Background: Palliative care (PC) is a multidisciplinary consult service aimed at ensuring goal-concordant medical care while optimizing patient/family support. Prior studies have illustrated that palliative care consultation can be associated with a significantly shorter length of stay but without significant change in mortality or disposition. Utilization of standard high-risk descriptive criteria or “triggers” that uniformly prompt palliative care involvement has been previously demonstrated as an effective approach to better integrate PC into the ICU.
Methods: We developed an ambidirectional study to evaluate the integration of trigger criteria. “Trigger criteria” were selected based on evidence from prior publications, including modified criteria from the Center to Advance Palliative Care. Our selected trigger criteria included NYHA Class 4, stage 4 malignancy, multiple organ failure (more than 3 systems), ICU admission greater than 7 days, and advanced dementia (FAST stage 6A or worse). In addition to developing a new order set to consult the PC service, we embedded a PC team member into daily multidisciplinary ICU rounds to improve real-time collaboration. We retrospectively analyzed ICU admission data from February 1st- August 1st, 2023. This cohort was compared to ICU admissions after initiating trigger criteria. To reduce selection bias we used the corresponding months of February 1st to August 1st, 2024. In addition to descriptive statistics, we compared palliative care consults, length of stay, and disposition using a paired T-test.
Results: 177 patients were admitted to the ICU February 1st to August 1st, 2023 prior initiation of trigger criteria. 239 patients were admitted to the ICU February 1st to August 1st, 2024 after utilization of trigger criteria (Table 1). The average age of patients was 68, and 56% were male. 28% of patients had a PC consult in the pretest, while 26% had a PC consult after trigger initiation. The average mortality was 32%. Average critical care length of stay prior was 4.2 days, and after initiation, 3.39 days; this was not statistically significant. Multiorgan failure was the major indication for PC consult, followed by ICU admission of more than 7 days (Image 1).
Conclusions: The initiation of trigger criteria did not significantly increase PC consults during the corresponding 6 month evaluation. Interestingly the ICU census increased by 25% and the PC consults percentage stayed consistent. Multiorgan failure and prolonged stay were the major triggers for consultation. We plan to complete a full-year review to determine if long-term change has occurred or if there are any other trends.
