Proximity of food retailers to schools and rates of overweight ninth grade students: an ecological study in California

The dependent variable for this study was the rate of overweight ninth grade students for public schools in California in 2007. The data were obtained from California Department of Education, which administers a physical fitness test (FITNESSGRAM^®) between February and May of each year. The test is required by state law for public school students in 5^th, 7^th and ninth grade, and in 2007 approximately 90% of schools participated[28]. The test reports students' body composition, as measured by skinfold (preferred method), body mass index (weight in kilograms divided by height in meters squared), or bioelectric impedance analyzers. These three different measurement options are provided to ensure broad participation, and while all are subject to error (typically 3 to 6%), comparisons of different methods report high test-retest reliability [29, 30]. The definition of overweight differs slightly from others, such as the Centers for Disease Control. Instead, classification in this category is determined by criterion-referenced gender-, age- and test-specific cut-offs (e.g. skinfold <32 for a 15 year old female; body mass index <24.5 for a 14 year old male). These standards were established by a national advisory panel, convened by the Cooper Institute of Dallas, Texas[29].

Our analysis is at the school-level because the physical fitness test results are publicly available only at this level of aggregation--individual results are confidential. The original dataset included all public schools in California that reported physical fitness test results. To obtain more stable rates of overweight students, we excluded schools with less than 100 students from the analysis. We focused on ninth graders because they are expected to have greater mobility, and greater financial ability to purchase food from nearby retailers, when compared to 7^th and 5^th graders.

The physical fitness dataset also provided information on the gender and ethnic composition of students tested in each school. We used this information to construct a variable for gender composition, with percentage of male students as the reference category. We also constructed variables for student composition by major ethnic groups, with percentage non-Hispanic white as the reference category, and additional variables for the percentage of students that were Hispanic/Latino, African-American, Asian, Native American (including Alaskan native), Pacific Islander, and those who declined to state their ethnicity.

The California Department of Education was the source for several additional datasets, all of which are also publicly available. One contained the street addresses of schools in the study, which we used in constructing the retailer proximity variables described below. We obtained two other datasets to construct independent variables for 1) percentage of all students (not just ninth graders) in each school receiving free & reduced price meals in 2007, and 2) urban or non-urban school location. The subsidized meal variable was used as a proxy for school-level socioeconomic status, which at the individual-level is known to be strongly associated with overweight. We defined urban schools as those located in large or mid-size cities (based on U.S. Census Bureau classifications) and compared them to schools that were not in these areas, defined as non-urban.

Another publicly available dataset was purchased from Environmental Systems Research Institute, Inc. (ESRI), It was used to construct variables for the presence/absence of three classes of retailers near schools: 1) fast food restaurants, 2) convenience stores, and 3) supermarkets. ESRI provided geocoded information for data that was first collected by the marketing firm InfoUSA, in 2007. We then selected retailers based initially on the following 8-digit National American Industry Classification System (NAICS) codes: Limited Service Restaurants (72221105), Convenience Stores (44512001), Supermarkets and Grocery Stores (44511001). As reported in another study however [31], the NAICS codes were inconsistently applied. For instance, there were many recognizable convenience stores identified with the NAICS code for Grocer Retail (44511003). To reduce misclassification, retail locations in this category that contained the terms 'Quick-,' 'Mini-,' and 'Liquor-' in the company name were recoded into the "convenience stores" category (n = 668). Supermarkets were defined as retailers in the Supermarket and Grocery Stores category with $2 million or more in annual sales, based on the Food Marketing Institute's definition [32]. Because Limited Service Restaurants (72221105) was also an overly broad category, those with five or more locations were selected as "fast food restaurants," or business chains that provide low price meals without table service. This category included all of the major fast food chains (e.g. McDonald's, Burger King, Taco Bell, Domino's). For the entire state of California we identified 3,646 supermarkets, 4,069 convenience stores, and 20,668 fast food restaurants.

The point locations of the schools were geocoded with the Streetmap USA (2006) dataset provided by ESRI, based on street addresses. We validated the accuracy of geocoded locations by spot-checking 10% of the schools. We found the addresses of these schools through web searches and re-geocoded them using Mapquest, an online geocoding service owned by America Online, Inc. [33]. Our spot-checking indicated more than 99% accuracy in our geocoded school locations.

Most previous studies of food proximity have not incorporated the actual pedestrian walking network into their analyses of proximity or food access. Instead they rely on Euclidean (straight-line, circular radius) buffers that do not account for the street network, sidewalks, and other elements of the areas surrounding schools where walking is most feasible [24, 31, 34, 35]. The result is that the area covered by Euclidean buffers can be substantially more than the area covered by equivalent distance network buffers. This can lead to erroneous or misleading results. To improve accuracy in this study we utilized network buffers along the actual street network for the entire state of California.

We used ESRI's Network Analyst extension to ArcView 9.1 to create 800 m network buffers around the final geocoded school points. This distance was selected because it is approximately a ten-minute walk, and is commonly used in other studies of retail food access near schools [24, 31, 35, 36]. These network buffers included both the street lines making up the 800 m street networks around each school, as well as polygons that contained the area encompassing these networked buffers. Once the network buffers were created, we performed a spatial join to calculate the number of fast food, convenience and supermarket retailers located within each of these network buffers. The final result of our GIS analysis was a matrix denoting the presence or absence of each of three classes of retailers within 800 m along the street network for each school. These data were exported out of the GIS for statistical analysis.

We analyzed all data with SPSS (version 15.0.1). Our analysis included descriptive statistics, correlations (Kendall's tau-b), and linear regression. Our first model regressed the dependent variable of school rate of overweight ninth grade students on the three types of nearby retailers assessed in the GIS analysis. Our second model included the additional school-level variables (ethnic, socioeconomic and gender composition, and urban/non-urban location) for comparison.

Prior to regression, we applied logarithm transformations to school ethnic composition variables, with the exceptions of the two largest groups (Hispanic/Latino and White, non-Hispanic), in order to meet assumptions of normality. We also employed multiple imputation as a strategy for dealing with missing data for two independent variables: percentage of students receiving free & reduced price meals, and urban/non-urban school location. This involved utilizing information from the existing data to generate plausible values for missing data, while also representing uncertainty by generating five different data sets with imputed values. These values were imputed with the software Amelia II (version 1.2), and combined for analysis using the procedures detailed by King et al.[37]

Table 1 shows the descriptive characteristics of the study. The sample consisted of 879 public schools in California with scores for 100 or more ninth grade students on the California Physical Fitness Test in 2007 (659 schools reporting scores for fewer ninth grade students were excluded from the analysis; 479 of these excluded schools reported overweight scores, with a mean rate of 33.8%). The number of students per school ranged from 100 to 1486. The percentage of students that were overweight in each school ranged from 3.0 to 75.4%, with a mean of 30.7%. Although this is a substantial range, it is not surprising given the heterogeneity in ethnic and income and composition of these schools (e.g. up to 91.4% white, non-Hispanic, or 100% Hispanic/Latino). Half of the study schools were located within an 800 m network buffer of a fast food restaurant. The other two types of retailers examined were less prevalent, with 28.6% of schools located near convenience stores, and 23.8% near supermarkets.

Rates of overweight ninth grade students were positively associated with the presence of nearby convenience stores and fast food restaurants, as shown in Table 2. Convenience stores demonstrated stronger correlations with school overweight rates (.138) than fast food restaurants (.091). Supermarkets demonstrated no association with school overweight rates, however.

Schools with higher percentages of Latino students were more likely to be near convenience stores (.137) and fast food restaurants (.101), but were not more likely to be near supermarkets (.020). Schools with more low-income students, as measured by the percentage receiving free & reduced price meals, demonstrated very similar correlations (.131 with convenience stores, .108 with fast food, .003 with supermarkets). Schools located in non-urban areas were less likely to be near fast food restaurants (-.137) and had lower rates of overweight students (-.083) when compared to urban schools.

The first model regresses the percentage of overweight ninth grade students on the three retail variables, as shown in Table 3. It indicates that controlling for the influence of these retailers simultaneously, the presence of a convenience store within an 800 m network buffer of a school is predicted to increase the percentage of overweight students by 3.5% (95% confidence interval 1.9, 5.2). For a typical-sized school (474 ninth grade students), for example, approximately 17 additional students in this grade would be expected to fall in the overweight category in comparison to school without a nearby convenience store. The presence of a fast food restaurant within this same distance is predicted to increase the percentage of overweight students by 1.8% (95% confidence interval 0.2, 3.4), approximately half the strength of the convenience store association. The presence of a supermarket within this buffer is predicted to decrease the number of overweight students by 1.9% (95% confidence interval -3.6, -0.1). The model itself explains 3% of the variation in rates of overweight students.

The second regression model is shown in Table 4. This model includes additional school-level variables that are known to be associated with overweight status. The variation explained was 51%, a substantial increase when compared to the first model. After controlling for these additional variables, the associations between retailers and school overweight rates were weaker, and only the association with convenience stores remained. The presence of a convenience store near a school is predicted to increase its overweight rate by 1.2%. This is approximately 1/3 of that reported in model 1, but the 95% confidence interval (0.03, 2.4) continues to indicate a positive relationship. For a school with the mean number of ninth graders (474), the model predicts that an additional 6 students would be overweight in comparison to a school without a nearby convenience store. The associations with fast food restaurants and supermarkets when controlling for additional variables are much smaller, and their 95% confidence intervals include zero.

Variables that had stronger positive associations with school overweight rates than nearby convenience stores included the percentage of students receiving free or reduced price meals, and percentage of Latino, African-American and Native American students. The percentage of Asian-American students demonstrated a stronger negative association with obesity rates. Variables in the model that were more weakly associated with school overweight rates than nearby convenience stores included gender composition, percent Pacific Islander or "declined to state" ethnicity, and urban/non-urban location.

Our study has a number of limitations. Its ecological design limits inferences about school food environments' potential relationships with individual-level behaviors/outcomes. Even at the school-level we lacked data for variables known to influence food/beverage consumption and physical activity, such as the type of food service [39], presence or absence of vending machines [40], open/closed campus [31], mode of transportation (e.g. buses), and physical education requirements [41]. In addition, the cross-sectional design did not allow for analysis of the time order of exposures and the outcome. Another potential limitation was the exclusion of schools with small numbers of students due to unstable rates. Our sources of data are prone to error, such as those we utilized to classify and geographically locate the retailers, as well as the methods and standards used to determine student overweight status. This error may bias the estimates toward or away from the null hypothesis.

Our study also has several strengths. We utilized network buffers to more accurately model retailer distance to schools when compared to studies that have employed Euclidean buffers. We simultaneously examined associations between three different retail types and the outcome of overweight, rather than focusing on a single type of retailer. Finally, our outcome measure of student overweight, although aggregated at the school-level, was based on criterion-referenced, test-specific measurements, instead of less accurate student self-reported height and weight.

Future studies could address some of the limitations of our analysis by ground-truthing retail location data, as well as incorporating additional school-level variables that may influence student energy intake and expenditure. Subsequent efforts could also benefit from characterizing the quantity and types of foods available within restaurants and stores, rather than relying on retailer categories, as otherwise similar establishments may differ greatly in their availability of healthy and less healthy foods [42, 43]. Additional methodological improvements could include longitudinal designs and/or cross-classified multilevel models [44]. The latter could incorporate the retail food environments of both schools and residences at the macro-level, and examine how these interact with individual student health outcomes and behaviors, as in a recent study conducted by Laska et al. with adolescents in Minneapolis/St. Paul [26].