Older Adults' End-of-Life Care: Are Retrospective Assessments Patterned by Proxy Relationship to Decedent?

Understanding the quality of older adults' end-of-life care is essential for informed decision-making and practitioner accountability. However, retrospective appraisals provided by proxies may be shaped by their relationship to the decedent. Failure to adjust for proxy relationship status may generate misleading characterizations of end-of-life care patterning. We examine how proxy reporters (spouse, daughter, son, other relative, other non-relative, paid caregiver) differ in their appraisals of the decedent's care in the last month of life.

Data are from 12 waves (2011-2022) of the National Health and Aging Trends Study (n = 3,229). We use binomial and multinomial logistic regression to evaluate proxy status differences regarding ten dimensions of care in the last month of life, reported by NHATS decedents' proxies: overall quality; adequately treated breathing problems, pain, and sadness/anxiety; care coordination; decisions made with patient input; care concordant with patient wishes; informed about care; personal care needs met; and respectful treatment. Models were adjusted for death context and decedent characteristics.

Paid caregivers are more likely than family members to offer positive appraisals on all ten outcomes. Paid caregivers differ from other non-family proxies with respect to care delivery, but not symptom management. Family members differ slightly from one another, with spouses offering more critical assessments of patient/family input, and receipt of care concordant with wishes. Daughters and other relatives offered superior overall appraisals, relative to sons.

Researchers, practitioners, and policymakers should recognize that the proxy's relationship to decedent may influence their appraisals of end-of-life care.