Mediators of the association between depression and role functioning

Abstract

Objective: While the adverse effect of Major Depressive Episode on role functioning is well established, the exact pathways remain unclear.

Method: Data from The European Study of the Epidemiology of Mental Disorders, a cross-sectional survey including 21 425 adults from six European countries, were used to assess 12-month depression (Composite International Diagnostic Interview), activity limitations and role functioning in the past 30 days (Disability Assessment Schedule). An a priori model based on the World Health Organization's International Classification of Functioning, Disability and Health was designed and a structural equation model for categorical and ordinal data was used (MPlus) to estimate the extent to which six limitations mediated the association between depression and role functioning.

Results: The unadjusted association between depression and role functioning was strong (0.43; SE = 0.04). In the best-fitting model, only concentration and attention problems and embarrassment mediated a significant amount of association (direct effect dropped to 0.17; SE = 0.10, which was no longer significant).

Conclusions: Targeting cognition and embarrassment in treatment could help reduce depression-associated role disfunctioning.

Fig. 1

Composition of the study sample, i.e. respondents who were administered questions about activity limitations and role functioning (the ESEMeD WHODAS). All analyses were weighted to produce estimates of statistics that would have been obtained if the entire sampling frame had participated and to restore the relative size of each country’s general population.

Fig. 2

The fully saturated mediation model (mutual associations between de mediators not depicted). Arrows represent the associations between MDE and activity limitations (a1–a6), the associations between activity limitations and role functioning (b1–b6) and the direct association of MDE and role functioning (c).

Fig. 3

The final model (CFI = 0.999, RMSEA: 0.010; Δχ² = 4.68, df = 3, P = 0.20). Path coefficients represent standardized partial regression coefficients and standard errors (in parentheses).