Comparative evaluation of relative fat mass and body mass index in predicting cardiometabolic multimorbidity in older adults: results from the English Longitudinal Study of Ageing.

Relative fat mass (RFM) is a more accurate measure of body fat percentage than body mass index (BMI). However, its association with cardiometabolic multimorbidity (CMM) and its predictive value have not been examined. This study evaluated and compared the associations and predictive utility of RFM and BMI for CMM. We analyzed data from 3,348 adults (mean age 64 years; 45.1% male) in the English Longitudinal Study of Ageing who were free of hypertension, coronary heart disease, diabetes, and stroke at wave 4 (2008-2009). RFM was derived from height and waist circumference. CMM was defined at wave 10 (2021-2023) as the presence of two or more of hypertension, cardiovascular disease, diabetes, or stroke. Odds ratios (ORs) with 95% confidence intervals (CIs) and measures of discrimination were estimated. Over 12-15 years of follow-up, 197 participants developed CMM. Restricted cubic spline models showed linear dose-response relationships for both RFM and BMI (p for nonlinearity > .05). Higher RFM was strongly associated with CMM (per 1-SD increase: OR 1.66, 95% CI 1.29-2.15; top vs bottom tertile: OR 2.70, 95% CI 1.46-4.99). Associations for BMI were weaker (per 1-SD increase: OR 1.28, 95% CI 1.12-1.47; top vs bottom tertile: OR 1.88, 95% CI 1.24-2.85). Adding RFM to a conventional risk model modestly increased discrimination (ΔC-index = 0.0088, p = .29) and significantly improved model fit (-2 log likelihood, p < .001). Corresponding values for BMI were ΔC-index = 0.0049 (p = .46) and -2 log likelihood (p < .001). The C-index gain from RFM was 0.0039 greater than BMI (p = .39). In an older UK population, RFM was a stronger indicator and predictor of CMM risk than BMI.