Abstract
Background
The body mass index (BMI) is widely used to assess obesity but fails to reflect fat distribution. The body roundness index (BRI) has been proposed as a better surrogate for central obesity, yet direct comparisons of their predictive power remain limited.
Methods
We analyzed 341 336 adults from the Taiwan MJ cohort, free of cardiovascular disease at baseline, with a mean follow‐up of 19.1 years. Anthropometric, laboratory, and lifestyle data were analyzed to compare the predictive performance of BRI and BMI for all‐cause and cardiovascular mortality. Mediation analysis using CRP (C‐reactive protein) was conducted to evaluate the role of inflammation in the relationship between these anthropometric indices and death.
Results
BRI showed a significant J‐shaped association with death, with increased risks at extremes. In multivariable models, BRI outperformed BMI in discriminating all‐cause (area under the receiver operating characteristic curve, 70.5% versus 59.3%) and cardiovascular death (area under the receiver operating characteristic curve, 75.4% versus 62.7%). Categorical net reclassification improvement analysis indicated that BRI significantly enhanced risk prediction for all‐cause (net reclassification improvement, 44.5%) and cardiovascular death (net reclassification improvement, 30.2%). Crucially, CRP levels mediated 23.9% of BRI's effect on all‐cause death and 13.2% on cardiovascular death, while no significant mediation was observed with BMI.
Conclusions
BRI provides superior mortality prediction compared with BMI, likely due to its association with visceral adiposity and inflammation. These findings highlight BRI's potential as a more accurate tool for risk stratification and the importance of targeting inflammation to mitigate obesity‐related risks.