Outcomes of informal and formal care models at the end of life: An AI-supported systematic review

BACKGROUND: Multimorbidity and functional decline become increasingly prevalent with age, particularly in those near the end of life. This poses critical challenges for both families and society in providing adequate care. However, it remains unclear which type of care yields better outcomes. AIM: To systematically assess the impact of informal care, formal care, and a combination of both on health outcomes, healthcare utilization, and costs of care for older adults near the end of life. DESIGN: A systematic review with narrative synthesis was registered on PROSPERO (CRD42024512003). Screening was assisted by machine learning tool ASReview, and study quality was assessed using the Mixed Methods Appraisal Tool, with support from large language model TXYZ.AI. DATA SOURCES: Articles published between database inception and June 4th, 2025 were identified in PubMed, Embase, and CINAHL. RESULTS: Thirty-six studies, including 20 observational and 16 interventional were included. Most observational studies examined place of death, with some associations involving informal care pointing towards lower home death or greater hospital death. In contrast, physician and nurse home visits were associated with a greater likelihood of home death. Specialized formal care interventions generally reported improvements in quality of life and reductions in healthcare use. Quality assessment showed 85.2% AI-human agreement, with 44.4% of studies meeting all seven quality criteria. DISCUSSION: Informal care did not necessarily facilitate home death, whereas formal home-based care was linked to a greater likelihood of home death. Interventional evidence also suggests broader benefits of specialized formal care, although certainty was limited by methodological concerns.