Evaluation and appraisal of quality indicators for end-of-life cancer care in an African context.

Abstract

1517 Background: Population-based quality indicators (QIs) for end-of-life (EoL) cancer care are used to monitor and improve care delivery. Most QIs measure overuse of anticancer therapy or acute care near death and underuse of palliative care. This enables optimization of patient-centered care and healthcare resource utilization, which is especially critical in low-resource settings. Yet, QIs developed in U.S. may not be directly transferable in Africa. We evaluated EoL cancer care in Rwanda using existing QIs and appraised their relevance and feasibility in Rwanda. Methods: We conducted a retrospective chart review of all patients who died or were referred for EoL care (i.e., hospice) in 2022 in a Rwandan cancer center. Missing death dates were sought from family and community health workers. Relevance and feasibility of established QIs were appraised by the study team. QI performance was compared with published U.S. benchmarks. Results: Of 297 patients (median age 43; 16% pediatric; 61% female), the most common cancers were breast (19%), cervical (9%), acute lymphoid leukemia (9%), gastric (8%), lymphoma (7%), rectal (5%). Nearly half (46%) were stage IV at diagnosis. Most (72%) received treatment; 61% had chemotherapy. After outreach, death dates were available for 95%. Of 18 QIs validated in U.S., 5 were retained or adapted for Rwanda and 13 excluded due to contextual irrelevance or inadequate documentation. Length of stay in EoL care achieved its benchmark, while performance on other QIs was suboptimal (see Table). Conclusions: This study is among the first to evaluate and appraise established EoL cancer care QIs in Africa and highlights opportunities to reduce overuse of non-beneficial chemotherapy near death, increase use of palliative care and hospice, and adapt or develop novel EoL QIs to best fit African contexts. Comparison of EoL Care in Rwanda with published benchmarks. Rwanda Benchmark % referred for EoL care before death (n=297) 45% >55% % with length of stay in EoL care 3 days (n=114) 5% <8% % who died in a hospital (n=281) 48% <17% % starting new chemo within 30 days of death (n=297) 12% <2% % receiving chemo within 14 days of death (n=297) 17% <10% % receiving chemo within 30 days of death (n=297) 24% N/A* *Not a published benchmark; 45% in a published report from Uganda.