Aims/hypothesis The aim of this study was to investigate the risks of all-cause and cause-specific mortality among participants with neither, one or both of diabetes and depression in a large prospective cohort study in the UK. Methods Our study population included 499,830 UK Biobank participants without schizophrenia and bipolar disorder at baseline. Type 1 and type 2 diabetes and depression were identified using self-reported diagnoses, prescribed medication and hospital records. Mortality was identified from death records using the primary cause of death to define cause-specific mortality. We performed Cox proportional hazards models to estimate the risk of all-cause mortality and mortality from cancer, circulatory disease and causes of death other than circulatory disease or cancer among participants with either depression ( n =41,791) or diabetes ( n =22,677) alone and with comorbid diabetes and depression ( n =3597) compared with the group with neither condition ( n =431,765), adjusting for sociodemographic and lifestyle factors, comorbidities and history of CVD or cancer. We also investigated the interaction between diabetes and depression. Results During a median of 6.8 (IQR 6.1–7.5) years of follow-up, there were 13,724 deaths (cancer, n =7976; circulatory disease, n =2827; other causes, n =2921). Adjusted HRs of all-cause mortality and mortality from cancer, circulatory disease and other causes were highest among people with comorbid depression and diabetes (HRs 2.16 95% CI 1.94, 2.42; 1.62 95% CI 1.35, 1.93; 2.22 95% CI 1.80, 2.73; and 3.60 95% CI 2.93, 4.42, respectively). The risks of all-cause, cancer and other mortality among those with comorbid depression and diabetes exceeded the sum of the risks due to diabetes and depression alone. Conclusions/interpretation We confirmed that depression and diabetes individually are associated with an increased mortality risk and also identified that comorbid depression and diabetes have synergistic effects on the risk of all-cause mortality that are largely driven by deaths from cancer and causes other than circulatory disease and cancer. Graphical abstract