Use of palliative care consultation in terminal gastrointestinal cancer hospitalizations.

Abstract

90 Background: Palliative care is central to high-quality end-of-life oncology care, yet inpatient use near death remains inconsistent. In a nationally representative U.S. inpatient sample, we evaluated palliative care consultation (PCC) among patients with gastrointestinal (GI) cancer who died during hospitalization, focusing on patient-, treatment-, and hospital-level gaps. Methods: Adult hospitalizations with a primary GI malignancy diagnosis and inpatient mortality were identified in the 2019–2023 National Inpatient Sample. PCC was identified using ICD-10 codes. Survey-weighted multivariable regression models evaluated factors associated with PCC and adjusted associations between PCC and end-of-life outcomes. Results: Among 8,915 GI cancer hospitalizations ending in inpatient death, only 2,312 patients (25.9%) received PCC. Patients receiving PCC more often had metastatic disease (68.5% vs 56.2%), were from the highest median household income quartile (24.2% vs 20.4%), were not receiving cancer-directed treatment (84.9% vs 79.5%), and were treated at urban teaching hospitals (82.8% vs 78.0%) (all P<0.001). After adjustment, PCC was more common among patients with mild comorbidity (adjusted odds ratio [aOR], 1.29; 95% CI, 1.12–1.50) and metastatic disease (aOR, 1.57; 95% CI, 1.39–1.77), but less likely among those receiving ongoing cancer-directed treatment (aOR, 0.74; 95% CI, 0.63–0.86), those from non-metropolitan areas (aOR, 0.80; 95% CI, 0.65–0.99), and those treated at private investor-owned hospitals (aOR, 0.36; 95% CI, 0.27–0.47). PCC was strongly associated with do-not-resuscitate documentation (64.1% vs 19.5%; aOR, 7.09; 95% CI, 6.29–7.98). PCC was also associated with lower odds of high-intensity acute care markers, including sepsis (aOR, 0.56; 95% CI, 0.39–0.80), shock (aOR, 0.72; 95% CI, 0.58–0.88), respiratory failure (aOR, 0.39; 95% CI, 0.26–0.57), acute kidney injury (aOR, 0.69; 95% CI, 0.58–0.82), mechanical ventilation (aOR, 0.61; 95% CI, 0.48–0.78), and GI bleeding (aOR, 0.60; 95% CI, 0.46–0.77). PCC was not associated with longer length of stay and was associated with lower costs ($20,400 vs $24,900) and charges ($70,300 vs $92,600) (both P<0.001). Conclusions: In this nationally representative U.S. cohort, three in four patients with GI cancer who died in the hospital had no documented PCC. Independent gaps persisted among patients receiving cancer-directed treatment, those from non-metropolitan areas, and those treated at private investor-owned hospitals. Standardized inpatient PCC triggers represent a measurable quality target to improve goal-concordant documentation, reduce high-intensity end-of-life care, and advance value-based oncology care.