Predictors of 15-year survival among Australian women with diabetes from age 76-81

Aims: To assess the impact of diabetes on the survival of older women, adjusted for other all-cause mortality predictors. Methods: Data were used from the 1921-26 cohort of the Australian Longitudinal Study on Women's Health, when the women were aged 76-81 years at baseline, with linkage to the National Death Index. Survival curves were plotted to compare the survival of women with no diabetes, incident diabetes and prevalent diabetes over 15 years. Cox proportional hazards models were used to examine the association between diabetes and all-cause mortality risks. Results: A total of 972 (11.7%) of 8296 eligible women reported either incident, 522 (6.3%) or prevalent, 450 (5.4%) diabetes. The median survival times were 10.1, 11.4 and 12.7 years among women with prevalent, incident and no diabetes, respectively. The risks of death were 30% [HR: 1.30 (95% CI: 1.16-1.45)] and 73% [HR: 1.73 (CI: 1.57-1.92)] higher for women with incident and prevalent diabetes compared to women without diabetes. These associations were sustained after controlling for demographics, body mass index, smoking status, comorbidities and health care use. Conclusions: This study revealed that diabetes is associated with reduced survival probabilities for older women with minimal moderation after adjustment for other predictors. Our findings suggest that diabetes management guidelines for older women need to integrate factors such as comorbidities, smoking and being underweight to reduce the risk of mortality.