Household Trajectories and Depressive Symptoms in Later Life: Evidence from Sequence Analysis and Marginal Structural Models

• In Event: Aging and the Life Course Section Roundtables
  In Refereed Roundtable (90 minute): 6: Family Relationships and Dynamics

Abstract

Abstract: This study examines how dynamic household structures across later adulthood shape depressive symptoms in later life. Using data from the U.S. Health and Retirement Study (1992–2022), we applied sequence analysis to classify adults aged 51–68 (N = 14,394) into nine distinct household trajectories, capturing long-term patterns such as living only with a spouse, living alone, or co-residing with children or kin. To address time-varying confounding between household changes, prior health, and socioeconomic status, we employed Marginal Structural Models (MSM) with stabilized inverse probability weights. After adjustment, most apparent associations between household structure and depressive symptoms (CESD) disappeared. The only trajectory showing a significant difference was living only with children, which predicted lower depressive symptoms compared to living only with a spouse. These results suggest that mental health differences across household structure types largely reflect selection and confounding, rather than causal effects. Methodologically, this study integrates sequence analysis with weighted causal modeling, offering a rigorous framework to evaluate how household trajectories influence later-life mental health.

Authors

• Huimin Zhang, University of Florida

• Won-tak Joo, University of Florida

• Jiahui Lyu, University of Florida