after 2015, except for the maternal age
group 15–19, in which the rate increases to 1.6 percent in 2100 due to the
sustained growth of the fertility rate. For all women of childbearing age, an
estimated 3.8 million LBW infants are projected to be born between 2015
and 2100 because of the mother’s prenatal smoking.
By raising the MLA to 19, SimSmoke estimates that there will be a
cumulative total of 2,000 LBW outcomes averted (range 1,200–2,800)

APPENDIX D 349
in the first 5 years (2015–2020), 60,700 (range 44,600–79,000) within

50 years, and 122,800 (range 90,700–159,200) within 85 years. If the
MLA is increased to 21, the number of averted cases each year will be
more than twice as high as for MLA 19. For all women of childbearing
age, an MLA of 21 is predicted to avert about 217,900 LBW cases (range
176,700–267,000) from 2015 to 2065 and about 435,100 cases (range
353,500–532,600) between 2015 and 2100. Increasing the MLA to 25 is
predicted to avert a total of 593,000 LBW cases averted (range 419,100–
842,800) within 85 years.
PTB Under the status quo, the smoking-attributable PTB incidence rate
for mothers of age 15–19 increases slightly, from 0.94 percent in 2015 to
0.95 percent in 2100, while the rates for other age groups all show slight
declines. For all women of childbearing age, the incidence rate is 0.51 per-
cent in 2015, decreasing to 0.49 percent by 2100. Because of escalating
birth rates, however, the number of smoking-attributable PTBs is estimated
to increase from 20,800 in 2015 to 28,200 in 2065 and 33,500 in 2100. A
total of about 2,307,000 smoking-attributable PTBs are predicted between
2015 and 2100.
An MLA of 19 is estimated to prevent a total of 1,300 smoking-
attributable PTBs (range 810–1,880) for all women of childbearing age
compared to the status quo level over the first 5 years, an additional 39,000
(range 29,500–52,000) within 50 years, and a total of 81,000 (range
60,000–105,000) for the entire span from 2015 to 2100. By increasing the
MLA to 21, the number of averted cases is predicted to be more than two
times higher than for an MLA of 19 for the age group 15–19 and more than
three times higher for age groups 20–24, 25–34, and 35–49. For all women,
MLA 21 is predicted to prevent a total of about 142,000 PTB cases (range
116,000–174,000) by 2065 and 283,300 cases (range 231,000–346,000) by
2100. Increasing the MLA to 25 is predicted to prevent a total of 385,000
PTBs (range 273,000–543,000) between 2015 and 2100.
SIDS Under the status quo policy, the incidence rate of smoking-attribut-
able SIDS for the maternal age group 15–49 is 0.008 percent in 2015. Since
birth rates are projected to increase, the estimated smoking-attributable
SIDS cases will slightly increase over time. From 2015 to 2100, the total
number of annual SIDS cases will increase by about 200 (from 320 to 520),
with a total of 35,600 smoking-attributable deaths over that period.
Over the period from 2015 to 2100, SimSmoke predicts a total of 1,100
(range 832–1,455) SIDS deaths would be averted by raising the MLA to
19; a total of 3,980 (range 3,200–4,900) deaths would be averted (range
3,200–4,900) under MLA 21; and 5,400 (range 3,800–7,700) deaths would
be averted under MLA 25.

In summary, raising the MLA to 19, 21, or 25 is projected to have an

increasingly larger impact on LBW, PTB, and SIDS, especially in raising the
MLA to 21. Between 2015 and 2100, about 122,800 LBW cases, 80,900
PTBs, and 1,100 SIDS cases are projected to be averted under MLA 19;
this would increase to 435,100 LBW cases, 283,300 PTBs, and 3,980 SIDS
cases under MLA 21; and it would become 593,000 LBW cases, 384,600
PTBs, and 5,400 SIDS cases under MLA 25. Thus, about three times more
cases would be averted under MLA 21 than MLA 19, and about 1.35 times
more cases would be prevented under MLA 25 than MLA 21.
In applying SimSmoke to estimate adverse birth outcomes, five limita-
tions merit consideration: (1) The analysis does not distinguish the overlap
in diagnosis between LBW babies and PTBs. Consequently, the sum of the
two outcomes is an overstatement. (2) The analysis does not specifically
incorporate the time quit or the amount of cigarettes smoked by those who
continue smoking while pregnant. The analyses can be extended to consider
these factors as well as to allow for age-specific variations in relative risks as
better information becomes available. (3) In examining maternal smoking
over time, important differences in smoking behaviors by socioeconomic
status were not considered. (4) The under-diagnosis and underreporting
of adverse MCH outcomes merit further consideration as they apply to
estimating smoking-attributable risks. (5) The model does not directly in-
corporate changes in policies that target pregnant smokers such as smoking
cessation and other health care–related programs.
RESULTS FROM UPPER AND LOWER SCENARIOS

FOR INITIATION RATE ASSUMPTIONS FOR
DIFFERENT MLA POLICY OPTIONS
We present smoking prevalence, mortality, and health outcome projec-
tions from both models under the upper and lower initiation scenarios.
Corresponding figures and tables for the mid-initiation scenario are shown
in Chapter 8. Figures D-2 through D-5 show projections of smoking preva-
lence from 2015 to 2100. Tables D-6 through D-9 show premature deaths
prevented for selected years for both models. Tables D-10 through D-13
show projected years of life lost for the CISNET model. Figures D-6 and
D-7 show projected cumulative lung cancer deaths prevented according to
the CISNET model.
ADDITIONAL MODEL OUTPUTS

In this section we present additional outcomes from the CISNET model.
Figures D-8 and D-9 show projections of premature deaths due to smoking
from 2000 to 2100 by gender for all initiation scenarios. The figures also

APPENDIX D 351

8 1 12 1 16 1
Status Quo
MLA 19
MLA 21
MLA25
6
2020 2040 2060 2080 2100

Year
Women Men
10 14

8 12 16 20
8
6
2020 2060 2100 2020 2060 2100

Year Year
FIGURE D-2 CISNET model–projected smoking prevalence for the upper scenarios
of the three MLA policy options for adults (18+), adult women, and adult men in
the United States for 2014–2100.
show projections under an idealized scenario where all smoking initiation

stops in 2015 (Ideal).
Figures D-10 and D-11 show projected mean-pack years for adults ages
40 or older for all initiation scenarios.


Status Quo
8 1 12 1 16 1
MLA 19
MLA 21
MLA25
6
2020 2040 2060 2080 2100

Year
Women Men
10 1 14 1

8 12 16 20
8
6
2020 2060 2100 2020 2060 2100

Year Year
FIGURE D-3 CISNET model–projected smoking prevalence for the lower scenarios
of the three MLA policy options for adults (18+), adult women, and adult men in
the United States for 2014–2100.

APPENDIX D 353
10 12 14 16 18 Adult Smoking Prevalence

Status Quo
MLA 19
MLA 21
MLA25
8
2020 2040 2060 2080 2100

Year
Women 10 12 14 16 18 20 Men
16

14
12
10
8
2020 2060 2100 2020 2060 2100

Year Year
FIGURE D-4 SimSmoke model–projected smoking prevalence for the upper sce-
narios of the three MLA policy options for adults (18+), adult women, and adult
men in the United States for 2014–2100.

10 12 14 16 18 Adult Smoking Prevalence

Status Quo
MLA 19
MLA 21
MLA25
8
2020 2040 2060 2080 2100

Year
Women 10 12 14 16 18 20 Men
16

14
12
10
8
2020 2060 2100 2020 2060 2100

Year Year
FIGURE D-5 SimSmoke model–projected smoking prevalence for the lower sce-
narios of the three MLA policy options for adults (18+), adult women, and adult
men in the United States for 2014–2100.

APPENDIX D 355
TABLE D-6 Cumulative Premature Deaths Expected and Prevented by

Period: CISNET Upper Scenarios
MLA/Outcome 2020–2039 2040–2059 2060–2079 2080–2099 2015–2100
Status Quo
expected
MLA 19
MLA 21
MLA 25
NOTE: Assumes upper scenarios and that the policy is implemented in 2015. Although the
table carries many significant figures to aid in reproducibility, precision is limited to one or
two digits.

Period: CISNET Lower Scenarios
MLA/Outcome 2020–2039 2040–2059 2060–2079 2080–2099 2015–2100
Status Quo
expected
MLA 19
MLA 21
MLA 25
NOTE: Assumes lower scenarios and that the policy is implemented in 2015. Although the
two digits.


Period: SimSmoke Upper Scenarios
MLA/Outcome 2020–2039 2040–2059 2060–2079 2080–2099 2015–2100
Status Quo
expected
MLA 19
MLA 21
MLA 25
Deaths prevented 5,000 139,000 528,000 873,000 1,546,000
NOTE: Assumes upper scenarios and that the policy is implemented in 2015. Although the
two digits.

Period: SimSmoke Lower Scenarios
MLA/Outcome 2020–2039 2040–2059 2060–2079 2080–2099 2015–2100
Status Quo
expected
MLA 19
MLA 21
MLA 25
NOTE: Assumes lower scenarios and that the policy is implemented in 2015. Although the
two digits.

TABLE D-10 Years of Life Lost (YLL) by Period: CISNET Model, Upper Scenario
MLA 19 Upper MLA 19 % MLA 21 Upper MLA 21 % MLA 25 MLA 25 %
Status Quo Scenario Reduction Scenario Reduction Upper Scenario Reduction
2000–2019 134,823,000 — 0.0% — 0.0% — 0.0%
2020–2039 106,126,000 — 0.0% — 0.0% — 0.0%
2040–2059 68,217,000 128,000 0.2% 429,000 0.6% 662,000 1.0%
2060–2079 46,490,000 7,212,000 1.6% 2,416,000 5.2% 3,731,000 8.0%
2080–2099 36,688,000 1,240,000 3.4% 4,152,000 11.3% 6,416,000 17.5%
TABLE D-11 Lifetime Years of Life Lost (YLL) by Cohort: CISNET Model, Upper Scenario
MLA 19 Upper MLA 19 % MLA 21 Upper MLA 21 % MLA 25 Upper MLA 25 %
Status Quo Scenario Reduction Scenario Reduction Scenario Reduction
2000–2019 40,116,000 1,518,000 3.8% 5,082,000 12.7% 7,855,000 19.6%
2020–2039 36,447,000 1,459,000 4.0% 4,884,000 13.4% 7,547,000 20.7%
2040–2059 36,084,000 1,445,000 4.0% 4,837,000 13.4% 7,475,000 20.7%
2060–2079 37,412,000 1,498,000 4.0% 5,015,000 13.4% 7,750,000 20.7%

2080–2099 38,874,000 1,557,000 4.0% 5,211,000 13.4% 8,053,000 20.7%

357
TABLE D-12 Years of Life Lost (YLL) by Period: CISNET Model, Lower Scenario
358

MLA 19 Lower MLA 19 % MLA 21 Lower MLA 21 % MLA 25 MLA 25 %
Status Quo Scenario Reduction Scenario Reduction Lower Scenario Reduction
2000–2019 134,823,000 — 0.0% — 0.0% — 0.0%
2020–2039 106,126,000 — 0.0% — 0.0% — 0.0%
2040–2059 68,217,000 83,000 0.1% 292,000 0.4% 333,000 0.5%
2060–2079 46,490,000 467,000 1.0% 1,645,000 3.5% 1,875,000 4.0%
2080–2099 36,688,000 803,000 2.2% 2,827,000 7.7% 3,224,000 8.8%
TABLE D-13 Lifetime Years of Life Lost (YLL) by Cohort: CISNET Model, Lower Scenario
MLA 19 Lower MLA 19 % MLA 21 Lower MLA 21 % MLA 25 Lower MLA 25 %
Status Quo Scenario Reduction Scenario Reduction Scenario Reduction
2000–2019 40,116,000 982,000 2.4% 3,459,000 8.6% 3,946,000 9.8%
2020–2039 36,447,000 944,000 2.6% 3,324,000 9.1% 3,792,000 10.4%
2040–2059 36,084,000 935,000 2.6% 3,292,000 9.1% 3,755,000 10.4%
2060–2079 37,412,000 970,000 2.6% 3,414,000 9.1% 3,894,000 10.4%

2080–2099 38,874,000 1,007,000 2.6% 3,547,000 9.1% 4,046,000 10.4%

APPENDIX D 359
MLA19
MLA21
MLA25
100,000
60,000
20,000
0
2020 2040 2060 2080 2100

Year
FIGURE D-6 CISNET model–estimated number of cumulative lung cancer deaths
prevented per year for the three MLA policy options: Upper scenarios.

MLA19
MLA21
MLA25
100,000
60,000
20,000
0
2020 2040 2060 2080 2100

Year
FIGURE D-7 CISNET model–estimated number of cumulative lung cancer deaths

prevented per year for the three MLA policy options: Lower scenarios.

APPENDIX D 361
140,000
Number of Premature Deaths
120,000
100,000
80,000
60,000
40,000
20,000
0
2000 2020 2040 2060 2080 2100
Year
Status quo Estimate Ideal
FIGURE D-8 CISNET model–projected number of female deaths prevented per

year for MLA 21. Ideal represents a scenario where no smoking initiation occurs
after 2015.

350,000
300,000
Number of Premature Deaths
250,000
200,000
150,000
100,000
50,000
0
2000 2020 2040 2060 2080 2100
Year
Status quo Esmate Ideal
FIGURE D-9 CISNET model–projected number of male deaths prevented per year
for MLA 21. Ideal represents a scenario where no smoking initiation occurs after
2015.
12.0
Mean Pack-Years for Ages 40+
10.0
8.0
6.0
4.0
2.0
0.0
2000 2020 2040 2060 2080 2100
Year
Status quo Estimate Ideal
FIGURE D-10 CISNET model–projected mean smoking pack-years for women age
40 or older for MLA 21.

APPENDIX D 363
20
Mean Pack-Years for Ages 40+
18
16
14
12
10
8
6
4
2
0
2000 2020 2040 2060 2080 2100
Year
Status quo Es mate Ideal
FIGURE D-11 CISNET model–projected mean smoking pack-years for men age
40 or older for MLA 21.

REFERENCES
Aliyu, M. H., O. Lynch, R. Saidu, A. P. Alio, P. J. Marty, and H. M. Salihu. 2010. Intrauter-
ine exposure to tobacco and risk of medically indicated and spontaneous preterm birth.
American Journal of Perinatology 27(5):405–410.
Aliyu, M. H., O. Lynch, R. E. Wilson, A. P. Alio, S. Kristensen, P. J. Marty, V. E. Whiteman,
and H. M. Salihu. 2011. Association between tobacco use in pregnancy and placenta-
associated syndromes: A population-based study. Archives of Gynecology and Obstetrics
283(4):729–734.
Anderka, M., P. A. Romitti, L. Sun, C. Druschel, S. Carmichael, G. Shaw, and National Birth
Defects Prevention Study. 2010. Patterns of tobacco exposure before and during preg-
nancy. Acta Obstetricia et Gynecologica Scandinavica 89(4):505–514.
Anderson, C. M., D. M. Burns, K. W. Dodd, and E. J. Feuer. 2012. Chapter 2: Birth-cohort-
specific estimates of smoking behaviors for the U.S. population. Risk Analysis 32(Suppl
1):S14–S24.
Azagba, S., and M. F. Sharaf. 2013. The effect of graphic cigarette warning labels on smok-
ing behavior: Evidence from the Canadian experience. Nicotine & Tobacco Research
15(3):708–717.
Burns, D., L. Garfinkel, and J. Samet (eds.). 1997. Changes in cigarette-related disease risks and
their implication for prevention and control. Bethesda, MD: National Cancer Institute.
CDC (Centers for Disease Control and Prevention). 2014. CDC WONDER. http://wonder.
cdc.gov (accessed January 16, 2015).
Cnattingius, S. 2004. The epidemiology of smoking during pregnancy: Smoking prevalence,
maternal characteristics, and pregnancy outcomes. Nicotine & Tobacco Research 6(Suppl
2):S125–S140.
de Koning, H. J., R. Meza, S. K. Plevritis, K. ten Haaf, V. N. Munshi, J. Jeon, S. A. Erdogan, C.
Y. Kong, S. S. Han, J. van Rosmalen, S. E. Choi, P. F. Pinsky, A. Berrington de G onzalez,
C. D. Berg, W. C. Black, M. C. Tammemagi, W. D. Hazelton, E. J. Feuer, and P. M.
McMahon. 2014. Benefits and harms of computed tomography lung cancer screening
strategies: A comparative modeling study for the U.S. Preventive Services Task Force.
Annals of Internal Medicine 160(5):311–320.
Dietz, P. M., L. J. England, C. K. Shapiro-Mendoza, V. T. Tong, S. L. Farr, and W. M.
Callaghan. 2010. Infant morbidity and mortality attributable to prenatal smoking in the
U.S. American Journal of Preventive Medicine 39(1):45–52.
Friend, K., and D. T. Levy. 2001. Smoking treatment interventions and policies to promote
their use: A critical review. Nicotine & Tobacco Research 3(4):299–310.
———. 2002. Reductions in smoking prevalence and cigarette consumption associated with
mass-media campaigns. Health Education Research 17(1):85–98.
Gilpin, E. A., J. P. Pierce, and A. J. Farkas. 1997. Duration of smoking abstinence and success
in quitting. Journal of the National Cancer Institute 89(8):572–576.
Hammond, D., G. T. Fong, R. Borland, K. M. Cummings, A. McNeill, and P. Driezen.
2007. Text and graphic warnings on cigarette packages: Findings from the Interna-
tional Tobacco Control Four Country Study. American Journal of Preventive Medicine
32(3):202–209.
Hazelton, W. D., J. Jeon, R. Meza, and S. H. Moolgavkar. 2012. Chapter 8: The FHCRC lung
cancer model. Risk Analysis 32(Suppl 1):S99–S116.
HHS (Department of Health and Human Services). 1990. The health benefits of smoking
cessation: A report of the Surgeon General. Atlanta Georgia: U.S. Department of Health
and Human Services, Public Health Service, Centers for Disease Control, Office on
Smoking and Health.

APPENDIX D 365
———. 2004. The health consequences of smoking: A report of the Surgeon General. Atlanta,
HHS, CDC, NCHS, and DVS (Department of Health and Human Services, Centers for Disease
Control and Prevention, National Center for Health Statistics, and Department of Vital
Statistics). 2014a. CDC WONDER: Linked birth/infant death records for 2011. http://
wonder.cdc.gov/lbd.html (accessed February 25, 2015).
———. 2014b. CDC WONDER: Natality public-use data for 2012. http://wonder.cdc.gov/
natality-current.html (accessed February 25, 2015).
Holford, T. R., and L. Clark. 2012. Chapter 4: Development of the counterfactual smoking
histories used to assess the effects of tobacco control. Risk Analysis 32(Suppl 1):S39–S50.
Holford, T. R., D. T. Levy, L. A. McKay, L. Clarke, B. Racine, R. Meza, S. Land, J. Jeon,
and E. J. Feuer. 2014a. Patterns of birth cohort-specific smoking histories, 1965–2009.
American Journal of Preventive Medicine 46(2):e31–e37.
Holford, T. R., R. Meza, K. E. Warner, C. Meernik, J. Jeon, S. H. Moolgavkar, and D. T. Levy.
2014b. Tobacco control and the reduction in smoking-related premature deaths in the
United States, 1964–2012. JAMA 311(2):164–171.
Huang, J., F. J. Chaloupka, and G. T. Fong. 2014. Cigarette graphic warning labels and
smoking prevalence in Canada: A critical examination and reformulation of the FDA
regulatory impact analysis. Tobacco Control 23(Suppl 1):i7–i12.
Hughes, J. R., E. N. Peters, and S. Naud. 2008. Relapse to smoking after 1 year of abstinence:
A meta-analysis. Addictive Behaviors 33(12):1516–1520.
Jeon, J., R. Meza, M. Krapcho, L. D. Clarke, J. Byrne, and D. T. Levy. 2012. Chapter 5: Actual
and counterfactual smoking prevalence rates in the U.S. population via microsimulation.
Risk Analysis 32(Suppl 1):S51–S68.
Levin, M. 1953. The occurrence of lung cancer in man. Acta Unio Internationalis Contra
Cancrum 9(3):531–541.
Levy, D. T., and K. B. Friend. 2000. A simulation model of tobacco youth access policies.
Journal of Health Politics, Policy and Law 25(6):1023–1050.
———. 2001. A computer simulation model of mass media interventions directed at tobacco
use. Preventive Medicine 32(3):284–294.
———. 2002a. Examining the effects of tobacco treatment policies on smoking rates and
smoking related deaths using the SimSmoke computer simulation model. Tobacco Con-
trol 11(1):47–54.
———. 2002b. A simulation model of policies directed at treating tobacco use and depen-
dence. Medical Decision Making 22(1):6–17.
Levy, D. T., K. M. Cummings, and A. Hyland. 2000a. Increasing taxes as a strategy to
reduce cigarette use and deaths: Results of a simulation model. Preventive Medicine
31(3):279–286.
———. 2000b. A simulation of the effects of youth initiation policies on overall cigarette use.
Levy, D. T., K. B. Friend, H. Holder, and M. Carmona. 2001a. Effect of policies directed
at youth access to smoking: Results from the SimSmoke computer simulation model.
Tobacco Control 10(2):108–116.

Levy, D. T., K. B. Friend, and E. Polishchuk. 2001b. Effect of clean indoor air laws on s mokers:
The clean air module of the SimSmoke computer simulation model. Tobacco Control
10(4):345–351.
Levy, D. T., F. Chaloupka, and J. G. Gitchell. 2004. The effects of tobacco control policies on
smoking rates: A tobacco control scorecard. Journal of Public Health Management and
Practice 10(4):338–353.
Levy, D. T., N. Nikolayev, and E. A. Mumford. 2005. Recent trends in smoking and the role
of public policies: Results from the SimSmoke tobacco control policy simulation model.
Addiction 10(10):1526–1537.
Levy, D. T., A. Hyland, C. Higbee, L. Remer, and C. Compton. 2007. The role of public poli-
cies in reducing smoking prevalence in California: Results from the California tobacco
policy simulation model. Health Policy 82(2):167–185.
Levy, D. T., C. Tworek, E. J. Hahn, and R. E. Davis. 2008. The Kentucky SimSmoke tobacco
policy simulation model: Reaching Healthy People 2010 goals through policy change.
Southern Medical Journal 101(5):503–507.
Levy, D. T., S. I. Cho, Y. M. Kim, S. Park, M. K. Suh, and S. Kam. 2010a. SimSmoke model
evaluation of the effect of tobacco control policies in Korea: The unknown success story.
Levy, D. T., A. L. Graham, P. L. Mabry, D. B. Abrams, and C. T. Orleans. 2010b. Modeling
the impact of smoking cessation treatment policies on quit rates. American Journal of
Preventive Medicine 38(3S):S364–S372.
Levy, D., L. M. de Almeida, and A. Szklo. 2012a. The Brazil SimSmoke policy simulation
model: The effect of strong tobacco control policies on smoking prevalence and smoking-
attributable deaths in a middle income nation. PLoS Medicine 9(11): e1001336.
Levy, D. T., R. G. Boyle, and D. B. Abrams. 2012b. The role of public policies in reducing
smoking: The Minnesota SimSmoke tobacco policy model. American Journal of Preven-
tive Medicine 43(5 Suppl 3):S179–S186.
Levy, D. T., K. Blackman, L. M. Currie, and U. Mons. 2013a. Germany SimSmoke: The
effect of tobacco control policies on future smoking prevalence and smoking-attributable
deaths in Germany. Nicotine & Tobacco Research 15(2):465–473.
Levy, D. T., L. Currie, and L. Clancy. 2013b. Tobacco control policy in the UK: Blueprint for
the rest of Europe? European Journal of Public Health 23(2):201–206.
Levy, D., R. L. Rodriguez-Buno, T. W. Hu, and A. E. Moran. 2014. The potential effects of
tobacco control in China: Projections from the China SimSmoke simulation model. BMJ
348:g1134.
Levy, D., K. Blackman, G. T. Fong, and F. Chaloupka (eds.). In press. The role of tobacco
control policies in reducing smoking and deaths in the eighteen heavy burden nations:
Results from the MPOWER SimSmoke tobacco control policy model. In National Cancer
Institute monograph Tobacco control policies in low and middle income nations, edited
by G. Fong, F. Chaloupka, and A. Yurekli. Rockville, MD: National Cancer Institute.
Lewit, E. M., D. Coate, and M. Grossman. 1981. The effects of government regulation on
teenage smoking. Journal of Law and Economics 24(3):545–569.
Lilienfeld, A. M. 1980. Advances in quantitative methods in epidemiology. Public Health
Reports 95(5):462–469.
Matthews, T. J., and M. F. MacDorman. 2013. Infant mortality statistics from the 2010 period
linked birth/infant death data set. National Vital Statistics Reports 62(8):1–26.

APPENDIX D 367
McMahon, P. M., R. Meza, S. K. Plevritis, W. C. Black, C. M. Tammemagi, A. Erdogan, K.

ten Haaf, W. Hazelton, T. R. Holford, J. Jeon, L. Clarke, C. Y. Kong, S. E. Choi, V. N.
Munshi, S. S. Han, J. van Rosmalen, P. F. Pinsky, S. Moolgavkar, H. J. de Koning, and
E. J. Feuer. 2014. Comparing benefits from many possible computed tomography lung
cancer screening programs: Extrapolating from the National Lung Screening Trial using
comparative modeling. PLoS ONE 9(6):e99978.
Melvin, C. L., E. K. Adams, and V. Miller. 2000. Costs of smoking during pregnancy: Devel-
opment of the Maternal and Child Health Smoking Attributable Mortality, Morbidity
and Economic Costs (MCHSAMMEC) software. Tobacco Control 9(Suppl 3):iii12–iii15.
Meza, R., W. D. Hazelton, G. A. Colditz, and S. H. Moolgavkar. 2008. Analysis of lung cancer
incidence in the Nurses’ Health and the Health Professionals’ Follow-Up Studies using a
multistage carcinogenesis model. Cancer Causes and Control 19(3):317–328.
Meza, R., K. ten Haaf, C. Y. Kong, A. Erdogan, W. C. Black, M. C. Tammemagi, S. E. Choi,
J. Jeon, S. S. Han, V. Munshi, J. van Rosmalen, P. Pinsky, P. M. McMahon, H. J. de
Koning, E. J. Feuer, W. D. Hazelton, and S. K. Plevritis. 2014. Comparative analysis of 5
lung cancer natural history and screening models that reproduce outcomes of the NLST
and PLCO trials. Cancer 120(11):1713–1724.
Moolgavkar, S. H., T. R. Holford, D. T. Levy, C. Y. Kong, M. Foy, L. Clarke, J. Jeon, W. D.
Hazelton, R. Meza, F. Schultz, W. McCarthy, R. Boer, O. Gorlova, G. S. Gazelle, M.
Kimmel, P. M. McMahon, H. J. de Koning, and E. J. Feuer. 2012. Impact of reduced
tobacco smoking on lung cancer mortality in the United States during 1975–2000. Jour-
nal of the National Cancer Institute 104(7):541–548.
Orzechowski, W., and R. C. Walker. 2012. The tax burden on tobacco: Historical compilation,
Vol. 47, 2012. Arlington, VA: Orzechowski and Walker. http://www.taxadmin.org/fta/
tobacco/papers/tax_burden_2012.pdf (accessed February 23, 2015).
———. 2013. The tax burden on tobacco: Historical compilation, Vol. 48, 2013. Arlington, VA:
Orzechowski and Walker. http://www.taxadmin.org/fta/tobacco/papers/tax_burden_2013.
pdf (accessed March 20, 2015).
Rosenberg, M. A., E. J. Feuer, B. Yu, J. Sun, S. J. Henley, T. G. Shanks, C. M. Anderson, P. M.
McMahon, M. J. Thun, and D. M. Burns. 2012. Chapter 3: Cohort life tables by smoking
status, removing lung cancer as a cause of death. Risk Analysis 32 (Suppl 1):S25–538.
Sprague, W. W. 2009. LCFIT: The Lee–Carter fitter (Web-based software for estimating the
Lee–Carter method for modeling and forecasting mortality). http://lcfit.demog.berkeley.
edu (accessed January 16, 2015).
Steyn, K., T. De Wet, Y. Saloojee, H. Nel, and D. Yach. 2006. The influence of maternal ciga-
rette smoking, snuff use and passive smoking on pregnancy outcomes: The Birth to Ten
Study. Paediatric and Perinatal Epidemiology 20(2):90–99.
Tong, V. T., P. M. Dietz, S. L. Farr, D. V. D’Angelo, and L. J. England. 2013. Estimates of
smoking before and during pregnancy, and smoking cessation during pregnancy: Com-
paring two population-based data sources. Public Health Reports 128(3):179–188.
USCB (U.S. Census Bureau). 2013a. Annual estimates of the resident population by single
year of age and sex for the United States, states, and Puerto Rico commonwealth:
April 1, 2010 to July 1, 2012. http://factfinder.census.gov/faces/tableservices/jsf/pages/
productview.xhtml?src=bkmk (accessed November 26, 2014).
———. 2013b. National estimates by age, sex, race: 1900–1979. https://www.census.gov/
popest/data/national/asrh/pre-1980/PE-11.html (accessed November 26, 2014).
———. 2014. Methodology, assumptions, and inputs for the 2014 national projections. http://
www.census.gov/population/projections/files/methodology/methodstatement14.pdf (ac-
cessed February 23, 2015).

van den Berg, G., M. van Eijsden, F. Galindo-Garre, T. G. Vrijkotte, and R. J. Gemke. 2013.
Smoking overrules many other risk factors for small for gestational age birth in less
educated mothers. Early Human Development 89(7):497–501.
Warner, K. 1989. Effects of the antismoking campaign: An update. American Journal of Public
Health 79(2):144–151.
Warner, K. E., and H. A. Murt. 1983. Premature deaths avoided by the antismoking campaign.
WHO (World Health Organization). 2008. WHO report on the global tobacco epidemic,
2008: The MPOWER package. Geneva: World Health Organization.
———. 2013. WHO report on the global tobacco epidemic, 2013: Enforcing bans on tobacco
advertising, promotion and sponsorship. Geneva: World Health Organization.
Zhang, K., and X. Wang. 2013. Maternal smoking and increased risk of sudden infant death
syndrome: A meta-analysis. Legal Medicine 15(3):115–121.

Appendix E
Open Meeting Agendas
MEETING ONE
Tuesday, February 4, 2014
National Academy of Sciences, Board Room
2101 Constitution Avenue, NW
Washington, DC
10:30 a.m. Convene Open Session
Introductions
Richard Bonnie
Presentation of the Charge to the Committee and Discussion

Eric Lindblom
Director of the Office of Policy
Center for Tobacco Products, Food and Drug Administration
11:30 a.m. State and Local Efforts in Raising the Minimum Purchase Age
Peter H. Fisher
Vice President, State Issues
Campaign for Tobacco-Free Kids
12:30 p.m. Adjourn Open Session
369

MEETING TWO
Thursday, April 10, 2014
Keck Center, Room 100
500 Fifth Street, NW
Washington, DC
OPEN SESSION
8:30 a.m. Adolescent and Young Adult Brain Development
Jay Giedd, M.D.
Chief, Brain Imaging Section
Child Psychiatry Branch, National Institute of Mental Health
9:30 a.m. Adolescent and Young Adult Cognitive and Psychosocial

Development and Decision Making
Laurence Steinberg, Ph.D.
Professor, Department of Psychology
Temple University
10:30 a.m. Break
10:45 a.m. Effect of Nicotine on the Developing Brain

Neal Benowitz, M.D.
Professor, Department of Medicine
University of California, San Francisco, School of Medicine
11:45 a.m. Emerging and Alternative Nicotine and Tobacco Products

David Abrams, Ph.D.
Executive Director, Schroeder Institute for Tobacco Research
and Policy Studies
Legacy for Health
12:45 p.m. Lunch
Advertising, Marketing, and Messaging Strategies for

Adolescents and Young Adults
1:45 p.m. Tobacco Industry Messaging Strategies

Pam Ling, M.D., M.P.H.
Associate Professor, Department of Medicine
University of California, San Francisco

APPENDIX E 371
2:45 p.m. The Role of Media Channels and Messages in Shaping U.S.
Tobacco Use Patterns
Donna Vallone, Ph.D., M.P.H.
Senior Vice President, Research and Evaluation
Legacy for Health
3:45 p.m. Reflections: Clinical Pediatric Perspective

Jonathan Winickoff, M.D., M.P.H.
Associate Professor of Pediatrics
Massachusetts General Hospital for Children
4:15 p.m. Public Comment
Closing Comments
Richard Bonnie, Committee Chair
(When no additional public comments, adjourn open session.)


Appendix F
Committee Biographical Sketches
RICHARD J. BONNIE, L.L.B. (Chair), is the director of the Institute of

Law, Psychiatry, and Public Policy at the University of Virginia. He is also
a charter fellow of the College on the Problems of Drug Dependence, where
he has served twice on the board of directors. In addition to these positions,
Bonnie is the Harrison Foundation Professor of Medicine and Law in the
School of Law and a professor of psychiatry and neurobehavioral sciences
in the School of Medicine at the University of Virginia. His research focuses
on criminal law, bioethics, and public policies relating to mental health,
substance abuse, aging, and public health. Mr. Bonnie received his law
degree from the University of Virginia. He has been a major contributor to
the Institute of Medicine (IOM) and the National Research Council (NRC),
where he has served on multiple committees, including the Workshop for
Understanding the Demand for Illegal Drugs (2010); the Committee on
Reducing Tobacco Use: Strategies, Barriers and Consequences (chair, 2007);
the Committee on Developing a Strategy to Prevent and Reduce Underage
Drinking (chair, 2004); the Committee on Data and Research for Policy on
Illegal Drugs (2001); and the Committee on Preventing Nicotine Depen-
dence in Children and Adolescents (vice chair, 1994). He is currently the
chair of the Committee on Improving the Health, Safety and Well-Being of
Young Adults (2013–2015). Mr. Bonnie was elected to the IOM in 1991.
ANTHONY J. ALBERG, Ph.D., M.P.H., is the Blatt Ness Endowed Chair

in Oncology and a professor of public health sciences and currently serves
as the interim director of cancer control of the Hollings Cancer Center at
the Medical University of South Carolina. He is an epidemiologist whose
373

research focuses on non-melanoma skin cancer, cigarette smoking, health

effects of secondhand smoke, etiology of tobacco-associated malignancies,
and tobacco prevention and control. Dr. Alberg is a member of the editorial
board for cancer screening and prevention of the National Cancer Insti-
tute’s Physician Data Query (PDQ), editor for the epidemiology section of
the American College of Chest Physician’s Lung Cancer Guidelines III, and
associate editor of the American Journal of Epidemiology. He was a stand-
ing member of the National Institutes of Health’s epidemiology of cancer
study section, and he has been a contributing author to two U.S. Surgeon
General’s reports on the health consequences of smoking. He is a member
of South Carolina’s Cancer Control Advisory Committee (CCAC) and
chair of the CCAC’s Cancer Surveillance Committee. Dr. Alberg received
his M.P.H. from the Yale School of Medicine and his Ph.D. from the Johns
Hopkins Bloomberg School of Public Health.
REGINA BENJAMIN, M.D., M.B.A., is the NOLA.com/Times Picayune

Endowed Chair of Public Health Sciences at Xavier University of Louisiana
and from 2009 to 2013 served as the 18th Surgeon General of the United
States. She specializes in prevention policies and health promotion among
both individuals and large populations, especially concerning obesity, child-
hood obesity, and children’s health. She has special interests in rural health
care, health disparities among socioeconomic groups, suicide, violence,
and mental health. Prior to her role as Surgeon General, Dr. Benjamin
founded and directed a rural primary care health clinic in Bayou La Batre,
Alabama, where she administered health care to residents without easy
access to doctors. Dr. Benjamin attended Morehouse School of Medicine,
and she received her M.D. from the University of Alabama, Birmingham,
and her M.B.A. from Tulane University. She has served on several IOM
and NRC committees, including the Committee on Health Threats and
Workforce Resilience (2009); the Committee on a Comprehensive Review
of the DHHS Office of Family Planning Title X Program (2007–2009); the
Committee on Cancer Survivorship: Improving Care and Quality of Life
(2004–2005); the Committee on Crossing the Quality Chasm-Next Steps
Summit (2003–2004); and the Committee on Cancer Research Among Mi-
norities and the Medically Underserved (1997–1999). She was elected to
the IOM in 1997. She became a MacArthur Fellow in 2008.
JONATHAN CAULKINS, Ph.D., is the H. Guyford Stever Professor of

Operations Research at Carnegie Mellon University’s Heinz College of Pub-
lic Policy and Management. Dr. Caulkins was also the founding director of
the Pittsburgh branch of RAND, a corporation designed to improve deci-
sion making and to implement better policy worldwide. His main research
methods are focused on mathematical problem solving and model develop-

APPENDIX F 375
ment related to social policy and interventions. His interests are focused on
drugs, crime, delinquency, and prevention. He received his M.S. in systems
science and mathematics from Washington University and his S.M. in elec-
trical engineering and computer science from the Massachusetts Institute
of Technology (MIT). He holds a Ph.D. in operations research, also from
MIT. For the IOM and the NRC, he has served on the Committee on Re-
ducing Tobacco Use: Strategies, Barriers, and Consequences (2007); the
Committee on Immunotherapies and Sustained-Release Formulations for
Treating Drug Addiction (2004); and the Committee on Estimating Costs
to the Department of Justice of Increased Border Security Enforcement by
the Department of Homeland Security (2010–2011), and, since 2013, the
Committee on Modernizing the Nation’s Crime Statistics.
BONNIE HALPERN-FELSHER, Ph.D., is a professor of pediatrics, the

director of research, and the associate director of the Adolescent Medicine
Fellowship Program in the Division of Adolescent Medicine, Department of
Pediatrics at Stanford University. Prior to holding these positions, she was the
co-director of research for the Adolescent Medicine Fellowship, co-director
of the General Pediatric Fellowship, and a professor in the Division of Ado-
lescent Medicine, Department of Pediatrics at the University of California.
She was a faculty member in the following programs: the Psychology and
Medicine Postdoctoral Program at the University of California, San Francisco
(UCSF), the Center for Health and Community, the Center for T obacco Con-
trol Research and Education, and the UCSF Heller Diller Family Compre-
hensive Cancer Center. Dr. Halpern-Felsher’s main research areas are child,
adolescent, and emerging adult development; adolescent and young adult
health; risk behavior and risk perceptions; decision making; risk communica-
tion; tobacco control among adolescents and young adults; and t obacco pre-
vention. Dr. Halpern-Felsher received her M.A. in psychology and her Ph.D.
in developmental psychology from the University of C alifornia, R
iverside.
She has contributed her knowledge and time to several IOM and NRC
studies, including the following committees: the Committee on Scientific
Standards for Studies on Modified Risk Tobacco Products (2011); the Com-
mittee on Contributions for the Behavioral and Social Sciences in Reduc-
ing and Preventing Teen M otor Vehicle Crashes (2007); the Committee on
Reducing T obacco Use: Strategies, Barriers and Consequences (2007); and
the Committee on Developing a Strategy to Prevent and Reduce Underage
Drinking (2004).
SWANNIE JETT, Dr.P.H., is the executive director of the Seminole County

Department of Health in Florida. Dr. Jett has been influential in improv-
ing county health efforts by strengthening infrastructure through funding
improvements, increasing workforce competencies, and creating strong

partnerships that are aimed at improving overall population health out-

comes. With more than 18 years of public health experience, he has worked
to promote health awareness and public health policy, mainly concerning
health disparities and air pollution. He was previously the public health
director for Bullitt County Health Department in Kentucky, the manager
of clinical operations for the Louisville Metro Department of Public Health
and Wellness, and the public health officer for the U.S. Air Force National
Guard. Dr. Jett received his M.S. in biosystems engineering and environ-
mental science from the University of Tennessee and his doctorate in public
health with a preventative medicine/environmental health emphasis from
the University of Kentucky.
HARLAN JUSTER, Ph.D., is the director of the New York State Depart-
ment of Health’s Bureau of Tobacco Control. The bureau administers
the statewide tobacco control program, which uses a population-oriented,
policy, and systems change approach to altering the tobacco environment
in New York. The program relies on evidence-based and promising inter-
ventions to reduce youth initiation, promote adult cessation, and eliminate
exposure to secondhand smoke. Prior to his role as director, he served as
manager of the tobacco surveillance, evaluation, and research team for the
same program. In that role he was responsible for program evaluation,
local and statewide surveillance, and contributing to the science of tobacco
control. Dr. Juster earned his Ph.D. in psychology from the University at
Albany and is a licensed psychologist in New York State.
JONATHAN D. KLEIN, M.D., M.P.H., is the associate executive director

of the American Academy of Pediatrics and the director of the academy’s
Julius B. Richmond Center of Excellence for Children. He is also a profes-
sor of adolescent medicine in the Department of Pediatrics at the University
of Rochester in New York, where he previously served as the associate
chair of the Department of Pediatrics. Dr. Klein’s research at the Richmond
Center is focused on secondhand smoke exposure, health systems and be-
havior change for adolescent tobacco cessation, and education to engage
pediatricians and other clinicians in helping eliminate childhood exposure
to tobacco. Dr. Klein’s other research areas are in the access, quality, and
effectiveness of child and adolescent preventive services and in related
survey methods. He joined the academy in 2009, and his oversight respon-
sibilities there include research, tobacco control, membership and strategic
planning, and international health. Dr. Klein received his M.P.H. in health
policy and management from the Harvard School of Public Health and his
M.D. from the University of Medicine and Dentistry of New Jersey.

APPENDIX F 377
PAULA M. LANTZ, Ph.D., is a professor and the chair of the Department

of Health Policy and Management at George Washington University as
well as a professor of public health policy and public administration. Her
research interests are focused on the role of public health in health care
reform; clinical preventive services; and health disparities as a result of
social inequalities. Prior to her role as chair and professor, she was a pro-
fessor at the University of Michigan’s School of Public Health and Gerald
R. Ford School of Public Policy, and she served as chair of the Department
of Health Management and Policy. Her published research on tobacco
includes Radon, Smoking and Lung Cancer: The Need to Refocus Radon
Control Policy (2012). Dr. Lantz received her M.S. in preventive medicine
epidemiology and her Ph.D. in sociology, both from the University of
Wisconsin–Madison. She was elected to the IOM in 2012 and since 2013
has served on the IOM Roundtable on Population Health Improvement.
ROBIN MERMELSTEIN, Ph.D., is the director of the Institute of Health

Research and Policy. She is also a professor of psychology and a clinical pro-
fessor of community health sciences at the University of Illinois at C
hicago.
Dr. Mermelstein has more than 25 years of tobacco-based research experi-
ence ranging from longitudinal studies of the etiology of youth smoking to
cessation interventions for adult smokers. She is nationally recognized for
her research approaches to studying contextual factors related to the devel-
opment of nicotine addiction and the development of clinical intervention
methods for adolescent and adult smokers. Dr. Mermelstein received her
Ph.D. in clinical and community psychology from the University of Oregon.
RAFAEL MEZA, Ph.D., is an assistant professor of epidemiology at the

University of Michigan School of Public Health. His research focuses lie
in cancer risk assessment and cancer epidemiology data analysis, using
mathematically based models to assess disease dynamics. Dr. Meza’s cur-
rent research centers on developing models that will be used to evaluate the
public health impact of screening strategies for lung, colon, and esophagus
cancer risk. Some of his recently published peer-reviewed studies concern-
ing tobacco usage are “Lung Cancer in Never Smokers: Epidemiology
and Risk Prediction Models” (2012) and “Impact of the Reduction in
Tobacco Smoking on Lung Cancer Mortality in the U.S. During the Period
1975–2000” (2012). In addition to his duties as a professor, Dr. Meza is
a member of the Cancer Intervention and Surveillance Modeling Network
(CISNET) and the Cancer Prevention and Control Program at the University
of M
ichigan’s Comprehensive Cancer Center. Dr. Meza attended the Uni-
versity of W
ashington, where he received his Ph.D. in applied mathematics.

PATRICK O’MALLEY, Ph.D., is a research professor at the Survey Research

Center of the Institute for Social Research at the University of M
ichigan. He
specializes in the epidemiology of drug use, with a special focus on tobacco
usage in youth populations. He is a co–principal investigator of Monitoring
the Future: A Continuing Study of the Lifestyles and Values of American
Youth, funded by the National Institute on Drug Abuse. The study focuses
on substance abuse, including tobacco abuse, and related behaviors among
secondary school students, college students, and young adults. He is also a
co–principal investigator on the Youth, Education, and Society study, funded
by the Robert Wood Johnson Foundation. Dr. O’Malley received his Ph.D.
from the University of Michigan. He has served on committees for the NRC,
including the Committee on Opportunities in Drug Abuse Research (1995)
and the Committee on Drug Abuse Prevention Research (1992).
KIMBERLY THOMPSON, Sc.D., is the president of Kid Risk, Inc., and

a professor of preventive medicine and global health at the University of
Central Florida’s College of Medicine. She previously served on the faculty
of the Harvard School of Public Health in the departments of health policy
and management and maternal and child health, as well as at the Children’s
Hospital Boston in adolescent medicine. She founded and directed the
Harvard Kids Risk Project, which she later developed into a nonprofit or-
ganization called Kid Risk, Inc. Her research interests and teaching focus on
developing and applying economic, dynamic, risk, decision, and integrated
mathematical models to examine and connect the benefits of preventive
medicine interventions and global health policies. She leads global health
research related to managing infectious diseases, including polio, measles,
rubella, and cholera. Her research on policy development and implementa-
tion assesses the effects of methodological choices on modeling decisions
and economics (including cost-effectiveness, benefit–cost, decision, value-
of-information, and risk analysis) and how a consideration of uncertainty,
variability, and time affect policy outcomes. Dr. Thompson received her
M.S. in chemical engineering from MIT and her Sc.D. in environmental
health from Harvard University. She has served on several IOM and NRC
committees, including the Committee on Ranking FDA Product Categories
Based on Health Consequences (2009).

Legal Access To Tobacco Products

Uploaded by

Copyright:

Available Formats

Legal Access To Tobacco Products

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

Legal Access To Tobacco Products

Uploaded by

Copyright:

Available Formats

This PDF is available from The National Academies Press at http://www.nap.edu/catalog.php?

Public Health Implications of Raising the Minimum Age of Legal

Visit the National Academies Press online and register for...

Instant access to free PDF downloads of titles from the

NATIONAL ACADEMY OF SCIENCES

NATIONAL ACADEMY OF ENGINEERING

NATIONAL RESEARCH COUNCIL

10% off print titles

Custom notification of new releases in your field of interest

Special offers and discounts

Copyright © National Academy of Sciences. All rights reserved.

Committee on the Public Health Implications of

Board on Population Health and Public Health Practice

Richard J. Bonnie, Kathleen Stratton, and Leslie Y. Kwan, Editors

Copyright © National Academy of Sciences. All rights reserved.

THE NATIONAL ACADEMIES PRESS 500 Fifth Street, NW Washington, DC 20001

This study was supported by Contract/Grant No. HHSF22301031T between the

International Standard Book Number 13: 978-0-309-31624-8

Copyright 2015 by the National Academy of Sciences. All rights reserved.

Printed in the United States of America

Suggested citation: IOM (Institute of Medicine). 2015. Public health implications

Copyright © National Academy of Sciences. All rights reserved.

“Knowing is not enough; we must apply.

Advising the Nation. Improving Health.

Copyright © National Academy of Sciences. All rights reserved.

The National Academy of Sciences is a private, nonprofit, self-perpetuating society

The Institute of Medicine was established in 1970 by the National Academy of

Copyright © National Academy of Sciences. All rights reserved.

COMMITTEE ON THE PUBLIC HEALTH

RICHARD J. BONNIE (Chair), Harrison Foundation Professor of

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

ANNETTE M. BACHAND, Colorado State University

Copyright © National Academy of Sciences. All rights reserved.

KENNETH W. WACHTER, University of California, Berkeley

Although the reviewers listed above have provided many constructive

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

Drug Administration (FDA) Commissioner David Kessler characterizing

Copyright © National Academy of Sciences. All rights reserved.

and creative. Adult prevalence remains about 18 percent, and smoking-

Richard J. Bonnie, Chair

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

2 PATTERNS OF TOBACCO USE BY ADOLESCENTS AND

3 THE DEVELOPMENTAL AND ENVIRONMENTAL CONTEXT

4 THE EFFECTS OF TOBACCO USE ON HEALTH 91

Copyright © National Academy of Sciences. All rights reserved.

Impact of Cigarette Smoking on Mortality, 121

5 RESTRICTIONS ON YOUTH ACCESS TO TOBACCO

6 EVIDENCE ON THE EFFECTS OF YOUTH ACCESS

Copyright © National Academy of Sciences. All rights reserved.

Underage Access Restrictions in the Context of Other

7 THE EFFECT ON TOBACCO USE OF RAISING THE

8 HEALTH BENEFITS OF RAISING THE MINIMUM AGE OF

Copyright © National Academy of Sciences. All rights reserved.

2 PATTERNS OF TOBACCO USE BY ADOLESCENTS AND

3 THE DEVELOPMENTAL AND ENVIRONMENTAL CONTEXT

5 RESTRICTIONS ON YOUTH ACCESS TO TOBACCO

6 EVIDENCE ON THE EFFECTS OF YOUTH ACCESS

7 THE EFFECT ON TOBACCO USE OF RAISING THE

8 HEALTH BENEFITS OF RAISING THE MINIMUM AGE OF