Impact of a novel advanced support unit on intensive care unit mortality at a comprehensive cancer center.

Abstract

573 Background: Many patients with advanced cancer die shortly after an intensive care unit (ICU) admission. For these patients, aggressive critical care may offer limited benefit, may not align with patients’ goals, and may prolong suffering. ICU admissions near the end of life (EOL) may be related to prognostic uncertainty, limited prognostic awareness or acceptance, preference for life-sustaining therapies, and delayed goals-of-care (GOC) discussions. To address these challenges, The University of Texas MD Anderson Cancer Center (MDACC) implemented a novel Advanced Support Unit (ASU), an intermediate care unit designed to promote multidisciplinary care team communication with patients with limited prognosis and support goal-concordant care. This study examined ICU outcomes before and after ASU implementation. Methods: This retrospective study compared consecutive ICU admissions 18 months before (March 1, 2023 to August 31, 2024) and 18 months after (September 1, 2024 to February 28, 2026) ASU opening. The ASU aimed to provide a patient-centered environment with a dedicated interdisciplinary care team to provide focused GOC discussions, and psycho-social support for patients preparing for EOL. Available therapies included limited vasopressors, non-invasive ventilation, and de-escalation of invasive mechanical ventilation. Plan-Do-Study-Act cycles optimized operations, eligibility, and ASU utilization. The primary outcome was ICU mortality. The secondary outcomes were (1) ICU mortality in the first 48 hours, (2) ICU utilization, defined as the proportion of hospital encounters requiring ICU admission and (3) ICU length of stay (LOS). Results: There were 3,991 and 3,841 ICU admissions during the pre-implementation and post-implementation periods, respectively. ICU mortality decreased from 17.82% (711/3991) to 15.05% (578/3841) (P = 0.001). Mortality in the first 48 hours of ICU admission decreased from 7.6% (302/3991) to 5.5% (210/3841) (P=0.0002). ICU utilization decreased from 8.63% (3847/44599) to 8.03% (3605/44904) (P = 0.00001). ICU LOS decreased from 4.13 days to 3.91 days (Rate ratio = 0.947; p < 0.0001). During the post-implementation period, there were 262 ASU admissions with 109 (40.8%) from the ICU, 44 (16.7%) from the emergency department, 107 (41.6%) from the medical/surgical units, and 2 (<1%) from other areas. ASU admissions resulted in 137 (52%) deaths in the ASU, 63 (24%) transfers to the general medical/surgical units, 42 (16%) transfers to the palliative/supportive care unit,15 (6%) direct discharges (13 to home hospice), and only 5 (2%) requiring escalation to the ICU. Conclusions: Implementation of a novel ASU was associated with significant reductions in ICU mortality. ICU utilization outcomes were consistent with our program goal. Further studies are needed to evaluate long-term patient-centered outcomes and generalizability to other institutions.