Objectives: Accumulating evidence shows depression as a risk factor for frailty, but studies are mainly population-based and widely differ in their assessment of either depression or frailty. We investigated the association between depression and frailty among geriatric outpatients using different assessment instruments for both conditions.
Method: Among 315 geriatric outpatients (mean age 72.1 years, 68.3% female sex) participating the MiMiCS-FRAIL cohort study, major and subthreshold depression were measured with psychiatric diagnostic interview according to DSM-5 criteria (SCID-5) as well as with instruments to screen and measure severity of depressive symptoms (GDS-15 and PHQ-9). Frailty was assessed according to a screening instrument (FRAIL-BR) and a multidimensional Frailty Index (FI-36 items). Multiple logistic and linear regression were performed to assess the association between depression (independent variable) and frailty (dependent variable) adjusted for confounders.
Results: Frailty prevalence in patients with no, subthreshold or major depressive disorder increases from either 14.5%, 46.5% to 65.1% when using the FRAIL-BR questionnaire, and from 10.2%, 20.9%, to 30.2% when using the FI-36 index. These association remain nearly the same when adjusted for covariates. Both the FRAIL-BR and the FI-36 were strongly associated with major depressive disorder, subthreshold depression, and depressive symptoms by PHQ-9 and GDS-15.
Conclusion: Late life depression and frailty are associated in a dose-dependent manner, irrespective of the used definitions. Nonetheless, to avoid residual confounding, future research on underlying biological mechanisms should preferably be based on formal psychiatric diagnoses and objectively assessment frailty status.
Keywords: Elderly; cohort; depression; frailty; prognosis.