Association of the Frequency of In-Home Care Services Utilization and the Probability of In-Home Death

Importance: The provision of end-of-life care is an important policy issue associated with population aging around the world. Yet it is unclear whether the provision of in-home care services can allow patients the option of in-home death at end of life. Objective: To assess whether the frequent use of in-home care services can assist recipients to stay at home at the end of life. Design, Setting, and Participants: This cohort study of older adults in Japan's long-term care insurance system used national claims data. Participants were long-term care insurance beneficiaries aged 65 years or over who died in 2015, excluding those who died due to external causes such as accidents. Data analyses were conducted from October to December 2020. Exposures: Mean days of in-home care service used per week from the first day of the month before the month of death to the date of death. Main Outcomes and Measures: Primary outcome was whether the older person died at home (or not). To address lack of information on individual preference for place of death, we used an instrumental variable estimation with the full-time equivalent number of care workers providing in-home care services per older population at the municipality level in 2014. Results: Of the 572 059 decedents included in the study, 314 743 (55.0%) were women (median IQR age, 87 81-91 years). The proportion of in-home deaths was 10.5% (60 175 decedents), and 81 675 decedents (14.3%) used in-home care services at least once prior to their death. Ordinary least squares and 2-stage least squares analyses both indicated that more frequent use of in-home care was associated with a higher probability of in-home deaths (ordinary least squares estimate, 5.0 percentage points; 95% CI, 4.9-5.1 points vs 2-stage least squares estimate, 3.6 percentage points; 95% CI, 2.3-4.9 points). Conclusions and Relevance: This retrospective cohort study using an instrumental variable approach demonstrated that more frequent use of in-home care services at the end of life was associated with a higher probability of in-home death. One policy implication of these results is that in order to meet the end-of-life preferences of patients, it is not only necessary to promote the provision of medical services at home but also to ensure an adequate supply of care workers.