Neighborhood Characteristics of Fast Food Restaurant Locations Jennifer R. Bonds and Dominic Farris Harvard School of Public Health June 2005 Brisa N.

Neighborhood Characteristics of Fast Food Neighborhood Characteristics of Fast Food Restaurant LocationsRestaurant Locations

Jennifer R. Bonds and Dominic FarrisJennifer R. Bonds and Dominic FarrisHarvard School of Public HealthHarvard School of Public Health

June 2005June 2005Brisa N. Sanchez M.Sc. And Steven Gortmaker, PhDBrisa N. Sanchez M.Sc. And Steven Gortmaker, PhD

The Obesogenic Environment

Obesogenic environments encourage consumption of food and/or discourage physical activity (Chisolm et al., 1998, Price and Gottesman, 1991; Stunkard, 1991; Weinsier et al., 1998)

Studies have shown associations between fast food intake and increased BMI and weight gain (French, Harnack, and Jeffery, 2000;French and Jeffery, 1998)

Cost is the most significant predictor of dietary choices (Sooman, Macintyre, Anderson, 1993; Foley, Pollard, 1998; Mackerras, 1997)

Fast food consumption is related to obesity and this relationship is strongest among low-income individuals (French, Harnack, Jeffrey, 2000; Jeffrey, French, 1998)

Fast Food Restaurant Placement

Easy access to restaurants is a concern for customers and businesses strive to meet this demandBusiness planners consider neighborhood demographics when determining the placement of new restaurantsBusiness planners may also consider a wide variety of factors including property taxes, zoning, and business permitsNeighborhood racial segregation in Chicago

Research Inquiries

There are more fast food restaurants per person in low income census tracts (excluding census tracts in the lowest 10 percentile) than middle/high income census tractsEthnic specific restaurants targeted at Blacks and Hispanics in majority Black and Hispanic census tracts than census tracts of predominately other racial/ethnic groups

Operational Definitions

Fast food restaurant – eating places where customers order items and pay before eating and where food can be eaten on the premises or taken outCensus tract – the unit of measure for the US census; it is small and its boundaries are drawn along visible features such as roads and are always nested within counties Poverty status – income thresholds determined by the census bureau based on family size and total family income in the last 12 months; this value changes each year based on inflation

Operational Definitions

Housing value – an estimate of how much a house and lot would sell for if it were for sale (excluding properties that were renter occupied)Shopping area – one-half square mile boundaries around each census tractCommunity area –defined by sociologists at the University of Chicago during the 1920s, and at that time corresponded to neighborhoods; there are 77 and they are used for political purposes by the city of Chicago

Data Characteristics

Wealth and Ethnicity in Chicago Community Areas

Population Density

Low Low-medium Medium High-medium(n=162) (n=172) (n=161) (n=162)

Characteristics Mean SD Mean SD Mean SD Mean SDPopulation 2345 1742 3852 2682 3997 2835 4151 2735Area (sq mi) 0.1833 0.2815 0.1968 0.1392 0.2425 0.1876 0.3305 0.3688Population density (persons/sq mi) 17124 10372 22498 12630 19266 10775 17048 11901% black residents 0.8543 0.2674 0.5343 0.4212 0.3506 0.4167 0.2668 0.387% hispanic residents 0.0958 0.2193 0.304 0.3579 0.3685 0.3212 0.2538 0.2284% white residents 0.0655 0.1232 0.226 0.217 0.3624 0.2707 0.5259 0.3074% renters 0.6477 0.1217 0.5893 0.1186 0.5216 0.1577 0.4381 0.1939% residents below poverty line* 0.4508 0.1448 0.2764 0.0692 0.1839 0.0726 0.1145 0.0505

Neighborhood Wealth- Median Income

Wealth as Median Income**

Wealth as Median Housing Value

Characteristics Mean SD Mean SD Mean SD Mean SD Mean SDPopulation 2986.71 2215.65 3634.64 2514.87 4107.33 2868.50 3987.83 2926.81 2930.57 2138.66Area (sq mi) 0.26 0.32 0.32 0.53 0.28 0.23 0.29 0.31 0.16 0.16Population density (persons/sq mi) 15110.05 9062.37 16253.23 9824.09 18366.52 11953.12 17768.75 11990.68 22664.93 13164.92Percentage of residents who are black 0.79 0.34 0.62 0.43 0.22 0.34 0.26 0.37 0.17 0.27Percentage of residents who are hispanic 0.16 0.29 0.27 0.35 0.39 0.31 0.20 0.21 0.15 0.14percentage of residents who are white 0.10 0.17 0.20 0.25 0.47 0.27 0.54 0.32 0.68 0.25percentage of residents who are renters 0.53 0.14 0.47 0.19 0.45 0.21 0.50 0.21 0.59 0.16percentage of residents below the poverty line*0.34 0.13 0.23 0.12 0.18 0.13 0.17 0.16 0.14 0.13

* sample size slightly smaller (n=158, 161,151, 154, 159 )

Median Housing ValueLow

(n=163)Low-medium

(n=171)Medium(n=154)

High-medium(n=161)

High(n=162)

Community Areas by Major Ethnic Group

Majority Hispanic Majority Black Majority White# census tracts (n=159) (n=326) (n=342)

Mean SD Mean SD Mean SDCharacteristics%commercial 34.35% 23.30% 24.76% 20.22% 29.65% 23.76%income 34,358 7,556 27,332 11,707 50,485 16,440pop density 3,976 2,681 2,821 2,331 3,755 2,661%renter 54.42% 14.28% 53.83% 18.46% 45.96% 20.42%%owner 37.51% 16.35% 32.52% 21.90% 48.07% 21.95%college exposure 27.38% 11.33% 39.43% 15.82% 58.13% 22.71%no high school 50.01% 11.92% 33.80% 13.28% 21.61% 14.48%

Census Tract Characteristics by Ethnicity

**Data Analysis

Census tract data based on 2000 Census and 2004 American Housing Survey Restaurants were categorized by ethnic group: Black, Hispanic, and otherUsed two determinants of wealth: median income and median housing valueDid not consider census tracts with median income and median housing value in the lowest 10 percentileShopping areas where developedUsed Poisson regression

Results: Restaurants and Wealth

Income: Adjusted for commercialization and the percentage of renters

The number of restaurants increases as income and commercialization increasesThe percentage of renters decreases as income and commercialization increasesConclusion: Our hypothesis was not supported

Median Income

Results: Restaurants and Wealth

Housing Value: Adjusted for commercialization and the percentage of renters

The number of restaurants increases as housing value and commercialization increasesThe percentage of renters decreases as housing value and commercialization increasesConclusion: our hypothesis was not supported

HousingValue

Results: Black American - targeted

RestaurantsAdjusted for median income, median housing value, commercialization, population density, and total number of restaurantsThere are more Black American - targeted restaurants in majority Black census tracts compared to census tracts that are majority Hispanic and other ethnic groupsCommercialization, housing value, and the total number of restaurants in the census tracts have a positive association to the number of Black American – targeted restaurantsConclusion: Our hypothesis was supported

Black American targeted

Restaurants and their Location

Results: Hispanic - targeted Restaurants

Adjusted for median income, median housing value, commercialization, population density, and total number of restaurantsThere are less Hispanic – targeted restaurants in majority Hispanic census tracts compared to census tracts that are majority Black and other ethnic groupsCommercialization, total number of restaurants, housing value and income have negative association to the number of Hispanic – targeted restaurants in census tractsConclusion: Our hypothesis was not supported

Hispanic-targeted

Restaurants and their Location

**Limitations

Limited information on housing value, income, and poverty status for some census tractsOverlap in census tracts???Incomplete list of Hispanic and Asian targeted restaurantsOmitted information from census tracts with very small populations (less than 200 people)

**Implications

Can be generalized to other large, diverse metropolitan areas Public health interventions in majority Black neighborhoods can reflect knowledge of fast food restaurant placementHispanic targeted fast food restaurant chains are not popular among HispanicsPopulation within Chicago city limits does not differ greatly based on wealth as compared to census tracts in the greater Chicago area*

**Future Research

Census tracts within the city limits should be compared to census tracts in the greater metropolitan area outside the city limitsAsian targeted restaurants should be studiedOther factors of fast food restaurant placement should be studiedWhy are more/less in certain areas*Is the appearance of restaurants in lower income areas in response to local demand or does their appearance drive demand?*Access*

Acknowledgements

We would like to thank the following people:

Bryn AustinSteve MellyDr. Steven GortmakerBrisa Sanchez, M. Sc.Dr. Louise RyanIsabelle Angelouski

Question & Answer