Impact of hospice enrollment timing on overall survival and quality of life indicators in patients with pancreatic ductal adenocarcinoma (PDAC).

Abstract

243 Background: This study aimed to examine the effects of hospice enrollment timing on survival and end-of-life quality of life indicators for pancreatic ductal adenocarcinoma (PDAC) patients. Methods: A retrospective chart review of 152 patients with biopsy-proven PDAC from August 2009 to December 2022 at a single center was conducted. Patients were categorized based on hospice enrollment timing: Early-Hospice (EH) (≥ 60 days before death), Late-hospice (LH) (< 60 days), and No Hospice (NH). Parameters assessed included overall survival (OS), palliative care service interaction, treatment lines, time from last chemotherapy treatment to death, hospital utilization after last chemotherapy treatment, documentation of advance care directives (ACD), and place of death. Statistical analysis was performed using Chi-squared test with Yates' continuity correction and Kruskal-Wallis test. The time-to-event endpoint was estimated using the Kaplan-Meier method. Results: Analysis of 152 patients showed that hospice enrollment timing did not significantly influence OS in PDAC patients (p=0.6). Median OS for EH (n=10), LH (n=84), and NH (n=58) groups were 10.1 months, 15.7 months, and 12.5 months, respectively. Subgroup analysis of patients with unresectable tumors at diagnosis (EH [n=7], LH [n=61], and NH [n=40]) showed no statistically significant difference in survival (9.19 vs. 12.22 vs. 9.08, p=0.08). The time from first interaction with the Palliative Care team to death was not different between groups. However, patients in EH had a notably longer median duration between their last chemotherapy session and death compared to LH and NH (201 vs. 43 vs. 31.5 days, p<0.01), as well as fewer days hospitalized after the last chemotherapy session (0.5 vs. 5 vs 8 days, p=0.02). Hospice enrollees had a substantially higher rate of documented ACD, and were far more likely to die at home. Conclusions: Although the timing of hospice enrollment did not significantly impact overall survival in PDAC patients, it played a significant role in influencing end-of-life quality of life indicators. Interaction with the palliative care team did not vary across groups, implying that hospice referral by oncology is acceptable and effective. However, documentation of ACDs for non-hospice enrollees was very low, a clear area for improvement.[Table: see text]