ABSTRACT
Importance: Patients often visit the emergency department (ED) near the end of life. Their common disposition is inpatient hospital admission, which can result in a delayed transition to hospice care and, ultimately, an inpatient hospital death that may be misaligned with their goals of care. Objective: To assess the association of hospice use with a novel multidisciplinary hospice program to rapidly identify and enroll eligible patients presenting to the ED near end of life. Design, Setting, and Participants: This pre-post quality improvement study of a novel, multifaceted care transitions program involving a formalized pathway with email alerts, clinician training, hospice vendor expansion, metric creation, and data tracking was conducted at a large, urban tertiary care academic medical center affiliated with a comprehensive cancer center among adult patients presenting to the ED near the end of life. The control period before program launch was from September 1, 2018, to January 31, 2020, and the intervention period after program launch was from August 1, 2021, to December 31, 2022. Main Outcome and Measures: The primary outcome was a transition to hospice without hospital admission and/or hospice admission within 96 hours of the ED visit. Secondary outcomes included length of stay and in-hospital mortality. Results: This study included 270 patients (median age, 74.0 years [IQR, 62.0-85.0 years]; 133 of 270 women [49.3%]) in the control period, and 388 patients (median age, 73.0 years [IQR, 60.0-84.0 years]; 208 of 388 women [53.6%]) in the intervention period, identified as eligible for hospice transition within 96 hours of ED arrival. In the control period, 61 patients (22.6%) achieved the primary outcome compared with 210 patients (54.1%) in the intervention period (P < .001). The intervention was associated with the primary outcome after adjustment for age, race and ethnicity, primary payer, Charlson Comorbidity Index, and presence of a Medical Order for Life-Sustaining Treatment (MOLST) (adjusted odds ratio, 5.02; 95% CI, 3.17-7.94). In addition, the presence of a MOLST was independently associated with hospice transition across all groups (adjusted odds ratio, 1.88; 95% CI, 1.18-2.99). There was no significant difference between the control and intervention periods in inpatient length of stay (median, 2.0 days [IQR, 1.1-3.0 days] vs 1.9 days [IQR, 1.1-3.0 days]; P = .84), but in-hospital mortality was lower in the intervention period (48.5% [188 of 388] vs 64.4% [174 of 270]; P < .001). Conclusions and Relevance: In this quality improvement study, a multidisciplinary program to facilitate ED patient transitions was associated with hospice use. Further investigation is needed to examine the generalizability and sustainability of the program.
