Background <p>The association between body roundness index (BRI) and all-cause mortality and the occurrence of cardiovascular disease (CVD) in patients with diabetes is unknown. This study aimed to determine the association between BRI and all-cause/CVD mortality in a diabetic cohort.</p> Methods <p>A total of 8227 individuals with diabetes from the 1999–2018 National Health and Nutrition Examination Survey&#xa0;(NHANES) database were included. Multifactorial Cox regression models were used to analyze the association between BRI and mortality in patients with diabetes. Multivariate-adjusted restricted cubic spline (RCS) regression was used to test for nonlinearity.</p> Results <p>During a median follow-up of 7.25&#xa0;years, a total of 27.22% died, with 9.18% of these deaths due to CVD. After fully adjusting for potential confounders, BRI remained significantly associated with all-cause/CVD mortality in the diabetic population. The restricted cubic splines revealed no significant nonlinear relationship between BRI and all-cause mortality (<i>P</i> = 0.29) or CVD mortality (<i>P</i> = 0.73). BRI was better associated with all-cause/CVD mortality in patients with diabetes compared to other body metabolic indices.</p> Conclusions <p>In patients with diabetes, we found an association between BRI and all-cause/CVD mortality.</p>