Bidirectional associations between memory and depression moderated by sex and age: Findings from the CLSA

Research has struggled to understand the temporal relationship between cognition and depression. Some literature suggests that depression may be a risk factor for memory decline, while other work indicates that memory decline may precede depression symptoms. The purpose of this study was to clarify the temporal relationship between memory and depression, examining the moderating role of sex and age.

Data were drawn from two time points in the Canadian Longitudinal Study on Aging (CLSA). Memory was measured using a composite of immediate and delayed verbal recall scores, and depressive symptoms were measured using the Center for Epidemiologic Studies Short Depression Scale (CESD-10). Separate cross-lagged panel models (CLPMs) were run based on age (i.e., ages 45-64; ages 65+) and sex (n = 51,338).

Results indicated bidirectional associations between depressive symptoms and memory such that depressive symptoms at baseline predicted memory at follow-up (β= 0.029-0.068, with all p-values <0.01) and memory at baseline predicted depressive symptoms at follow-up (β= 0.025-0.033, with all p-values <0.05). The only exception was in the older female group, where memory did not predict depressive symptoms (β= -0.006, p = 0.543). Depressive symptoms at baseline were a stronger predictor of memory at follow-up than memory at baseline was for depressive symptoms at follow-up in all groups except for older males.

The findings suggest small but consistent bidirectional associations between depression and memory in almost all sex/age groupings. Depressive symptoms tended to be a stronger predictor of memory than memory was for future depressive symptoms.