JAMA Netw Open. 2026 Jul 1;9(7):e2621697. doi: 10.1001/jamanetworkopen.2026.21697.
ABSTRACT
IMPORTANCE: Changes in family structure in the US may be a driving force behind the growing need for friend caregivers. Yet, this segment of caregivers is often overlooked in health care settings and related research.
OBJECTIVE: To provide national estimates of friend caregivers, compare the characteristics of care provided by friend vs family caregivers, and identify characteristics of older adults associated with receiving care from friends.
DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional analysis of the 2023 National Health and Aging Trends Study included data on older adults (aged ≥65 years).
EXPOSURE: Sociodemographic and health-related characteristics of care recipients.
MAIN OUTCOMES AND MEASURES: The primary outcome was having a friend who provided assistance in the past month due to health or functional limitations. Adjusted Wald tests were used for continuous variables and Rao-Scott χ2 tests for categorical variables to compare characteristics of care provided by friend versus family caregivers and of care recipients receiving friend-involved versus family-only care. Multivariable logistic regression was used to identify older adult characteristics independently associated with friend involvement in care networks.
RESULTS: This study included 451 friend and 4985 family caregivers identified by 2619 older adults with activity limitations (weighted mean [SE] age, 78.7 [0.25] years; 63.2% [95% CI, 60.5%-65.9%] female) representing 2.4 million friend caregivers and 21.4 million family caregivers providing care to approximately 12.2 million recipients. Compared with family caregivers, friend caregivers were less likely to coreside with the care recipient (3.3% [95% CI, 1.6%-6.7%] vs 46.4% [95% CI, 44.2%-48.6%]; P < .001) or serve as the sole caregiver (11.1% [95% CI, 7.7%-15.9%] vs 21.2% [95% CI, 19.4%-23.3%]; P < .001). Although they provided fewer care hours per month than family caregivers (mean [SE], 18.1 [3.8] vs 66.6 [2.7] hours; P < .001), friends were more likely to help with transportation (65.7% [95% CI, 59.8%-71.2%] vs 52.7% [95% CI, 50.6%-54.8%]; P < .001), specialize in a single domain of care (66.4% [95% CI, 60.6%-71.9%] vs 33.6% [95% CI, 31.7%-35.5%]; P < .001), and share care with others. In multivariable models, being older (odds ratio [OR], 0.97; 95% CI, 0.95-0.99; P = .002) and married (OR, 0.52; 95% CI, 0.31-0.97; P = .01) were associated with a lower likelihood of having friends involved in the care network, whereas having a college degree (OR, 1.66; 95% CI, 1.14-2.43; P = .01) and living alone (OR, 2.29; 95% CI, 1.50-3.50; P < .001) were associated with a greater likelihood of having friends involved in the care network.
CONCLUSIONS AND RELEVANCE: In this cross-sectional study of Medicare beneficiaries aged 65 years and older, friend caregivers represented a specialized segment of the unpaid caregiving workforce, particularly for older adults lacking family resources. Understanding how friends provide care can help health care professionals better recognize them as an integral part of the caregiving network.
PMID:42390861 | DOI:10.1001/jamanetworkopen.2026.21697
