Measuring the Health Effects of SPRAWL

Measuring the Health Effects ofSPRAWL

A National Analysis of Physical Activity, Obesity and Chronic Disease

Barbara A. McCann

Reid Ewing

Smart Growth America

Surface Transportation Policy Project

SEPTEMBER 2003

Measuring the Health Effects ofSPRAWL

Barbara A. McCann

Reid Ewing

Smart Growth America

Surface Transportation Policy Project

SEPTEMBER 2003

A National Analysis of Physical Activity, Obesity and Chronic Disease

Acknowledgements

Measuring the Health Effects of Sprawl: A National Analysis of Physical Activity,

Obesity, and Chronic Disease was written by Barbara A. McCann and Reid Ewing.Additional data analysis was provided by Michelle Ernst, Linda Bailey, and John Pucher.The sprawl index used in this study was developed with assistance from Rolf Pendall ofCornell University and John Ottensmann of Indiana University. Editorial suggestions andgeneral advice were provided by Don Chen, David Goldberg, Thomas Schmid, RichardKillingsworth, Teri Larson, Chuck Alexander, Jennifer Hudman, Susan Krutt, KathrynThomas, and Kate Kraft. Images were provided by Lawrence Frank and Michael Ronkin.The authors would also like to thank Rutgers University, the Surface TransportationPolicy Project, and Smart Growth America (SGA) for their longstanding support.

Support for this project was provided by a grant from The Robert Wood JohnsonFoundation in Princeton, New Jersey. Additional funding and resources were alsoprovided by Smart Growth America, which is supported by the Surdna Foundation, theGeorge M. Gund Foundation, the Ford Foundation, the Fannie Mae Foundation, theDavid and Lucile Packard Foundation, the William and Flora Hewlett Foundation, theHenry M. Jackson Foundation, the Turner Foundation, the Prince Charitable Trusts, theJoyce Foundation.

Copies of Measuring the Health Effects of Sprawl can be obtained at Smart GrowthAmerica’s website, www.smartgrowthamerica.org. Hard copies can be obtained for $15by calling or writing SGA, 1200 18th St. NW Suite 801, Washington DC 20036, (202) 207-3350, or by emailing [email protected].

Barbara A. McCann, former Director of Information and Research at Smart GrowthAmerica, is a writer and public policy expert on the impact of the built environment onpublic health and quality of life. She can be reached at [email protected].

Reid Ewing is Research Professor at the National Center for Smart Growth, University ofMaryland. This research was conducted while Professor Ewing was Director of theVoorhees Transportation Center at Rutgers University.

Table of Contents

Executive Summary .................................................................................................................1

1. Introduction ..........................................................................................................................5

II. How the Study Was Done ..................................................................................................9

III. Findings: How Sprawl Relates to Weight,

Physical Activity, and Chronic Disease ............................................................................ 13

IV. The Need for Further Research .....................................................................................23

V. Considering Health When We Plan Our Communities .............................................25

Appendix .................................................................................................................................. 31

Executive Summary

Health experts agree: most Americans are too sedentary and weigh too much.Obesity has reached epidemic levels, and diseases associated with inactivityare also on the rise. What is creating this public health crisis? Much of the

focus to date has been on whether Americans are eating too much fattening food. Butresearchers are starting to pay attention to the other half of the weight-gain equation:Americans’ low levels of physical activity. A pressing question for public health officialsis whether the design of our communities makes it more difficult for people to getphysical activity and maintain a healthy weight.

This report presents the first national study to show a clear association between thetype of place people live and their activity levels, weight, and health. The study,Relationship Between Urban Sprawl and Physical Activity, Obesity, and Morbidity,

found that people living in counties marked by sprawling development are likely towalk less and weigh more than people who live in less sprawling counties. Inaddition, people in more sprawling counties are more likely to suffer fromhypertension (high blood pressure). These results hold true after controlling forfactors such as age, education, gender, and race and ethnicity.

Researchers measured the degree of sprawl with a county ‘sprawl index’ that used dataavailable from the US Census Bureau and other federal sources to quantifydevelopment patterns in 448 counties in urban areas across the United States.Counties with a higher degree of sprawl received a lower numerical value on the index,and county sprawl index scores range from 63 for the most sprawling county to 352 forthe least sprawling county. Sprawling counties are spread-out areas where homes arefar from any other destination, and often the only route between the two may be on a

The findings presented here are from the article, Relationship Between Urban Sprawl

and Physical Activity, Obesity and Morbidity, by Reid Ewing, Tom Schmid, RichardKillingsworth, Amy Zlot, and Stephen Raudenbush, published in the September 2003issue of the American Journal of Health Promotion.

This report is intended to make this important piece of research more accessible to thegeneral public. In addition to presenting research findings, this report summarizesrecent research done by others on the links between the way we’ve built ourcommunities, physical activity, and health. It also includes recommendations for changeand resources for those interested in further exploration of this topic.

2 • Measuring the Health Effects of Sprawl: A National Analysis

busy high-speed arterial road that is unpleasant or even unsafe for biking or walking.People who live in these areas may find that driving is the most convenient way to geteverything done, and they are less likely to have easy opportunities to walk, bicycle, ortake transit as part of their daily routine.

Indeed, previous research has shown that people living in sprawling areas drive more,while people living in compact communities are more likely to walk. Medical researchhas shown that walking and similar moderate physical activity is important tomaintaining healthy weight and bestows many other health benefits. What isgroundbreaking about this study is that it is the first national study to establish adirect association between the form of the community and the health of the peoplewho live there.

Analysis shows sprawl is linked to healthThe study compared the county sprawl index to the health characteristics of more than200,000 individuals living in the 448 counties studied, using a large national healthsurvey, the Behavioral Risk Factor Surveillance System (BRFSS), which is maintained bythe Centers for Disease Control and Prevention (CDC).

The results show that people in moresprawling counties are likely to have a higherbody mass index (BMI), a standard measureof weight-to-height that is used to determineif people are overweight or obese. A 50-pointincrease in the degree of sprawl on thecounty sprawl index was associated with aweight gain of just over one pound for theaverage person. Looking at the extremes, thepeople living in the most sprawling areas arelikely to weigh six pounds more than peoplein the most compact county. Expecteddifferences in weight for an average personliving in different counties are shown inFigure 1, left. Obesity, defined as a BMI of 30or higher, followed a similar pattern. Theodds that a county resident will be obeserises ten percent with every 50-pointincrease in the degree of sprawl on thecounty sprawl index.

The study also found a direct relationshipbetween sprawl and chronic disease. The oddsof having hypertension, or high bloodpressure, are six percent higher for every 50-point increase in the degree of sprawl. The 25most sprawling counties had average

0

50

100

150

200

250

300

350

400

New York County 161.1

Kings County 163.1Bronx County 163.3

Queens County 164.0San Francisco County 164.2

Suffolk County 164.9

Cook County 165.5

Delaware County 166.1

McHenry County 166.6Clay County 166.9 El Dorado County 167.0

Hanover County 167.2Isanti County 167.3Walton County 167.3 Geauga County 167.5

Cou

nty

Spra

wl S

core

more sprawling

less sprawling

FIGURE 1. Sprawl and WeightExpected Weight for a 5’7” Adult (lbs.)

Smart Growth America • STPP •3

hypertension rates of 25 per 100 while the 25least sprawling had hypertension rates of 23 per100. The researchers did not find anystatistically significant association betweencommunity design and diabetes orcardiovascular disease. While all three chronicconditions are associated with being inactiveand overweight, many other factors includingheredity may moderate the relationshipbetween sprawl and chronic diseases.

People in sprawling areas walk less forexercise, which may help explain the higherobesity levels. But routine daily activity, suchas walking for errands, may have a bigger role.When the researchers controlled for theamount of walking for exercise that peoplereported, they found that people in moresprawling counties weigh more whether or notthey walk for exercise. This suggests thatpeople in sprawling areas may be missing out on significant health benefits that areavailable simply by walking, biking, climbing stairs, and getting physical activity aspart of everyday life.

These results point toward the need to continue investigating how our communities maybe affecting our health. Additional studies are needed to better understand therelationship between sprawling development and the risk of being overweight, and tomore precisely measure physical activity.

Creating Healthy CommunitiesWe know that people would like to have more opportunities to walk and bicycle: recentnational polls found that 55 percent of Americans would like to walk more instead ofdriving, and 52 percent would like to bicycle more. Leaders looking to reshape theircommunities to make it easier to walk and bicycle have many options. They can investin improved facilities for biking and walking, install traffic calming measures to slowdown cars, or create Safe Routes to School programs that focus on helping kids walkand bike to school. They also can create more walkable communities by focusingdevelopment around transit stops, retrofitting sprawling neighborhoods, and

25 Most Compact Counties 25 Most Sprawling Counties0%

5%

10%

15%

20%

25%

30%

22.8%

25.3%

FIGURE 2. Sprawl and Blood Pressure Percent of Adult Population with Hypertension

Source: BRFSS Hypertension rates, weighted by county (1998-2000).

People living in counties marked by sprawling development are likely to walk less, weigh

more, and are more likely to have high blood pressure.


revitalizing older neighborhoods that are already walkable. When paired withprograms that educate people about the benefits of walking, these changes can helpincrease physical activity.

Addressing these issues is essential both for personal health and for the long-termhealth of our communities. Physical inactivity and being overweight are factors inover 200,000 premature deaths each year. The director of the CDC recently saidobesity might soon overtake tobacco as the nation’s number-one health threat.Meanwhile, rising health care costs are threatening state budgets. Getting decisionmakers to consider how the billions spent on transportation and development canmake communities more walkable and bikeable is one avenue to improving the healthand quality of life of millions of Americans.

Getting decision makers to consider how the billions spent on transportation

and development can make communities more walkable and bikeable is one

avenue to improving the health and quality of life of millions of Americans.

Oregon Department of Transportation, Pedestrian & Bicycle Program

1. Introduction

More Than a Personal Problem

W eight loss is an American obsession, one that has played out almostexclusively at the individual level. Diet books and programs are ubiquitous,and each January gyms burst with new members determined to stick to their

New Year’s resolutions. Yet the American waistline has continued to expand at analarming rate, and obesity has been declared an epidemic.1 Recent data from a nationalsurvey found that almost 65 percent of the adult population is overweight and almostone in three people is obese.2 In the past 25 years, the portion of children 6-11 who areoverweight has doubled, while the portion of overweight teens has tripled: now 15percent of children and teenagers aged 6 to 19 are overweight.3 And this epidemic is farfrom a cosmetic concern: being overweight is a contributing factor to many chronic

10%-14% 15%-19% 20%-24% ≤ 24%

Portion of Adult Population Obese

FIGURE 3. Obesity* Among U.S. Adults*BMI>30 or ~30lbs overweight for a 5’4” woman

Source: Mokdad A H, et al. J Am Med Assoc 2001;286:10.

–


diseases and conditions, including hypertension, type-2 diabetes, colon cancer,osteoarthritis, osteoporosis, and coronary heart disease. 4 The director of the CDCrecently said that obesity and physical inactivity are gaining on tobacco and may soonovertake tobacco as the nation’s number one health threat.5

While Americans traditionally have seen weight as a personal concern, public healthadvocates have begun looking at how factors beyond our personal control may bemaking it harder to stay fit. Much of the public debate around obesity has focusedon the constant availability of fattening snacks, the ‘super sizing’ of portions, andthe marketing practices of fast-food restaurants. Now, health advocates are lookingto our physical surroundings as a contributor to weight gain as well: If theenvironment is making it too easy to overeat, might there be something about ourcommunities that is making it too difficult to get the physical activity needed tostay fit?

Physical Activity and SprawlThere is good reason to suspect that a lack of physical activity is contributing toobesity and other health problems. Three in four Americans report that they do notget enough exercise to meet the recommended minimum of either 20 minutes ofstrenuous activity three days a week or 30 minutes of moderate activity five days aweek.6 About one in four Americans remains completely inactive during their leisuretime. Yet these alarming statistics are not new. Reported exercise levels have remainedsteady for decades.7

What may be changing is the amount of physical activity people get in the course ofeveryday life. People move about as part of doing their jobs, taking care of their homes

and families, and especially asthey travel from place to place.One hint that this type ofmovement may be in declinecomes from a recent poll thatfound that while 71 percent ofparents of school-aged childrenwalked or biked to school whenthey were young, only 18 percentof their children do so.8 Also,according to the US Census,between 1990 and 2000 theportion of working Americanswho walked to work droppedfrom 3.9 to 2.9 percent.

Preliminary studies show thatthe movement from compactneighborhoods to spread-out,automobile-dependent

1991 1995 1999 20010%

5%

10%

15%

20%

25%

12.0%

15.3%

18.9%

20.9%

Perc

enta

ge o

f adu

lts w

ho a

re o

bese

FIGURE 4. Trend in Adult Obesity, 1991-2001

Source: National Center for Chronic Disease Prevention and Health Promotion, CDC.BRFSS data 1991-2001.


communities has meant a declinein daily physical activity. Acommon denominator of modernsprawling communities is thatnothing is within easy walkingdistance of anything else. Housesare far from any services, stores,or businesses; wide, high-speedroads are perceived as dangerousand unpleasant for walking; andbusinesses are surrounded byvast parking lots. In suchenvironments, few people try towalk or bicycle to reachdestinations. Urban planningresearch shows that ‘urban form’ –the way streets are laid out, thedistance between destinations,the mix of homes and stores — islinked to physical activity because it influences whether people must drive or areable to choose more physically active travel, such as walking.9

Even as routine physical activity seems to be declining, recognition of itsimportance is growing in the public health community. Evidence is mounting thatmoderate activity can have a significant impact on health, an impact that goes farbeyond weight control. People who are active are less likely to suffer from coronaryheart disease, non-insulin dependent diabetes, high blood pressure, or to get colon

Percentage of adults who walked or bicycled to school

Percentage of children who now walk or bicycle to school

0% 10% 20% 30% 40% 50% 60% 70% 80%

18%

71%

FIGURE 5. Fewer Children are Walking to School

A common denominator of modern sprawling communities is that nothing is

within easy walking distance of anything else. Houses are far from any services,

stores, or businesses; wide, high-speed roads are perceived as dangerous

and unpleasant for walking; and businesses are surrounded by vast parking lots.

Source: Surface Transportation Policy Project. American Attitudes Toward Walking and

Creating Better Walking Communities. April 2003.



Moderate Physical ActivityUS Health and Human Services Secretary TommyThompson recently released a report that urgedAmericans to get moving: “Simply walking 30minutes a day can have a measurable impact on aperson’s health and in preventing diseases such asdiabetes. You don’t need to join a gym or be a greatathlete to get active and make a difference in yourhealth.”1 Public health advocates have coined theterm “active living” to describe a way of life thatintegrates physical activity into daily routines. It can mean walking to the store or towork, climbing stairs instead of taking the elevator, or biking to school. For decades,health experts have advocated getting physical activity through vigorous aerobicexercise, but recent research shows that even moderate activity yields significantbenefits, especially for those who are generally inactive.

or pancreatic cancer. Physical activity helps relieve the symptoms of arthritis, andcan help lift depression, relieve anxiety, and result in an overall improvement inmood and well-being.

Public health researchers have just begun to conduct research on how the builtenvironment affects physical activity.10-15 They are asking: Have we “engineered”movement out of our daily lives to such a degree that our neighborhoods are nowcontributing to the obesity epidemic and other health problems?16 This study is animportant step toward answering this question because it is the first rigorousnationwide investigation of the potential relationship between urban form, physicalactivity, and health.

U.S. Department of Health and Human Services.

II. How the Study Was Done

This report is based on a study that required an intensive and unusualcollaboration between urban planning researchers and public health researchers.Reid Ewing, then at Rutgers University and now Research Professor at the

National Center for Smart Growth, University of Maryland, was the principalinvestigator and spent many months in Atlanta working with researchers from theCDC, finding ways for the two very different fields to speak a common language anddesign a rigorous research methodology. He worked with Tom Schmid and Amy Zlotof the Physical Activity and Health Branch of the CDC and Rich Killingsworth ofActive Living by Design at the University of North Carolina. Statistician StephenRaudenbush of the University of Michigan, the nation’s foremost authority onhierarchical modeling, provided valuable assistance in the statistical analysis. Thefirst peer-reviewed article based on the study was published in the September/October 2003 issue of the American Journal of Health Promotion. The study was anoutgrowth of a project begun at the Surface Transportation Policy Project andcontinued at Smart Growth America to quantify sprawl and its impact on quality oflife. This report is intended to make an important piece of research more accessibleto the general public. It addition to the study findings, it includes other recentresearch on the degree of sprawl, physical activity, and health. For a detailedmethodology of the original study, please refer to the published article(www.HealthPromotionJournal.com).

Urban Form Data: The county sprawl indexThe study’s urban form data is derived from a landmark study of metropolitan sprawlthat Rutgers and Cornell Universities conducted for Smart Growth America (SGA), anational public interest group working for smart growth policies. Unlike previousstudies, which attempted to evaluate sprawl based on one or two statistics, the SGAmetropolitan sprawl index uses 22 variables to characterize four ‘factors’ of sprawl for83 of the largest metropolitan area in the US for the year 2000. The sprawl ‘scores’ foreach metropolitan area show how much they spread out housing, segregate homesfrom other places, have only weak centers of activity, and have poorly connected streetnetworks. The factor scores, along with an overall sprawl index for the metro areas,represent the most comprehensive, academically rigorous quantification of sprawl in

This is the first rigorous nationwide investigation of the potential

relationship between urban form, physical activity, and health.


FACTOR VARIABLE SOURCE

Residential Density Gross population density in persons per US Censussquare mile

Percentage of population living at densities US Censusless than 1,500 persons per square mile(low suburban density)

Percentage of population living at densities US Censusgreater than 12,500 persons per squaremile (urban density that begins to betransit supportive)

Net population density of urban lands USDA NaturalResourcesInventory

Connectivity of the Average block size in square miles Census TIGERStreet Network files

Percentage of small blocks Census TIGER(< 0.01 square mile) files

the United States. The first report based on this research, Measuring Sprawl and Its

Impact, was released in October 2002 and can be found at Smart Growth America’swebsite, www.smartgrowthamerica.org.

For this study, however, researchers wanted a finer grain of information: while thesprawl index measures sprawl across an entire metropolitan region, residential andhealth data are available at the county level. So they used relevant data from themetropolitan sprawl study to create a county-level index that scores 448 counties.Because fewer data are available at the county level, the index is less comprehensivethan the metropolitan index, but is nevertheless the most complete measurement ofsprawl available at the county level. The county sprawl index uses six variables fromthe US Census and the Department of Agriculture’s Natural Resources Inventory toaccount for residential density and street accessibility (for more information, seethe published paper).

A review of the county sprawl index shows that the most sprawling counties in urbanregions in the US tend to be outlying counties of smaller metropolitan areas in theSoutheast and Midwest. Goochland County in the Richmond, Virginia metro area, andClinton County in the Lansing, Michigan region, received very low numerical scores onthe index, indicating a high degree of sprawl. At the most compact end of the scale arefour New York City boroughs; San Francisco County; Hudson County (Jersey City, NJ);Philadelphia County; and Suffolk County (Boston). Falling near the median are centralcounties of low-density metro areas, such as Mecklenburg County in the Charlotte, NCarea; counties of small metro areas, such as Allen County in the Fort Wayne, IN area;

TABLE 1. County Sprawl Index Variables


and inner suburban counties in large metros such as Washington County in theMinneapolis-St. Paul area.

Counties with a higher degree of sprawl received a lower numerical value on the index.County sprawl scores ranged from the highly compact 352 for Manhattan, to a verysprawling score of 63 for Geauga County outside of Cleveland, Ohio. But Manhattan, andto a lesser degree Geauga, are outliers: most counties are clustered near the middle ofthe index, around the average score of 100. A complete listing of the county sprawlscores is provided in the Appendix.

Health Data: The Behavioral Risk Factor Surveillance SystemThe Behavioral Risk Factor Surveillance System, or BRFSS, is the primary US source ofscientific data on adult behaviors that can endanger health. The survey collects self-reported information about current health risk factors and status, and is the largestcontinuous telephone survey in the world. The BRFSS allows the CDC, which conductsthe survey, to monitor national and state trends in health risk and health outcomes. (Formore information about the BRFSS, see http://www.cdc.gov/brfss/about.htm.) For thisstudy, data from 1998 to 2000 were pooled to create a database of 206,992 respondentsfrom 448 counties.

Researchers looked at the eight BRFSS variables that are believed to be part of thecausal chain between the physical environment and health, including health risk factorssuch as obesity, behaviors such as leisure-time walking, and chronic health problemssuch as hypertension. People responding to such surveys tend to underestimate theirweight, so the overweight and obesity levels reported may be low. Respondents wereconsidered to have a health condition if their doctor or other health professional had

The Importance of Streets that ConnectOne factor used to assess the degree of sprawl in acommunity is the degree to which streets form a gridthat provides many alternate routes. This is especiallyimportant for encouraging bicycling and walkingbecause a lack of direct routes will discourage peoplefrom walking. These two neighborhoods in Atlanta showa one-kilometer “as the crow flies” circle from a home,and then the one-kilometer distance the resident of thathome could travel on the road network. In the sprawlingneighborhood, travel is dramatically constricted by thelack of through streets.

From Health and Community Design by Lawrence D. Frank, Island Press June 2003.


diagnosed it. Researchers also used six additional variables from the database in orderto control for gender, age, race and ethnicity, smoking, diet, and education (as a proxy forincome and access to health information).

Analyzing the DataThe analysis conducted to relate the BRFSS data to the county sprawl index was farmore than a simple correlation. This study linked the thousands of individualrespondents to their home county. This allowed researchers to evaluate each individualin relation to the degree of sprawl where they live. To account for both personal andplace-related influences on behavior and health, researchers used multi-level modeling.The level 1 model looks within each county and relates the characteristics of the peoplesurveyed (such as their age, gender, etc.) to their behavior and health characteristics. Thelevel 2 model takes the level 1 relationships for each county and explains them in termsof the county sprawl index. This kind of modeling is often referred to as hierarchical.Hierarchical or multi-level modeling is used in cases like this where respondents are notindependent of one another (as assumed in ordinary modeling) but instead sharecharacteristics of a given place. A more detailed description of the methodology can befound in the published paper.

This study evaluated thousands of individual respondents in relation

to the degree of sprawl in their home counties.

III. FindingsHow Sprawl Relates to Weight, Physical Activity, and Chronic Disease

The researchers found that people living in sprawling places were likely to weighmore, walk less, and have a greater prevalence of hypertension than people livingin counties with more compact development patterns.

Sprawl Is Linked to WeightThis study used data on body mass index (BMI) to determine if the degree of sprawl hadany influence on weight. BMI is a common measurement of weight to height that reliablypredicts levels of body fat (see box).

The study found that people who live in more sprawling counties were likely to beheavier than people who live in more compact counties. For every 50-point increasein sprawl as measured by the sprawl index, the BMI of residents would be expectedto rise by .17 points. This translates into an increase in weight of just over one poundfor the average person.

Body Mass IndexBody Mass Index measures weight in relation to height. It is a mathematical formula thatdivides a person’s body weight in kilograms by the square of his or her height in meters.BMI is highly correlated with body fat, and can indicate that a person is overweight orobese. People with a Body Mass Index of 25 or higher are considered overweight, whilethose with a BMI of 30 or higher are considered obese. According to the NationalInstitutes of Health, all adults who have a BMI of 25 or higher are considered at risk forpremature death and disability as a consequence of being overweight.

The average BMI of the more than 200,000 people in this study was 26.1. In generalwithin this sample, BMI was higher among men. Both men and women tend to getheavier through middle age, and BMI tends to decline after age 64. African Americansand Hispanics tend to have a higher BMI than whites, while Asians are apt to have a lowerBMI. Also, people who are college educated, or who eat three or more servings of fruitsand vegetables in a day tend to have lower BMIs. All of these factors were controlled forin this study so that the association between weight and the degree of sprawl might bebetter isolated. To learn more about BMI and to calculate your own, visit http://www.cdc.gov/nccdphp/dnpa/bmi/bmi-adult.htm.


Table 2, above, places a person of average BMI at the center of the sprawl index – whichhappens to be McHenry County outside of Chicago – in order to show how expected BMIdiffers for selected counties according to their sprawl ranking. The average BMI for allrespondents in the study is 26.1, and the average height is 5’7”. The study itself wasbased on individuals, not on averages, so these figures are provided to illustrate thedifference in weight expected for persons of the same gender, age, and othercharacteristics living in different places.

The sprawl scale shows us that Hanover County, near Richmond Virginia, is 50 pointsmore sprawling than Delaware County, outside of Philadelphia. An average person

People living in sprawling areas may be missing out on significant

health benefits that are available simply by walking, bicycling,

climbing stairs, and getting physical activity as part of everyday life.

COUNTY EXPECTED WEIGHT OFCOUNTY SPRAWL SCORE EXPECTED BMI AVG PERSON (5’7”)

New York, NY 352.07 25.23 161.1

San Francisco, CA 209.27 25.72 164.2

Suffolk, MA 179.37 25.83 164.9

Cook, IL 150.15 25.93 165.5

Delaware, PA 125.34 26.01 166.1

McHenry, IL 100.08 26.10 166.6

Clay, FL 87.51 26.14 166.9

El Dorado, CA 85.67 26.15 167.0

Hanover, VA 74.97 26.19 167.2

Isanti, MN 70.12 26.20 167.3

Walton, GA 69.61 26.20 167.3

Geauga, OH 63.12 26.23 167.5

Table 2. Sprawl, BMI and Expected Weight



living in Hanover County would beexpected to have a BMI of 26.19; for anotherwise identical person who is 5’7”this translates into a weight of justover 167 pounds. His or hercounterpart in less-sprawlingDelaware County would be expected tohave a BMI of 26.01, and would weighin at 166.1 pounds, or about one poundless. This would be true even aftercontrolling for gender, age, diet, andother factors.

Looking at extremes, the differencein BMI between people living in themost and least sprawling countieswas just under 1 BMI unit. Thatmeans a person living in the mostsprawling county, Geauga Countyoutside Cleveland Ohio, would beexpected to weigh 6.3 pounds morethan a person living in the most compact county, New York County (Manhattan).However, Manhattan is an exceptional example in that it is far more compact thanany other county in the United States. A more typical compact county is SuffolkCounty in central Boston. A person living in Suffolk County would be expected toweigh about 2.6 pounds less than a person living in Geauga County, Ohio. A tablewith the expected BMI for each county is located in the Appendix.

Sprawl and Obesity

The researchers found a similar pattern for adult obesity. Regardless of gender, age,education levels, and smoking and eating habits, the odds of being obese were higherin more sprawling counties. For example, they were ten percent higher for a personliving in Hanover County, with above-average sprawl, than in Delaware County, withbelow-average sprawl. While this study examined the health conditions at theindividual level, the weighted averages for entire counties illustrate the relationship. Inthe 25 most-sprawling counties, 21 percent of the population was obese; in the 25 leastsprawling, 19 percent of the population was obese.

25 Most Compact Counties 25 Most Sprawling Counties0%

5%

10%

15%

20%

25%

19.2%

21.2%

FIGURE 6. Sprawl and ObesityPercent of Adult Population Who Are Obese

Source: BRFSS obesity rates, weighted by county (1998-2000).

The odds that a county resident will be obese rises ten percent

with every 50-point increase in the degree of sprawl.


Evidence from Other StudiesWhile this study used data at the county level to look at relationships across theUnited States, another research project is underway in Atlanta that looks at healthstatus at the neighborhood level.17 While most of the results have not yet beenreleased, the study has found that the proportion of white men who are overweightdeclined from 68 to 50 percent as housing density in neighborhoods increased fromtwo units per acre to eight units per acre, and the proportion of obese men declinedfrom 23 to 13 percent in those more compact neighborhoods.18 Similar relationshipshold for white women and African American men, but the sample size on AfricanAmerican women was too small to determine a relationship.

Sprawl is Linked to Physical ActivityThe most likely way that the design of our communities may influence weight is byencouraging or discouraging physical activity, particularly routine physical activitythat is involved in daily life – what is referred to as ‘active living.’ For most people, thismeans the simple act of walking to the store, to work, or to other places that are a partof their daily routine.

This study tested this idea by analyzing some of the physical activity data from theBehavioral Risk Factor Surveillance System. The survey asks whether people got anyleisure-time physical activity within the past month, and if so, what kind of activity,how often they participated in it, and how long they spent on each occasion. It isimportant to note that these questions focus on intentional exercise during leisure-time, as opposed to routine daily activity. The BRFSS has not measured routinephysical activity such as walking to the store or to a transit stop, climbing stairs in abuilding, or bicycling to work. A recent federal survey found that more than 40

Recommended Physical Activity: Who is getting it?The US Surgeon General now recommends getting 30 minutes of moderate activity atleast five days a week to maintain a basic level of health. Almost two-thirds of Americansdon’t reach this goal. Men are more likely to be physically active than women, and non-Hispanic whites report more activity than people of other races. Younger people andthose with higher education levels also are more likely to be active. But people over 65are more likely to say they get the recommended amount of exercise, mainly becausethey walk so much. Women walk for exercise more than men, with walking increasing upto 75 years of age. Walking is almost equally popular among all races, but people withhigher education levels tend to walk more.

Men are more likely to report having high blood pressure, diabetes, and coronary heartdisease, and older people and those with lower education levels are also more likely tohave these conditions.


percent of walking trips fall into this category.19 Later in the report, we’ll explain whythis study shows that such routine activity deserves much closer examination.

Sprawl and Walking for ExerciseThe study suggests that the degree of sprawl does not influence whether people get anyexercise in their leisure hours. When asked about running, golf, gardening, walking, orany other leisure-time physical activity in the past month, people in sprawling andcompact areas were equally likely to report that they had exercised in some way. Whilemore people in compact areas reported reaching the recommended level of physicalactivity, this result was not statistically significant.

However, the study did show that the degree of sprawl makes a difference in howmuch people engaged in the most common form of exercise – walking. People inmore sprawling places reported that they spent less time walking in their leisuretime than people living in compact locations. For every 50-point increase in thecounty sprawl index, people were likely to walk fourteen minutes less for exercise ina month. This result is not a consequence of different demographics; theresearchers controlled for gender, age, education, ethnicity, and other factors. Thismeans that between the extremes of Manhattan and Geauga County Ohio, NewYorkers walked for exercise 79 minutes more each month. Looking at the weightedaverages for the population as a whole, people in the 25 most sprawling countieswalked an average of 191 minutes per month, compared to 254 minutes per monthamong those who live in the 25 most compact counties.

Sprawl and Walking for TransportationFurther analysis of the relationships between walking, weight, and location points tothe probability that routine physical activity is a significant factor in the lower BMIsof people who live in more compact communities. The researchers found that the lowerlevels of walking for exercise among those living in sprawling counties only accountedfor a small fraction of the higher BMIs in these areas. Both body mass index andobesity levels were higher in the more sprawling counties, independent of how muchpeople walked in their leisure time.

The Question of Self-SelectionPeople walk more in more walkable neighborhoods, but is this just a matter of self-selection? Do people who want to get that type of physical activity choose to live inplaces that provide it? Some studies indicate that walking and biking facilitiesactually encourage people to be more active. In a survey of US adults using a park orwalking and jogging trail, almost 30 percent reported an increase in activity sincethey began using these facilities.2 A recent poll found that, if given a choice, 55percent of Americans would rather walk than drive to destinations. And most peoplesaid inconvenience (61%) and time pressures (47%) kept them from walking more.3


In this study, urban form has astronger relationship to BMI thandoes leisure-time walking. Sprawlmay be affecting other types ofphysical activity, such as walking fortransportation, that are, in turn,influencing weight. This study wasunable to directly measure othertypes of walking that may becontributing to better fitness andlower weight: the walking trips thatpeople take to go to the store, tovisit friends, or to get to work. Manypeople may not consider suchmoderate activity, taken in thecourse of the day, as part of theirexercise regimen. Yet medicalresearch shows such modest exerciseis important, and this may helpexplain why, regardless of how muchthey walked in their leisure time,those living in sprawling countieswere likely to weigh more. Itsuggests people living in sprawlingareas may be missing out onsignificant health benefits that areavailable simply by walking, biking,

climbing stairs, and getting other types of physical activity as part of everyday life.

Other Evidence of the Link Between Sprawl & Walking

A closely related study found that the degree of sprawl influences how much peoplewalk in everyday life. The first study using the metropolitan-level sprawl index foundthat in more compact places, people are far more likely to walk to work.20 Theportion of commuters walking to work is one-third higher in more compactmetropolitan areas than in metro areas with above average sprawl. Public transittrips also typically involve some routine physical activity because most transit tripsinclude a walk to or from a train or bus stop. The metropolitan sprawl study foundthat in the top 10 most sprawling metropolitan regions an average of just twopercent of residents took a bus or train to work, while in the ten most compactregions (excluding the extreme cases, New York and Jersey City), an average of sevenpercent took the bus or train.21

Many transportation studies show that in places with a better pedestrian environment —with sidewalks, interconnected streets, and a mix of businesses and homes, people tendto drive less. For example, a study of two pairs of neighborhoods in the San FranciscoBay Area found that people walked to shopping areas more frequently in olderneighborhoods with nearby stores and a well-connected grid street network.22 Another

10 Most Sprawling Metro Areas 10 Most Compact Metro Areas0%

1%

2%

3%

4%

5%

2%

7%

Perc

enta

ge o

f all

com

mut

e tr

ips

6%

7%

8%

FIGURE 7. Share of Commute Trips by Transit

Source: Ewing R, Chen D. Measuring Sprawl and it’s Impact. Smart GrowthAmerica. October 2002.


The study discussed in this report found that in more compact counties in

the United States, people tend to walk more and weigh less. However, in the

US, most counties are quite spread out, and truly compact counties are very

few. A simple comparison with places that tend to be far more compact –

European cities – shows striking differences in physical activity and obesity

levels, even though the data cannot control for the many other variables that

influence activity and health on the two continents.4

Transportation systems in Europe do far more to provide for and encourage

people to walk or bicycle to get around. The density of housing and jobs in a

sampling of European Union cities are, on average, three times higher than

in a sample of American cities. Consequently, the levels of walking and

bicycling for daily transportation are about five times higher in European

Union countries than in the United States. In Europe, people make 33

percent of their trips by foot or bicycle, while in the United States the

portion is about 9.4 percent. The difference in bicycling is particularly

stark: rates in Europe average about 11 percent, while in the US less than

one percent of trips are made by bicycle. The travel habits of older people,

who would be expected to be especially sensitive to safety and comfort, are

revealing as well. Americans over 75 years old take six percent of their trips

by foot or bicycle, while Dutch and German citizens of the same age make

about half their trips by foot and bicycle.5 The difference is clearly not in

the physical and mental limitations that come with age.

While Europeans engage in more physically active travel, they also have

much lower rates of obesity, diabetes, and hypertension than in the US.

A recent study found that obesity rates in the Netherlands, Denmark, and

Sweden are one-third of the American rate, and Germany’s rate is one-half

the American rate.6

Obviously, many factors besides physical activity influence weight and

health. Research shows that Americans consume eight percent more food

each day than do Europeans. Other factors may include differences in

dietary customs, health care systems, genetic predisposition, and the

ability to afford health care and a nutritious diet. Europeans and

Americans also smoke, drink, and consume caffeine and drugs at different

rates. But the raw figures do suggest that the Europeans may have

something to teach us about controlling weight and improving health

through routine physical activity, particularly by walking and biking to

get where we are going.

THE EUROPEAN EXPERIENCE


study that focused specifically on physical activity found that urban and suburbanresidents who lived in older neighborhoods (measured by whether their homes were builtbefore 1946) were more likely to walk long distances frequently than people living innewer homes.23 A recent poll asking people about their walking habits found that just 21percent of self-described suburban residents walked to a destination in the previousweek, while 45 percent of city residents had taken a walk to get somewhere.24

A recent review that evaluated results from studies of neighborhoods in fourmetropolitan areas estimated that communities designed for walking encourage anextra 15 to 30 minutes of walking per week. For a 150-pound person, that extra exercisecould mean losing – or keeping off – between one to two pounds each year.25 Anotherstudy also confirmed the importance of this type of activity: it found that walking orbicycling to work was associated with lower weight and less weight gain over timeamong middle-aged men, whether or not they engaged in more vigorous exercise.26

Sprawl and Chronic DiseaseExtensive medical research shows that physical inactivity contributes to a variety ofchronic health conditions in addition to obesity. Since this study found that a county’sdevelopment pattern is associated with higher weights and lower levels of physicalactivity, could urban form also be associated with higher levels of disease? This studyused BRFSS data to explore that question, looking at the prevalence of hypertension,diabetes, and coronary heart disease.

People who live in more sprawling counties were more likely to suffer from hypertensionthan people in more compact counties, even after controlling for age, education, gender,and other demographic factors. Hypertension, commonly known as high blood pressure,

increases the risk of heart attack and stroke. Both obesity and physical inactivity are riskfactors for hypertension. The odds of a resident having high blood pressure are about 6percent higher in a county that is less sprawling than average than in a county moresprawling than average (25 units above and below the mean sprawl index, respectively).Comparing the most and least compact places, the odds of having high blood pressurewere 29 percent lower in Manhattan than in Geauga County, Ohio. While this studyexamined the health conditions at the individual level, the weighted averages for entirecounties illustrate the differences found: the 25 most-sprawling counties had averagehypertension rates of 25 per 100 while the 25 least sprawling had hypertension rates of 23per 100. Just as the tendency toward obesity may be exacerbated by a sedentary lifestylein sprawling places, so may the tendency toward high blood pressure. The relationship

The odds that a resident will have high blood pressure increases six

percent for every 50-point increase in the degree of sprawl.


between hypertension and sprawl is not as strong as the association between obesity andsprawl, but the existence of any relationship between urban form and a disease associatedwith physical inactivity is still noteworthy when so many other factors impinge on health.

The researchers found weak associations between diabetes and urban form, and betweencoronary heart disease and urban form, but these associations did not reach the level ofstatistical significance. Factors outside the scope of this study may obscure anyrelationship between admittedly complex diseases and the degree of sprawl.

Why Sprawl May be Linked to Chronic DiseaseThe potential relationship between community design and chronic disease is most likelythrough sprawl’s impact on physical activity, a proven factor in many chronic diseases. A1996 Surgeon General’s report cited hundreds of studies showing the link betweenphysical activity and health.27 Inactivity contributes to being overweight or obese, and isalso connected to a host of health problems. Diseases associated with being overweightand physically inactive reportedly account for over 200,000 premature deaths each year,second only to tobacco-related deaths.28-29

Sprawl and Health at the Metropolitan LevelIn addition to the analysis using the county sprawl index, the researchers relatedhealth data to the metropolitan level sprawl index to see if any relationships held atthe much larger regional level. This analysis found only one statistically significantassociation at the metropolitan level – people walk less for exercise in more sprawlingmetropolitan areas.

The fact that sprawl measured at the county level is significant in many cases, andsprawl measured at the metropolitan level is not, suggests that the built environment“close to home” is most relevant to public health. The association may be even strongerat the neighborhood level. This project has been on the macro scale; other studies ofsprawl and health at a finer scale are showing strong associations between the builtenvironment, physical activity, and health.

IV. The Need for Further Research

The goal of this study was to explore the possibility that the way we’ve built ourcommunities could have a direct impact on health. As a broad national study, itdoes not give a definitive answer in several areas, but points the way toward

research that is needed to show whether these relationships hold true.

Does sprawling development actually cause obesity, disease, or lower rates of

walking? Since a cross-sectional study of this sort cannot control for all the possibledifferences between people living in different places, it is premature to say that sprawlcauses obesity, high blood pressure, or other health conditions. These results show thatsprawl is associated with these conditions, but studies using control groups or that lookat changes in individuals’ weight and health over time are needed to explore causality.

What is the impact of physical activity that falls outside the definition of leisure-

time exercise? As mentioned above, the BRFSS only measures walking as a leisure-timeactivity. Other types of physical activity include walking for transportation; performingphysical labor on the job; or doing work around the house, such as cleaning or gardening.Future studies should look for greater precision in characterizing physical activity. Thenewest version of the BRFSS asks questions about these many types of physical activity.Similarly, researchers need better measures of walking. In looking at minutes walked,this study only included those who listed walking as one of their top two forms ofexercise, missing those who walk for exercise less frequently.

Are there any threshold effects in changing physical activity rates? Does a change

from one level of compactness to another yield major differences in activity levels

or health? It may well be that the relationship between sprawl and physical activity orhealth is not linear: that communities must reach certain thresholds of compactness inorder to make any significant difference in physical activity. For example, moving from aneighborhood with one or two houses per acre to one with three or four houses per acremay not be enough to trigger any changes in behavior.

What does research at the neighborhood level tell us? This study looks at counties andeven metropolitan regions, large areas compared to the living and working environmentsof most people. If the effect of the built environment is strongest on a smaller scale, weneed studies done at that level. The Active Living Research program at San Diego StateUniversity is sponsoring such studies, and the “SMARTRAQ” research project at GeorgiaTech is starting to show results for neighborhoods in Atlanta.

How do other factors in the environment influence physical activity, weight, and

disease? Because they are not directly measured in the sprawl index, this study does notaccount for many things that may influence physical activity, such as the availability of


parks, sidewalks, or multi-use trails, or even climate, topography, and crime. Futureresearch is needed to fill this void.

Is there any relationship between location and what or how much people eat? Thisstudy only partially accounted for food intake — the other half of the weight equation.The only diet-related variable available was the number of fruit and vegetable servingsper day. It may be that people in compact and sprawling places eat differently. Futureresearch may, for example, relate the density of fast food restaurants and availability offood choices to diet and obesity.

Did the sampling design of the BRFSS have any influence over the results? Thecomplex nature of the BRFSS study reinforces the need to be cautious in interpretingthese early findings. The CDC is in the process of developing methods to adjust state-based weights so the BRFSS can be used with more confidence at the local level.

The strong and growing interest in this field is an encouraging sign that research toanswer these questions is on the way. Two notable efforts are coming from the CDC andThe Robert Wood Johnson Foundation (RWJF). The CDC has established an ActiveCommunities Research Group, which is investigating some of these connections. RWJF hasdevoted $70 million to academic research and on-the-ground strategies to encouragephysical activity. For example, Active Living Research has supported a series of carefullytargeted research grants to expand understanding of what makes a community activity-friendly. The RWJF also has established Active Living By Design, a program that will awardgrants to 25 communities to plan and modify the built environment to support andpromote increased physical activity. Information about these programs and more can befound through the Active Living Network, at www.activeliving.org.

The strong and growing interest in this field is an encouraging sign

that more research to answer these questions is on the way.

PHOTOS: www.pedbikeimages.org/Dan Burden

V. Considering health when we plan our communities

This study shows that the way our communities are built – the urban form – maybe significantly associated with some forms of physical activity and with somehealth outcomes. After controlling for demographic and behavioral

characteristics, these results show that residents of sprawling places are likely towalk less, weigh more, and are more likely to have high blood pressure than residentsof compact counties. The way that communities are built appears to have an impacton health. Public health research shows that even a small shift in the health of theoverall population can have important public health implications.30 In addition,changes to the built environment can have an effect that lasts far beyond individualresolve to diet or exercise.

Increasing Physical Activity: Benefits for individuals

and the communityThe potential for improving health through physical activity is enormous. A majorNational Institutes of Health study of more than 3,200 patients at high risk for type-2diabetes found that by losing weight and increasing exercise (primarily throughwalking), participants reduced their risk of getting diabetes by 58 percent. Among olderpeople, the risk was reduced by 71 percent. The study was halted early because thefindings were so dramatic and conclusive that researchers felt they had to be shared.31

Perhaps most fundamentally, physically fit people simply live longer. A landmark studypublished in the New England Journal of Medicine in 2001 found that physical fitness isa better predictor of the risk of death than smoking, hypertension, heart disease, andother risk factors. Physical activity improved survival for people with every diseasestudied. “No matter how we twisted it, exercise came out on top,” said lead authorJonathan Myers of Stanford University.32

Beyond the obvious benefits to individuals, finding ways to help more people be moreactive could have benefits for the entire health care system. A new analysis found thattreatment of conditions tied to being overweight or obese costs an estimated $78 billionannually.33 Health-care costs associated with obesity are estimated to be higher thanthose associated with either smoking or drinking,34 and another study found thathelping people lose weight and become more active could save more than $76 billion inhealth care costs annually.35 These savings are desperately needed: health care costs areaccelerating rapidly,36 with costs related to caring for people who are overweight orobese accounting for an estimated 37 percent of the increase.37 Up to 75 percent ofhealth care costs are associated with chronic diseases, many of which are tied to obesityand physical inactivity.


Meanwhile, governments and private-sector developers are spending billions to build theinfrastructure that shapes our communities – the roads, homes, offices, and buildingswhere people spend their daily lives. For example, the federal government spent about$35 billion in 2001 on transportation, and the transportation funding legislation is oneof the largest spending bills passed by Congress. State and local governments spendanother $124 billion on transportation infrastructure. Using just a small fraction of suchinvestments to create more walkable and bikeable communities is an efficient way toincrease physical activity and improve health.

Seeking SolutionsThe good news is that the potential for getting exercise as part of daily life is alreadyenormous. More than a quarter of all trips in urbanized areas are a mile or less, and fullyhalf of all trips are under three miles, an easy bicycling distance.38 Yet most of thosetrips are now made by automobile.

Converting more trips to biking and walking is possible, as evidenced by the experiencein Europe (see special section, page 19). Recent research identifies six ways that theNetherlands and Germany have achieved their high rates of biking and walking: heavyinvestment in better walking and biking facilities; traffic calming of residentialneighborhoods; urban design sensitive to the needs of non-motorists; restrictions onautomobile use in cities, rigorous traffic education and strict enforcement of strongtraffic laws protecting pedestrians and cyclists.

How can communities in the United States re-shape themselves to promote physicalactivity? The CDC is developing a Guide for Community Preventive Services, which isgathering evidence from case studies and other research to highlight some of the mosteffective. Smart Growth America’s website (www.smartgrowthamerica.org) serves asportal to many groups and activities. A few primary strategies are listed below.

Narrowing streets at intersections, creating raised crosswalks, and installing

traffic circles makes streets safer and more pleasant for pedestrians.



Invest in Bicycle and Pedestrian InfrastructureIn many states, sidewalks and bicycle lanes or wide shoulders are not routinelyincluded when a road is built or improved.39 But many communities are creatingnetworks of sidewalks and bike lanes that help people on foot and bicycle get wherethey are going safely. To learn about creating bike- and pedestrian-friendly streets, seeIncreasing Physical Activity through Community Design by the National Center forBicycling and Walking (www.bikewalk.org), or visit the Pedestrian and BicycleInformation Center at www.walkinginfo.org.

Calm TrafficTraffic engineers are using a variety of new techniques to slow traffic and givepedestrians and cyclists priority on neighborhood streets. Narrowing streets atintersections, creating raised crosswalks, and installing traffic circles makes streetssafer and more pleasant for pedestrians. In Seattle, for example, engineers installedhundreds of traffic circles on neighborhood streets, decreasing traffic crashes byroughly 77 percent. Learn about traffic calming approaches by visiting the Institute ofTraffic Engineers at www.ite.org.

Create Safe Routes to SchoolThe trip to school can be one of the first places to help kids get active, every day.Childhood obesity and inactivity have reached epidemic proportions, andtransportation studies show that young children are spending more time in carsthan ever before. Communities across the country are trying to change that throughSafe Routes to School programs that create a safe walking and biking environmentfor the trip to school, and encourage children and their parents to get in the habit ofwalking. In California, one-third of federal traffic safety funds are devoted tocreating Safe Routes to School. A bill has been introduced in Congress to create anationwide program; for information visit http://www.americabikes.org/saferoutes.asp. The National Highway Traffic Safety Administration has created atoolkit for communities interested in creating Safe Routes to School programs. Formore information, http://www.nhtsa.dot.gov/people/injury/pedbimot/ped/saferouteshtml/overview.html.

Build Transit-Oriented DevelopmentMany communities around the country are concentrating a mix of housing andbusinesses around train or bus stations. This makes it more convenient for people towalk to and from transit, and to pick up a quart of milk or drop off dry cleaning alongthe way. For example, Dallas, Texas is using its new light-rail line as a launching pointfor creating new, walkable neighborhoods. Overall community design is alsoimportant, especially in developing places where walking and bicycling is convenient.See the book Solving Sprawl by Kaid Benfield (Natural Resources Defense Council,2001) for a wealth of examples of these types of projects.

Also, Reconnecting America's Center for Transit-Oriented Development has conductedinnovative research and developed numerous tools to help communities pursue suchdevelopment solutions. See www.reconnectingamerica.org for more details.


Retrofit Sprawling CommunitiesMillions of Americans live in places where it is difficult to walk anywhere. A recent pollfound that 44 percent of those surveyed said it was difficult for them to walk to anydestination from their home.40 Communities can create pedestrian cut-throughs that allowpeople who live on cul-de-sacs to reach shops, parks and offices on foot. Founderingshopping malls, isolated from neighborhoods by expansive parking lots, are being rebornas developers cut new streets through the once-massive buildings, remodeled to holdapartments and businesses as well as shops. The Congress for the New Urbanism’s web sitegives many good examples of these types of projects (www.cnu.org).

Revitalize Walkable NeighborhoodsMany cities and towns have downtowns and main streets with the basic attributes of awalkable and bikeable community, but they lack economic investment. Thesestruggling communities may have dozens, if not hundreds, of vacant buildings; a lackof good retail outlets; and high crime rates. Local governments are concentrating onrevitalizing these neighborhoods through commercial investment, bringing vacantproperty back to productive use, and creating new housing for a mix of income levels.Smart Growth America and several partners have formed a national Vacant PropertiesCampaign to address some of these issues. See www.vacantproperties.org.

Historic preservation has also proven to be an effective strategy for revitalizing Main Streets,traditional downtowns and historic corridors. The National Trust for Historic Preservationoffers many tools to local practitioners through their network and web site at www.nthp.org.

Educate and Encourage

While changing community design is critical, making sure that people understand thebenefits of physical activity and seek it out is also essential. Many programs combineenvironmental changes with outreach to inform and motivate people. For example, manycommunities undertaking Safe Routes to School programs celebrate ‘Walk a Child toSchool Day’ in October. In addition, the CDC has launched a national youth mediacampaign aimed at helping young teenagers make healthy choices that include physicalactivity (http://www.cdc.gov/youthcampaign/index.htm).

PHOTOS: www.pedbikeimages.org/Dan Burden


ConclusionThe way we build our communities appears to affect how much people walk, how much theyweigh, and their likelihood of having high blood pressure. These findings are in line with agrowing body of research which shows that community design influences how peopletravel and how physically active they are in the course of the day. While more research isneeded, urban planners, public health officials, and citizens are already looking to changecommunities to make it easier to get out on a bicycle or on foot. Ultimately, such long-termchanges may help more Americans lead healthier and happier lives.

Endnotes1 Mokdad AH, Serdula MK, Dietz WH, Bowman BA, Marks JS, Koplan JP. The spread of the obesity epidemic in the United

States, 1991-1998. JAMA. 1999;282:1519-1522.2 Flegal K, Carroll M, Ogden C, Johnson C. Prevalence and trends in obesity among US adults, 1999-2000. JAMA.

2002;288:1723-1727.3 National Center for Health Statistics, Centers for Disease Control and Prevention. Prevalence of overweight among

children and adolescents: United States, 1999-2000. Available at: http://www.cdc.gov/nchs/products/pubs/pubd/hestats/overwght99.htm. Accessed on July 15, 2003.

4 Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The Disease Burden Associated with Overweight andObesity. JAMA. 1999;282:1523-1529.

5 Orr A. CDC: Obesity fastest-growth health threat. Reuters. June 5, 2003. Available at: http://www.ucsfhealth.org/childrens/health_library/reuters/2003/06/20030605elin026.html.

6 Pratt M, Macera CA, Blanton C. Levels of physical activity and inactivity in children and adults in the United States:Current evidence and research issues. Med & Science in Sports & Exercise. 1999;31(11 Suppl): S526-S533. (Physical ActivityLevels for U.S. Overall. Available at: http://apps.nccd.cdc.gov/dnpa/piRec.asp?piState=us&PiStateSubmit=Get+Stats.Accessed October 31, 2002).

7 Centers for Disease Control and Prevention. Physical Activity Trends—United States, 1990-1998. Morbidity and

Mortality Weekly Report. March 9, 2001;50(9):166-169.8 Surface Transportation Policy Project. American attitudes toward walking and creating better walking communities.

April 2003. Available at: http://www.transact.org/report.asp?id=205. Accessed July 15, 2003.9 Ewing R, Cervero R. Travel and the built environment. Transportation Research Record 1780. 2001:87-114.10 Sallis JF, Owen N. Physical Activity and Behavioral Medicine. Thousand Oaks, CA: Sage Publications; 1999. Footnotes

13-18 in paper.11 Humpel N, Owen N, Leslie E. Environmental factors associated with adults’ participation in physical activity. Am J Prev

Med. 2002;22:188-199.12 King AC, Jeffery RW, Fridinger F, Dusenbury L, Provence S, Hedlund SA, Spangler K. Environmental and policy

approaches to cardiovascular disease prevention through physical activity: Issues and opportunities. Health Ed

Quarterly. 1995; 22(4): 499-511.13 Schmid TL, Pratt M, Howze E. Policy as intervention: Environmental and policy approaches to the prevention of

cardiovascular disease. Am J Pub Health. 1995; 85(9): 1207-1211.14 Sallis JF, Owen N. Ecological models. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education:

Theory, Research, and Practice. Second edition. San Francisco: Jossey-Bass, 1997:403-424.15 Sallis JF, Bauman A, Pratt M. Environmental and policy interventions to promote physical activity. Am J Prev Med.

1998;15(4): 379-397.16 Killingsworth RE, Lamming J. Development and public health—could our development patterns be affecting our

personal health. Urban Land. July 2001:12-17.17 Strategies for Metropolitan Atlanta’s Regional Transportation and Air Quality (SMARTRAQ). Available at: http://

www.smartraq.net.18 Frank L, Engelke P, Schmid T. Health and Community Design. Washington, DC: Island Press; 2003:187.19 National Highway Traffic Safety Administration, Bureau of Transportation Statistics. National Survey of Pedestrian

and Bicyclist Attitudes and Behaviors. May 2003. Available at: http://www.bicyclinginfo.org/survey2002.htm.20 Ewing R, Pendall R, Chen, D. Measuring sprawl and its impact. Smart Growth America. October 2002. Available at:

http://www.smartgrowthamerica.org.21 Ewing, Pendall and Chen, 19.22 Handy, SL. Understanding the link between urban form and nonwork travel behavior. J. Planning Ed and Res 1996; 15:

183-198.


23 Berrigan D, Troiano RP. The association between urban form and physical activity in U.S. adults. Am J Prev Med.

2002;23(2S):74-79.24 Surface Transportation Policy Project. American attitudes toward walking and creating better walking communities.

Additional data analysis by Belden, Russonello and Stewart.25 Saelens B, Sallis J, Frank L. Environmental correlates of walking and cycling: Findings from the transportation, urban

design, and planning literatures. Annals of Behavioral Medicine; Mar/Apr. 2003.26 Wagner, A, Simon C, Ducimetiere P, et al. Leisure-time physical activity and regular walking or cycling to work are

associated with adiposity and 5 y weight gain in middle-aged men: The PRIME study. International Journal of Obesity

and Related Metabolic Disorders. 2001, 25:940-948.27 U.S. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General.

Atlanta, GA: Centers for Disease Control and Prevention; 1996.28 McGinnis JM, Foege WH. Actual causes of death in the United States. JAMA. 1993;270:2207-2212.29 Allison DB et al. Annual deaths attributable to obesity in the United States. JAMA. 1999;282:1530-1538.30 Rose G. Sick individuals and sick populations. Int J of Epidemio 1985; 14(1):32-38.31 National Institutes of Health. Diet and exercise dramatically delay type 2 diabetes. Available at: http://www.nih.gov/

news/pr/aug2001/niddk-08.htm; and Squires S. Exercise, diet cut diabetes risk by 58%. Washington Post. August 9,2001.

32 Reid B. High risk inactivity. Washington Post. March 26, 2002:F1.33 Finkelstein EA, Fiebelkorn IC, Wang G. National medical spending attributable to overweight and obesity. Health

Affairs web exclusive. May 14, 2003. Available at: http://www.healthaffairs.org/WebExclusives/Finkelstein_Web_Excl_051403.htm. Accessed July 15, 2003.

34 Sturm R. The effects of obesity, smoking, and drinking on medical problems and costs. Health Affairs March/April

2002;21(2):245-253.35 Pratt M, Macera CA, Wang G. Higher direct medical costs associated with physical inactivity. The Physician and

Sportsmedicine. 2000; 28:63-70.36 Matthews T. Health care’s perfect storm. State Government News. February 1, 200337 Finkelstein, Fiebelkorn, Wang, National medical spending.38 Clarke A. National Household Transportation Survey, original analysis. Data available at: http://nhts.ornl.gov/2001/

index.shtml.39 Wilkinson B, Chauncey B. Are we there yet? Assessing the performance of state departments of transportation on

accommodating bicycles and pedestrians. National Center for Bicycling & Walking. February, 2003. Available at: http://www.bikewalk.org/assets/pdf/AWTY031403.pdf. Accessed July 15, 2003.

40 Surface Transportation Policy Project. American Attitudes Toward Walking.

Endnotes from Boxes1 Office of the Assistant Secretary for Planning and Evaluation, US Department of Health and Human Services. Physical

Activity Fundamental to Preventing Disease. June 20, 2002. Available at http://aspe.hhs.gov/health/reports/physicalactivity/.

2 Brownson R et al. Environmental and policy determinants of physical activity in the United States. American Journal

of Public Health. 2001;91(12): 1995-2003.3 Surface Transportation Policy Project. American Attitudes Toward Walking.4 Pucher J. Comparison of public health indicators in Europe, Canada, and the United States as a function of transport

and land use characteristics: An exploratory analysis.” New Brunswick, NY Bloustein School of Planning and PublicPolicy, Nov. 30, 2001. Unpublished working paper commissioned to supplement this study.

5 Pucher J, Dijkstra L. Promoting safe walking and cycling to improve public health: Lessons from the Netherlands andGermany. American Journal of Public Health. July 2003;93(7).

6 Pucher J, Dijkstra L. Promoting Safe Walking and Cycling.


Sta

teC

ount

yM

etro

Are

aS

praw

lEx

pect

edEx

pect

edP

erce

ntP

erce

ntIn

dex

BM

IW

eig

htdi

ffer

ence

indi

ffer

ence

Sco

reod

ds o

fin

odd

shy

pert

ensi

onof

obe

sity

from

Ave

rag

efr

om A

vera

ge

ALA

BA

MA

Bal

dwin

Mob

ile,

AL

83.1

626

.16

167.

01

2.0

2%3.

63%

Jeff

erso

nB

irm

ingh

am, A

L10

8.45

26.0

716

6.46

-1.0

0%

-1.7

8%M

obil

eM

obil

e, A

L98

.85

26.1

016

6.67

0.1

4%0

.24%

She

lby

Bir

min

gham

, AL

87.1

626

.14

166.

921.

54%

2.76

%S

t. C

lair

Bir

min

gham

, AL

83.7

626

.16

167.

00

1.95

%3.

50%

Wal

ker

84.9

826

.15

166.

971.

80%

3.24

%A

RIZ

ON

AM

aric

opa

Ph

oeni

x-M

esa,

AZ

111.

5126

.06

166.

39-1

.36%

-2.4

1%

Pim

aTu

cson

, AZ

101.

7326

.09

166.

60-0

.21%

-0.3

7%A

RK

AN

SA

SC

ritt

ende

nM

emph

is, T

N-A

R-M

S94

.07

26.1

216

6.77

0.7

1%1.

27%

Faul

kner

Litt

le R

ock-

Nor

th L

ittl

e R

ock,

AR

83.4

526

.16

167.

01

1.99

%3.

57%

Lon

oke

Litt

le R

ock-

Nor

th L

ittl

e R

ock,

AR

81.2

226

.16

167.

05

2.26

%4.

06%

Pul

aski

Litt

le R

ock-

Nor

th L

ittl

e R

ock,

AR

108.

04

26.0

716

6.47

-0.9

5%-1

.69%

Sal

ine

Litt

le R

ock-

Nor

th L

ittl

e R

ock,

AR

82.0

026

.16

167.

04

2.17

%3 .

89%

CA

LIFO

RN

IAA

lam

eda

Oak

lan

d, C

A13

6.64

25.9

716

5.84

-4.2

7%-7

.47%

Con

tra

Cos

taO

akla

nd,

CA

115.

7726

.05

166.

3 0-1

.86%

-3.2

9%

El D

orad

oS

acra

men

to, C

A85

.67

26.1

516

6.96

1.72

%3 .

09%

Fres

noFr

esno

, CA

98.0

226

.11

166.

690

.24%

0.4

2%

Ker

nB

aker

s fie

ld, C

A95

.07

26.1

216

6.75

0.5

9%1.

05%

Los

Ang

eles

Los

Ang

eles

-Lon

g B

each

, CA

141.

7425

.96

165.

73-4

.85%

-8.4

7%

Mar

inS

an F

ran

cisc

o, C

A11

1.80

26.0

616

6.3 8

-1.3

9%-2

.47%

Nap

aS

an F

ran

cisc

o-O

akla

nd-

San

Jos

e, C

A10

7.0

126

.08

166.

49-0

.83%

-1.4

8%

Ora

nge

Ora

nge

Cou

nty,

CA

131.

7425

.99

165.

94-3

.71%

-6.5

1%P

lace

rS

acra

men

to, C

A95

.58

26.1

216

6.74

0.5

3%0

.94%

Riv

ers i

deR

iver

side

-San

Ber

nard

ino,

CA

101.

3 426

.10

166.

61-0

.16%

-0.2

8%S

acra

men

toS

acra

men

to, C

A11

6.3 5

26.0

416

6.28

-1.9

3 %-3

.41%

San

Ber

nard

ino

Riv

ersi

de-S

an B

erna

rdin

o, C

A10

0.4

926

.10

166.

63-0

.06%

-0.1

0%

San

Die

go

San

Die

go,

CA

119.

7326

.03

166.

21-2

.32%

-4.1

0%

San

Fra

nci

sco

San

Fra

nci

sco,

CA

209.

2725

.72

164.

24-1

2.19

%-2

0.6

8%S

an J

oaqu

inS

tock

ton

-Lod

i, C

A11

0.9

426

.06

166.

40-1

.29%

-2.2

9%

San

Mat

eoS

an F

ran

cisc

o, C

A13

2.0

925

.99

165.

94-3

.75%

-6.5

8%S

anta

Bar

bara

San

ta B

arba

ra-S

anta

Mar

ia-L

ompo

c, C

A11

5.84

26.0

5-0

.35

-1.8

7%-3

.30

%

San

ta C

lara

San

Jos

e, C

A12

7.28

26.0

1-0

.60

-3.1

9%-5

.62%

San

ta C

ruz

San

Fra

nci

sco-

Oak

lan

d-S

an J

ose,

CA

111.

6326

.06

-0.2

5-1

.37%

-2.4

4%

Sol

ano

San

Fra

nci

sco-

Oak

lan

d-S

an J

ose,

CA

110

.41

26.0

6-0

.23

-1.2

3 %-2

.18%

Son

oma

San

Fra

nci

sco-

Oak

lan

d-S

an J

ose,

CA

101.

8426

.09

-0.0

4-0

.22%

-0.3

9%

Sta

nisl

aus

Mod

esto

, CA

109.

9126

.07

-0.2

2-1

.17%

-2.0

8%V

entu

raLo

s A

ngel

es-R

iver

side

-Ora

nge

Cou

nty,

CA

1 12.

7226

.06

-0.2

8-1

.50

%-2

.66%

Yolo

Sac

ram

ento

-Yol

o, C

A1 0

5.60

26.0

8-0

.12

-0.6

6%-1

.18%

CO

LOR

AD

OA

dam

sD

enve

r, C

O1 2

5.56

26.0

1-0

.56

-3.0

0%

-5.2

7%

Ara

pah

oeD

enve

r, C

O1 1

4.56

26.0

5-0

.32

-1.7

2%-3

.04%

Bou

lder

Den

ver-

Bou

lder

-Gre

eley

, CO

1 08.

1 526

.07

-0.1

8-0

.96%

-1.7

1 %

App

endi

x

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


Sta

teC

ount

yM

etro

Are

aS

praw

lEx

pect

edEx

pect

edP

erce

ntP

erce

ntIn

dex

BM

IW

eig

htdi

ffer

ence

indi

ffer

ence

Sco

reod

ds o

fin

odd

shy

pert

ensi

onof

obe

sity

from

Ave

rag

efr

om A

vera

ge

Dou

gla

sD

enve

r, C

O94

.83

26.1

216

6.76

0.6

2%1.

10%

El P

aso

Col

orad

o S

prin

gs,

CO

105.

2926

.08

166.

53-0

.63%

-1.1

2%

Jeff

erso

nD

enve

r, C

O11

2.59

26.0

616

6.37

-1.4

9%-2

.63%

CO

NN

ECTI

CU

TFa

irfi

eld

New

Yor

k-N

orth

ern

New

Jer

sey-

Long

Isla

nd,

NY-

NJ-

CT-

PA10

7.34

26.0

716

6.48

-0.8

7%-1

.54%

Har

tfor

dH

artf

ord,

CT

102.

7526

.09

166.

58-0

.33%

-0.5

8%M

iddl

esex

Har

tfor

d, C

T90

.28

26.1

316

6.86

1.16

%2.

08%

New

Hav

enN

ew Y

ork-

Nor

ther

n N

ew J

erse

y-Lo

ng Is

lan

d, N

Y-N

J-C

T-PA

107.

1026

.08

166.

49-0

.84%

-1.4

9%To

llan

dH

artf

ord,

CT

81.7

726

.16

167.

04

2.19

%3.

94%

DE

LAW

AR

EN

ew C

astl

eP

hila

delp

hia-

Wilm

ing

ton

-Atl

anti

c C

ity,

PA

-NJ-

DE

-MD

114.

8326

.05

166.

32-1

.75%

-3.0

9%FL

OR

IDA

Bre

vard

Mel

bou

rne-

Titu

svil

le-P

alm

Bay

, FL

104.

7726

.08

166.

54-0

.57%

-1.0

1%

Bro

war

dFo

rt L

aude

rdal

e, F

L12

7.0

126

.01

166.

05

-3.1

6%-5

.57%

Cla

yJa

ckso

nvil

le, F

L87

.51

26.1

416

6.92

1.50

%2.

68%

Dad

eM

iam

i, FL

136.

1725

.98

165.

85-4

.21%

-7.3

8%D

uval

Jack

sonv

ille

, FL

114.

7426

.05

166.

3 2-1

.74%

-3.0

8%

Her

nan

doTa

mpa

-St

Pet

ersb

urg

-Cle

arw

ater

, FL

94.9

726

.12

166.

750

.60

%1.

07%

Hil

lsbo

roug

hTa

mpa

-St

Pet

ersb

urg

-Cle

arw

ater

, FL

114.

3 126

.05

166.

3 3-1

.69%

-2.9

9%

Nas

sau

Jack

sonv

ille

, FL

80.3

126

.1716

7.0

72.

3 7%

4.26

%O

rang

eO

rlan

do, F

L11

5.0

526

.05

166.

3 1-1

.78%

-3.1

4%

Osc

eola

Orl

ando

, FL

105.

5526

.08

166.

52-0

.66%

-1.17

%Pa

lm B

each

Wes

t P

alm

Bea

ch-B

oca

Rat

on, F

L11

0.3

826

.06

166.

41-1

.23 %

-2.1

8%

Pasc

oTa

mpa

-St

Pet

ersb

urg

-Cle

arw

ater

, FL

110

.22

26.0

616

6.42

-1.2

1%-2

.14%

Pin

ella

sTa

mpa

-St

Pet

ersb

urg

-Cle

arw

ater

, FL

126.

6426

.01

166.

06

-3.1

2%-5

.49%

Pol

kLa

kela

nd-

Win

ter

Hav

en, F

L10

5.24

26.0

816

6.53

-0.6

2%-1

.10

%S

emin

ole

Orl

ando

, FL

112.

1326

.06

166.

3 8-1

.43 %

-2.5

4%

St.

Joh

ns

Jack

sonv

ille

, FL

99.3

326

.10

166.

660

.08%

0.1

4%V

olu

sia

Day

tona

Bea

ch, F

L10

4.77

26.0

816

6.54

-0.5

7%-1

.01%

GEO

RG

IAB

arro

wA

tlan

ta, G

A77

.67

26.1

816

7.13

2.69

%4.

85%

Bu

tts

78.7

126

.1716

7.11

2.57

%4.

62%

Cat

oosa

Cha

ttan

oog

a, T

N-G

A85

.61

26.1

516

6.96

1.73

%3 .

10%

Che

roke

eA

tlan

ta, G

A85

.22

26.1

516

6.97

1.77

%3 .

18%

Cla

yton

Atl

anta

, GA

99.6

126

.10

166.

650

.05%

0.0

8%C

obb

Atl

anta

, GA

101.

01

26.1

016

6.62

-0.1

2%-0

.21%

Col

um

bia

Aug

ust

a-A

iken

, GA

-SC

87.3

026

.14

166.

921.

52%

2.73

%C

owet

aA

tlan

ta, G

A80

.87

26.17

167.

06

2.3 0

%4.

14%

Dad

eC

hatt

anoo

ga,

TN

-GA

77.6

126

.18

167.

132.

70%

4.86

%D

e K

alb

Atl

anta

, GA

103 .

9426

.09

166.

56-0

.47%

-0.8

3%

Dou

gla

sA

tlan

ta, G

A80

.29

26.17

167.

07

2.3 7

%4.

27%

Faye

tte

Atl

anta

, GA

75.7

426

.18

1 67.

1 72.

93%

5.28

%

Fors

yth

Atl

anta

, GA

72.0

426

.20

1 67.

263.

38%

6.1 1

%Fu

lton

Atl

anta

, GA

1 05.

4626

.08

1 66.

52-0

.65%

-1.1

5%

Gw

inn

ett

Atl

anta

, GA

93.7

626

.12

1 66.

780

.75%

1 .33

%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.

32 • Measuring the Health Effects of Sprawl: A National AnalysisS

tate

Cou

nty

Met

ro A

rea

Spr

awl

Expe

cted

Expe

cted

Per

cent

Per

cent

Inde

xB

MI

Wei

ght

diff

eren

ce in

diff

eren

ceS

core

odds

of

in o

dds

hype

rten

sion

of o

besi

tyfr

om A

vera

ge

from

Ave

rag

e

Hen

ryA

tlan

ta, G

A74

.13

26.1

916

7.21

3.13

%5.

64%

McD

uff

ieA

ugu

sta-

Aik

en, G

A-S

C78

.00

26.1

816

7.13

2.65

%4.

78%

New

ton

Atl

anta

, GA

79.4

426

.1716

7.0

92.

48%

4.46

%

Paul

ding

Atl

anta

, GA

82.1

026

.16

167.

03

2.15

%3.

87%

Ric

hm

ond

Aug

ust

a-A

iken

, GA

-SC

102.

4726

.09

166.

59-0

.29%

-0.5

2%

Roc

kdal

eA

tlan

ta, G

A82

.82

26.1

616

7.0

22.

07%

3.71

%S

pald

ing

Atl

anta

, GA

85.1

926

.15

166.

971.

78%

3.19

%

Wal

ker

Cha

ttan

oog

a, T

N-G

A81

.26

26.1

616

7.0

52.

25%

4.0

5%W

alto

nA

tlan

ta, G

A69

.61

26.2

016

7.31

3.68

%6.

66%

HA

WA

IIH

onol

ulu

Hon

olul

u, H

I12

6.76

26.0

116

6.0

5-3

.13%

-5.5

2%IL

LIN

OIS

Cli

nton

St

Lou

is, M

O-IL

86.6

926

.15

166.

931.

60%

2.86

%

Coo

kC

hica

go,

IL15

0.1

525

.93

165.

54-5

.79%

-10

.09%

Du

Pag

eC

hica

go,

IL11

4.64

26.0

516

6.32

-1.7

3%-3

.06%

Gru

ndy

Ch

icag

o, IL

87.7

426

.14

166.

911.

47%

2.63

%K

ane

Chi

cag

o, IL

108.

5326

.07

166.

45-1

.01%

-1.7

9%

Ken

dall

Chi

cag

o, IL

90.3

726

.13

166.

851.

15%

2.0

6%La

keC

hica

go,

IL10

8.92

26.0

716

6.45

-1.0

6%-1

.87%

Mad

ison

St

Lou

is, M

O-IL

102.

6226

.09

166.

58-0

.31%

-0.5

5%M

cHen

ryC

hica

go,

IL10

0.0

826

.10

166.

64-0

.01%

-0.0

2%

Mon

roe

St

Lou

is, M

O-IL

85.6

426

.15

166.

961.

72%

3 .0

9%S

t. C

lair

St

Lou

is, M

O-IL

104.

4126

.08

166.

54-0

.52%

-0.9

3%

Wil

lC

hica

go,

IL98

.81

26.1

016

6.67

0.1

4%0

.25%

IND

IAN

AA

llen

Fort

Way

ne,

IN97

.44

26.1

116

6.70

0.3

1%0

.55%

Boo

ne

Indi

anap

olis

, IN

78.1

026

.18

167.

122.

64%

4.75

%C

lark

Lou

isvi

lle,

KY-

IN10

2.63

26.0

916

6.58

-0.3

1%-0

.56%

De

Kal

bFo

rt W

ayn

e, IN

80.3

426

.1716

7.0

72.

3 7%

4.26

%D

earb

orn

Cin

cin

nati

, OH

-KY-

IN84

.42

26.1

516

6.98

1.87

%3 .

3 6%

Floy

dLo

uis

vill

e, K

Y-IN

97.7

126

.11

166.

690

.27%

0.4

9%H

amil

ton

Indi

anap

olis

, IN

93.9

026

.12

166.

780

.73 %

1.3 0

%

Han

cock

Indi

anap

olis

, IN

82.4

426

.16

167.

03

2.11

%3 .

79%

Har

riso

nLo

uis

vill

e, K

Y-IN

74.3

726

.19

167.

203 .

10%

5.58

%

Hen

dric

ksIn

dian

apol

is, I

N85

.56

26.1

516

6.96

1.73

%3 .

11%

Joh

nso

nIn

dian

apol

is, I

N96

.44

26.1

116

6.72

0.4

2%0

.76%

Lake

Chi

cag

o-G

ary-

Ken

osha

, IL

-IN-W

I11

0.9

926

.06

166.

40-1

.30

%-2

.30

%M

ario

nIn

dian

apol

is, I

N11

3 .13

26.0

516

6.3 5

-1.5

5%-2

.74%

Mor

gan

Indi

anap

olis

, IN

88.0

626

.14

166.

901.

43%

2.56

%P

orte

rC

hica

go-

Gar

y-K

enos

ha, I

L-IN

-WI

94.0

926

.12

166.

770

.71%

1.26

%

She

lby

Indi

anap

olis

, IN

88.1

826

.14

166.

901.

42%

2.54

%W

hitl

eyFo

rt W

ayn

e, IN

72.3

826

.20

1 67.

253.

34%

6.0

3%IO

WA

Dal

las

Des

Moi

nes

, IA

81.6

826

.16

1 67.

04

2.20

%3.

96%

Pol

kD

es M

oin

es, I

A1 0

5.34

26.0

81 6

6.52

-0.6

3%-1

.12%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


Sta

teC

ount

yM

etro

Are

aS

praw

lEx

pect

edEx

pect

edP

erce

ntP

erce

ntIn

dex

BM

IW

eig

htdi

ffer

ence

indi

ffer

ence

Sco

reod

ds o

fin

odd

shy

pert

ensi

onof

obe

sity

from

Ave

rag

efr

om A

vera

ge

Pot

taw

atta

mie

Om

aha,

NE

-IA93

.73

26.1

216

6.78

0.7

5%1.

34%

War

ren

Des

Moi

nes

, IA

79.8

626

.1716

7.0

82.

43%

4.36

%K

AN

SA

SB

utl

erW

ichi

ta, K

S79

.81

26.17

167.

09

2.43

%4.

37%

Har

vey

Wic

hita

, KS

73.8

926

.19

167.

223.

16%

5.69

%Jo

hn

son

Kan

sas

Cit

y, M

O-K

S10

3.50

26.0

916

6.57

-0.4

2%-0

.74%

Leav

enw

orth

Kan

sas

Cit

y, M

O-K

S94

.46

26.1

216

6.76

0.6

6%1.

18%

Mia

mi

Kan

sas

Cit

y, M

O-K

S71

.03

26.2

016

7.28

3.51

%6.

33%

Sed

gw

ick

Wic

hita

, KS

106.

1826

.08

166.

51-0

.73%

-1.3

0%

Wya

ndo

tte

Kan

sas

Cit

y, M

O-K

S11

1.25

26.0

616

6.39

-1.3

3%-2

.36%

KE

NT

UC

KY

Boo

ne

Cin

cin

nati

, OH

-KY-

IN92

.11

26.1

316

6.82

0.9

4%1.

69%

Bul

litt

Lou

isvi

lle,

KY-

IN86

.26

26.1

516

6.94

1.65

%2.

96%

Cam

pbel

lC

inci

nna

ti, O

H-K

Y-IN

109.

5726

.07

166.

43-1

.13%

-2.0

1%Je

ffer

son

Lou

isvi

lle,

KY-

IN11

0.0

826

.07

166.

42-1

.19%

-2.1

1%

Ken

ton

Cin

cin

nati

, OH

-KY-

IN10

8.82

26.0

716

6.45

-1.0

4%-1

.85%

Old

ham

Lou

isvi

lle,

KY-

IN84

.27

26.1

516

6.99

1.89

%3 .

3 9%

She

lby

87.6

126

.14

166.

911.

48%

2.66

%LO

UIS

IAN

AA

scen

sion

Bat

on R

oug

e, L

A87

.25

26.1

416

6.92

1.53

%2.

74%

Eas

t B

aton

Rou

ge

Bat

on R

oug

e, L

A10

5.67

26.0

816

6.52

-0.6

7%-1

.20

%Je

ffer

son

New

Orl

ean

s, L

A12

4.23

26.0

216

6.11

-2.8

4%-5

.01%

Livi

ngst

onB

aton

Rou

ge,

LA

82.7

626

.16

167.

02

2.0

7%3 .

72%

Orl

ean

sN

ew O

rlea

ns,

LA

149.

4725

.93

165.

56-5

.72%

-9.9

6%

St.

Ber

nard

New

Orl

ean

s, L

A11

3 .59

26.0

516

6.3 4

-1.6

0%

-2.8

4%S

t. C

harl

esN

ew O

rlea

ns,

LA

91.1

626

.13

166.

841.

06%

1.89

%

St.

Joh

n t

he B

apti

stN

ew O

rlea

ns,

LA

98.1

326

.11

166.

680

.22%

0.4

0%

St.T

amm

any

New

Orl

ean

s, L

A96

.48

26.1

116

6.72

0.4

2%0

.75%

Wes

t B

aton

Rou

ge

Bat

on R

oug

e, L

A91

.68

26.1

316

6.82

1.0

0%

1.78

%M

AR

YLA

ND

An

ne

Aru

nde

lB

alti

mor

e, M

D10

7.75

26.0

716

6.47

-0.9

2%-1

.63 %

Bal

tim

ore

Bal

tim

ore,

MD

107.

02

26.0

816

6.49

-0.8

3%-1

.48%

Bal

tim

ore

city

Bal

tim

ore,

MD

162.

7625

.88

165.

26-7

.20

%-1

2.46

%

Cal

vert

Was

hin

gto

n, D

C-M

D-V

A-W

V90

.84

26.1

316

6.84

1.10

%1.

96%

Car

roll

Bal

tim

ore,

MD

81.9

226

.16

167.

04

2.17

%3 .

91%

Cec

ilP

hila

delp

hia-

Wilm

ing

ton

-Atl

anti

c C

ity,

PA

-NJ-

DE

-MD

86.8

726

.15

166.

931.

57%

2.82

%C

harl

esW

ashi

ng

ton

, DC

-MD

-VA

-WV

89.7

226

.14

166.

871.

23%

2.20

%

Fred

eric

kW

ashi

ng

ton

, DC

-MD

-VA

-WV

87.0

926

.14

166.

931.

55%

2.77

%H

arfo

rdB

alti

mor

e, M

D92

.47

26.1

316

6.81

0.9

0%

1.61

%

How

ard

Bal

tim

ore,

MD

93.6

526

.12

166.

780

.76%

1.3 6

%M

ontg

omer

yW

ashi

ng

ton

, DC

-MD

-VA

-WV

112.

7026

.06

166.

3 6-1

.50

%-2

.66%

Pri

nce

Geo

rge’

sW

ashi

ng

ton

, DC

-MD

-VA

-WV

1 12.

4226

.06

1 66.

37-1

.47%

-2.6

0%

Que

en A

nn

e’s

Bal

tim

ore,

MD

77.2

426

.18

1 67.

1 42.

75%

4.94

%

MA

SS

AC

HU

SET

TSB

rist

olB

osto

n, M

A-N

H1 1

3.62

26.0

51 6

6.34

-1.6

1 %-2

.85%

Esse

xB

osto

n, M

A-N

H1 1

8.56

26.0

41 6

6.23

-2.1

8%-3

.86%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


tate

Cou

nty

Met

ro A

rea

Spr

awl

Expe

cted

Expe

cted

Per

cent

Per

cent

Inde

xB

MI

Wei

ght

diff

eren

ce in

diff

eren

ceS

core

odds

of

in o

dds

hype

rten

sion

of o

besi

tyfr

om A

vera

ge

from

Ave

rag

e

Ham

pden

Spr

ing

fiel

d, M

A N

ECM

A10

8.58

26.0

716

6.45

-1.0

2%-1

.80

%

Ham

pshi

reS

prin

gfi

eld,

MA

NEC

MA

87.7

726

.14

166.

911.

47%

2.63

%M

iddl

esex

Bos

ton

, MA

-NH

121.

5626

.03

166.

17-2

.53%

-4.4

7%

Nor

folk

Bos

ton

, MA

-NH

113.

6426

.05

166.

34-1

.61%

-2.8

5%P

lym

outh

Bos

ton

, MA

-NH

100

.26

26.1

016

6.64

-0.0

3%-0

.06%

Su

ffol

kB

osto

n, M

A-N

H17

9.37

25.8

316

4.90

-9.0

1%-1

5.49

%W

orce

ster

Bos

ton

, MA

-NH

99.4

426

.10

166.

650

.07%

0.1

2%

MIC

HIG

AN

Bay

Sag

inaw

-Bay

Cit

y-M

idla

nd,

MI

95.1

026

.12

166.

750

.59%

1.0

5%C

lint

onLa

nsi

ng-E

ast

Lan

sing

, MI

66.6

326

.21

167.

374.

05%

7.33

%

Eat

onLa

nsi

ng-E

ast

Lan

sing

, MI

77.3

626

.18

167.

142.

73%

4.92

%G

enes

eeD

etro

it-A

nn

Arb

or-F

lint

, MI

99.0

426

.10

166.

660

.11%

0.2

0%

Ing

ham

Lan

sing

-Eas

t La

nsi

ng, M

I10

3.26

26.0

916

6.57

-0.3

9%-0

.69%

Ken

tG

ran

d R

apid

s-M

usk

egon

-Hol

lan

d, M

I98

.44

26.1

116

6.68

0.1

9%0

.33%

Lape

erD

etro

it, M

I71

.56

26.2

016

7.27

3.44

%6.

22%

Livi

ngst

onD

etro

it-A

nn

Arb

or-F

lint

, MI

82.7

226

.16

167.

02

2.0

8%3 .

73%

Mac

omb

Det

roit

, MI

107.

2726

.07

166.

48-0

.86%

-1.5

3 %M

idla

nd

Sag

inaw

-Bay

Cit

y-M

idla

nd,

MI

82.2

526

.16

167.

03

2.14

%3 .

84%

Mon

roe

Det

roit

, MI

83.4

526

.16

167.

01

1.99

%3 .

57%

Oak

lan

dD

etro

it, M

I10

5.71

26.0

816

6.52

-0.6

8%-1

.20

%

Ott

awa

Gra

nd

Rap

ids-

Mu

skeg

on-H

olla

nd,

MI

87.3

626

.14

166.

921.

52%

2.72

%S

agin

awS

agin

aw-B

ay C

ity-

Mid

lan

d, M

I96

.22

26.1

116

6.72

0.4

5%0

.80

%

St.

Cla

irD

etro

it, M

I88

.30

26.1

416

6.90

1.40

%2.

51%

Was

hten

awD

etro

it-A

nn

Arb

or-F

lint

, MI

99.2

726

.10

166.

660

.09%

0.1

6%

Way

ne

Det

roit

, MI

123 .

2226

.02

166.

13-2

.72%

-4.8

0%

MIN

NES

OTA

Ano

kaM

inn

eapo

lis-

St

Pau

l, M

N-W

I95

.92

26.1

116

6.73

0.4

9%0

.87%

Car

ver

Min

nea

poli

s-S

t P

aul,

MN

-WI

85.6

626

.15

166.

961.

72%

3 .0

9%C

hisa

go

Min

nea

poli

s-S

t P

aul,

MN

-WI

79.3

926

.1716

7.0

92.

48%

4.47

%

Dak

ota

Min

nea

poli

s-S

t P

aul,

MN

-WI

98.0

926

.11

166.

680

.23 %

0.4

1%H

enn

epin

Min

nea

poli

s-S

t P

aul,

MN

-WI

119.

7426

.03

166.

21-2

.32%

-4.1

0%

Isan

tiM

inn

eapo

lis-

St

Pau

l, M

N-W

I70

.12

26.2

016

7.3 0

3 .62

%6.

54%

Ram

sey

Min

nea

poli

s-S

t P

aul,

MN

-WI

123 .

09

26.0

216

6.13

-2.7

1%-4

.78%

Sco

ttM

inn

eapo

lis-

St

Pau

l, M

N-W

I90

.36

26.1

316

6.85

1.15

%2.

07%

Was

hing

ton

Min

nea

poli

s-S

t P

aul,

MN

-WI

96.8

026

.11

166.

710

.38%

0.6

8%

Wri

ght

Min

nea

poli

s-S

t P

aul,

MN

-WI

79.8

526

.1716

7.0

82.

43%

4.3 6

%M

ISS

ISS

IPP

ID

e S

oto

Mem

phis

, TN

-AR

-MS

82.0

126

.16

167.

04

2.16

%3 .

89%

Hin

dsJa

ckso

n, M

S97

.30

26.1

116

6.70

0.3

2%0

.58%

Mad

ison

Jack

son

, MS

80.1

626

.1716

7.0

82.

3 9%

4.3 0

%

Ran

kin

Jack

son

, MS

81.6

626

.16

167.

04

2.21

%3 .

96%

MIS

SO

UR

IC

ass

Kan

sas

Cit

y, M

O-K

S83

.70

26.1

61 6

7.0

01 .

96%

3.52

%

Cla

yK

ansa

s C

ity,

MO

-KS

98.3

926

.11

1 66.

680

.19%

0.3

4%Fr

ankl

inS

t Lo

uis

, MO

-IL83

.89

26.1

61 6

7.0

01 .

94%

3.48

%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a


Sta

teC

ount

yM

etro

Are

aS

praw

lEx

pect

edEx

pect

edP

erce

ntP

erce

ntIn

dex

BM

IW

eig

htdi

ffer

ence

indi

ffer

ence

Sco

reod

ds o

fin

odd

shy

pert

ensi

onof

obe

sity

from

Ave

rag

efr

om A

vera

ge

Jack

son

Kan

sas

Cit

y, M

O-K

S11

3.43

26.0

516

6.35

-1.5

8%-2

.81%

Jeff

erso

nS

t Lo

uis

, MO

-IL93

.97

26.1

216

6.77

0.7

2%1.

29%

Lafa

yett

eK

ansa

s C

ity,

MO

-KS

85.0

026

.15

166.

971.

80%

3.23

%P

latt

eK

ansa

s C

ity,

MO

-KS

90.2

626

.13

166.

861.

17%

2.0

9%

Ray

Kan

sas

Cit

y, M

O-K

S74

.60

26.1

916

7.20

3.0

7%5.

53%

St.

Cha

rles

St

Lou

is, M

O-IL

108.

4526

.07

166.

46-1

.00

%-1

.77%

St.

Lou

isS

t Lo

uis

, MO

-IL11

8.14

26.0

416

6.24

-2.1

4%-3

.77%

NE

BR

AS

KA

Dou

gla

sO

mah

a, N

E-IA

118.

1026

.04

166.

24-2

.13%

-3.7

7%

Sar

pyO

mah

a, N

E-IA

101.

7626

.09

166.

60-0

.21%

-0.3

7%W

ashi

ngto

nO

mah

a, N

E-IA

76.5

126

.18

167.

162.

84%

5.11

%

NEV

AD

AC

lark

Las

Veg

as, N

V-A

Z11

4.46

26.0

516

6.32

-1.7

1%-3

.02%

NEW

JE

RS

EYB

erg

enB

erg

en-P

assa

ic, N

J13

0.4

126

.00

165.

97-3

.55%

-6.2

4%

Bu

rlin

gto

nP

hila

delp

hia,

PA

-NJ

101.

5826

.09

166.

61-0

.19%

-0.3

4%C

amde

nP

hila

delp

hia,

PA

-NJ

123 .

7626

.02

166.

12-2

.79%

-4.9

1%

Cu

mbe

r lan

dP

hila

delp

hia-

Wilm

ing

ton

-Atl

anti

c C

ity,

PA

-NJ-

DE

-MD

94.1

626

.12

166.

770

.70

%1.

25%

Esse

xN

ewar

k, N

J15

2.13

25.9

216

5.50

-6.0

2%-1

0.4

6%

Glo

uces

ter

Phi

lade

lphi

a, P

A-N

J10

1.53

26.0

916

6.61

-0.1

8%-0

.32%

Hu

dson

New

Yor

k-N

orth

ern

New

Jer

sey-

Long

Isla

nd,

NY-

NJ-

CT-

PA19

0.0

625

.79

164.

66-1

0.1

6%-1

7.3 8

%

Hu

nter

don

Mid

dles

ex-S

omer

set-

Hu

nter

don

, NJ

81.2

826

.16

167.

05

2.25

%4.

05%

Mer

c er

New

Yor

k-N

orth

ern

New

Jer

sey-

Long

Isla

nd,

NY-

NJ-

CT-

PA11

6.27

26.0

416

6.28

-1.9

2%-3

.39%

Mid

dles

exM

iddl

esex

-Som

erse

t-H

unt

erdo

n, N

J12

1.71

26.0

316

6.17

-2.5

5%-4

.50

%M

onm

outh

Mon

mou

th-O

cean

, NJ

110

.86

26.0

616

6.40

-1.2

8%-2

.28%

Mor

ris

New

ark,

NJ

101.

2026

.10

166.

62-0

.14%

-0.2

5%O

cean

Mon

mou

th-O

cean

, NJ

112.

1426

.06

166.

3 8-1

.43 %

-2.5

4%

Pass

aic

Ber

gen

-Pas

saic

, NJ

140

.41

25.9

616

5.75

-4.7

0%

-8.2

1%S

alem

Phi

lade

lphi

a, P

A-N

J89

.71

26.1

416

6.87

1.23

%2.

20%

Som

erse

tM

iddl

esex

-Som

erse

t-H

unt

erdo

n, N

J97

.01

26.1

116

6.71

0.3

6%0

.64%

Su

ssex

New

ark,

NJ

91.7

726

.13

166.

820

.98%

1.76

%

Uni

onN

ewar

k, N

J13

6.13

25.9

816

5.85

-4.2

1%-7

.37%

War

ren

New

ark,

NJ

96.6

626

.11

166.

720

.40

%0

.71%

NEW

MEX

ICO

Ber

nali

lloA

lbu

quer

que,

NM

112.

1026

.06

166.

3 8-1

.43 %

-2.5

3 %N

EW Y

OR

KA

lban

yA

lban

y-S

chen

ecta

dy-T

roy,

NY

105.

1326

.08

166.

53-0

.61%

-1.0

8%

Bro

nxN

ew Y

ork,

NY

250

.72

25.5

816

3 .3 3

-16.

42%

-27.

3 5%

Erie

Bu

ffal

o-N

iag

ara

Fall

s, N

Y10

6.81

26.0

816

6.49

-0.8

1%-1

.43 %

Gre

ene

76.6

726

.18

167.

152.

82%

5.0

7%K

ing

sN

ew Y

ork,

NY

263 .

6525

.54

163 .

05

-17.

70%

-29.

3 1%

Livi

ngst

onR

oche

ster

, NY

76.5

626

.18

167.

162.

83%

5.10

%M

adis

onS

yrac

use

, NY

75.5

726

.18

1 67.

1 82.

95%

5.31

%

Mon

roe

Roc

hest

er, N

Y1 0

3.62

26.0

91 6

6.56

-0.4

3%-0

.76%

Mon

tgom

ery

Alb

any-

Sch

enec

tady

-Tro

y, N

Y89

.71

26.1

41 6

6.87

1 .23

%2.

21%

Nas

sau

Nas

sau

-Su

ffol

k, N

Y1 3

6.56

25.9

71 6

5.84

-4.2

6%-7

.46%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


tate

Cou

nty

Met

ro A

rea

Spr

awl

Expe

cted

Expe

cted

Per

cent

Per

cent

Inde

xB

MI

Wei

ght

diff

eren

ce in

diff

eren

ceS

core

odds

of

in o

dds

hype

rten

sion

of o

besi

tyfr

om A

vera

ge

from

Ave

rag

e

New

Yor

kN

ew Y

ork,

NY

352.

07

25.2

316

1.11

-25.

92%

-41.

40%

Nia

gar

aB

uff

alo-

Nia

gar

a Fa

lls,

NY

98.5

226

.11

166.

670

.18%

0.3

1%O

non

dag

aS

yrac

use

, NY

101.

7326

.09

166.

60-0

.21%

-0.3

7%

Ont

ario

Roc

hest

er, N

Y79

.85

26.17

167.

08

2.43

%4.

36%

Ora

nge

New

Yor

k-N

orth

ern

New

Jer

sey-

Long

Isla

nd,

NY-

NJ-

CT-

PA98

.10

26.1

116

6.68

0.2

3%0

.40

%

Orl

ean

sR

oche

ster

, NY

79.6

626

.1716

7.0

92.

45%

4.41

%O

sweg

oS

yrac

use

, NY

83.9

326

.16

166.

991.

93%

3.47

%

Pu

tnam

New

Yor

k, N

Y92

.69

26.1

316

6.80

0.8

7%1.

56%

Que

ens

New

Yor

k, N

Y21

8.90

25.6

916

4.0

3-1

3.19

%-2

2.28

%

Ren

ssel

aer

Alb

any-

Sch

enec

tady

-Tro

y, N

Y99

.04

26.1

016

6.66

0.1

1%0

.20

%R

ich

mon

dN

ew Y

ork,

NY

162.

8925

.88

165.

26-7

.21%

-12.

48%

Roc

klan

dN

ew Y

ork,

NY

110

.19

26.0

616

6.42

-1.2

1%-2

.14%

Sar

atog

aA

lban

y-S

chen

ecta

dy-T

roy,

NY

88.9

026

.14

166.

891.

33%

2.38

%

Sch

enec

tady

Alb

any-

Sch

enec

tady

-Tro

y, N

Y10

8.87

26.0

716

6.45

-1.0

5%-1

.86%

Su

ffol

kN

assa

u-S

uff

olk,

NY

109.

8826

.07

166.

42-1

.17%

-2.0

7%

Way

ne

Roc

hest

er, N

Y74

.63

26.1

916

7.20

3 .0

6%5.

52%

Wes

t che

ster

New

Yor

k, N

Y12

8.3 7

26.0

016

6.0

2-3

.32%

-5.8

4%

NO

RT

H C

AR

OLI

NA

Cab

arru

sC

harl

otte

-Gas

toni

a-R

ock

Hil

l, N

C-S

C89

.47

26.1

416

6.87

1.26

%2.

26%

Dav

idso

nG

reen

sbor

o—W

inst

on-S

alem

—H

igh

Poi

nt, N

C85

.42

26.1

516

6.96

1.75

%3 .

14%

Dav

ieG

reen

sbor

o—W

inst

on-S

alem

—H

igh

Poi

nt, N

C70

.99

26.2

016

7.28

3 .51

%6.

3 4%

Du

rham

Ral

eig

h-D

urh

am-C

hape

l Hil

l, N

C99

.12

26.1

016

6.66

0.1

0%

0.1

9%

Fors

yth

Gre

ensb

oro—

Win

ston

-Sal

em—

Hig

h P

oint

, NC

96.5

826

.11

166.

720

.41%

0.7

3 %Fr

ankl

inR

alei

gh

-Du

rham

-Cha

pel H

ill,

NC

76.5

026

.18

167.

162.

84%

5.11

%

Gas

ton

Cha

rlot

te-G

asto

nia-

Roc

k H

ill,

NC

-SC

93.0

626

.12

166.

790

.83%

1.48

%G

uil

ford

Gre

ensb

oro—

Win

ston

-Sal

em—

Hig

h P

oint

, NC

97.2

626

.11

166.

700

.33 %

0.5

8%

Lin

coln

Cha

rlot

te-G

asto

nia-

Roc

k H

ill,

NC

-SC

78.5

626

.1716

7.11

2.58

%4.

65%

Mec

klen

burg

Cha

rlot

te-G

asto

nia-

Roc

k H

ill,

NC

-SC

96.8

226

.11

166.

710

.38%

0.6

8%

Ora

nge

Ral

eig

h-D

urh

am-C

hape

l Hil

l, N

C86

.21

26.1

516

6.94

1.65

%2.

97%

Ran

dolp

hG

reen

sbor

o—W

inst

on-S

alem

—H

igh

Poi

nt, N

C77

.32

26.1

816

7.14

2.74

%4.

93%

Row

anC

harl

otte

-Gas

toni

a-R

ock

Hil

l, N

C-S

C87

.22

26.1

416

6.92

1.53

%2.

75%

Sto

kes

Gre

ensb

oro—

Win

ston

-Sal

em—

Hig

h P

oint

, NC

71.2

626

.20

167.

273 .

48%

6.28

%

Uni

onC

harl

otte

-Gas

toni

a-R

ock

Hil

l, N

C-S

C75

.93

26.1

816

7.17

2.91

%5.

24%

Wak

eR

alei

gh

-Du

rham

-Cha

pel H

ill,

NC

95.8

926

.11

166.

730

.49%

0.8

7%

Yadk

inG

reen

sbor

o—W

inst

on-S

alem

—H

igh

Poi

nt, N

C69

.1726

.21

167.

3 23 .

74%

6.75

%O

HIO

Bu

tler

Cin

cin

nati

-Ham

ilto

n, O

H-K

Y-IN

102.

2926

.09

166.

59-0

.27%

-0.4

8%

Car

roll

Can

ton

-Mas

sillo

n, O

H79

.10

26.17

167.

102.

52%

4.53

%C

lark

Day

ton

-Spr

ing

fiel

d, O

H96

.10

26.1

116

6.73

0.4

6%0

.83%

Cle

rmon

tC

inci

nna

ti, O

H-K

Y-IN

86.9

026

.15

166.

931.

57%

2.82

%C

uyah

oga

Cle

vela

nd-

Lora

in-E

lyri

a, O

H1 1

5.84

26.0

51 6

6.29

-1.8

7%-3

.30

%

Del

awar

eC

olu

mbu

s, O

H81

.99

26.1

61 6

7.0

42.

1 7%

3.89

%Fa

irfi

eld

Col

um

bus,

OH

85.7

726

.15

1 66.

951 .

71%

3.0

6%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


Sta

teC

ount

yM

etro

Are

aS

praw

lEx

pect

edEx

pect

edP

erce

ntP

erce

ntIn

dex

BM

IW

eig

htdi

ffer

ence

indi

ffer

ence

Sco

reod

ds o

fin

odd

shy

pert

ensi

onof

obe

sity

from

Ave

rag

efr

om A

vera

ge

Fran

klin

Col

um

bus,

OH

116.

7226

.04

166.

27-1

.97%

-3.4

8%

Fult

onTo

ledo

, OH

66.8

326

.21

167.

374.

03%

7.29

%G

eaug

aC

leve

lan

d-Lo

rain

-Ely

ria,

OH

63.1

226

.23

167.

454.

49%

8.13

%

Gre

ene

Day

ton

-Spr

ing

fiel

d, O

H91

.03

26.1

316

6.84

1.0

7%1.

92%

Ham

ilto

nC

inci

nna

ti, O

H-K

Y-IN

112.

4526

.06

166.

37-1

.47%

-2.6

1%

Lake

Cle

vela

nd-

Lora

in-E

lyri

a, O

H96

.84

26.1

116

6.71

0.3

8%0

.67%

Lick

ing

Col

um

bus,

OH

84.5

626

.15

166.

981.

85%

3.33

%

Lora

inC

leve

lan

d-Lo

rain

-Ely

ria,

OH

94.5

026

.12

166.

760

.66%

1.17

%Lu

cas

Tole

do, O

H11

1.48

26.0

616

6.39

-1.3

6%-2

.40

%

Mad

ison

Col

um

bus,

OH

83.0

026

.16

167.

02

2.0

4%3.

67%

Mah

onin

gYo

ung

stow

n-W

arre

n, O

H98

.13

26.1

116

6.68

0.2

2%0

.40

%

Med

ina

Cle

vela

nd-

Lora

in-E

lyri

a, O

H76

.59

26.1

816

7.16

2.83

%5.

09%

Mia

mi

Day

ton

-Spr

ing

fiel

d, O

H86

.81

26.1

516

6.93

1.58

%2.

84%

Mon

tgom

ery

Day

ton

-Spr

ing

fiel

d, O

H10

8.47

26.0

716

6.46

-1.0

0%

-1.7

8%P

icka

way

Col

um

bus,

OH

84.7

426

.15

166.

981.

83%

3 .29

%

Por

tag

eC

leve

lan

d-A

kron

, OH

83.9

726

.16

166.

991.

93%

3 .46

%S

tark

Can

ton

-Mas

sillo

n, O

H10

6.62

26.0

816

6.50

-0.7

8%-1

.39%

Su

mm

itC

leve

lan

d-A

kron

, OH

106.

6226

.08

166.

50-0

.79%

-1.3

9%Tr

um

bull

You

ngst

own

-War

ren

, OH

93.5

926

.12

166.

780

.77%

1.3 7

%

Uni

on79

.11

26.17

167.

102.

52%

4.53

%W

arre

nC

inci

nna

ti, O

H-K

Y-IN

89.9

526

.13

166.

861.

20%

2.15

%

Woo

dTo

ledo

, OH

84.2

426

.15

166.

991.

89%

3 .40

%O

KLA

HO

MA

Can

adia

nO

klah

oma

Cit

y, O

K81

.11

26.1

616

7.0

62.

27%

4.0

9%

Cle

vela

nd

Okl

ahom

a C

ity,

OK

95.0

726

.12

166.

750

.59%

1.0

5%C

reek

Tuls

a, O

K91

.30

26.1

316

6.83

1.0

4%1.

86%

Log

anO

klah

oma

Cit

y, O

K80

.83

26.17

167.

06

2.3 1

%4.

15%

McC

lain

Okl

ahom

a C

ity,

OK

79.9

726

.1716

7.0

82.

41%

4.3 4

%

Okl

ahom

aO

klah

oma

Cit

y, O

K10

6.3 1

26.0

816

6.50

-0.7

5%-1

.33 %

Osa

ge

Tuls

a, O

K98

.63

26.1

016

6.67

0.1

6%0

.29%

Pot

taw

atom

ieO

klah

oma

Cit

y, O

K88

.26

26.1

416

6.90

1.41

%2.

52%

Rog

ers

Tuls

a, O

K87

.03

26.1

416

6.93

1.56

%2.

79%

Tuls

aTu

lsa,

OK

108.

6426

.07

166.

45-1

.02%

-1.8

1%W

agon

erTu

lsa,

OK

88.8

926

.14

166.

891.

3 3%

2.3 8

%

OR

EGO

NC

lack

amas

Por

tlan

d-Va

nco

uver

, OR

-WA

98.4

526

.11

166.

680

.18%

0.3

3 %M

ultn

omah

Por

tlan

d-Va

nco

uver

, OR

-WA

131.

4125

.99

165.

95-3

.67%

-6.4

4%

Was

hing

ton

Por

tlan

d-Va

nco

uver

, OR

-WA

108.

2926

.07

166.

46-0

.98%

-1.7

4%Ya

mhi

llP

ortl

and-

Van

couv

er, O

R-W

A98

.23

26.1

116

6.68

0.2

1%0

.38%

PE

NN

SYLV

AN

IAA

dam

s81

.01

26.1 7

1 67.

06

2.29

%4.

1 1%

All

eghe

nyP

itts

burg

h, P

A1 2

0.9

926

.03

1 66.

1 8-2

.47%

-4.3

5%

Bea

ver

Pit

tsbu

rgh

, PA

1 05.

5226

.08

1 66.

52-0

.66%

-1.1

6%B

uck

sP

hila

delp

hia,

PA

-NJ

1 00

.15

26.1

01 6

6.64

-0.0

2%-0

.03%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


tate

Cou

nty

Met

ro A

rea

Spr

awl

Expe

cted

Expe

cted

Per

cent

Per

cent

Inde

xB

MI

Wei

ght

diff

eren

ce in

diff

eren

ceS

core

odds

of

in o

dds

hype

rten

sion

of o

besi

tyfr

om A

vera

ge

from

Ave

rag

e

Car

bon

All

ento

wn

-Bet

hleh

em-E

asto

n, P

A93

.99

26.1

216

6.77

0.7

2%1.

28%

Che

ster

Phi

lade

lphi

a, P

A-N

J89

.84

26.1

316

6.87

1.22

%2.

18%

Col

um

bia

Scr

anto

n—

Wil

kes-

Bar

re—

Haz

leto

n, P

A92

.46

26.1

316

6.81

0.9

0%

1.61

%

Cu

mbe

rlan

dH

arri

sbu

rg-L

eban

on-C

arli

sle,

PA

97.6

426

.11

166.

690

.28%

0.5

0%

Dau

phin

Har

risb

urg

-Leb

anon

-Car

lisl

e, P

A11

3.77

26.0

516

6.34

-1.6

3%-2

.88%

Del

awar

eP

hila

delp

hia,

PA

-NJ

125.

3426

.01

166.

09

-2.9

7%-5

.23%

Faye

tte

Pit

tsbu

rgh

, PA

98.6

626

.10

166.

670

.16%

0.2

8%

Lack

awan

naS

cran

ton

—W

ilke

s-B

arre

—H

azle

ton

, PA

111.

01

26.0

616

6.40

-1.3

0%

-2.3

1%La

nca

ster

Lan

cast

er, P

A94

.09

26.1

216

6.77

0.7

1%1.

26%

Leba

non

Har

risb

urg

-Leb

anon

-Car

lisl

e, P

A10

2.33

26.0

916

6.59

-0.2

8%-0

.49%

Lehi

gh

All

ento

wn

-Bet

hleh

em-E

asto

n, P

A11

9.67

26.0

316

6.21

-2.3

1%-4

.08%

Luze

rne

Scr

anto

n—

Wil

kes-

Bar

re—

Haz

leto

n, P

A10

7.0

926

.08

166.

49-0

.84%

-1.4

9%M

onro

e86

.35

26.1

516

6.94

1.64

%2.

94%

Mon

tgom

ery

Phi

lade

lphi

a, P

A-N

J10

7.0

626

.08

166.

49-0

.84%

-1.4

9%N

orth

ampt

onA

llen

tow

n-B

ethl

ehem

-Eas

ton

, PA

110

.65

26.0

616

6.41

-1.2

6%-2

.23 %

Per

ryH

arri

sbu

rg-L

eban

on-C

arli

sle,

PA

82.9

126

.16

167.

02

2.0

5%3 .

69%

Phi

lade

lphi

aP

hila

delp

hia,

PA

-NJ

187.

7825

.80

164.

71-9

.92%

-16.

98%

Was

hing

ton

Pit

tsbu

rgh

, PA

100

.95

26.1

016

6.62

-0.1

1%-0

.20

%W

estm

orel

and

Pit

tsbu

rgh

, PA

100

.53

26.1

016

6.63

-0.0

6%-0

.11%

Wyo

min

gS

cran

ton

—W

ilke

s-B

arre

—H

azle

ton

, PA

78.6

426

.1716

7.11

2.57

%4.

63%

York

York

, PA

94.7

826

.12

166.

760

.62%

1.11

%

RH

OD

E IS

LAN

DB

rist

olP

rovi

den

ce-F

all R

iver

-War

wic

k, R

I-MA

118.

6626

.04

166.

23-2

.20

%-3

.88%

Ken

tP

rovi

den

ce-F

all R

iver

-War

wic

k, R

I-MA

115.

9926

.04

166.

29-1

.89%

-3.3

3 %

Pro

vide

nce

Pro

vide

nce

-Fal

l Riv

er-W

arw

ick,

RI-M

A13

0.5

625

.99

165.

97-3

.57%

-6.2

7%W

ashi

ngto

nP

rovi

den

ce-F

all R

iver

-War

wic

k, R

I-MA

92.4

526

.13

166.

810

.90

%1.

61%

SO

UT

H C

AR

OLI

NA

Aik

enA

ugu

sta-

Aik

en, G

A-S

C86

.39

26.1

516

6.94

1.63

%2.

93%

Ber

kele

yC

harl

esto

n-N

orth

Cha

rles

ton

, SC

90.1

226

.13

166.

861.

18%

2.12

%

Cha

rles

ton

Cha

rles

ton

-Nor

th C

harl

esto

n, S

C11

0.2

826

.06

166.

42-1

.22%

-2.1

6%D

orch

este

rC

harl

esto

n-N

orth

Cha

rles

ton

, SC

87.8

226

.14

166.

911.

46%

2.62

%

Gre

envi

lle

Gre

envi

lle-

Spa

rtan

burg

-An

ders

on, S

C94

.35

26.1

216

6.77

0.6

8%1.

21%

Lexi

ngto

nC

olu

mbi

a, S

C86

.41

26.1

516

6.94

1.63

%2.

92%

Pic

ken

sG

reen

vill

e-S

part

anbu

rg-A

nde

rson

, SC

83.7

826

.16

167.

00

1.95

%3 .

50%

Ric

hlan

dC

olu

mbi

a, S

C10

1.86

26.0

916

6.60

-0.2

2%-0

.39%

Spa

rtan

burg

Gre

envi

lle-

Spa

rtan

burg

-An

ders

on, S

C86

.73

26.1

516

6.93

1.59

%2.

85%

York

Cha

rlot

te-G

asto

nia-

Roc

k H

ill,

NC

-SC

84.1

126

.15

166.

991.

91%

3 .43

%

TE

NN

ESS

EE

An

ders

onK

noxv

ille

, TN

90.2

026

.13

166.

861.

17%

2.10

%B

lou

ntK

noxv

ille

, TN

89.5

126

.14

166.

871.

26%

2.25

%

Car

ter

Joh

nso

n C

ity-

Kin

gsp

ort-

Bri

stol

, TN

-VA

97.9

326

.11

166.

690

.25%

0.4

4%C

heat

ham

Nas

hvil

le, T

N74

.75

26.1

91 6

7.20

3.0

5%5.

50%

Dav

idso

nN

ashv

ille

, TN

1 01 .

1 726

.10

1 66.

62-0

.14%

-0.2

5%D

icks

onN

ashv

ille

, TN

80.9

226

.1 71 6

7.0

62.

30%

4.1 3

%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


Sta

teC

ount

yM

etro

Are

aS

praw

lEx

pect

edEx

pect

edP

erce

ntP

erce

ntIn

dex

BM

IW

eig

htdi

ffer

ence

indi

ffer

ence

Sco

reod

ds o

fin

odd

shy

pert

ensi

onof

obe

sity

from

Ave

rag

efr

om A

vera

ge

Gra

ing

er80

.10

26.17

167.

08

2.40

%4.

31%

Ham

ilto

nC

hatt

anoo

ga,

TN

-GA

99.8

326

.10

166.

650

.02%

0.0

4%H

awki

ns

Joh

nso

n C

ity-

Kin

gsp

ort-

Bri

stol

, TN

-VA

86.8

126

.15

166.

931.

58%

2.84

%

Jeff

erso

n82

.45

26.1

616

7.0

32.

11%

3.79

%K

nox

Kno

xvil

le, T

N99

.34

26.1

016

6.66

0.0

8%0

.14%

Rob

erts

onN

ashv

ille

, TN

77.3

226

.18

167.

142.

74%

4.93

%R

uth

erfo

rdN

ashv

ille

, TN

85.3

426

.15

166.

961.

76%

3.16

%

Sev

ier

Kno

xvil

le, T

N88

.16

26.1

416

6.90

1.42

%2.

54%

She

lby

Mem

phis

, TN

-AR

-MS

103.

9826

.09

166.

55-0

.47%

-0.8

4%

Sul

livan

Joh

nso

n C

ity-

Kin

gsp

ort-

Bri

stol

, TN

-VA

93.2

826

.12

166.

790

.80

%1.

43%

Su

mn

erN

ashv

ille

, TN

87.0

926

.14

166.

931.

55%

2.77

%

Tipt

onM

emph

is, T

N-A

R-M

S77

.54

26.1

816

7.14

2.71

%4.

88%

Uni

coi

Joh

nso

n C

ity-

Kin

gsp

ort-

Bri

stol

, TN

-VA

104.

1826

.09

166.

55-0

.50

%-0

.88%

Uni

onK

noxv

ille

, TN

84.3

926

.15

166.

981.

87%

3 .3 6

%W

ashi

ngto

nJo

hn

son

Cit

y-K

ing

spor

t-B

rist

ol, T

N-V

A92

.36

26.1

316

6.81

0.9

1%1.

63%

Wil

liam

son

Nas

hvil

le, T

N83

.12

26.1

616

7.0

12.

03 %

3 .64

%W

ilso

nN

ashv

ille

, TN

78.6

726

.1716

7.11

2.57

%4.

63%

TEX

AS

Bex

arS

an A

nton

io, T

X11

2.72

26.0

616

6.3 6

-1.5

0%

-2.6

6%B

razo

ria

Hou

ston

-Gal

vest

on-B

razo

ria,

TX

96.0

226

.11

166.

730

.48%

0.8

5%

Col

lin

Dal

las,

TX

101.

00

26.1

016

6.62

-0.1

2%-0

.21%

Com

alS

an A

nton

io, T

X92

.67

26.1

316

6.80

0.8

8%1.

57%

Dal

las

Dal

las,

TX

114.

5526

.05

166.

3 2-1

.72%

-3.0

4%D

ento

nD

alla

s, T

X98

.68

26.1

016

6.67

0.1

6%0

.28%

El P

aso

El P

aso,

TX

110

.26

26.0

616

6.42

-1.2

1%-2

.15%

Ell

isC

ount

yD

alla

s, T

X88

.64

26.1

416

6.89

1.3 6

%2.

44%

Fort

Ben

dH

oust

on, T

X10

0.6

326

.10

166.

63-0

.08%

-0.1

3 %G

alve

ston

Hou

ston

-Gal

vest

on-B

razo

ria,

TX

109.

9826

.07

166.

42-1

.18%

-2.0

9%

Gua

dalu

peS

an A

nton

io, T

X91

.01

26.1

316

6.84

1.0

8%1.

92%

Har

ris

Hou

ston

, TX

113 .

2526

.05

166.

3 5-1

.56%

-2.7

7%

Hay

sA

ust

in-S

an M

arco

s, T

X88

.93

26.1

416

6.88

1.3 3

%2.

3 7%

Hid

alg

oM

cAll

en-E

dinb

urg

-Mis

sion

, TX

100

.30

26.1

016

6.64

-0.0

4%-0

.06%

Joh

nso

nFo

rt W

orth

-Arl

ing

ton

, TX

89.9

426

.13

166.

861.

20%

2.16

%K

aufm

anD

alla

s, T

X88

.42

26.1

416

6.90

1.3 9

%2.

49%

Libe

rty

Hou

ston

, TX

85.0

026

.15

166.

971.

80%

3 .23

%M

ontg

omer

yH

oust

on, T

X88

.10

26.1

416

6.90

1.43

%2.

56%

Park

erFo

rt W

orth

-Arl

ing

ton

, TX

80.9

426

.1716

7.0

62.

29%

4.12

%R

ockw

all

Dal

las,

TX

90.9

826

.13

166.

841.

08%

1.93

%

Tarr

ant

Fort

Wor

th-A

rlin

gto

n, T

X1 1

0.6

226

.06

1 66.

41-1

.26%

-2.2

3%Tr

avis

Au

stin

-San

Mar

cos,

TX

1 06.

7926

.08

1 66.

49-0

.80

%-1

.43%

Wal

ler

Hou

ston

, TX

94.4

526

.12

1 66.

760

.66%

1 .1 8

%W

i lli

amso

nA

ust

in-S

an M

arco

s, T

X98

.61

26.1

01 6

6.67

0.1 7

%0

.30

%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.


tate

Cou

nty

Met

ro A

rea

Spr

awl

Expe

cted

Expe

cted

Per

cent

Per

cent

Inde

xB

MI

Wei

ght

diff

eren

ce in

diff

eren

ceS

core

odds

of

in o

dds

hype

rten

sion

of o

besi

tyfr

om A

vera

ge

from

Ave

rag

e

UTA

HD

avis

Sal

t La

ke C

ity-

Og

den

, UT

107.

2726

.07

166.

48-0

.86%

-1.5

3%S

alt

Lake

Sal

t La

ke C

ity-

Og

den

, UT

114.

4326

.05

166.

32-1

.70

%-3

.01%

Web

erS

alt

Lake

Cit

y-O

gde

n, U

T10

6.0

726

.08

166.

51-0

.72%

-1.2

8%V

IRG

INIA

Che

sape

ake

city

Nor

folk

-Vir

gin

ia B

each

-New

port

New

s, V

A-N

C10

3.17

26.0

916

6.57

-0.3

8%-0

.67%

Che

ster

fiel

dR

ich

mon

d-P

eter

sbu

rg, V

A93

.89

26.1

216

6.78

0.7

3%1.

30%

Din

wid

die

Ric

hm

ond-

Pet

ersb

urg

, VA

72.4

526

.19

167.

253.

33%

6.0

2%

Fair

fax

Was

hin

gto

n, D

C-M

D-V

A-W

V11

7.81

26.0

416

6.25

-2.1

0%

-3.7

0%

Glo

uces

ter

Nor

folk

-Vir

gin

ia B

each

-New

port

New

s, V

A-N

C82

.82

26.1

616

7.0

22.

07%

3.71

%

Goo

chla

nd

Ric

hm

ond-

Pet

ersb

urg

, VA

67.5

926

.21

167.

353.

93%

7.11

%H

anov

erR

ich

mon

d-P

eter

sbu

rg, V

A74

.97

26.1

916

7.19

3.0

2%5.

45%

Hen

rico

Ric

hm

ond-

Pet

ersb

urg

, VA

100

.73

26.1

016

6.63

-0.0

9%-0

.16%

Jam

es C

ity

Nor

folk

-Vir

gin

ia B

each

-New

port

New

s, V

A-N

C90

.41

26.1

316

6.85

1.15

%2.

05%

Lou

dou

nW

ashi

ng

ton

, DC

-MD

-VA

-WV

94.5

726

.12

166.

760

.65%

1.16

%N

ew K

ent

Ric

hm

ond-

Pet

ersb

urg

, VA

76.4

926

.18

167.

162.

84%

5.11

%

Nor

folk

cit

yN

orfo

lk-V

irg

inia

Bea

ch-N

ewpo

rt N

ews,

VA

-NC

131.

9225

.99

165.

94-3

.73 %

-6.5

4%P

orts

mou

th c

ity

Nor

folk

-Vir

gin

ia B

each

-New

port

New

s, V

A-N

C12

4.93

26.0

116

6.0

9-2

.92%

-5.1

5%

Pow

hata

nR

ich

mon

d-P

eter

sbu

rg, V

A72

.48

26.1

916

7.25

3 .3 3

%6.

01%

Pri

nce

Geo

rge

Ric

hm

ond-

Pet

ersb

urg

, VA

105.

8126

.08

166.

51-0

.69%

-1.2

3 %

Pri

nce

Wil

liam

Was

hin

gto

n, D

C-M

D-V

A-W

V99

.98

26.1

016

6.64

0.0

0%

0.0

0%

Ric

hm

ond

city

Ric

hm

ond-

Pet

ersb

urg

, VA

127.

1826

.01

166.

04

-3.1

8%-5

.60

%

Sco

ttJ o

hn

son

Cit

y-K

ing

spor

t-B

rist

ol, T

N-V

A89

.47

26.1

416

6.87

1.26

%2.

26%

Sta

ffor

dW

ashi

ng

ton

, DC

-MD

-VA

-WV

87.9

026

.14

166.

911.

45%

2.60

%

Su

ffol

k ci

tyN

orfo

lk-V

irg

inia

Bea

ch-N

ewpo

rt N

ews,

VA

-NC

89.1

626

.14

166.

881.

3 0%

2.3 2

%V

irg

inia

Bea

ch c

ity

Nor

folk

-Vir

gin

ia B

each

-New

port

New

s, V

A-N

C11

3 .91

26.0

516

6.3 4

-1.6

4%-2

.91%

Was

hing

ton

Joh

nso

n C

ity-

Kin

gsp

ort-

Bri

stol

, TN

-VA

94.3

826

.12

166.

770

.67%

1.20

%Yo

rkN

orfo

lk-V

irg

inia

Bea

ch-N

ewpo

rt N

ews,

VA

-NC

110

.62

26.0

616

6.41

-1.2

6%-2

.23 %

WA

SH

ING

TON

Cla

rkP

ortl

and-

Van

couv

er, O

R-W

A10

3 .44

26.0

916

6.57

-0.4

1%-0

.73 %

Kin

gS

eatt

le-B

elle

vue-

Ever

ett,

WA

118.

01

26.0

416

6.25

-2.1

2%-3

.75%

Pie

r ce

Sea

ttle

-Tac

oma-

Bre

mer

ton

, WA

107.

5926

.07

166.

48-0

.90

%-1

.60

%S

noh

omis

hS

eatt

le-B

elle

vue-

Ever

ett,

WA

100

.73

26.1

016

6.63

-0.0

9%-0

.15%

WIS

CO

NS

IND

ane

Mad

ison

, WI

102.

4626

.09

166.

59-0

.29%

-0.5

2%K

enos

haC

hica

go-

Gar

y-K

enos

ha, I

L-IN

-WI

106.

1626

.08

166.

51-0

.73 %

-1.3

0%

Milw

auke

eM

ilwau

kee-

Wau

kesh

a, W

I13

2.54

25.9

916

5.93

-3.8

0%

-6.6

7%O

zauk

eeM

ilwau

kee-

Wau

kesh

a, W

I88

.43

26.1

416

6.90

1.3 9

%2.

48%

Rac

ine

Milw

auke

e-R

acin

e, W

I10

3 .10

26.0

916

6.57

-0.3

7%-0

.66%

St.

Cro

ixM

inn

eapo

lis-

St

Pau

l, M

N-W

I76

.22

26.1

816

7.16

2.87

%5.

17%

Was

hing

ton

Milw

auke

e-W

auke

sha,

WI

80.7

526

.1716

7.0

62.

3 2%

4.17

%W

auke

sha

Milw

auke

e-W

auke

sha,

WI

90.4

426

.13

166.

851.

14%

2.0

5%

The

ave

rag

e B

MI f

or t

he p

opul

atio

n s

ampl

ed w

as 2

6.1 .

Thi

s ch

art

show

s th

e di

ffer

ence

in B

MI t

hat

is e

xpec

ted

due

to s

praw

ling

dev

elop

men

t pa

tter

ns,

hol

ding

all

oth

er f

acto

rs e

qual

. The

exp

ecte

d w

eig

ht is

bas

ed o

n a

5'7"

per

son

, the

ave

rag

e h

eig

ht f

or t

he m

en a

nd

wom

en s

ampl

ed in

the

stu

dy.

Smart Growth America is a coalition of nearly 100 advocacy organizations that have a stake in how

metropolitan expansion affects our environment, quality of life and economic sustainability. Our diverse

coalition partners include national, state and local groups working on behalf of the environment, historic

preservation, social equity, land conservation, neighborhood redevelopment, farmland protection, labor,

town planning, and public health. SGA’a website provides introductory and in-depth information on all

aspects of smart growth. Visit www.smartgrowthamerica.org

The goal of The Surface Transportation Policy Project is to ensure that transportation policy and

investments help conserve energy, protect environmental and aesthetic quality, strengthen the economy,

promote social equity, and make communities more livable. We emphasize the needs of people, rather than

vehicles, in assuring access to jobs, services, and recreational opportunities. www.transact.org

To order more copies of Measuring the Health Effects of Sprawl, please go towww.smartgrowthamerica.org for a free download or send $15 to Smart Growth America,1200 18th Street, NW Washington, DC 20036. You may also call us at 202.207.3350 or email usat [email protected].

Self Improvement

Measuring the Health Effects of SPRAWL