The temporal association between incident late-life depression and incident dementia

Abstract

Objective: There is an established association between depression and subsequent dementia. The present study examined temporal associations between incident late-life depression and subsequent dementia, also considering age and sex. Methods: We used longitudinal health claims data from the largest German health insurance provider (‘Allgemeine Ortskrankenkasse’) considering up to 9 follow-up years in piecewise exponential models. ICD-10 codes were used to define incident depression and dementia in individuals ≥65 years (n = 97 110). Results: Incident depression was associated with a higher risk of subsequent dementia (incidence rate ratios (IRR) adjusted for age and sex: IRR = 1.58, 95% CI = 1.51–1.64). The strongest association was found for the shortest interval of 1 quarter (IRR = 2.04, 95% CI = 1.88–2.21), with significant associations up to an interval of roughly 3 years. The association was more pronounced and lasted for more quarters in the younger portion of this study group (ages from 65–74: IRR = 2.00, 95% CI = 1.83–2.18; 75–84: IRR = 1.64, 95% CI = 1.55–1.73; ≥85: IRR = 1.19, 95% CI = 1.08–1.31). It was stronger among men than women (men: IRR = 1.98, 95% CI = 1.84–2.14; women: IRR = 1.44, 95% CI = 1.37–1.51) with no sex-specific temporal association. Conclusion: This large claims data study confirmed that incident late-life depression is associated with a higher risk of dementia within the 3 years following diagnosis. Hence, incident late-life depression should prompt further cognitive examinations and referrals to specialists. This might apply especially to younger seniors and men.