clasp-software-summary-final-april-2013

A Healthy Toronto By Design Report
A Health and Environment Enhanced
Land Use Planning Tool – Highlights
April 2013

i A Health and Environment-Enhanced Land Use Planning Tool: Highlights
Reference:
Toronto Public Health and Urban Design for Health, A Health and Environment-Enhanced
Land Use Planning Tool: Highlights, April 2013. City of Toronto
Authors:
Larry Frank, Jim Chapman, Suzanne Kershaw, Kim Perrotta and Monica Campbell
Acknowledgements:
This report provides the highlights of a technical report, A Health-Based Decision Tool: Enabling
Public Health, Transportation & Land Use Planners to Assess Key Health Impacts of Urban
Design Scenarios (March 2012), that was prepared for Toronto Public Health by Urban Design 4
Health Limited. Copies of that report can be accessed at:
http://www.toronto.ca/health/hphe/pdf/clasp_2012.pdf
This project has been made possible through financial and/or in-kind contributions from Health
Canada, through the Canadian Partnership Against Cancer’s CLASP initiative, the Heart and
Stroke Foundation and Toronto Public Health. The views expressed in this report represent the
views of Toronto Public Health and Urban Design for Health, Inc. (www.ud4h.com), but do not
necessarily represent the views of the project funders.
We would like to thank the following individuals for providing significant input into the development
of the software tool:
• Alice Miro, Project Manager, CLASP Initiative, Built Environment and Health, Heart and Stroke
Foundation
• Shawn Chirrey, Manager, Healthy Public Policy, TPH *
• Ronald Macfarlane, Supervisor, Healthy Public Policy, TPH
• Ned Sabev, Research Consultant, Healthy Public Policy, TPH *
• Dr. Brian Cook, Research Consultant, Healthy Communities, TPH
• Nancy Day, Supervisor, Health Status and Epidemiology, TPH
• Jane Speakman, Solicitor, Legal Services
• Sudha Sabanadesan, Research Consultant, TPH
• Dr. Loren Vanderlinden, Supervisor,Healthy Public Policy, TPH
• Cheryl Dow, Research Consultant, Healthy Public Polic, TPH *
• Stephen Samis, Project Manager, Heart and Stroke Foundation *
*No longer with Toronto Public Health or Heart and Stroke Foundation
We would also like to thank the following individuals for providing invaluable input or
technical/advisory support at the early stages of the project and software development:
• Allison Stewart, Senior Planner, Fraser Valley Regional District
• Ciara De Jong, Manager, Research and Policy, Toronto Environmental Office
• Dr. Helena Swinkels, Medical Health Officer, Fraser Health Authority
• Dr. Nazeem Muhajarine, Professor, Community & Health Epidemiology, University of
Saskatchewan

ii A Health and Environment-Enhanced Land Use Planning Tool: Highlights
• Erin Welk, Technical and Policy Advisor, Sustainable Communities Group, Pembina Institute
• Jason Emmert, Air Quality Planner, Metro Vancouver
• Peter Gozdyra, Medical Geographer, Centre for Research on Inner City Health, St. Michael’s
Hospital
• Tina Avta, Planner, Canadian Institute of Planners
• Dr. David Mowat, Medical Officer of Health, Peel Region
• Kacey Lizon, Senior Planner, Sacramento Area Council of Governments
• Raef Porter, Senior Analyst, Sacramento Area Council of Governments
Lastly, a special thanks to those who have helped with design and implementation of the Surrey
Central Station case study:
• Mary Beth Rondeau, Senior Urban Designer, City of Surrey
• Don Luymes, Manager, Community Planning, City of Surrey
• Philip Bellefontaine, Transportation Engineer, City of Surrey
• Stuart Jones, Senior Planner, City of Surrey
Data files from the following agencies were used in this project:
• Toronto Public Health
• City of Toronto Open Data Repository (Toronto.ca/open)
• The Toronto Transit Commission
• Municipal Property Assessment Corporation
• Statistics Canada
• Data Management Group at the University of Toronto Joint Program in Transportation
• City of Surrey Planning & Development Department
Copies:
Copies of the report can be downloaded at
http://www.toronto.ca/health/hphe/built_environment.htm
For Further Information:
Healthy Public Policy Directorate
Toronto Public Health
277 Victoria Street, 7th
Floor
Toronto, Ontario M5B 1W2
416-392-6788

iii A Health and Environment-Enhanced Land Use Planning Tool: Highlights
About the Healthy Toronto By Design Report Series
Healthy Toronto By Design was released by Toronto Public Health in October 2011 and was the
first in a series of reports on how local communities shape the health of their residents. The
report noted that healthy cities are cities that are liveable, prosperous and sustainable. They
are cities with high quality built and natural environments, public transit, housing, culture,
education, food and health care. Healthy cities don't just happen. They result from creative
vision, strategic decision-making and thoughtful implementation that respects the needs and
challenges of all residents. They happen by design – through intentional investment and
provision of infrastructure, programs and services with health in mind.
This report is one of a series which explore what makes a healthy city. Visit Toronto Public
Health's website at: http://www.toronto.ca/health/hphe/built_environment.htm for a list of
reports in the series. Some of the topic areas in the series include the following:
• Toward Healthier Apartment Neighbourhoods – This report synthesizes zoning barriers
and opportunities to promote healthy neighbourhoods, particularly in clusters of
residential apartment towers in low income areas and inner suburbs of Toronto.
• The Walkable City – This report summarizes the findings of a Residential Preferences
Survey that gauges public demand for walkable versus more auto-oriented
neighbourhoods, and links this information with travel choices, physical activity levels
and body weight.
• Inventory of Best Practices – This report showcases examples of innovative practices
and policies across city government in Toronto that promote healthy built
environments.
• Road to Health: Improving Walking and Cycling in Toronto - This report synthesizes
evidence on health benefits and risks associated with walking, cycling and physical
activity related to the use of public transit, as well as economic assessments and specific
strategies to increase the use and safety of active transportation in Toronto.
• Health and Environment-Enhanced Land Use Planning Tool - A software tool has been
developed to assist policy and decision-makers understand how different approaches to
neighbourhood design might impact health- and environment-related outcomes such as
levels of physical activity, active modes of transportation, vehicle-kilometres travelled,
body weight, and greenhouse gas emissions. A 400-page technical report describes the
development and pilot testing of the tool, while a highlights report synthesizes
information on the tool's development, use and applications.

iv A Health and Environment-Enhanced Land Use Planning Tool: Highlights
Executive Summary
Toronto Public Health (TPH) engaged Urban Design 4 Health, Ltd.(UD4H)(www.UD4H.com ) to
enhance an existing land-use planning tool to enable it to evaluate the health- and
environment-related outcomes associated with planning decisions in Toronto. The tool uses
scenario or “sketch planning” tools to show how changes in community design are likely to
impact a range of outcomes. Scenario planning tools have grown in use along with computing
power and geographical information system (GIS) capabilities, and are now regularly employed
in regional and local planning efforts. Scenario planning tools can be used to convey the likely
health, transportation, and environmental impacts associated with varying land use and
transportation investment decisions.
Using Toronto data, statistical relationships between built environment variables, travel choices
and physical activity were derived. These relationships were corrected for demographic factors
such as gender and income which also affect physical activity and travel choices. Overall, the
analysis found that walkable neighbourhoods near high quality transit service are associated
with increased levels of active transportation, transit use, physical activity, caloric expenditures.
They were also associated with lower body mass index (BMI), auto use, and transportation-
related greenhouse gas emissions. These neighbourhoods were also associated with higher
vehicle ownership and levels of education. These findings are consistent with the findings from
many studies conducted across North America over the last decade. However, since the study
area was limited to the City of Toronto the relationships found are likely not as strong as would
be observed if outlying areas where commutes are the longest and car dependence greatest
had also been considered.
The results of this analysis were programmed into CommunityViz, a visually engaging, GIS-based
scenario land use planning tool. The tool allows policy and decision makers to assess how alternative
approaches to neighbourhood design might affect a variety of health- and environment-related
outcomes.
The following health- and environment-related outcomes were added to the “off-the-shelf”
CommunityViz software:
Body mass index (BMI)
Daily energy expenditure (kcal/kg/day)
Walk trip occurrence for exercise
Walk trip occurrence to work/school
Bicycle trip occurrence for exercise
Bicycle trip occurrence to work/school
High blood pressure (likelihood)
Average daily active (walk and bike) trips per average person
Average daily transit trips per person
Average daily auto (driver and passenger) trips per person
Average daily vehicle kilometres travelled (VKT) per person
Average daily grams of carbon dioxide (CO2) per household from personal transport as a measure of
greenhouse gases (GHG).

v A Health and Environment-Enhanced Land Use Planning Tool: Highlights
The enhanced CommunityViz tool makes it possible to assess outcomes that are indicators of
health- and environment-related impacts for different land use planning scenarios for local
situations in Toronto. This enhanced CommunityViz software is a re-usable, freely transferable
CommunityViz Analysis template that includes a collection of formulas, assumptions, indicators
and charts that have been built into the CommunityViz open modelling framework. The base
software must be purchased from Placeways at www.placeways.com.
The initial CommunityViz Analysis Template was pilot tested on the redevelopment proposal for
the West Don Lands area in Toronto. This Template was built with a dataset which can be
reapplied to other scenarios in Toronto in the future. UD4H has provided Toronto with the
base datasets necessary to run the tool for subsequent study areas within the City of Toronto.
The CommunityViz Analysis Template developed for Toronto can also be applied to other urban
centres across Canada with similar urban forms. To assess its applicability to other cities in
Canada, the tool was also pilot tested on the development plan for the Surrey Central Station in
the City of Surrey which is part of the Greater Vancouver Regional District. This trial showed
how the tool could be used in a different region of Canada using local base data.
Both pilot studies estimated the impacts of varying land use and transportation planning
options on the levels of physical activity, use of active modes of transportation, use of vehicles,
emissions of transportation-related greenhouse gases, and the BMI for residents in the
neighbourhoods examined. The estimates produced can be used in the review and decision-
making processes related to those neighbourhoods. A number of new data sources or
refinements could be made to further increase the precision and application of the tool.

vi A Health and Environment-Enhanced Land Use Planning Tool: Highlights
Table of Contents
Page
Executive Summary........................................................................................................................iv
Project Summary............................................................................................................................. 1
Land Use Planning, Transportation Systems & Human Health ...................................................... 2
Deriving the Health- & Environment-Related Coefficients............................................................. 4
Health and Environment-Related Land Use Planning Tool........................................................... 11
Toronto Case Study - West Don Lands ......................................................................................... 14
Applying the Tool to Other Regions – Surrey Case Study............................................................. 21
CommunityViz Analysis Template, Data Sharing & Further Applications .................................... 27
References .................................................................................................................................... 31
List of Tables
Table 1: Data Sources – Travel Choices, Health and Climate-Related Indicators........................... 5
Table 2: Built Environment Variables included in the CommunityViz Analysis Template ............. 6
Table 3: Demographic and Household Variables Used in Regression Models............................... 8
Table 4: CommunityViz System Requirements............................................................................. 12
Table 5: File Types Contained in the CommunityViz Analysis Template...................................... 13
Table 6: West Don Lands Scenario Summary ............................................................................... 18
Table 7: Estimated Outcome Values for West Don Lands, Unit Level.......................................... 19
Table 8: Estimated Outcome Values for West Don Lands, Population-Level............................... 20
Table 9: Comparison of Existing Conditions and Plan Scenarios, Surrey...................................... 25
Table 10: Built Environment Values for Surrey Study Area Scenarios.......................................... 25
Table 11: Estimated Outcomes for Surrey Study Area for Scenarios, Unit Level......................... 26
Table 12: Estimated Outcomes for Surrey Study Area for Scenarios, Population Level .............. 26
List of Figures
Figure 1: Postal Code Neighbourhood Buffer ................................................................................ 6
Figure 2: Change Scenario 1: High Density Mixed Use ................................................................ 17
Figure 3: Change Scenario 2: Medium Density............................................................................ 17
Figure 4: Parcel-based Polygons in Surrey, BC.............................................................................. 22
Figure 5: Surrey Central Station Plan Assumptions...................................................................... 24

vii A Health and Environment-Enhanced Land Use Planning Tool: Highlights

1 A Health and Environment-Enhanced Land Use Planning Tool: Highlights
CommunityViz Analysis
Template refers to the empty
shell of formulas and
spatial/mathematical
calculations that can be applied
to other jurisdictions with
similar base datasets.
CommunityViz Analysis Project
File refers to the analysis
package for a given study area
which contains base datasets
and subsequent outcomes
calculated using the formulas
contained in the CommunityViz
Analysis Template.
Project Summary
Context
There is growing evidence that many features of the built environment can affect the health of
the public by influencing travel choices, levels of physical activity, and access to healthy foods,
which can in turn increase or decrease a person's risk for developing chronic diseases such as
diabetes and heart disease. Built environment variables such as population density, separation
of land uses, access to and quality of transportation services, and access to recreational
amenities, have been found to affect human health directly, by influencing behaviour, and
indirectly, by influencing air quality, air temperatures, water quality, and the climate, which can
in turn affect public health. While urban planners and elected officials must weigh many
considerations when making decisions about land use and transportation planning options,
they have not traditionally had access to quantifiable information about how these decisions
might impact public health and the environment. The health- and environment-enhanced land
use planning tool developed for the City of Toronto represents one step forward on that count.
Project Steps
For this project, UD4H conducted research on the relationships between the built environment and
travel choices, levels of physical activity, and emissions of GHG from transportation sources. The results
from the research were used to enhance the GIS-based scenario planning software called
CommunityViz. This enhanced software allows the users to measure the health- and environment-
related outcomes associated with built environment scenarios that could be applied to an area of the
community.
CommunityViz works with a wide range of spatial and tabular data sources and is built so the
user can create custom analyses and reports. This flexibility means the tool has the potential to
inform a number of planning processes at different levels of government. CommunityViz was
developed by Placeways, LLC. UD4H contracted with
Placeways to implement the CommunityViz enhancements
that were needed to incorporate the research results.
CommunityViz is an extension to ESRI’s ArcGIS software. It is
used to perform specialized spatial calculations and queries
to estimate the potential impacts of a plan, development or
other scenario. Users can indicate future changes to a study
area in a variety of ways, including providing a new built
environment GIS shapefile, ”sketching” future scenarios
directly onto a map, or using neighbourhood/place types
preset with various built environment attributes.
CommunityViz Analysis Template
The final product of this project is a re-usable, transferable
CommunityViz Analysis Template; a collection of formulas,
assumptions, indicators and charts that are built into the
CommunityViz open modelling framework.

The first CommunityViz Analysis Template was built with a case study dataset in Toronto, and
then everything except the data was saved as a template for application to similar datasets in
other parts of Toronto and in other regions of the country that have a similar urban form.
The user loads base datasets into the CommunityViz Analysis Template to create a local
CommunityViz Analysis Project File, which can then be used to assess the effect of different
community design scenarios for the selected study area.
The CommunityViz Analysis can be applied to many different questions but the template
applied to the two case studies relates to specific datasets and can be applied to another
jurisdiction with equivalent base data from that community.
The CommunityViz Analysis Template was piloted on two case studies: the West Don Lands
re-development plan in the City of Toronto; and the Surrey Central Station redevelopment plan
in the Greater Vancouver Regional District. In both cases, the CommunityViz Analysis Template
provided health- and environment-related outcome results which can be used in the review
and decision-making processes.
Scenario Planning Tools
Scenario planning tools, also known as “sketch planning” tools, allow users to test how changes
in community design might impact outcomes such as employment, transportation, energy use,
and emissions of greenhouse gases and pollutants. Scenario planning tools have grown in use
along with computing power and GIS capabilities, and are now regularly employed in regional
and local planning efforts.
Generally, scenario planning tools are based on empirically tested “elasticities” that relate land
use characteristics such as residential density, land use mix and street connectivity, to
commonly utilized planning metrics such as vehicle emissions, residential energy use, and
residential school and employment populations, and jobs/housing balance. Where research is
available, these tools can be enhanced to include health- and environment-related outcomes as
was done for this project.
Although scenario planning tools are not as sophisticated or detailed as full regional land use or
transportation model suites typically used by regional planning agencies, they are much quicker
to set up and run. This means that they can be used to generate ideas in public meetings and
for smaller efforts like development review. Although most tools will still require a certain
level of expertise and training, they are generally more visually oriented and appropriate for
use by less-technical audiences, especially once the initial setup is complete.
Land Use Planning, Transportation Systems & Human Health
The recent rapid increase in obesity and related chronic diseases has spurred a widespread
recognition of the built environment’s role in population health. Research on the topic has
expanded greatly over the last decade, re-establishing the historic connection between urban
planning and public health and causing both professions to re-examine the many complex
relationships between public health and the places in which people live, work and play.

The built environment is a heavily impacted by land use decisions and transportation
investment. There is a number of public health impacts associated with travel choices that
result from different built environment patterns, including the amount of active travel –
walking and cycling – that people do. Other health outcomes related to environmental noise,
vehicle-related safety, air pollution, and access to healthy foods, recreational facilities,
greenspace and jobs, are more complicated. But, with these outcomes, urban form,
community design and transportation facilities play an important role as well.
Neighbourhood design affects travel patterns primarily by influencing the proximity between
destinations and the directness of travel between these destinations. Proximity is a function of
both the density (compactness) of development and the diversity of land uses. Density and
diversity work in tandem to determine how many activities are within a convenient walking or
cycling distance1
. Connectivity determines how directly one can travel between activities on
the street or path network. Other factors can also complement or undermine a
neighbourhood’s walkability including the presence or lack of sidewalks and other
infrastructure for cycling and walking; building placement and site design, transit accessibility,
and visual appearance of the streetscape in terms of aesthetics, interest and safety2
Generally, land use and transportation planning decisions that shift travel from private vehicles
to active transportation and transit can provide multiple health and environmental benefits
including increased physical activity, less sedentary time spent sitting in cars, fewer emissions
of air pollutants and greenhouse gases per person, reduced risk of vehicle-related injuries and
deaths, and less environmental noise.
.
There continues to be some debate in the literature over the role that “self selection” plays in
the relationship between built environment and travel behavior. Recent research suggests
that both neighbourhood preferences and neighbourhood features have an impact on the
travel choices and behavior of residents3
. Cross-sectional studies conducted in many places
and at many scales of measurement have demonstrated a statistically significant relationship
between the “walkability” of a neighbourhood and the travel choices of residents in those
neighbourhoods4
. Unfortunately, only a few longitudinal studies have been completed but
these suggest that there is likely a causal relationship between the built environment and
health. For example, a recent study showed reduction in obesity levels after a light rail line
opened in Charlotte North Carolina. This study found that riding the rail line led to an average
weight loss of around 6 ½ pounds, and that Light Rail users are 81% less likely to be obese over
time5
Built Environment, Physical Activity & Body Weight
.
In general, the consensus in the research is that there is a connection between the built
environment, travel choices, active modes of transportation, and the resulting levels of physical
1
Frank 2000, Sallis et al. 2004, Frank and Engelke 2001
2
Toronto Public Health, 2012.
3
Bagley and Mokhtarian 2002; Frank et al. 2007a; Handy et al. 2006; Khattak and Rodriguez 2005; Kitamura et al.
1997; Schwanen and Mokhtarian 2004; Schwanen and Mokhtarian 2005a; Schwanen and Mokhtarian 2005b
4
Ewing and Cervero 2001, Frank 2000; Boarnet and Crane 2001, USEPA 2001, Kuzmyak and Pratt 2003, Bento et al.
2003; TRB/IOM 2005
5
Macdonald et al, 2010

activity and body weights6
. Urban form has been found to be associated with the total amount
of physical activity as well as with the amount of cycling and walking7
. Sprawling land use
patterns have also been correlated with higher body weights, obesity, and chronic diseases
associated with physical inactivity and/or obesity8
. Time spent driving has also been linked to
obesity, cardio-respiratory fitness, and other indicators of metabolic risk9
Built Environment, Air Quality & Human Health
. Demographic
variables such as income, gender, age and auto ownership, are also strongly related to physical
activity and body weight.
The relationships between land use patterns, vehicle emissions, air quality, and human health
are complex. Air pollution is made up of a variety of substances, each with different sources,
patterns of distribution, chemical reactions and health impacts. Each pollutant therefore has a
different association with land use patterns and transportation, making it difficult to determine
how a particular land use policy will affect air pollution levels or exposure risks.
Higher levels of VKT are associated with increased emissions of air pollutants and greenhouse
gases. Residents and workers in compact, walkable areas tend to drive less10
, and are
therefore responsible for fewer transportation-related air pollutants on a per person basis.11
However, denser areas tend to be more congested with vehicles and pedestrians which can increase
the levels of certain pollutants in the air adjacent to roadways and the number of people potentially
exposed. Therefore, in some circumstances, people living in more compact areas can be exposed to
elevated levels of air pollution, while generating less air pollution on a per person basis12
Deriving the Health & Environment-Related Coefficients
.
Research and Data Sources
The health- and environment-enhanced land use planning tool was developed with two types
of Toronto data; travel and health-related data and built environment data. The Canadian
Community Health Survey (CCHS) and the Transportation Tomorrow Survey (TTS) provided the
data to derive the health- and environment-related relationships. Census data was also used to
provide neighbourhood (dissemination area) level demographics.
6
Lopez, 2004; Papas et al, 2007
7
King et al. 2003; Saelens et al. 2003b
8
Ewing et al, 2003a; Frank et al. 2004; Giles-Corti et al., 2003; Saelens et al., 2003a; Frank et al, 2005; Sturm &
Cohen, 2004
9
Hoehner et al, 2012
10
LFC 2005a; Frank and Pivo 1995; Cervero 1991; Cervero and Kockelman 1997; McCormack et al. 1996, Frank et al.
2006; Ewing and Cervero 2001; Holtzclaw 1994; Dunphy and Fisher 1996; Frank et al. 2000; PBQD 1996; Ross and
Dunning 1997; Kitamura et al. 1997; Cervero and Gorham 1995; Cervero and Kockelman 1997.
11
LFC et al. 2005a; Frank et al. 2000; Frank et al. 2006.
12
Marshall et al., 2009.

The CCHS is conducted regularly by Statistics Canada to provide estimates of health
determinants for 10 age-sex groups within each health region in Canada. The 2007/2008 CCHS
data was selected for use in the analysis because it best matched the time periods of the built
environment and other data sets used in the project. It contained 4,077 participants;
representing 2,870 postal codes in the City of Toronto (6.1 percent of the total postal codes in
the City) (see Table 1).
The TTS survey was conducted in 2006 by the Data Management Group at the University of
Toronto Joint Program in Transportation. It sampled approximately 150,000 households across
the Greater Toronto Area. Participants were asked questions about trips on a particular day for
household members over 11 years of age as well as questions about household demographics.
The data used for this project was for participants in the City of Toronto. TTS provided
transportation outcomes related to travel behavior, such as number of trips, and distance and
minutes travelled by the various travel modes. The TTS survey was also used to estimate
transport-related carbon dioxide (CO2) emissions. Averaged TTS data was provided at the
postal code level. Survey data covered 22,091 (46.8% of total) postal codes across Toronto (see
Table 1).
13
Daily energy expenditure, expressed in kcal/kg/day (PACDTLE), was derived by Statistics Canada based on
participant responses to several activity questions. It is calculated by combining the time each participant spent
engaging in leisure (e.g. walking, cycling, or sports) and transportation (e.g. walking/cycling to work) activities in
the last three months. The total calories burned during all activities was calculated and converted into a daily value
based on the participant’s weight. Respondents are classified as follows: 3.0 kcal/kg/day or more = physically
active; 1.5 to 2.9 kcal/kg/day = moderately active; less than 1.5 kcal/kg/day = inactive. Source: Canadian
Community Health Survey (CCHS): 2008 (Annual component) and 2007-2008, Derived Variable (DV) Specifications,
Master and share file. http://www.statcan.gc.ca/imdb-bmdi/document/3226_D2_T9_V6-eng.pdf
Table 1: Data Sources – Travel Choices, Health and Climate-Related Indicators
Data Source for Relationships Outcome variables
2007/2008 Canadian Community
Health Survey (CCHS)
Body mass index (based on self-reported height and weight)
Daily energy expenditure (kcal/kg/day) 13
Walk trip occurrence for exercise
Walk trip occurrence to work or school
Bicycle trip occurrence for exercise
Bicycle trip occurrence to work or school
High blood pressure
2006 Transportation Tomorrow
Survey (TTS)
Average daily walk and/or bike trips per average person
Average daily transit trips per person
Average daily auto (driver and passenger) trips per person
Average daily trip kilometres per person
Average daily grams of CO2 per household from transportation

Buffered Postal Codes
Built environment measures were
calculated for each postal code in
the City of Toronto. Since the postal
code is the smallest geographic level
at which a CCHS and TTS reports
information for participants,
postal codes formed the base
geography for analysis and
subsequently for the software
tool enhancements.
In order to consider the spatial
context of each postal code, the
area surrounding it was included
in the calculation of built
environment variables. The area
around each postal code is
referred to as a buffer.
Buffers were created to cover the area that can be travelled from the centre of the postal code in
all directions along the street network, for a distance of 1 kilometre. These network buffers
establish a “walk shed”; the area people can actually access around their homes within a 6-10
minute walk. The street network used to create the buffers is modified so that it includes only
those streets on which pedestrians are allowed to travel (e.g. limited access highways and their on-
ramps are not included).
Table 2: Built Environment Variables included in the CommunityViz Analysis Template
Built Environment Variables
Road/Travel
Length of bicycle facilities (total metres)
Length of trails (total metres)
Length of bicycle facilities and trails (total metres)
Length of all roads (total metres)
Length of walkable roads (total metres)
Sidewalk coverage (ratio of metres of sidewalks to metres of sidewalk eligible
roads
Intersection density (count/sq km in buffer)
Distance to nearest major arterial. The crow-fly distance from postal code
centroid to nearest major arterial road (kilometres)
School
School density (count/sq km in buffer)
Distance from postal code centroid to nearest school (metres, measured on the
street network)
Figure 1: Postal Code Neighbourhood Buffer

These buffers were created for all 47,246 six-digit postal codes in the City of Toronto. GIS software
created by ESRI, called ArcGIS 9.3, and its network analyst extension was used to build the buffers
(see Figure 1). The final buffered postal code level built environment variables, found to be
significant in the statistical modelling process, were incorporated into the CommunityViz Analysis
Template (see Table 2). These variables were calculated at the buffered postal code level.
Demographic Variables
A number of demographic and socioeconomic variables are associated with built environment
variables, travel behavior, health- and environment-related outcomes. These variables, listed in
Table 3, have been included in the model as inputs because they can are confounding factors that
can affect the health- and environment-related outcomes even though they are not the variables
that are being examined.
Table 2: Built Environment Variables included in the CommunityViz Analysis Template
Built Environment Variables
Transit
Transit stop count
Transit stop density (count/sq km in buffer)
Distance from postal code centroid to nearest transit stop/station (metres,
measured on the street network)
Parks Park presence (one or more parks intersecting buffer)
Food
Farmers market density (count/sq km in buffer)
Restaurant density (count/sq km in buffer)
Supermarket density (count/sq km in buffer)
Takeout restaurant density (count/sq km in buffer)
Convenience store density (count/sq km in buffer)
Regional
Location/Access
Regional accessibility. Average road network based distance to 8 regional
destinations. See Appendix D for additional details.
Land Use
Residential density (dwelling unit count/acres of residential land in buffer).
Retail floor to land area ratio (FAR): total retail building floor area divided by the
total retail parcel land area.
Office floor to land area ratio (FAR): total office building floor area divided by the
total office parcel land area.
Park area (total acres in buffer)

Associations: Built Environment Variables & Health- and Environment-Related Outcomes
The relationships or associations between each built environment feature and a health- or
environment-related outcome were derived for Toronto using data from either the CCHS or the TTS
and various built environment features that were mapped for the City of Toronto (e.g., walkability
maps, park density maps).
The relationships between built environment variables and the outcome of interest are expressed
as coefficients that can be placed into an equation. These coefficients and equations have been
developed and built into the CommunityViz Analysis Template. Each coefficient provides the
change expected in an outcome per unit increase in a particular built environment variable. For
example, the coefficient of -0.0304 for the variable “vehicles per household” means that “an
increase of 1 vehicle per household is associated with a 0.0304 reduction in the average number of
walking trips per person, while holding all other variables in the model constant".
Details describing the derivation of built environment coefficients can be found in the background
technical report, A Health-Based Decision Tool: Enabling Public Health, Transportation & Land Use
Planners to Assess Key Health Impacts of Urban Design Scenarios (March 2012), which can be
accessed at: http://www.toronto.ca/health/hphe/pdf/clasp_2012.pdf.
Table 3: Demographic and Household Variables Used in Regression Models
Data Sources Demographic and Household Variables
2007/2008 Canadian Community
Health Survey (CCHS)
Age
Sex
Household size
Household income
Education level
Employed (1=yes; 0=no)
Ethnicity (1=white; 0=non-white)
Immigrated (1=yes; 0=no)
2006 Transportation Tomorrow
Survey (TTS)
% female
Average number of students per household (age 6-17 years)
Average number of persons per household
% employed
Average number of vehicles per household
Population count
Vehicle count
Census
% less than 15 years
Median household income
% with university degree
% visible minority population
% immigrated

Below are some of the highlights on the relationships found between the built environment
variables and the health- and environment-related outcomes for the City of Toronto:
• Body Mass Index (BMI) - Higher walkability index values were associated with lower BMI
values. For example, for every unit increase in the walkability index, it was found that the
BMI goes down by 0.3.
• Energy Expenditure - Higher residential density was significantly associated with increased
daily energy expenditures among adults, as was increased access to schools. For example, for
every 1 km decrease in the distance to a school, energy expenditures increased by 0.09
kcal/kg/day.
• Walking for Exercise/Leisure - Greater trail length and living less than 1 km from a park are
built environment variables that were both associated with higher odds of making a walking
trip each month. For example, if a park is added to an area within 1 km of a participant’s
postal code where no park previously existed, the odds of the participant making a trip on
foot are almost four times greater after the park is established. Higher walkability index
values were associated with a greater number of walking trips for both recreational and
utilitarian purposes. Living in a postal code where the walking distance to a park is less than 1
km was also associated with an increase in the number of walking trips taken by residents.
• Walking to Work/School - Higher walkability index values were significantly associated with a
higher number of walking trips to work and school. Distance to the nearest school was also
significantly associated with a higher number of walking trips. For example, an increase in
the distance to the nearest school of 1 km was associated with a 43 percent decrease in the
odds of a resident making at least one walking trip per month.
• Biking Trips for Exercise or Leisure - Higher residential density was associated with greater
odds of a resident making a bike trip. For example, an increase in residential density of 10
dwelling units was associated with a 4.4 percent increase in the odds of a resident making at
least one bike trip for exercise/leisure purposes. Higher walkability index values were also
associated with a greater number of bike trips, as was the distance “as the crow flies” to the
nearest major arterial.
• Biking Trips to Work or School - Higher walkability index values and kilometres of bicycle
facilities (e.g., bike lane or trail) were both associated with greater odds that residents would
make bike trips to work or school. For example, a 1 km increase in bicycle facilities was
associated with an 8.6 percent increase in the odds of a resident making a bike trip to work or
school.
• High Blood Pressure - Higher levels of physical activity, measured using monthly frequency of
walking and cycling trips for any purpose, were associated with lower odds of having high
blood pressure. For example, ten additional active trips (i.e., walking or bike trips) per month
were associated with an 8 percent decrease in the odds of having high blood pressure.
• Active Mode Trips (Cycling and Walking) - Higher walkability index values, residential
density, the log of park acres, sidewalk coverage, metres of bike lanes and trails, the log of
transit stop/station density, and school density, were all significantly associated with higher
odds of making at least one trip by an active mode of transportation.

Of the built environment variables examined, residential density and sidewalk coverage had
the strongest associations with the likelihood of a resident making at least one trip per day by
an active mode of transportation. For example, an increase in residential density of 10
dwelling units per acre was associated with a 10 percent increase in the odds of making at
least one trip by an active mode of transportation, and an increase in sidewalk coverage of 10
percent was associated with a 6 percent increase.
Higher walkability index values, shorter distance to the nearest school, more walkable roads,
and shorter average travel distances to major regional destinations, were significantly
associated with a higher number of trips by active modes of transportation as well.
Of the built environment variables examined, distance to school and metres of walkable
roads had the strongest associations with the number of trips per person by active modes of
transportation. For example, a decrease in the distance to the nearest school of 100 metres
was associated with a 2 percent increase in active mode trips per person, while an increase in
1 km of walkable road was associated with a 1 percent increase in active mode trips.
• Transit Trips - Higher walkability index values, a greater number of acres of park, increased
density of transit stops/stations, a greater number of metres of bike lanes and trails, and
shorter average travel distances to major destinations, were all significantly associated with
higher odds of making at least one trip by transit per day.
Of the built environment variables, walkability index values and transit density had the
strongest associations with the likelihood of making at least one transit trip. For example, an
increase in transit density of 1 stop per square kilometre was associated with a 1 percent
increase in the odds of making at least one daily transit trip.
Shorter distances to the nearest transit stop/station and higher walkability index values were
significantly associated with a higher number of transit trips per person. Shorter distances to
the nearest school were marginally associated with higher number of transit trips per person.
A decrease in the distance to the nearest transit stop/station of 100 metres was associated
with a 2 percent increase in the number of transit trips per person, while a decrease in the
distance to the nearest school of 100 metres was associated with a 0.5 percent increase in
transit trips.
• Auto Trips - Lower walkability index values, fewer residential dwelling units, fewer metres of
bike lanes, longer distance to the nearest school, and longer average travel distances to
major regional destinations, were all significantly associated with a higher number of auto
trips per person.
Of the built environment variables examined, the walkability index values and the number of
residential dwelling units per postal code had the strongest associations with the average
number of auto trips per person. For example, an increase in 1000 residential units within the
postal code buffer was associated with a 2 percent decrease in the number of auto trips per
person. Lower walkability values, fewer metres of all road types, longer distance to the nearest
school, and longer average travel distances to major regional destinations were significantly
associated with higher daily VKT per person.

Of the built environment variables, regional accessibility and metres of all road types had the
strongest associations with daily kilometres of travel per person. For example, an increase in 1
kilometre of road length within the postal code buffer was associated with a 1 percent decrease
in the daily VKT per person.
Health and Environment-Related Land Use Planning Tool
Selecting CommunityViz as a Software Tool
The selected software, CommunityViz, is owned and administrated by a private company,
Placeways LLC. CommunityViz is an extension to the GIS platform created by ESRI, called ArcGIS
(www.esri.com). CommunityViz, with ArcGIS, can calculate impacts of geographic decisions in real
time and communicate the results of those impacts to decision makers. CommunityViz can create
3D scenes, maps and reports, charts, graphs, and interactive scenarios.
CommunityViz was selected because:
• It can visualize, analyze, and interpret spatial data;
• It can be enhanced/modified to add health impact and climate modules;
• It is user-friendly and is based on the widely used ArcGIS software platform;
• The enhancements made as part of this project can be freely shared with others who also
have copies of the ArcGIS and CommunityViz software;
• It can replicate UD4H’s research methods, particularly the measurement of urban form;
• It can work with a variety of existing datasets;
• It has a low, one-time cost;
• It can calculate results in real time.
What Is Needed to Use CommunityViz?
CommunityViz allows the user to create different scenarios in GIS for purposes of evaluating
potential changes in investments or policies. For example, a proposed development, a change in
the development codes, a transportation plan, or the implementation of a conservation district.
CommunityViz (version 4.2) can be purchased from Placeways, LLC (www.placeways.com). As of
March 2012 the cost for one license with support was US$850.
ArcGIS/CommunityViz can be run on a desktop or laptop. The software/hardware requirements for
CommunityViz are identified in Table 4 and below:
• ESRI ArcMap™ (ArcView, ArcEditor, or Arc7uInfo) 10
• ESRI Network Analyst extension14
• Windows XP, Windows Vista, or Windows 7
• Microsoft .Net Framework 2.0
• DirectX 9.0c
14
Network Analyst information -- http://www.esri.com/software/arcgis/extensions/networkanalyst/index.html

Example Regression Equation
Outcome X = a +
(b1 x net residential density) +
(b2 x land use mix) +
(b3 x intersection density) +
(b4 x transit density) +
(b5 x distance to convenience
store) +
(b6 x park in buffer) +
(b7 x transit access) +
(b8 x auto access) +
(b9 x household income) +
(b10 x # household workers) +
(b11 x # household non-workers) +
(b12 x # household children) +
(b13 x # household cars) +
(b14 x gender)
Table 4: CommunityViz System Requirements
Minimum Preferred Ideal
RAM 512 MB 1 GB 1+ GB
Processor 750 MHz 1 GHz 2+ GHz
Available Hard Disk Space 1 GB 5 GB 5+ GB
Standard 3-button Mouse Yes Yes Yes
DirectX 9 capable graphics card with the
following amount of texture memory
64 MB 128 MB 256+ MB
Building the CommunityViz Analysis Template
The CommunityViz tool built for Toronto Public Health is a re-usable CommunityViz Analysis
Template; a collection of formulas, assumptions, indicators and charts built into the CommunityViz
open modelling framework. The CommunityViz Analysis project file created by the user for a given
study area uses data, standard GIS software, and formulas
programmed into CommunityViz Analysis Template to
conduct all calculations.
UD4H provided Placeways, LLC with the final regression
equations for each outcome, including instructions for
applying variable transformations (e.g. log transformation).
In addition, UD4H gave Placeways formulas to calculate all
of the independent variables in the regression equations,
definitions for any index variables, and instructions on
which data fields to use in the calculations. Placeways, LLC
incorporated all of these into the CommunityViz tool.
The key pieces of necessary information are the multi-
variate regression equations derived by UD4H for each
outcome. Each outcome has its own regression equation
with unique values for ‘a’ (the constant) and for each of the
‘b’s (i.e., the co-efficients that mathematically describe the
relationship between each of the built environment
variables and each of the health- and environment-related
outcomes).
CommunityViz uses base data (i.e., values for built environment measures) and assumptions (using
default values or those provided by the user) to populate the regression equation for each
outcome.
To create a value for an outcome, CommunityViz draws from data provided by the user to calculate the
values for each independent variable for each buffered postal code in the study area for both “existing
conditions” and “change scenarios”. Those buffered postal code level built environment values are
inserted into the regression equation, multiplied by the coefficients, and summed along with the
constants, to generate an estimate for each outcome. Values are averaged across multiple postal
codes to create estimates for each outcome for each study area.

Running the CommunityViz Analysis Template
The initial CommunityViz Analysis Template was built with a specific case study dataset. Everything
in the CommunityViz Analysis project file except the data was saved in such a way that it can be
reapplied to similar datasets in future efforts. This template or “shell” includes all of the regression
equations and related spatial and mathematical calculations that were programmed into
CommunityViz as part of this project. The saved template can be applied to a different geographic
area where similar base datasets exist. The user loads the template into a new empty analysis, and
links the base datasets in the new geographic area to the corresponding layers in the
CommunityViz Analysis Template that are shown in Table 5. The regression equations and dynamic
attributes can then be calculated for the new study area.
Table 5: File Types Contained in the CommunityViz Analysis Template
Type Description
Polygons Boundaries of base spatial unit (e.g. postal code, other parcel aggregation)
Polygon centroids Spatial center of polygons (point location)
Buffers
Polygons created using buffering technique (1km travel distance along walkable
road, from polygon centroids)
Road network Pedestrian-accessible roads and trails
Major arterials Major roads (e.g. Yonge Street in Toronto)
Bicycle facilities
Bike lanes, major/minor multi-use pathways, park roads, signed routes, suggested
on-street connections, suggested on-street routes
Trails Pedestrian-only walkways, park pathways, recreation trails, etc.
Schools Elementary and secondary schools
Transit stops Includes all types of transit stops (e.g. bus, streetcar, light rail, subway)
Intersections Points representing 3+ leg intersections
Food locations
Dine-in restaurants; Take-out restaurants; Convenience/variety stores;
Supermarkets; Farmers’ markets
Study area boundary Polygon delimiting boundaries of analysis

Toronto Case Study - West Don Lands
Introduction
A case study site in Toronto was chosen to test the CommunityViz Analysis Template that was
developed for Toronto. The West Don Lands redevelopment plan was selected. As part of the
long term process to revitalize Toronto's waterfront, significant changes are planned for the 80
acre West Don Land site, including:
• 6,000 to 6,500 housing units, 1,300 of which will be affordable rental housing
• Residences in a mix of housing types from townhouses to mid-rise buildings and towers
• 1 million square feet of office and retail space
• New streets improving connectivity
• New parks including an 18-acre park immediately adjacent to the Don River
• A new streetcar line
• A new school.
Over half of the West Don Lands site will be developed for the Pan Am Games as the Athletes’
Village for athletes and officials. After the Games, which will last approximately eight weeks in
the summer of 2015, the Athletes’ Village will be converted into the housing in the plan.
The case study compared the existing conditions of the site to two redevelopment scenarios:
Change Scenario 1 will be the West Don Lands redevelopment plan which proposes high
density, mixed use development for the site; and Change Scenario 2 will be a medium density
residential development (~6 dwelling units/acre) for the site.
Steps in the Process
Create Buffered Postal Codes
A file of base case (existing conditions) built environment measures was created for every
postal code in the City of Toronto to support this case study. These measures are calculated at
the buffered postal code level; the area within each postal code plus the area within a 1 km
walking distance from the centroid of each postal code in all directions along the street
network. The buffered postal code is referred to as the “buffer” or the “measurement area” in
this document.
The buffer area around each postal code is included because it puts the postal code in the
context of the neighbourhood that surrounds it. This is important because a person’s travel
patterns do not depend upon the postal code boundary alone; they are influenced by the area
that can be accessed within a reasonable time and distance.
The GIS software created by ESRI, called ArcGIS 9.3, and its Network Analyst extension, were
used to build the buffers. These buffers were created for all 47,246 six-digit postal codes in the
City of Toronto.

Insert Built Environment and Demographic Data
All the built environment measures used in the regression equations, described earlier, were
created for these buffered postal codes. Demographics and populations counts for each postal
code were calculated at the non-buffered postal code level using estimated ratios of people
and vehicles per household.
Paint in Place Types
To create the land use patterns in CommunityViz, the user “paints” Place Types onto postal
codes in the study area. Place Types are development or neighbourhood types, each with its
own set of assumptions about density and distribution of different land uses. A menu of Place
Types could include, for example, “Low Density Single Family,” “Medium Density Mixed Use,” or
“Mixed Use Employment Centre,” among others. A default set of Place Types are provided in
CommunityViz, but these can be modified and added to as needed.
Reporting Areas
Results in CommunityViz can be reported for two geographies:
• For the study area which includes only the postal codes within the study area itself. The
built environment measures for the study area postal codes are calculated at the buffered
postal code level; and
• For the impacted area which includes any postal code whose buffer intersects the study
area. The impacted area is included because the built environment measures within the
study area can also affect the health- and environment-related outcomes in the areas
adjacent to it.
The health- and environment-related outcomes are calculated for each postal code and then
averaged to create estimates for the entire study area or impacted area. Even though the
buffers used to calculate the outcomes overlap, the results presented in CommunityViz can be
summed to calculate the total effect on the population living in the study area and the
impacted area.
Reporting Formats
The estimated health- and environment-related outcomes can be reported for the existing
conditions, for the change scenarios, and on a jurisdiction-wide basis. They can be reported on
a per-person or per-household basis or on a percentage basis or at a study or impacted area
population level. The results can be reported on a daily or monthly time period or for an
extended period such as a year.
West Don Lands Case Study
Existing Conditions Scenario
According to 2007 parcel data, there are currently 96 residential units on 0.34 acres of
residential land in the West Don Lands today. There are an additional 3.4 acres of vacant multi-

family development and 0.3 acres of vacant single family land. The parcel data indicates no
retail and less than an acre of parks. Most of the land is classified, according to 2007 Municipal
Property Assessment Corporation (MPAC) parcel data, as property in process of redevelopment
utilizing existing structure(s), vacant commercial land, auto dealership, railway right-of-way,
and automotive fuel station with or without service facilities, or there is no indication of use.
Some civic, parking and multi-family uses are also present.
The outcome calculations performed by the CommunityViz tool are done at the postal code
geography. The area currently captures 15 spatially unique postal codes. A multi-use trail is
located along the eastern edge of the study area adjacent to the Don River, which provides
pedestrian access to the downtown core and the Beach neighbourhood to the east. There is
currently one supermarket and one take-out restaurant in the study area. Four streetcar/bus
stops are located along the perimeter of the study area on King Street East and Mill Street.
Change Scenario 1: West Don Lands Plan – High Density Mixed Use
The full anticipated development of the West Don Lands was used for the first change scenario.
This scenario includes permanent buildings and built environment changes anticipated for the
Pan Am Athlete’s Village as represented in the final build-out of the West Don Lands Plan. Any
temporary buildings or other changes were ignored. A significant amount of work has already
been done by the City of Toronto to plan the future of the West Don Lands area. This work is
described in a number of documents which served as valuable references for the scenario
development. In most cases there was sufficient detail in the provided documents to represent
the planned development in CommunityViz. However, in some cases, it was necessary to make
some assumptions to achieve the required level of detail regarding the built environment
changes.
The West Don Lands Plan anticipates 36.7 acres of new development on the 80-acre site,
including 6,000 to 6,500 housing units and 1 million square feet of office and retail space. The
West Don Lands Block Plan described the form, density, and amount of development for each
block in the redevelopment area, and was used to develop the land use assumptions for the
area. The Plan gave no specific guidance as to the location of office or retail development, but
generally showed focus areas for each type. The information from the existing West Don Land
planning documents was used to develop five Place Types specifically for use in the case study.
These Place Types and the applications to the case study are shown on Figure 2. n in the
estimated outcome values.

Figure 2: Change Scenario 1: High Density Mixed Use
Figure 3: Change Scenario 2: Medium Density

Additional transportation network and destination changes were made as part of the West Don
Lands Plan scenario. These are indicated in Table 6 and described in detail in the background
report.
Change Scenario 2: Medium Density Residential Only
For comparative purposes, a second West Don Lands Change Scenario was created and tested.
For this scenario, it was assumed that the entire site would contain only medium-density, single
family residential development, similar to that found in a 1950s middle class suburban
development. Change Scenario 2 uses a residential density of 6 dwelling units/acre. This is a
level similar to many of the residential areas in Scarborough neighbourhoods northeast of
downtown. These scenario assumptions are illustrated in Figure 3 and included in Table 6.
Scenario 3: Low Density Residential-Only Comparison
The West Don Lands results were also compared against estimates derived for an existing low
density, residential-only neighbourhood in Toronto that is surrounded by a buffer area that is
also of low density, residential development. The low density study area selected contains
twenty-one postal codes with an average net residential density of 2.4 units per acre and a
buffered postal code net residential density of 2.7 units per acre. This neighbourhood has very
different demographic and socioeconomic characteristics than that anticipated for the West
Don Lands Study Area. It has:
Table 6: West Don Lands Scenario Summary
Variables
Existing Conditions Scenario 1:
Plan
Scenario 2:
Medium Density
City-wide
Average
Postal
codes
(n=15)
Buffered
postal
codes
(n=15)
Postal
codes
(n=15)
Buffered
postal
codes
(n=15)
Postal
codes
(n=15)
Buffered
postal
codes
(n=15)
Buffered
postal codes
(n=43,091)
Net residential density
(units per residential acres)
282 72.4 221 119.0 6.0 36.8 23.8
Land use mix (0-1) 0 0.5 0.5 0 0.5 0.3
Retail floor area ratio 0 0.8 1.7 1.2 0 0.8 0.4
Intersection density
(count/sq km)
67.5 134.8 175 152.0 67.5 134.8 85.0
Transit density
(count/sq km)
17.5 40.4 27.5 42.0 17.5 40.4 30.1
Number of intersections 27 337 70 380 27 337
Since these
are not
averages, city
level values
are not
provided
Number of transit stops 7 101 11 105 7 101
Pedestrian-accessible
roads (km)
5.7 58.4 9.5 62.2 5.7 58.4
Bicycle Facilities (km) 1.5 11.8 8.2 18.5 1.5 11.8
Trails (km) 0.7 4.8 1.5 2.3 0.7 4.8
Schools 0 12 1 13 0 12
Food locations 3 159 37 193 3 159

• More people per household than West Don Lands and the City average
• More vehicles per household than West Don Lands and the City average
• A slightly higher education levels than West Don Lands and much higher levels than the
City average
• Dramatically higher household incomes than West Don Lands and the City average and
• A higher percentage of non-white people than West Don Lands and the City average.
Because demographic values are included in the formulas used to estimate outcomes, these
differences will account for some of the differences seen in the estimated outcome values.
Results – West Don Lands Change Scenarios
As illustrated in Tables 7 and 8, across all outcomes, Change Scenario 1 (the proposed
development) resulted in the greatest estimated positive impacts on physical activity and GHG
emissions from vehicles, with substantial increases in walking and cycling outcomes and
significant decreases in GHG emissions, vehicle trips, and VKT predicted, relative to the Existing
Conditions Scenario.For many outcomes, the West Don Lands redevelopment plan (Change
Scenario 1) produced estimated values that were very different from the City average as well;
active trips per person were almost 3.5 times greater, transit trips per person were 1.5 times
higher, GHG emissions were almost halved, and cycling to school and work on a monthly basis
were 10 times higher. The BMI and high blood pressure outcomes showed less change because
these outcomes are impacted by confounding factors that were beyond the scope of the study.
The medium density residential Change Scenario 2 produced the opposite results of Change
Scenario 1. Health- and environment-related outcomes estimated were poorer than the
Existing Conditions Scenario, with substantial decreases in active transportation and walking for
exercise, slight increases in BMI and high blood pressure, and substantial increases in vehicle
use and VKT. With Change Scenario 2, the impact on health- and environment-related
outcomes is less negative than it could be because the surrounding area, which is higher in
density and mixed land uses, serves to increase the overall walkability of the buffer area used
to estimate outcomes.
Table 7: Estimated Outcome Values for West Don Lands, Unit Level
Travel, Health- & Climate-
Related Outcomes
West Don Lands
Existing
Conditions
Change
Scenario 1:
West Don
Lands Plan
Change Scenario
2:
West Don Lands
Medium Density
Scenario 3:
Low Density
Study Area
City Level
Average
average active
trips/person/day
0.23 0.48 0.17 0.05 0.14
average transit
trips/person/day
0.60 0.79 0.51 0.32 0.49
average automobile
trips/person/day
1.00 0.52 1.20 2.31 1.33

Table 7: Estimated Outcome Values for West Don Lands, Unit Level
Travel, Health- & Climate-
Related Outcomes
West Don Lands
Existing
Conditions
Change
Scenario 1:
West Don
Lands Plan
Change Scenario
2:
West Don Lands
Medium Density
Scenario 3:
Low Density
Study Area
City Level
Average
average trip
kilometers/person/day
18.17 15.43 20.02 29.19 22.58
average CO2 generated from
vehicles (kg/HH/day)
3.38 2.39 4.28 5.92 4.21
walking for exercise monthly
freq.
14.25 15.57 13.66 9.52 10.12
walk to work/school monthly
freq.
7.79 10.94 7.57 2.24 5.58
bicycle for exercise monthly
freq.
1.08 1.53 0.93 0.49 0.63
bicycle to work/school
monthly freq.
0.80 2.71 0.74 0.05 0.25
daily energy expenditure
(kcal/kg/day)
2.28 2.73 2.02 2.29 2.04
body mass index 24.31 24.14 24.36 24.51 24.64
high blood pressure
(likelihood)
9.58% 9.11% 9.66% 7.82% 7.38%
As can be seen in Table 7, the estimates suggest that the Low Density Study Area (Scenario 3)
has the highest level of vehicle travel and the highest GHG emissions and the poorest health-
related outcomes of all the scenarios modelled, with fewer active trips, fewer walking trips for
exercise or to work or school, fewer transit trips, and higher BMIs.
Table 8: Estimated Outcome Values for West Don Lands, Population-Level
Outcome Existing
Conditions
Change
Scenario 1
Change
Scenario 2
Population 202 13,474 1,307
active trips/day 47 6,486 223
transit trips/day 121 10,663 661
automobile trips/day 201 7,062 1,566
trip kilometers/day 3,663 207,873 26,166
walking for exercise (times/month) 2,874 209,733 17,853
walk to work/school (times/month) 1,571 147,369 9,893
bicycle for exercise (times/month) 218 20,613 1,218
bicycle to work/school (times/month) 161 36,454 967

One exception to this pattern is the likelihood of having high blood pressure. The Low Density
Study Area reports the lowest likelihood of high blood pressure for all of the scenarios
examined. This outcome is attributed to the differences in demographic factors in the study
area. The Low Density Study Area is home to people with substantially higher income levels
and slightly lower ages than the West Don Lands study area; these factors decrease the
likelihood that people in this study area will have high blood pressure. With most outcomes,
the Low Density Study Area also performs more poorly than the City average.
Applying the Tool to Other Regions – Surrey Case Study
Introduction
To evaluate the feasibility of applying the tool to another urban region outside of Toronto and
describe the process and methods of using the tool in other parts of Canada, the CommunityViz
tool was also tested on a study area in British Columbia – the Central Station area of Surrey, a
previously suburban city Southeast of Vancouver that is rapidly becoming an urban centre in its
own right.
This study area was selected because Surrey Central Station is a major transit hub for the
region’s rapid transit system and is part of the Surrey City Centre. The type, scale and support
for the planned development around the station made it a viable pilot study site for application
of the software tool. The plan for this area is to:
• Introduce 42,000 more residents and 21,500 more employees by 2031 into the City
Centre, of which the Central Station is only a part
• Increase the residential and commercial density in the study area
• Potentially increase transit service in the area
• Increase street connectivity and pedestrian-oriented street design as part of the City
Centre plan
• Make the Surrey City Centre a Regional Town Centre in the Greater Vancouver Regional
District’s Livable Region Strategic Plan
Capitalize on the location of the Simon Fraser University Surrey Campus within the station area
to develop the area as an activity hub.
The plans for the Surrey City Centre were the basis for a “change scenario” to be tested in
CommunityViz. The current conditions were used as the baseline scenario.
Steps in the Process
Create Buffered Polygons
The postal code geography method used in the West Don Lands case study was adapted for
Surrey as postal code boundaries were not available. An automated process was developed to
use Surrey parcel data. This method can be used in any case where postal code polygons are
not available.

Using GIS, contiguous parcels smaller than 20,000 square metres were aggregated into larger
polygons. In many cases, the results were polygons that included all the parcels within a city
block. Large parcels over 20,000 square metres were treated as distinct polygons as these
parcels were often equivalent in
size to city blocks. Any very small
remaining polygons were merged
into a neighbouring polygon. The
aggregated small parcels and the
separate large parcels were then
merged into a single polygon GIS
shapefile.
The final result was 603 polygons,
with a mean polygon size of 20,000
square metres, within which nested
7804 individual parcels. Though
this polygon size was still larger
than the Toronto postal codes,
which are about 10,000 square
metres, smaller polygon sizes
would only be possible using an
entirely manual aggregation
process which was not within scope
of this project15
Insert Built Environment and Demographic Data
.
As in Toronto, built environment variables were calculated for Surrey for the area included in
the 1 km buffered polygon level. The buffer-level built environment measures created for the
City of Surrey, as well as their source layers, were loaded into the CommunityViz Analysis
Template developed for the City of Toronto case study. Dissemination area level 2006 Census
data were obtained from Statistics Canada via the City of Surrey. Dissemination area level
values were assigned to each polygon based on the location of the centre point for each
polygon. Once the Surrey datasets were loaded into the CommunityViz Analysis Template,
health- and environment-related outcomes were estimated for the existing and future
scenarios.
Reporting Areas
The results in CommunityViz can be reported for:
• The Study Area, which includes only the polygons within the study area itself. However,
15
In Toronto, a city block is often composed of 2-4 postal codes, while only one postal code is used for the same
sized city block in Surrey.
Figure 4: Parcel-based Polygons in Surrey, BC
Polygons coloured in grey (n=39) constitute the Surrey Central Station
study area. The black rectangle shows the extent of built environment
data that were provided by the City of surrey

the built environment measures for the study area polygons are calculated at the
buffered polygon level (i.e., the 1 km network buffer surrounding each polygon); and
• The Impacted Area, which is the area surrounding the study area that will be affected by
the changes in the study area. Any polygon whose buffer intersects the study area is part
of the impacted area.
Health- and environment-related outcomes are calculated for each polygon and then averaged
to create estimates for the entire study area or impacted area. Even though the measurement
areas (buffers) used to calculate the outcomes overlap, the results presented in CommunityViz
can be summed to calculate the total effect on the population living in the study area and the
impacted area.
Toronto CommunityViz Analysis Template Applied to Surrey
The equations and relationship coefficients used to estimate outcome values in Surrey are
based on data and research from Toronto, while the data used to apply them to Surrey are
based on Surrey built environment and demographic data. Individual estimates should be
reviewed carefully, and compared to other sources such as transportation model outputs, for
validation.
Vehicle ownership per household rates for each Surrey Place Type was based on rates originally
calculated in Toronto using the TTS. These rates are typically fairly constant between urban
regions, so they were not updated for Surrey. If local data were available to estimate vehicle
per household rates by housing type for Surrey, the assumptions in the CommunityViz Analysis
could be updated using those.
People per household rates for each place type in the Surrey existing conditions scenario were
based on rates originally calculated in Toronto based on the Canadian Census data. Again,
because these rates are typically fairly constant between urban regions, they were not updated
for Surrey. The City of Surrey provided the number of people per household for the change
scenario place types; these rates were consistent with the base Toronto rates.
Surrey Central Station Case Study
Existing Conditions
The Surrey Central Station study area is currently characterized by low-density commercial
development. Specific land uses include: a recreation centre and library, a large complex
including a regional shopping mall with a high-rise office tower, and the Simon Fraser University
Surrey Campus. The commercial area is surrounded by a mix of residential densities including
older single family detached housing, older low-rise apartments, and newer low-rise and high-
rise apartments. According to 2011 parcel data, there are currently 2,216 residential units in
this area. There is currently a large amount of vacant land within the study area boundary.
Much of this land is awaiting redevelopment at higher densities. The parcel data indicates that
this area includes 1.2 million square feet of retail, 1.1 million square feet of office space, and no
parks in the Surrey Central Station.

Transit service currently exists along 104 Ave, King George Blvd, Old Yale Road, and 132nd
Street. The Surrey Central Skytrain stop is located in the centre of the area, which services
neighbourhoods to the north and south of the study area. There are several food outlets, which
are mostly concentrated in the eastern part of the study area along King George Blvd. There are
currently no parks located within the study area. Several of the major arterials are designated
as bicycle facilities, and connect to parks just beyond the study area boundaries.
Built environment measures include net residential density, intersection density, retail floor to
land area ratio and land use mix. The addition of housing, office and commercial space, and
roads within the study area will increase housing unit counts, residential land area, net-
residential density, land use mix, retail floor area ratio, intersection density, and utilitarian
walkability in the area. The relative amount of change in these built environment measures will
not be as great at the larger 1 km buffered polygon level as it will be for the area within the
smaller case study boundaries. Further, the buffers overlap each other substantially since they
are drawn for a set of small, contiguous polygons. This is appropriate and needed to place each
polygon in the larger context of its neighbourhood.
Change Scenario: Surrey Central Station Plan
Planners and decision-makers in
Surrey have been reviewing the
City Centre’s existing conditions
and developing plans and
strategies for future development,
in order to accommodate
anticipated growth in a manner
that reinforces the area’s role as a
centre to serve the local
community, a City
Centre/downtown for Surrey, and
a Regional Town Centre for the
area south of the Fraser River.
The land use plans for the Surrey
Central Station (SCS) area include
high density mixed-use
development, including a number of new institutional land uses; a new City Hall, an arts centre,
and an expansion of the Simon Fraser University Surrey campus. Transit plans include
additional rapid transit and a reconfigured bus exchange. In addition, significant improvements
are planned for pedestrian and cycling infrastructure including a finer-grained street pattern.
The resources referenced provided general guidance, but did not provide sufficient detail to
develop the existing conditions and change scenarios needed to apply the software tool. In
some cases it was be necessary to make some assumptions to achieve the required level of
detail regarding built environment changes. Because of the stage of the City Centre planning
Figure 5: Surrey Central Station Plan Assumptions

process, these assumptions were often not finalized or documented. Discussions with City of
Surrey planners helped to identify appropriate assumptions (see Figure 5 and Table 9).
Scenario Built Environment Measures
Table 9 provides a comparison of the existing conditions and the results of applying the
CommunityViz place types to represent the Surrey Central Station Plan conditions. The major
changes are associated with housing density, roads and food locations. Table 10 compares the
built environment values for the existing conditions and the Surrey Central Station Plan.
16
Only calculated at the buffered polygon level in the change scenario.
Table 9: Comparison of Existing Conditions and Plan Scenarios, Surrey
Existing Conditions Surrey Central Station Plan
Housing Units 2,216 14,175
Retail/office floor area (sq ft) 2,379,163 2,494,094
Civic/institutional floor area (sq ft) 340,000 1,286,515
Roads 14.2 km 20.0 km
Intersections 47 91
Transit Stops 33 33
# of schools 0 0
# of food locations 53 73
Parks (acres) 0 1.8
Population 4,899 23,676
Table 10: Built Environment Values for Surrey Study Area Scenarios
Built Environment Variables Existing Conditions Surrey Central Station Plan
Net residential density (residential
units per residential acres)
17.9 101.0
Land use mix (0-1)16 0.11
Retail floor area ratio 0.20
Intersection density (count/sq km) 47.0 91.0
Transit density (count/sq km) 33.0 33.0
Number of intersections 47 91
Number of transit stops 33 33
Pedestrian-accessible roads (km) 14.2 20.0
Bicycle Facilities (km) 10.6 21.6
Trails (km) 6.7 7.2
Schools 0 0
Food locations 53 73

Results – Surrey
Table 11 compares the health- and environment-related outcomes for the two scenarios for
the study area on a unit basis. The analysis estimated that, if implemented as proposed, the
Surrey Central Station Plan could double the number of cycling trips to work or school per
person on a monthly basis, double the trips made with active transportation per person on a
daily basis, and nearly halve the trips taken by vehicles on a daily basis in the study area (see
Table 11). To create a clearer picture of the total impact, the unit level estimates were
multiplied by the study area population to create the population level estimated provided in
Table 12. The CommunityViz Analysis Template estimated that, across all outcomes, the Surrey
Central Station scenario would result in positive health- and environment-related outcomes
with decreases in vehicle use and increases in active transportation.
Table 11: Estimated Outcomes for Surrey Study Area for Scenarios, Unit Level
Outcomes Existing Conditions Surrey Central
Station Plan
average active trips/person/day 0.24 0.46
average transit trips/person/day 0.61 0.74
average automobile trips/person/day 0.89 0.52
average trip kilometers/person/day 23.77 22.40
average CO2 generated from vehicles (kg/household/day) 4.74 3.73
walking for exercise monthly freq. 13.91 14.23
walk to work/school monthly freq. 7.97 8.44
bicycle for exercise monthly freq. 0.57 0.65
bicycle to work/school monthly freq. 1.26 2.54
daily energy expenditure (kcal/kg/day) 1.40 1.55
body mass index 24.92 24.86
high blood pressure (likelihood) 7.82% 7.68%
Table 12: Estimated Outcomes for Surrey Study Area for Scenarios, Population Level
Outcome Existing Conditions Surrey Central
Station Plan
Population 4,899 23,676
active trips/day 1,199 10,778
transit trips/day 2,991 17,481
automobile trips/day 4,366 12,356
trip kilometers/day 116,453 530,392
walking for exercise (times/month) 68,132 336,794
walk to work/school (times/month) 39,032 199,913
bicycle for exercise (times/month) 2,791 15,320
bicycle to work/school (times/month) 6,167 60,243

CommunityViz Analysis Template, Data Sharing & Further
Applications
The CommunityViz Analysis template developed as part of this project can be used again and
again to evaluate health- and environment-related outcomes in other areas where changes in
the built environment are being considered. This CommunityViz Analysis template is a
computer file which contains the programmed equations for the statistical models, as well as
the formulas for the various built environment variables and default values needed by the
models. The template does not contain any disaggregated Census or other built environment
related data that would present confidentiality or data-sharing concerns.
Within Toronto
TPH has the most recent version of the CommunityViz software and the computer file
containing the reusable CommunityViz Analysis Template. In addition, Toronto has the GIS
shapefile containing postal-code-level built environment measures for existing conditions in the
City of Toronto and dissemination-area-level Census demographic data. This file is needed to
supply data on existing conditions anywhere in the city as one of the first steps in using the
CommunityViz tool. It is also desirable, but not required, to make use of parcel data within a
CommunityViz Analysis. In order to use these datasets (Census and MPAC parcel data), data-
sharing agreements are needed between the data providers and the data users. As a result of
this, the use of the GIS shapefile, containing the Toronto Census data, and the auxiliary use of
the parcel data will be restricted to those with the necessary permissions. Toronto Public
Health currently has the necessary data-sharing agreements in place with Statistics Canada and
MPAC to use the data provided.
Beyond Toronto
The Surrey case study demonstrated that the CommunityViz Analysis template is adaptable for
use in other regions. The same template is used whether inside or outside the City of Toronto,
but the data loaded into it will be different. While the statistical equations and coefficients
were developed using disaggregated Census and parcel data for the City of Toronto, the
CommunityViz Analysis template contains no disaggregated Census or parcel data for Toronto.
Users in other regions will use the same CommunityViz Analysis template but they will need to
use their own built environment and demographic data to populate the tool. Users outside the
City of Toronto will also need postal code boundaries or can develop their own polygon
geography similar in scale to the postal code as was done in the Surrey case study.
Limitations beyond Toronto
As with any professional planning exercise, results should be reviewed carefully, particularly in
smaller regions that differ substantially in urban form from Toronto or in studies testing urban
form changes well outside the spectrum of that present in Toronto today. Place Type
assumptions for people per residential dwelling type, square feet per residential dwelling type,
and vehicles per residential dwelling type currently contain City of Toronto-based assumptions.

These assumptions would ideally be replaced with locally available data if they are found to be
substantially different from the Toronto assumptions.
Requirements for Using the Tool
The basic requirements for using the tool are:
• ArcGIS Version 10 software and network analyst extension which can be purchased from
ESRI at www.esri.com and a proficient user;
• CommunityViz software which can be purchased from Placeways, LLC 2012 with one
license, with support, costing US$850 in 2012;
• The CommunityViz Analysis Template, developed as part of this project, which is free;
• Census dissemination area level demographic data and parcel data with attributes such as
land use type, building floor area, residential unit counts;
• Data availability and staff time to acquire, format and use it to create files in the
format/structure required by CommunityViz; and
• A study area with enough details about future changes to represent it in CommunityViz.
It is estimated that staff would require about 120 to 160 hours to create the data files needed
to use this tool in a municipality or region. This time does not include the time required to
acquire data from data providers. Once this work is completed for one study area within a
municipality or region, it does not need to be redone each time a new study area within that
larger area is evaluated. Each new study area will, however, require time to acquire and input
the data for the future conditions into CommunityViz.
To What Situations Can the Tool Be Applied?
The CommunityViz Analysis Tool is best applied to secondary plans and block plans but it could
be adapted or enhanced to be applied to other situations. The tool can be used to assess:
• The impact of arterial road redesign but it would need to be updated with additional data.
The current models do not differentiate between different road network configurations
but these data are available and could be readily integrated into the model.
• The impacts associated with the replacement of buses with Light Rail Transit or the
replacement of on-street buses with an underground subway, but it would require
additional data that captures the relative service characteristics across and between
different modes of travel. This includes factors impacting travel choice that are distinct
across competing modes of transit such as travel time, perceptions of safety, reliability,
comfort, and convenience. This information could be added to the model already
developed as an enhancement.
• The health benefits of separated bike lanes but this would require primary data collection
on air pollution exposure, and health benefits and safety impacts associated with
integrated and shared versus separated bike facilities.

Future Development of the Tool
During the data development and analysis phase of this project, a number of potential
improvements were noted that would help further refine the analysis and the tool. When data
and resources become available, the following set of potential improvements could be
considered:
• Detailed Locational Information – Precise address information was not available for TTS
or CCHS participants, which necessitated using built environment data created at the
postal code level. Address level participant data would increase the precision of the
analysis, as urban form measures could then be developed for the unique neighbourhood
accessible from each participant’s address location. Address level data would also enable
the creation of specific distance-based measures to different types of amenities.
• Expanded Geography for the Base Analysis - Currently, the analysis of the relationships
between built environment variables and health- and environment-related outcomes is
only based on the City of Toronto; it excludes the municipalities surrounding Toronto in
the Greater Toronto Area. As previously noted, having a wider range of urban forms
including exurban areas is critical to be able to detect associated differences in
behaviours and health outcomes. If additional areas with different development patterns
were included in the base analysis, the tool could be applied to a greater number of
communities in Canada.
• More Detailed Trip Data - Due to confidentiality issues, trip and person level data were
not available from the TTS. Data were aggregated to the average person or household at
the postal code level. If these data could be broken out by the trip per person or
household, more precise measurements of travel time and distance could be made. The
tool could then make more precise estimates of CO2 emissions. It could also be used to
estimate emissions of air pollutants that affect human health directly.
• More Precise CO2 Estimates – The CO2 estimates created for the tool could be refined in a
number of ways. Having detailed trip end address information, as noted above, would
make it possible to develop detailed “shortest time paths” which take into account
vehicle congestion between trip origins and destinations at the time of the trip. CO2
estimates could then be based on mode-specific shortest time routes, rather than the
postal code level average distances that were used in the analysis. Other improvements
that could be made with trip/person-level data include accounting for the type of vehicle,
vehicle occupancy, road- segment level congested-flows, and modelled speeds as noted
above.
• Expand the Built Environment Variables – Additional built environment datasets have
been made available since the data development stage of this project.
For example, bicycle post and ring locations are now available from the City of Toronto
open data website (http://www1.toronto.ca/wps/portal/open_data). In addition, the Bixi
Toronto (bike share) program that began in May 2011 could help characterize how cycling

behaviour varies across the City of Toronto. It is also potentially a rich source of cycling
data; Bixi locations could be incorporated into the analysis and the tool in the future.
Lastly, two datasets available from City of Toronto open data website may improve the
modelling of leisure or recreation-based outcomes; land cover data on tree cover from
Erdas Imagine raster imagery and the City of Toronto street tree dataset.
• Expand the Health Outcomes Included – The analysis did not include all of the health-
related outcomes that are potentially related to the built environment. For example,
datasets with locations of pedestrian/vehicle collisions in Toronto have recently become
available for the years 2000-2009. It would be worthwhile to explore how this type of
information could be incorporated into health/travel models in the future. Data on crime
and neighbourhood disorder could also provide insights into other factors which affect
travel choices and levels of physical activity among residents.
• Upgrade Physical Activity Data – Objectively measured data on physical activity, such as
data from accelerometers, is known to be significantly different, and presumably more
accurate, than self-reported data provide by survey participants. A comparatively small
sample of accelerometer data on physical activity could be used to correct the CCHS and
to re-calibrate the CommunityViz Analysis template.
• Validation of the Tool in Other Regions - The CommunityViz Analysis template should be
validated in other regions using CCHS data and transportation survey data for those
regions.
• Monetization of Changes in Health Outcomes – It would be possible to estimate the
costs and cost savings associated with different scenarios based on estimated health
impacts. This cost information could be implemented into CommunityViz as well, so that
in addition to health outcomes, the model could also report estimated health-related
costs.

References
Bagley, MN, Mokhtarian, PL (2002). The impact of residential neighborhood type on travel
behavior: A structural equation modeling approach. Annals of Regional Science, 36, 279–297.
Bento AM, Cropper ML, Mobarak AM and Vinha K (2003), The Impact of Urban Spatial Structure
on Travel Demand in the United States, World Bank Group Working Paper 2007, World Bank
(http://econ.worldbank.org/files/24989_wps3007.pdf).
Boarnet, M and Crane R (2001). “The Influence of Land Use on Travel Behavior: A Specification and
Estimation Strategies,” Transportation Research A, Vol. 35, No. 9 (www.elsevier.com/locate/tra),
November 2001, pp. 823-845.
Booth MN, Owen A, Bauman A, Clavisi O, Leslie E (2000). Social-cognitive and perceived
environmental influences associated with physical activity in older Australians. Prev Med 31: 15-
22.
Burns NM, Baker CA, Simonson E, Keiper C (1996). Electrocardiogram changes in prolonged
automobile driving. Perceptual & Motor Skills 23(1):210.
Cervero, R (1991). “Congestion Relief: The Land Use Alternative.” The Journal of Planning
Education and Research Vol. 10, pages 119-129.
Cervero, R, Gorham R (1995). Commuting in Transit versus Automobile Neighborhoods. Journal of
the American Planning Association Vol. 61, p. 210-225.
Cervero R and Kockelman K (1997). “Travel demand and the 3Ds: density, diversity, and Design.”
Transportation Research Part D 2(3), 199-219.
City of Surrey. (July 25, 2011). Corporate Report-- Surrey City Centre Plan Update – Stage 2 - Status
Report. File: 6520-20 (Surrey City Centre).
http://www.surrey.ca/bylawsandcouncillibrary/CR_2011_R151.pdf
City of Surrey. (October 23, 2006). Corporate Report-- Surrey City Centre Plan Update - Status
Report on Phase I Results. File: 6520-20 (Surrey City Centre).
http://www.surrey.ca/files/CRC020.pdf
Cohen S, Underwood LG, Gottlieb BH. 2000. Social Support Measurement and Intervention. Oxford
University Press. New York..
Diette GB, Lechtzin N, Haponik E, Devrotes A, Rubin HR (2003). “Distraction therapy with nature
sights and sounds reduces pain during flexible bronchoscopy: a complementary approach to
routine analgesia.” Chest 123(3):941-8.

Dumbaugh, E (2005). “Safe Streets, Liveable Streets.” Journal of the American Planning
Association, Vol. 71, No. 3.
Dunphy, R, Fisher K (1996). Transportation, Congestion, and Density: New Insights.
Transportation Research Record 1552, TRB, National Research Council, Washington D.C., pp. 89-
96.
Ewing R, Schmid T, Killingsworth R, Zlot A, Raudenbush S (2003a). “Relationship Between Urban
Sprawl and Physical Activity, Obesity, and Morbidity,” American Journal of Health Promotion, Vol.
18, No. 1 pp. 47-57.
Ewing R, Cervero R (2001). Travel and the Built Environment-A Synthesis. Transportation Research
Record 1780. TRB, National Research Council, Washington D.C., pp. 87-114.
Fontana AF, Kerns RD, Rosenberg RL, Colonese KL. (1989) Support, stress, and recovery from
coronary heart disease: a longitudinal causal model. Healthy Psychology 8(2):175-93.
Frank L (2000). Land Use and Transportation Interaction: Implications on Public Health and Quality
of Life. Journal of Planning, Education, and Research 20, 1: 6-22.
Frank L and Engelke P (2001). “The Built Environment and Human Activity Patterns: Exploring the
Impacts of Urban Form on Public Health.” Journal of Planning Literature 16, 2: 202-18.
Frank L and Pivo G (1995). “Impacts of Mixed Use and Density on Utilization of Three Modes of
Travel: SOV, Transit and Walking,” Transportation Research Record 1466, pp. 44-55.
Frank L, Andresen M and Schmid T (2004). Obesity Relationships With Community Design, Physical
Activity, and Time Spent in Cars. American Journal of Preventive Medicine Vol 27. No 2. June,
2004, pp. 87-97.
Frank L, Stone B and Bachman W (2000). “Linking Land Use with Household Vehicle Emissions in
the Central Puget Sound: Methodological Framework and Findings.” Transportation Research Part
D 5, 3: 173-96.
Frank LD, Saelens B, Powell KE, Chapman JE (2007a). Disentangling Urban Form Effects on Physical
Activity, Driving, and Obesity from Individual Pre-Disposition for Neighborhood Type and Travel
Choice: Establishing a Case for Causation. Social Science and Medicine, Volume 65, Issue 9: pp.
1898-1914.
Frank L, Sallis JF, Conway T, Chapman J, Saelens B, Bachman W (2006). “Multiple Pathways from
Land Use to Health: Walkability Associations With Active Transportation, Body Mass Index, and Air
Quality.” Journal of the American Planning Association Vol. 72 No. 1.

Frank LD, Schmid T, Sallis JF, Chapman J, Saelens B (2005). “Linking Objective Physical Activity
Data with Objective Measures of Urban Form.” American Journal of Preventive Medicine. Volume
28, No. 2S.
Frumkin H, 2001. “Beyond toxicity: The greening of environmental health.” Am J Prev Med
20:47-53.
Giles-Corti B, Macintyre S, Clarkson JP, Pikora T and Donovan RJ (2003). Environmental and
lifestyle factors associated with overweight and obesity in Perth, Australia. American Journal of
Health Promotion, 18, 93-102.
Guite HF, Clark C, Ackrill G. 2006. The impact of physical and urban environment on mental well-
being. Public Health 120:1117-1126.
Handy SL and Boarnet MG et al. (2002). “How the Built Environment Affects Physical Activity:
Views from Urban Planning.” In American Journal of Preventive Medicine, Vol. 23, No. 2
(supplement), pp. 64–73.
Handy S, Cao X, & Mokhtarian PL (2006). Does self selection explain the relationship between built
environment and walking behavior? Empirical evidence from Northern California. Journal of the
American Planning Association, 72(1), 55–74.
Hennessy DA, Wiesenthal DL (1997). “The relationship between traffic congestion, driver stress,
and direct versus indirect coping behaviours.” Ergonomics 40:348-61.
Hoehner, C.M., C.E. Barlow, P. Allen, M. Schootman. (2012). “Commuting Distance,
Cardiorespiratory Fitness, and Metabolic Risk (DOI: 10.1016). American Journal of Preventive
Medicine, Volume 42, Issue 6 (June 2012).
Hoffman H (1965). “Medizinisch-Psychologische Untersuchungen Zum Fahren im Verkehrsfluss”
[Medical-psychological studies on driving in traffic]. Ztschr f Verkehrssicherh 11:145-55.
Hoffman H, Reygers W (1960). “Kreislaufuntersuchungen bei Kraftfahrzeugfahrern unter
variierten fahrbedingungen” [Studies on the circulation of drivers under varying driving
conditions]. Zentralbl f Verkehrs Med 1960;3:131-151.
Holtzclaw, J (1994), Using Residential Patterns and Transit to Decrease Auto Dependence and
Costs. National Resources Defense Council.
Kaplan R, Kaplan S, Ryan RL (1998). With People in Mind: Design and Management of Everyday
Nature. Washington: Island Press.
Khattak AJ, Rodriguez D (2005). Travel behavior in neotraditional neighborhood developments: A
case study in USA. Transportation Research Part A, 481–500.

clasp-software-summary-final-april-2013

clasp-software-summary-final-april-2013

Related slideshows

More Related Content

clasp-software-summary-final-april-2013