Abstract 85: Food Swamps Are Associated With Incident Stroke In The Health And Retirement Study

TitleAbstract 85: Food Swamps Are Associated With Incident Stroke In The Health And Retirement Study
Publication TypeConference Paper
Year of Publication2023
AuthorsYang, D, Naqvi, IA, Gutierrez, J, Tom, S
Keywordsfood, Stroke

Introduction: The role of “food swamps”, an area characterized by a high-density of establishments selling fast-food and junk food relative to healthier options, on incident stroke is not well studied.

Hypothesis We hypothesized that a higher retail food environment index (RFEI), indicative of food swamps, would be associated with greater odds of incident stroke.

Methods: The sample comprised of community-dwelling stroke-free participants aged ≥50 years who enrolled in the 2010 epoch of the Health and Retirement Study (HRS), which is representative of the US population. If a participant moved to a new area during follow-up through 2016, we only considered incident strokes reported before relocation. The traditional RFEI is a county’s ratio of the number of fast-food restaurants and convenience stores to the number of grocers. The expanded RFEI additionally includes full-service restaurants as unhealthy food options, and farmers’ markets and specialized food stores as healthy food retailers. We averaged RFEI across all included follow-up years and dichotomized RFEI using a threshold of 5, previously shown to discriminate obesity rates. We used logistic regression models to assess the association between RFEI groups and incident stroke, adjusting for key covariates and weighting to account for survey design.

Results: Among 84,023,542 participants (mean age 64±10 years, 54% female, 84% white) in weighted analysis, 3,224,378 (3.8%) reported an incident stroke during follow-up. The average traditional and expanded RFEI were 6.5±2.7 (72% in ≥5 group) and 6.9±2.3 (84% in ≥5 group), respectively. In fully adjusted weighted analyses, the higher traditional RFEI group had greater odds of incident stroke compared to the lower group (OR [95% CI]: 1.135 [1.132-1.138]). We found a similar association with expanded RFEI groups and incident stroke (OR [95% CI]: 1.095 [1.092-1.098]).

Conclusions: Among community-dwelling adults in HRS, RFEI was associated with incident stroke, independent of demographics and health characteristics. Results highlight the potential importance of an area’s food options as a structural determinant in stroke, especially given most participants resided in areas with 6 times the amount of relative unhealthy to healthy food choices.

Citation Key13131