Abstract DP237: Association between the food environment and ischemic stroke risk: A population-based study

Background: A healthy diet is critical for stroke prevention. Dietary patterns can be influenced by the food environment. We evaluated the association between the modified retail food environment index (mRFEI), a measure that reflects the ratio of healthy vs unhealthy food sources in a neighborhood, and stroke incidence in a large population-based study. Methods: The Brain Attack Surveillance in Corpus Christi (BASIC) Project is an ongoing study that identifies all strokes among residents of Nueces County, Texas. All first-ever ischemic strokes from 2000-2023 were geocoded to census tracts by patient home address at the time of stroke. The primary exposure was the mRFEI (range 0-1; higher=healthier), obtained at the census-tract level from the National Neighborhood Data Archive for each year (2000-2023). Population counts and covariate data were obtained from the US Census (2000) and American Community Survey (2009-2023), with linear interpolation from 2001-2008. Stroke incidence rates for each census tract were estimated using Poisson regression, both unadjusted and after adjustment for age, sex, ethnicity, a neighborhood socioeconomic status (nSES) composite score (reflecting aggregate income, wealth, education, and occupation), and a mRFEI*nSES interaction term. Results: There were 6,586 first-ever ischemic strokes in the study region from 2000-2023. Figure 1 displays the mRFEI for each neighborhood. Notably, higher mRFEI (indicating a healthier mix of food sources) was associated with lower nSES (see Table 1). Poisson regression model parameters are shown in Table 2. In the unadjusted model, there was a marginal association between higher mRFEI and greater stroke incidence (β=1.24; 95%CI 0.98, 1.57, p=0.08). After adjustment for covariates, this association was attenuated and not statistically significant (β=1.01, 95%CI 0.85, 1.21, p=0.89). Higher nSES was associated with lower stroke incidence (β=0.95, 95%CI 0.93, 0.97, p<0.01). There was no interaction between nSES and mRFEI (p=0.47). Conclusions: In this large population-based study in an urban community, we found no association between the food environment (mRFEI) and stroke incidence. There was an unexpected inverse association between mRFEI and nSES, whereby more affluent neighborhoods, which had lower stroke rates, tended to have a less healthy mix of food sources. Further work is needed to disentangle the links between the food environment, neighborhood socioeconomic factors, and stroke risk.