Body roundness index trajectories and incident type 2 diabetes: a longitudinal analysis from the China Health and Retirement Longitudinal Study

Body Roundness Index (BRI) has emerged as a superior anthropometric measure for assessing visceral adiposity compared with traditional body mass index. However, the longitudinal relationship between BRI trajectory patterns and type 2 diabetes development remains underexplored, particularly in Chinese populations. We analyzed 4,965 participants aged ≥ 45 years from the China Health and Retirement Longitudinal Study. BRI trajectories were characterized using group-based trajectory modeling across 2011–2016. Incident diabetes was identified through self-reported physician diagnosis during 2017–2020 follow-up. Cox proportional hazards regression examined associations between BRI trajectory patterns and diabetes risk. Receiver operating characteristic (ROC) curve analysis, continuous Net Reclassification Improvement (NRI), and Integrated Discrimination Index (IDI) were used to evaluate the incremental predictive value of BRI trajectories. Three distinct BRI trajectory patterns emerged: low-stable (43.8%), moderate-stable (37.8%), and high-stable (18.4%) groups. During follow-up (mean 4.92 years), 387 incident diabetes cases occurred (7.79%). Compared with the low-stable group, moderate-stable and high-stable trajectories demonstrated progressively elevated diabetes risk with fully adjusted hazard ratios of 1.71 (95% CI 1.29–2.26) and 2.61 (95% CI 1.78–3.84), respectively (P for trend < 0.001). BRI trajectory incorporation significantly enhanced predictive capacity beyond conventional risk factors (area under the curve [AUC] improvement from 0.613 to 0.635, P = 0.011; continuous NRI = 27.62%, P < 0.001). Distinct longitudinal BRI trajectory patterns independently predict diabetes incidence with clear dose-response relationships, suggesting long-term BRI monitoring may enhance diabetes risk stratification and prevention strategies.