To investigate the association of body mass index with all-cause, cardiovascular and cancer mortality in individuals with and without diabetes.

We used data on 490,852 participants from the UK Biobank, with linkage to national mortality data between 2006 and 2016. Using Cox regression, we calculated hazard ratios (HRs) and 95% confidence intervals (95%CI) for all-cause, cardiovascular and cancer mortality within body mass index categories in people with and without diabetes adjusting for potential confounders. 24,789 (5.0%) participants reported having diabetes at baseline. Over a median follow-up of 6.9 years, 13,896 participants died, of which 1800 had diabetes. Compared with normal body mass index (18.5–24.9 kg/m²), mortality risk in the overweight group (25.0–29.9 kg/m²) was 33% lower in people with diabetes (HR 0.67, 95%CI 0.62–0.73) and 12% lower in participants without (HR 0.88, 95%CI 0.85–0.90). For class I obesity (30.0–34.9 kg/m²), mortality risk was 35% lower in participants with diabetes (HR 0.65, 95%CI 0.59–0.71) and 5% lower in participants without (HR 0.95, 95%CI 0.91–0.99). For class III obesity (≥40 kg/m²), there was a 10% non-significant lower mortality risk compared to normal body mass index in people with diabetes (HR 0.90, 95%CI 0.77–1.05); in contrast, the risk was 29% higher in people without diabetes (HR 1.29, 95%CI 1.13–1.45). Similar patterns were observed for cardiovascular mortality but not for cancer mortality.

The impact of obesity on the risk of mortality was dependent on the presence of diabetes: for the same level of obesity, mortality risk was higher in people without diabetes compared to those with diabetes.