Life-course pathways from childhood starvation to late-life depressive symptoms: a double machine learning approach.

BACKGROUND AND OBJECTIVES: Although previous research has documented the adverse effects of childhood starvation as a profound early-life adversity, most studies have focused on physical health outcomes or mortality, with limited integration of life-course theory and Bourdieu's capital-based framework. This study examined whether childhood starvation predicted depressive symptoms in later life and further investigated whether these associations were mediated by economic, social, and cultural capital. RESEARCH DESIGN AND METHODS: Data were drawn from the 2018, 2020, and 2023 waves of the China Longitudinal Aging Social Survey. An unbalanced panel dataset comprising 25,852 observations was constructed, with participants being older adults in China. The double machine learning approach was applied to estimate the causal impact of childhood starvation on late-life depressive symptoms and analyze the mediating roles of the 3 types of capital. RESULTS: The findings indicated that (a) individuals who experienced frequent hunger in childhood reported significantly higher levels of depressive symptoms in old age and (b) economic, social, and cultural capital each partially mediated the association between childhood starvation and late-life depressive symptoms. DISCUSSION AND IMPLICATIONS: This study advanced life-course scholarship by integrating a capital-based framework to understand the enduring psychological consequences of childhood starvation, linking early adversity to late-life mental health through cumulative depletion of key resources. It identifies capital erosion as a key mechanism and offers implications for policies aimed at reducing childhood food insecurity, strengthening life-course resources, and expanding mental health services.