- Four relatively stable spousal depressive symptom trajectories identified, including concordant and discordant patterns, with little evidence of widespread convergence between partners.
- Incident MCI risk was driven by the respondent's depressive symptoms; both-depressive and respondent-only groups had higher MCI risk; spouse-only group did not.
- No significant association with all-cause mortality; depressive symptoms related to poorer baseline cognition but not consistently faster cognitive decline.
Int J Geriatr Psychiatry. 2026 Oct;41(10):e70258. doi: 10.1002/gps.70258.
ABSTRACT
OBJECTIVE: To examine the longitudinal patterning of depressive symptoms within couples and clarify whether subsequent health risks associated with spousal depressive-symptom concordance reflect the respondent’s own symptoms, the spouse’s symptoms, or an additional effect of concordance.
METHODS: Using five waves of the China Health and Retirement Longitudinal Study (CHARLS; 2011-2020), we identified longitudinal patterns of spousal depressive symptoms using group-based trajectory modeling. Couples were further classified into four baseline depressive-symptom groups according to the depressive-symptom status of the respondent and spouse. Associations with incident mild cognitive impairment (MCI) and all-cause mortality were examined using Cox proportional hazards models. Additional analyses using continuous CES-D-10 scores, Fine-Gray competing-risk models, and linear mixed-effects models were conducted to evaluate individual contributions of respondent and spouse symptoms, the robustness of MCI findings, and longitudinal changes in cognitive domains, respectively.
RESULTS: Four relatively stable patterns of spousal depressive symptoms were identified, including concordant and discordant patterns, with no evidence of widespread convergence between spouses over time. Compared with the neither-depressive-symptoms group, both-depressive-symptoms participants (HR = 1.22, 95% CI: 1.08-1.38) and respondent-only participants (HR = 1.25, 95% CI: 1.10-1.42) had higher risks of incident MCI, whereas the spouse-only group did not (HR = 0.94, 95% CI: 0.82-1.07). The risk did not differ significantly between the both-depressive-symptoms and respondent-only groups. Continuous CES-D-10 analyses showed that respondent symptom severity, but not spouse symptom severity or the respondent-by-spouse interaction, was associated with incident MCI. No significant association was observed between depressive-symptom groups and all-cause mortality. Cognitive-domain analyses indicated that depressive-symptom groups were associated with poorer baseline cognitive performance but not consistently faster cognitive decline.
CONCLUSION: Spousal depressive symptoms may show concordance, but discordant patterns can also remain stable over time, providing little evidence of a general pattern of convergence between partners. The association with incident MCI was primarily related to the respondent’s own depressive symptoms, with no evidence that the spouse’s symptoms independently increased risk or that elevated symptoms in both partners conferred additional risk beyond the respondent’s own symptoms.
PMID:42895199 | DOI:10.1002/gps.70258
Share Evidence Blueprint
Save to Google Notes
Search Google Scholar
Save as PDF
⭐ My Revision List
