Objective: It has been suggested that self-rated health (SRH), a construct of overall subjective health, is predicted by depressive symptoms. However, depressive symptoms were also found to predict SRH. The present study aimed to simultaneously examine these alternatives as well as test the potential moderating role of age. Methods: Data concerning self-reported measures from two consecutive waves of the Survey of Health, Aging and Retirement in Europe (SHARE) were examined by using a longitudinal cross-lagged panel analysis. Results: The model that included the effects of both paths, i.e. from SRH to depressive symptoms and from depressive symptoms to SRH, demonstrated better fit than models including one or no cross-lagged paths. However, the longitudinal effect of SRH on depressive symptoms was stronger than the longitudinal effect of depressive symptoms on SRH, particularly among adults aged 65–79. The longitudinal effect of depressive symptoms on SRH was stronger among individuals aged 80 and older, compared with younger ages. Conclusions: Bidirectional effects may exist between SRH and depressive symptoms among adults aged 50 and older. Special attention should be paid to the unique effects of SRH on depressive symptoms in the years following retirement and also to depressive symptoms on SRH among the old-old.
- Cross-lagged analysis
- Depressive symptoms
- Self-rated health
- Survey of Health Ageing and Retirement in Europe (SHARE)