Abstract WP250: The Changing Impact of Neighborhood Socioeconomic Status on Stroke Incidence in the BASIC Project

Abstract

Background: Poverty is associated with increased stroke risk, but it is unclear how this has changed over time. We sought to evaluate temporal trends in stroke incidence by neighborhood socioeconomic status (nSES) in a south Texas community. Methods: Within the BASIC project, we identified all first-ever ischemic and hemorrhagic strokes among Mexican American or non-Hispanic White persons, age >45, from 2010-2021, geocoded to census tract by patient home address (n=4,137). Using 5 year rolling average annual data from the American Community Survey, a composite nSES score was assigned to each census tract, reflecting aggregate neighborhood income, wealth, education, and occupation (range -1.60, 1.95; lower=poorer). For each year, the nSES percentile of each tract was determined. Incidence rates were estimated using Poisson regression, with annual population estimates as offsets, including a quadratic term for time to account for nonlinear trends. The model included annual nSES percentile, age, sex, ethnicity, and nSES-time and ethnicity-time interactions. Results: Incidence rates for high vs low nSES neighborhoods for the 45-65-year age group are shown in the Figure. Incidence rate ratios for nSES changed over time (p<0.01 for interaction) in a u-shaped fashion, with no association present in 2010, a significant association in 2015 (RR for 90th vs 10th percentile nSES=0.59, 95% CI=0.49-0.70), and no association in 2021 (Figure). Conclusions: The gap in stroke incidence between the richest and poorest neighborhoods peaked in 2015 and was no longer present in 2021, due to declining rates in low nSES neighborhoods. Future work should explore the causes underlying these trends.