Mortality attributable to diabetes in Cuba: Estimates for 2019.

Objective: To estimate the national and provincial number of excess deaths due to diabetes across Cuba in 2019. Study design: Cross-sectional design with secondary data. Methods: We used DISMODII, a computerized generic disease model, to assess disease burden by modelling the relationships between incidence, prevalence, and disease-specific mortality. Baseline input data included population structure, total mortality, and age- and sex-specific estimates for diabetes prevalence from the Cuban National Health Survey 2019, and available published estimates of the relative risk of death for people with diabetes compared to people without diabetes. The results were internally validated with DISMODII output for duration of diabetes (years). Results: In 2019, we estimated an excess of mortality attributable to diabetes of 7.5 times the diabetes mortality reported by the National Death Registry, which is equivalent to 16.4 % of all deaths in Cuba. The percentages of all-cause mortality among provinces varied between 10.7 % in Villa Clara and 24.5 % in Ciego de Avila. Conclusions: These are the first estimates of mortality attributable to diabetes in Cuba and its provinces. Diabetes is likely to be a much more prominent leading cause of death than the 9th ranking reported by the Cuban National Death Registry 2019. Disease models similar to DISMODII are important tools to validate the epidemiologic indicators used in the burden of disease calculations.