Depressive symptoms are associated with hospitalization, but not with mortality in the elderly: a population-based study.

OBJECTIVES: Subjects with depression might request more healthcare services. The aim of this study was to evaluate the association of depressive symptoms with hospitalization and mortality rates in a general unselected older population. METHODS: We assessed the association of the 30-items Geriatric Depression Scale (GDS) score with one-year hospitalization and mortality in all 344 subjects aged 75+ living in Tuscania (Italy). This population had been enrolled in a national study of the genetic determinants of health status. Analyses were conducted using continuous, as well as categorical GDS score levels. RESULTS: After adjusting for potential confounders, depressive symptoms were associated in Cox regression modelling with hospitalization (relative risk, RR = 1.05; 95% confidence interval, CI = 1.01-1.09; p = 0.016), but not with mortality (RR = 1.06; 95% CI = 0.96-1.16; p = 0.246). Increasing GDS score levels were associated with increasing risk of hospitalization (p for trend = 0.033). Up to 39% of hospitalizations might be attributed to depression. CONCLUSION: Depressive symptoms are not associated with increased mortality rates in general older populations. However, depressive symptoms represent a potentially reversible determinant of increased hospitalization rates in these subjects, independent of the presence and severity of other medical conditions.