Objectives: To investigate multiple mediations of the association between education and depressive symptoms (BDI-V) by work-related stress (ERI) and social isolation, the regional variation of the first mediation and a potential moderating effect of regional unemployment rate.
Methods: 6339 employees born in 1959 and 1965 were randomly recruited from 222 sample points in a German cohort study on work, age, health and work participation. A multilevel model of moderated lower-level mediation was used to investigate the confirmatory research question. Multiple mediations were tested corresponding to Baron and Kenny. These analyses were stratified for age and adjusted for sex, negative affectivity and overcommitment.
Results: In the association between education and depressive symptoms, indirect effects of work-related stress and social isolation were significant in both age cohorts whereas a direct association was observable in the younger cohort, only. The significant regional variation in the association between work-related stress and depressive symptoms was not statistically explained by regional unemployment rate.
Conclusions: Our findings point out that work-related stress and social isolation play an intermediary role between education and depressive symptoms in middle-aged employees.