Comparative predictive values of anthropometric indices for cardiometabolic multimorbidity in middle-aged and older adults: a prospective study from the CHARLS study

Abstract Objectives This study aimed to compare the predictive performance of seven anthropometric indices—body mass index (BMI), waist-to-height ratio (WHtR), body roundness index (BRI), weight-adjusted waist index (WWI), a body shape index (ABSI), conicity index (CI) and waist circumference (WC)—for cardiometabolic multimorbidity (CMM) in middle-aged and older Chinese adults. Methods This study conducted a prospective study using data from the China Health and Retirement Longitudinal Study (CHARLS) 2011–2018. Propensity score matching (PSM) was utilized to control for biases induced by age and gender, with these two factors as the core matching variables and sample matching conducted at a 1:1 ratio. Multivariable logistic regression models were used to examine associations between anthropometric indices and CMM. Restricted cubic splines explored dose-response relationships between anthropometric indices and CMM. Receiver operating characteristic (ROC) curves evaluated discriminative performance of anthropometric indices in predicting CMM and specific types of CMM. Results Before PSM, a total of 7,469 participants were included, 554 participants (7.42%) developed CMM. In Model II, BMI, WHtR, BRI, CI and WC maintained significant associations across higher quartiles. Compared with the BMI Q 1 group, the risk of CMM in Q 2 group increased by 1.55 times (OR = 2.55, 95%CI = 1.65, 3.93, P < 0.001); the risk in Q 3 group increased by 2.04 times (OR = 3.04, 95%CI = 1.93, 4.81, P < 0.001); and the risk in Q 4 group increased by 4.89 times (OR = 5.89, 95%CI = 3.62, 9.57, P < 0.001). Compared with the WHR Q 1 group, the risk of CMM in Q 2 group increased by 1.26 times (OR = 2.26, 95%CI = 1.49, 3.42, P < 0.001); the risk in Q 3 group increased by 1.54 times (OR = 2.54, 95%CI = 1.65, 3.91, P < 0.001); and the risk in Q 4 group increased by 3.54 times (OR = 4.54, 95%CI = 2.87, 7.16, P < 0.001). Similar results were found in BRI. Compared with the CI Q 1 group, the risk in Q 3 group increased by 0.59 times (OR = 1.59, 95%CI = 1.04, 2.44, P = 0.032); and the risk in Q 4 group increased by 0.73 times (OR = 1.73, 95%CI = 1.11, 2.70, P = 0.015). Compared with the WC Q 1 group, the risk of CMM in Q 2 group increased by 0.82 times (OR = 1.82, 95%CI = 1.19, 2.80, P = 0.006); the risk in Q 3 group increased by 1.36 times (OR = 2.36, 95%CI = 1.51, 3.68, P < 0.001); and the risk in Q 4 group increased by 4.63 times (OR = 5.63, 95%CI = 3.46, 9.15, P < 0.001). WHtR, BRI, WWI, CI and WC all showed a U-shaped association with CMM risk. BMI demonstrated a linear relationship with CMM risk. BMI achieved the highest performance with identical AUC values of 0.720 (0.690–0.749), followed by WC with an AUC of 0.712 (0.682–0.742). BMI and WC exhibited superior predictive performance whether in predicting those specific types of CMM. Conclusion BMI and WC were superior to novel anthropometric indices for CMM risk prediction in middle-aged and older Chinese adults. The finding supports their value in identifying high-risk CMM individuals and reinforces their role as practical tools.