Abstract
IMPORTANCE: Evidence is limited regarding whether end-of-life care for individuals with dementia varies by race and ethnicity, and whether observed variations can be explained by differences in the physicians providing their care. OBJECTIVE: To evaluate end-of-life care among individuals with dementia across racial and ethnic groups, and to investigate whether care variations are explained by differences in treating physicians. DESIGN, SETTING, AND PARTICIPANTS: This cohort study used national, population-based claims data from a 20% random sample of Medicare fee-for-service beneficiaries aged 66 years or older with a diagnosis of dementia who died between 2016 and 2019. Data were analyzed from January 2024 to June 2025. MAIN OUTCOMES AND MEASURES: Emergency department, hospital, and intensive care unit use, mechanical ventilation or cardiopulmonary resuscitation, and feeding tube placement in last 30 days of life; death in acute care hospital; hospice use and palliative care counseling in last 180 days of life; and any billed advance care planning before death. RESULTS: Among 259 945 decedents with dementia (mean [SD] age, 85.8 [8.0] years; 60.4% female), 8.3% were non-Hispanic Black, 4.4% were Hispanic, and 87.3% were non-Hispanic White. Compared with non-Hispanic White decedents, non-Hispanic Black decedents were more likely to receive emergency department (difference, 5.7 percentage points [pp]; 95% CI, 5.0-6.4 pp), hospital (difference, 4.0 pp; 95% CI, 3.3-4.7 pp), intensive care unit (difference, 4.3 pp; 95% CI, 3.7-4.9 pp), mechanical ventilation or cardiopulmonary resuscitation (difference, 3.8 pp; 95% CI, 3.3-4.3 pp), and feeding tube placement (difference, 1.8 pp; 95% CI, 1.5-2.1 pp) care, as well as die in a hospital (difference, 3.5 pp; 95% CI, 2.9-4.1 pp). Non-Hispanic Black decedents were less likely to use hospice (difference, -6.1 pp; 95% CI, -6.8 to -5.4 pp) and more likely to receive palliative care counseling (difference, 3.2 pp; 95% CI, 2.6-3.9 pp) and billed advance care planning (difference, 1.8 pp; 95% CI, 1.2-2.3 pp) than non-Hispanic White decedents. Similar patterns were observed among Hispanic decedents. Variations in end-of-life care remained qualitatively unchanged when comparing decedents treated by the same physician. CONCLUSIONS AND RELEVANCE: Findings from this cohort study suggest that non-Hispanic Black and Hispanic decedents with dementia received more intensive end-of-life care despite higher rates of billed advance care planning and palliative care counseling than non-Hispanic White decedents. Observed racial and ethnic variations were not explained by differences in the physicians treating them.