Healthcare utilization and end-of-life care in pancreatic cancer: A single-institution cohort study.

Abstract

e23206 Background: Pancreatic cancer carries high symptom burden and frequent acute care use, yet real-world utilization and end-of-life care patterns are incompletely described. Methods: We conducted a retrospective cohort study of adults with pancreatic cancer treated at the Mitchell Cancer Institute (2018–2024) under institutional review board approval. Encounters from the tumor registry, electronic health record, and billing data included emergency department visits, hospitalizations, ICU admissions, hospice enrollment, palliative care consultation, and chemotherapy within 30 days of death. Total utilization was defined as the sum of encounters across settings. Early palliative care was defined as consultation ≥90 days before death or within 60 days of metastatic diagnosis. Utilization was summarized using medians (IQR) and compared by chemotherapy receipt using the Mann–Whitney U test. Results: Among 583 patients (median age 67 years; 51% male), 297 (50.9%) received chemotherapy and 210 (36.0%) died during follow-up. Median total utilization was 20 encounters [6.5–45]. Chemotherapy receipt was associated with higher utilization (39 [21–65] vs 7 [2–18], p < 0.001). ICU admission occurred in 49 patients (8.4%). Among decedents, hospice enrollment occurred in 94 (44.8%) and early palliative care in 46 (21.9%). Chemotherapy within 30 days of death occurred in 119 (56.7%), exceeding commonly used quality benchmarks ( < 20%). Conclusions: Care was characterized by high utilization and frequent intensive treatment near the end of life. Low hospice and early palliative care use highlight opportunities for earlier supportive care integration. Findings are limited by single-institution design and reliance on registry and billing data. Healthcare utilization and end-of-life care metrics. Measure Overall (N=583) Benchmark/Notes Age, median (IQR) 67 (58–73) Baseline Male sex, n (%) 298 (51.1) Baseline Total encounters, median (IQR) 20 (6.5–45) Chemo 39 vs 7; p<0.001 ICU admission, n (%) 49 (8.4) Care intensity Chemo ≤30d of death* 119 (56.7) QOPI <20% Hospice enrollment* 94 (44.8) QOPI >50% Early palliative care* 46 (21.9) — *Among decedents (N=210).