Predictive Value of Traditional and Novel Obesity Indices for Stroke and Its Subtypes Across Sexes and Glucose Status: Toward Precision Prevention Strategies

Background This study compared the predictive capacity of traditional and novel obesity indices for stroke subtype risks across sexes and glucose metabolic statuses, addressing limitations of current anthropometric tools. Methods A retrospective cohort of 398 270 UK Biobank participants who were stroke free was analyzed. Eight novel indices—weight‐adjusted waist index, a body shape index, body roundness index, visceral adiposity index, conicity index, lipid accumulation product, waist‐to‐hip ratio, and waist‐to‐height ratio—were compared with traditional measures (body mass index, waist circumference). Cox proportional hazards and Fine–Gray competing risk models generated hazard ratios (HRs) with 95% CIs, stratified by sex and glucose status. Predictive discrimination was assessed via Harrell’s C‐statistic. Results Novel indices outperformed traditional metrics for total and ischemic stroke. Weight‐adjusted waist index showed the strongest associations: men in the highest quartile had a 25% increased ischemic stroke risk (HR, 1.25 [95% CI, 1.11–1.36]); women had a 36% increased risk (HR, 1.36 [95% CI, 1.20–1.57]). For ischemic stroke prediction, weight‐adjusted waist index significantly surpassed body mass index in both sexes (C‐statistics: men, 0.712 [95% CI, 0.705–0.720] versus body mass index, 0.688 [95% CI, 0.680–0.696], P