Search

Abstract Background Depression is associated with an increased risk for type 2 diabetes mellitus. However, depression may take different courses, and it is not fully understood how these affect the development of diabetes. It is further to be determined whether sex modifies the association between depression and type 2 diabetes. Methods We analyzed data from the Gutenberg Health Study, a longitudinal and population-based cohort study (N = 15,010) in Germany. Depressive symptoms (measured by PHQ-9), history of depression, diabetes mellitus, and relevant covariates were assessed at baseline, and the outcomes of prediabetes and type 2 diabetes mellitus were evaluated 5 years later. Logistic regression was used to estimate odds ratios of incident prediabetes and type 2 diabetes mellitus, adjusting for potential confounders as identified in a Directed Acyclic Graph. Results In the confounder adjusted model, current depression (PHQ-9 ≥ 10 at baseline; OR = 1.79, 95% CI = 1.11 to 2.74, p = 0.011), and persistent depression had a statistically significant (OR = 2.44, 95% CI = 1.62 to 3.54, p = 0.005) effect on incident type 2 diabetes mellitus. A history of depression without current depression had no statistically significant effect on type 2 diabetes (OR = 1.00, 95% CI = 0.68 to 1.43, p = 0.999). The effect of depression on incident diabetes did not differ significantly between women (OR = 2.02; 95% CI = 1.32 to 3.09) and men (OR = 2.16; 95% CI = 1.41 to 3.31; p-value for interaction on the multiplicative scale p = 0.832 and on the additive scale p = 0.149). Depression did not have a significant effect on incident prediabetes. Conclusion This study shows how the history and trajectory of depression shape the risk for diabetes. This raises interesting questions on the cumulative effects of depression trajectories on diabetes and body metabolism in general. Depression can negatively affect physical health, contributing to increased morbidity and mortality in people with mental disorders.