The impact of age-friendly smart homes on care burden, out-of-pocket costs and their distributional inequality among urban older households in China

Population ageing and ageing-in-place policies leave more frail older adults at home, raising family care burden and costs and calling for more equitable home-based care. Age-friendly smart homes are increasingly installed to share this care. We ask whether such technology substitutes for care uniformly and how its benefits are distributed across households. Using a prospective matched cohort of 640 older households in 24 urban communities in Hefei, China, we estimate the average treatment effect on the treated (ATT) of installation with propensity-score weighting, test its variation with interaction models, and relate the passive and active functions configured in each treated home to outcomes in within-treatment dose models. Installation is associated with lower care burden and out-of-pocket costs and with fewer fall-related visits. The gains are unevenly captured. Hospital-day reductions concentrate among older adults living alone, a gradient that survives false-discovery-rate control across 15 moderator tests. Out-of-pocket savings narrow as human care availability rises, a gradient that survives the same control under the normal reference distribution and falls short of it under a small-cluster t reference. No gradient is detected for care-burden relief. Within treated homes, wider passive-monitoring coverage goes with caregivers withdrawing from device-mediated supervision. The out-of-pocket difference lies mainly in recorded inpatient and other medical spending, adjusting for hospital days changes it little, and the medical estimates are sensitive to assumptions about households without records. Allocation can combine need-based eligibility with priority for households facing care scarcity, with the balance set by the policy goal.