Temporal Impact of Hospice and Palliative Medicine Consults on End-of-Life Outcomes in Emergency Department and Hospitalized Patients

Document Type

Conference Proceeding

Publication Date

10-1-2024

Publication Title

Ann Emerg Med

Abstract

Study Objectives: In recent years, there has been growing recognition of the benefits associated with early engagement of hospice and palliative medicine (HPM) resources for patients nearing the end-of-life. Early access to primary and specialized palliative care, notably in the emergency department (ED) and inpatient settings, facilitates essential goals of care conversations, updates patients' code status preferences, and explores comfort care options while continuing disease targeted therapies. Despite the expanding evidence base supporting early HPM interventions, questions persist regarding the optimal timing and clinical setting of such consultations. This retrospective cohort study aims to address this gap by examining the outcomes associated with different timing intervals for HPM consultations, whether initiated in the ED, within the first 48 hours of an inpatient stay, or after 48 hours of hospitalization. Methods: We conducted a multicenter retrospective cohort study using electronic health records from five hospital based EDs within a large urban and suburban metropolitan health system. The study period ranged from January 1, 2018, to December 31, 2022, and included patients aged >18 years who had HPM consults ordered during ED or inpatient encounters. Patients were categorized into three cohorts: those who had HPM consults ordered in the ED, within the first 48 hours of admission (early), and after 48 hours of hospitalization (late). Patient data collected included demographics, inpatient hospital length of stay (LOS), ICU LOS, inpatient mortality, and final hospital dispositions. In cases where patients received multiple HPM consults per encounter, cohort assignment was determined based on the timing of their earliest HPM consult order. The three cohorts underwent an analysis of variance (ANOVA) to assess baseline and outcome differences among the groups. Descriptive statistics were employed to offer a synopsis of the characteristics and outcomes within each cohort. Results: The study analyzed 45,710 HPM consultations involving 25,609 unique patients across 31,072 encounters. Consultation distribution varied, with 6,220 initiated in the ED, 12,162 within 48 hours of hospitalization, and 12,690 after 48 hours of hospitalization. The mean age of the ED cohort was 77.7 years old (SD=13.88), statistically older than both the early (74.99, SD=14.86)) and late (74.36, SD=13.93) HPM consult groups (p < .001). The mean ED emergency severity index (ESI) was identical for all three groups at 2.12, p = 0.55. We observed significant associations between consult timing and various outcomes, including ICU length of stay, total hospital length of stay, and mortality rates in both ED and inpatient settings. For ICU length of stay, ED consults averaged 0.82 days, early inpatient consults 1.39 days, and late inpatient consults 4.99 days (p < .001). Similarly, for total hospital length of stay, ED consults averaged 4.75 days, early inpatient consults 5.69 days, and late inpatient consults 12.72 days (p < .001). Additionally, mortality rates varied across consult timings, with ED consults experiencing a mortality rate of 52.91% (n=3291), early inpatient consults 61.97% (n=7537), and late inpatient consults 69.61% (n=8833) (p < .001). Graphical summary of these comparisons is displayed in Figure 1. Conclusion: Our findings provide valuable preliminary insights into the temporal dynamics of HPM consultations in end-of-life hospital care. Early consultations, especially those initiated in the ED, were linked to shorter ICU and total hospital length of stay, as well as lower mortality rates. To advance these findings into practice, further efforts are needed to enhance primary palliative care skills among clinical teams and prioritize initiatives that enable early HPM consults in both the ED and inpatient setting. [Formula presented] No, authors do not have interests to disclose

Volume

84

Issue

4

First Page

S87

Last Page

S88

Find It @ Sladen

Share

COinS