Background: Increased longevity and reductions in diabetes(DM) complication rates have altered patterns of DM-related mortality in several countries. It is unclear whether this diversification of mortality has occurred in the UK.

Methods: With linked hospital and mortality data from 320,613 adults aged 26 - 91 from the 1998-2018 UK Clinical Practice Research Datalink, we used discrete Poisson regression to estimate death rates and proportional mortality from all causes, vascular diseases, cancers, and other (non vascular, non cancer) causes in DM patients. We compared this to age and sex matched individuals without DM.

Findings: From 1998-2018 all-cause death rates per 100 person-years declined by 36% in individuals with DM (4.42 (4.37-4.48) to 2.82 (2.79-2.85)). There were substantial reductions (57%, 1.84 (1.80-1.88) to 0.80 (0.78-0.81)) in vascular, moderate reductions in ’other’ (31%, 1.77 (1.73-1.80) to 1.21(1.19-1.23)), and minor reductions in cancer (8%, 0.92(0.89-0.94) to 0.84(0.82-0.86)) death rates. Within the ’other’ category, exceptions included increases in dementia (468%) and liver disease (86%) with static respiratory and accident death rates. These differential rates of decline resulted in increases in the proportion of all deaths due to cancer (20% to 29%) and ’other’ causes (39% to 43%). Rates of decline in all-cause and cancer mortality were greatest among young (<55 years) and middle aged (55-75 years), whereas those >75 years had no reduction in cancer mortality. All cause death rates also declined substantially among the non-DM population (from 2.78 (2.75-2.82) to 1.77 (1.75-1.79)/100 PYs) and across all major cause groupings.

Conclusions: Reductions in vascular mortality in patients with DM have not been matched by improvements in cancer mortality, while death rates for specific ’other’ causes (dementia and liver) have increased. This reflects a diversification of DM-related mortality with profound implications for management and prevention of DM.


J. Pearson-Stuttard: Advisory Panel; Self; Novo Nordisk A/S. Consultant; Self; DnaNudge Ltd. Y.J. Cheng: None. E.P. Vamos: None. J.E. Bennett: None. A.J. Cross: None. M. Ezzati: Consultant; Self; Prudential, Scor, Third Bridge. Research Support; Self; AstraZeneca Youth Health Programme. E. Gregg: None.


UK Wellcome Trust

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at