Abstract WE569: Multimorbidity Burden Differs by Stroke Subtype and Race/Ethnicity: Insights from the All of Us Research Program

Abstract

Background: Clustering of vascular risk factors shapes stroke outcomes, but its variation by race/ethnicity and stroke subtype at presentation in diverse populations is poorly understood. We hypothesized that clustering would differ across these groups. Methods: We conducted a cross-sectional analysis of 4,954 adult stroke patients with electronic health record (EHR) and survey data from the All of Us Research Program (2016-2023). Stroke subtype (ischemic vs. subarachnoid hemorrhage [SAH]) was identified using standardized EHR diagnosis code groupings. Vascular risk factors included smoking, obesity (BMI ≥30 kg/m 2 ), hypertension (SBP ≥140 or DBP ≥90 mmHg), diabetes (HbA1c ≥6.5%), dyslipidemia (LDL ≥100 mg/dL), and statin use. We summarized the distribution of risk factor counts, including the proportion with ≥3, by stroke subtype and race/ethnicity. We then used Poisson regression to estimate relative risks (RRs) of higher clustering, with models including age, sex, race/ethnicity, and stroke subtype as predictors. Results: The cohort (mean age 62.9 ±12.1; 53.5% female; 49% White, 26% Black, 15% Hispanic, 9% Other, 1% Asian) included 95.9% ischemic strokes and 4.1% SAH. Clustering was frequent: 33% of ischemic and 25% of SAH patients had ≥3 risk factors. By subtype, ischemic patients more often had 3, 4, or 5 risk factors (24%, 9%, 1%) than SAH patients (19%, 5%, 1%). By race/ethnicity, ≥3 risk factors occurred in 39% of Black, 34% of Hispanic, 33% of Other, 27% of White, and 22% of Asian patients. In models adjusted for age, sex, and subtype, clustering was higher among Black (RR 1.21, 95% CI 1.15-1.27), Hispanic (RR 1.09, 95% CI 1.03-1.16), and Other patients (RR 1.16, 95% CI 1.08-1.24) versus White, and lower among Asian patients (RR 0.75, 95% CI 0.60-0.91). Female patients had modestly higher clustering than males (RR 1.06, 95% CI 1.01-1.10). SAH patients had fewer clustered risk factors than ischemic patients (RR 0.88, 95% CI 0.79-0.97). Conclusion: In this diverse cohort of stroke survivors, vascular risk factor clustering was common and varied by race/ethnicity, sex, and subtype. Absolute prevalence was highest among Black and Hispanic patients and lowest among Asian patients. SAH survivors exhibited fewer comorbidities than ischemic stroke survivors, reflecting both etiologic differences and survivor bias. These findings underscore the high comorbidity burden among stroke survivors and the importance of addressing disparities in post-stroke care.