Background Diabetes is associated with an increased risk of dementia and disability. However, the implications of future trends in diabetes for the burden of these conditions are unclear. The aim of our study is to estimate the potential impact of trends in diabetes prevalence upon the future burden of dementia and disability in England & Wales by 2060.

Methods We used a probabilistic multi-state, open-cohort, Markov model to integrate observed trends in Type 2 diabetes, cardiovascular disease and dementia to forecast the occurrence of disability and dementia to 2060. The model incorporated English Longitudinal Study of Ageing (ELSA) data, published effect estimates for state transition probabilities, trends in mortality and dementia incidence.

The baseline scenario assumed that the recently observed trends in obesity would continue, resulting in a 26% increase in Type 2 diabetes cases by 2060. Against this baseline, we compared three other scenarios reflecting alternative projected trends in diabetes suggested by Public Health England models: increases of 7%, 20% and 49%. For each scenario, we then calculated the cumulative number of dementia and disability cases and number of life years lost or gained by 2060, in comparison to the baseline scenario.

We used probabilistic sensitivity analysis to estimate 95% uncertainty intervals (UI).

Results If the relative prevalence of Type 2 diabetes increases 49% by 2060, we might expect approximately 106,000 (95%UI 97,500 to 112,800) cumulative additional cases of disability, some 86,000 (95%UI: 80,000 to 92,500) additional cases of dementia and approximately 2,570,000 (95%UI: 2,500,000 to 2,660,000) life years lost by 2060.

If prevention policies succeed in slowing down the increase in Type 2 diabetes to 7% by 2060, we might expect approximately 94,000 (87,000 to 100,400) fewer new cases of disability, 77,000 (95%UI: 71,800 to 82,900) fewer cases of dementia and approximately 2,300,000 (95%UI: 2,220,000 to 2,370,000) life years gained by 2060. However, large benefits would only be seen after a substantial lag-time: only 4,700 (95%CI: 4,300 to 5,100) new cases of disability and 3,200 (95%CI: 2,900 to 3,500) new cases of dementia would be avoided by 2030.

Conclusion Substantial reductions in the future burden of dementia and disability appear eminently achievable if effective prevention policies succeed in halting the ongoing epidemic of obesity and associated Type 2 diabetes.

However, these reductions might only become visible after a substantial lag-period.