Relationship Between Obesity and Built Environment in the Chicago Area

Several tasks were completed:

1. Obtain, review, and clean Illinois Secretary of State driver’s license data.
2. Obtain census land-use, socioeconomic, transportation and demographic data from U.S. Census Bureau at the ZIP code level.
3. Geocode the driver’s license data at the ZIP code level.
4. Conduct statistical analysis and spatial statistical analysis at the ZIP code level to identify the impact of different factors on obesity.
5. Geocode the driver’s license data at the census tract level.
6. Integrate the geocoded driver’s license data with the Spatial Decision Support System database under development in the University of Illinois at Chicago Urban Transportation Center.
7. Conduct various queries and statistical analyses to investigate transportation policy issues.

Principal Investigator(s):

Paul Metaxatos


Paul Metaxatos
Siim Sööt
Lise Dirks
Kouros Mohammadian
George Yanos




Investigate the impact of various land-use, socioeconomic, transportation and demographic factors on obesity rates in the Chicago region and determine to what extent urban sprawl contributes to obesity. Some suggest that highways cause urban sprawl and therefore expansion of the highway network is at the root cause of obesity. While we have shown in our earlier reports that there is only a tenuous connection between highways and sprawl and other factors seem to be much more important, the perception that highways cause sprawl persists.


Driver’s license data were obtained from the Illinois Secretary of State’s office. The approximately seven million records received provide three critical pieces of information, height, weight and address. The height and weight allows a computation of the BMI index. No other information is needed. The address indicates where in the Chicago metropolitan area the person resides.

While we have very detailed geography of the place of residence, we aggregated the BMI data by ZIP code. This allowed us to bring extensive socioeconomic and demographic data into the analysis. Data on income, education, ethnicity, race, commuting modes, age and population density were used. From other sources we also included road density and intersection density as well as distance from downtown Chicago.

Expected Results or Products:

The primary result is that there are many sociodemographic variables that exhibit a stronger association with obesity than sprawl. Disadvantaged neighborhoods have the highest BMI levels. In particular, neighborhoods with low education levels, low income and high proportions of African-Americans and Latinos are most likely to have high BMI levels. Conversely, the North Shore communities of Kenilworth, Glencoe and Winnetka have the lowest BMIs. In the city of Chicago, the Lincoln Park and Gold Coast areas have the lowest BMIs.

Generally the communities that are ten to twenty miles from the Chicago downtown have the lowest levels and the city of Chicago neighborhoods have higher levels than suburbs in the 20 to 50 mile radius from Chicago. Moreover, walking to work is negatively associated with BMI, i.e., as walking increases BMI decreases. Oddly, the same is true for driving to work, though the regression coefficient is lower. Lastly, population density is also negatively associated with BMI though the strength of the relationship is not as pronounced as it is for the socioeconomic variables.

NOTE: A report or paper from this research is not immediately available.


Paul Metaxatos
Urban Transportation Center
University of Illinois at Chicago
412 South Peoria Street, Suite 340
Chicago, IL 60607
Voice: (312) 996-4713
Fax: (312) 413-0006


Illinois Department of Transportation
Metropolitan Transportation Support Initiative (METSI)
Idot logo 2