Long-term care insurance pilot exposure and severe-depression risk among older adults with baseline severe IADL impairment: a panel difference-in-difference-in-differences analysis
摘要
In the context of rapid population ageing, long-term care insurance (LTCI) has attracted growing attention for its potential effects on health and well-being. However, most existing studies focus on average changes in depressive symptoms and pay limited attention to severe-depression risk at the upper tail of the mental health distribution. This study examines whether city-level exposure to the LTCI pilot is associated with severe-depression risk among older adults, with particular attention to those with high care needs.
MethodsUsing four waves of panel data from the China Health and Retirement Longitudinal Study (CHARLS; 2013, 2015, 2018, and 2020), we constructed a sample of adults aged 60 years and older. The final analytical sample included 27,967 person-year observations from 8,258 respondents, with a mean baseline age of 65.98 years (SD 5.76). We used baseline severe impairment in instrumental activities of daily living (IADL) to identify the high-care-needs population, defined severe depression tail risk using CESD-10 scores > = 16, and estimated intention-to-treat (ITT) associations with city-level LTCI pilot exposure using a difference-in-difference-in-differences (DDD) model.
ResultsIn the baseline specification, city-level exposure to the LTCI pilot was associated with a higher upper-tail risk of severe depression among older adults with baseline severe IADL impairment. The baseline ITT estimate corresponded to an approximately 12.45-percentage-point higher probability of being in the severe-depression range among this high-care-needs group relative to comparison groups. Additional diagnostics, including exclusion of the 2020 wave, province-level clustering, wild cluster bootstrap inference, and leave-one-treated-city-out checks, supported the positive direction and city-level stability of the estimate. The sex-specific signal appeared stronger among women in exploratory analyses.However, statistical significance weakened under some sample-restriction specifications, and the findings should be interpreted as suggestive rather than conclusive.
ConclusionsThe findings suggest that the psychological consequences of LTCI exposure may be more visible in the upper tail of the depression distribution among older adults with high care needs than in average mental-health indicators, with the signal appearing stronger in women in exploratory analyses. Policy evaluation should therefore move beyond mean-based indicators and strengthen needs-matched, psychologically supportive long-term care for vulnerable older adults.