Layers of Support: Family Caregivers, Personal Care Aides, and Mortality in Patients Receiving Home-Based Primary...

We examined the association between caregiving arrangements and mortality risk in patients enrolled in home-based primary care. Patients were categorized as having a family caregiver (43%), both family caregiver and personal care aide (PCA; 36%), PCA only (13%), or no caregiver (8%). We used multinomial logistic regression to model caregiving arrangements, then applied inverse probability weighting to account for baseline sociodemographic, clinical, and functional differences between groups in a Cox proportional hazards model. Compared to patients with no caregiver, all other groups had significantly lower mortality risk: PCA only (HR = 0.25, 95% CI: 0.11-0.60, p < .01), family caregiver only (HR = 0.29, 95% CI: 0.14-0.61, p < .01), and both PCA and family caregiver (HR = 0.26, 95% CI: 0.12-0.58, p < .01). These findings demonstrate that when accounting for selection into caregiving arrangements, having any caregiver type substantially reduces mortality risk among homebound older adults.