Abstract
BACKGROUND: Stroke remains a major cause of disability and death. While central adiposity may contribute to vascular risk, the role of the body roundness index (BRI) (a waist-height-derived anthropometric indicator) in community screening populations is not well defined. We therefore investigated the association between BRI and prevalent stroke in a community-based screening sample. METHODS: Using data from a ChinaHEART cohort branch (6,858 adults), BRI was calculated from anthropometric measurements. Prevalent stroke was ascertained by self-reported physician diagnosis. Receiver operating characteristic (ROC) analysis was used to determine the optimal BRI cutoff. Logistic regression, adjusted for age, sex, marital status, smoking, alcohol use, hypertension, diabetes, blood pressure, fasting glucose, and lipid parameters, was performed to examine the association between BRI and prevalent stroke. Restricted cubic spline (RCS) analysis and subgroup/interaction analyses were further conducted. RESULTS: Among 6,858 participants, 192 (2.8%) reported prior stroke. The bootstrap-derived BRI cutoff was 4.597 (95% CI 4.149-4.798). High BRI (≥4.6) was associated with higher odds of prevalent stroke in the fully adjusted model (OR 1.766, 95% CI 1.279-2.438). BRI alone showed limited discrimination (AUC 0.584), while multivariable models achieved higher AUC (0.740 in the fully adjusted model). Adding BRI produced ΔAUC increases. RCS showed no evidence of nonlinearity. CONCLUSIONS: Our findings support BRI as a simple, non-invasive anthropometric measure that may offer complementary discriminatory information for community screening/triage; prospective studies with validated outcomes are warranted to confirm clinical utility.