Abstract
Background: For older groups, being overweight [body mass index (BMI; in kg/m2): 25 to <30] is reportedly associated with a lower or similar risk of mortality than being normal weight (BMI: 18.5 to <25). However, this “risk paradox” is partly explained by smoking and disease-associated weight loss. This paradox may also arise from BMI failing to measure fat redistribution to a centralized position in later life.Objective: This study aimed to estimate associations between combined measurements of BMI and waist-to-hip ratio (WHR) with mortality and incident coronary artery disease (CAD).Design: This study followed 130,473 UK Biobank participants aged 60-69 y (baseline 2006-2010) for ≤8.3 y (n = 2974 deaths). Current smokers and individuals with recent or disease-associated (e.g., from dementia, heart failure, or cancer) weight loss were excluded, yielding a “healthier agers” group. Survival models were adjusted for age, sex, alcohol intake, smoking history, and educational attainment. Population and sex-specific lower and higher WHR tertiles were <0.91 and ≥0.96 for men and <0.79 and ≥0.85 for women, respectively.Results: Ignoring WHR, the risk of mortality for overweight subjects was similar to that for normal-weight subjects (HR: 1.09; 95% CI: 0.99, 1.19; P = 0.066). However, among normal-weight subjects, mortality increased for those with a higher WHR (HR: 1.33; 95% CI: 1.08, 1.65) compared with a lower WHR. Being overweight with a higher WHR was associated with substantial excess mortality (HR: 1.41; 95% CI: 1.25, 1.61) and greatly increased CAD incidence (sub-HR: 1.64; 95% CI: 1.39, 1.93) compared with being normal weight with a lower WHR. There was no interaction between physical activity and BMI plus WHR groups with respect to mortality.Conclusions: For healthier agers (i.e., nonsmokers without disease-associated weight loss), having central adiposity and a BMI corresponding to normal weight or overweight is associated with substantial excess mortality. The claimed BMI-defined overweight risk paradox may result in part from failing to account for central adiposity, rather than reflecting a protective physiologic effect of higher body-fat content in later life.