Abstract
BACKGROUND: Although racial disparities in end-of-life care exist, whether the effect of specialist palliative care (PC) on end-of-life outcomes differs by race is unknown. METHODS: We created a propensity-matched cohort of Black and non-Hispanic White (NHW) older adults with metastatic cancer who did and did not receive specialist PC in the last 6 months of life. End-of-life outcomes included hospice use, hospice enrollment ≥ 3 days, intensive care unit (ICU) use in the last 30 days of life, and use of chemotherapy in the last 14 days of life. We used Cox regression models to evaluate for the presence of interaction between specialist PC and race for all outcomes on the multiplicative scale using interaction terms, and on the additive scale using the relative risk due to interaction (RERI), attributable proportion (AP) and synergy index (SI). RESULTS: After 1:1 matching, 13,931 exposed and 13,931 unexposed older adults were included, of which 13.4% were Black. In comparison to those who did not receive specialist PC, both Black and NHW older adults who received specialist PC were more likely to use hospice (71.0% versus 52.4% for Black, p < 0.0001; 80.4% versus 63.2% for NHW, p < 0.0001), have hospice enrollment ≥ 3 days (65.5% versus 44.0% for Black, p < 0.0001; 74.2% versus 52.6% for NHW, p < 0.0001), and have less use of the ICU (16.4% versus 19.9% for Black, p = 0.0058; 11.3% versus 14.0% for NHW, p < 0.0001) or chemotherapy at the end-of-life (1.2% versus 2.4% for Black, p = 0.0075, 1.4% versus 3.4% for NHW, p < 0.0001). There was no evidence of multiplicative interaction or additive interaction for any outcome. CONCLUSION: In older adults with metastatic cancer, there was no evidence for heterogeneity of effect for specialist PC between Black and NHW patients, suggesting that differences in the efficacy of specialist PC are not responsible for racial disparities in end-of-life care.