The aim of the study was to investigate the association between body roundness index (BRI) and all-cause mortality and cardiovascular mortality in general population.

A retrospective cohort study.

The status of cardiovascular mortality and all-cause mortality of participants were followed through 31 December 2015. Multivariate adjusted Cox restricted cubic spline regression models and Kaplan–Meier survival curves were used to evaluate the relationship between BRI and cardiovascular mortality and all-cause mortality.

A sample of 47 356 participants from the National Health and Nutrition Examination Surveys 1999–2014 with aged ≥18 years.

Mean age was 47 years and female were 49·9 %. During a median follow-up of 92 months, 4715 participants died from any cause, with 985 died of CVD. In multivariate adjusted Cox regression, compared with the lowest quartile of Body roundness index (BRI), the hazard ratios (HR) for all-cause mortality from other quartiles were 0·83, 95 % CI (0·75, 0·92), 0·73, 95 % CI (0·65, 0·81) and 0·80, 95 % CI (0·72, 0·89), respectively (Pfor trend < 0·05) and the HR for cardiovascular mortality from other quartiles were 0·79, 95 % CI (0·62, 1·00), 0·78, 95 % CI (0·62, 0·99) and 0·79, 95 % CI (0·62, 1·01), respectively (P for trend > 0·05). In the restricted cubic spline regression models, the relationship was showed U-shaped between BRI and all-cause mortality and cardiovascular mortality. In Kaplan–Meier survival curves, the lowest cumulative survival rate of cardiovascular mortality and all-cause mortality was recorded in the highest BRI quartile.

The U-shaped association between BRI and all-cause mortality and cardiovascular mortality in a large population-based cohort was observed.