RPD117 - Screening Tool Identifies Patient Characteristics and Outcomes Linked to High Palliative Care Needs

Key Message : Analysis of a screening tool describes patient characteristics/outcomes associated with high palliative care (PC) need scores, including Black race, longer hospitalizations, and higher rates of PC consultation and in-hospital death. Results support the tool’s ability to identify patients with PC needs and potential to expand equitable access to PC services.

Abstract: Background/Rationale:
Proactively screening inpatients for palliative care (PC) needs can target interventions and promote equitable access. We evaluated patient characteristics and outcomes associated with electronic health record (EHR)-based PC needs assessment.

Objectives/

Purpose:
To describe patient characteristics and outcomes associated with PC need scores generated by the Pro-PC screening tool (1,2,3).

Methods/Approach:
We applied the Pro-PC tool, which includes domains for symptom, psychosocial, spiritual, and advance care planning needs, to the EHRs of hospitalized patients with serious illness in an academic medical center from July 1, 2023-June 30, 2024. We divided peak PC need score per admission into three categories (low quartile, medium two quartiles, high quartile) and assessed associations between low and high categories and demographics, diagnoses, and outcomes.

Results:
Of 29,426 records of seriously-ill inpatients, those with high PC need scores were more likely to have lung disease (23.9%) and/or cancer (22.8%). Black patients were most likely to have high need scores (36% vs 25%; p< 0.001). There were no clinically meaningful differences in patient age/sex.

Patients with high scores had longer hospital stays (high: 40 days, low: 6.6 days; p< 0.001) and were more likely to receive a PC consultation (high: 16.9%, low: 0.7%; p< 0.001). They were more likely to have a discharge code status of DNR/DNI (high: 18.6%, low: 7.5%; p< 0.001), to be referred to hospice (high: 4.7%, low: 0.7%; p< 0.001), and to die in the hospital (high: 8.9%, low: 0.7%; p< 0.001).

Conclusion/Implications:
Our findings show that Pro-PC need scores identify sicker inpatients with greater PC needs. Among the highest scoring patients, only 16.9% received PC consults, suggesting opportunities to improve care through increased consultation or other need-based interventions. These results can be used to advocate for additional resources for PC teams and to develop and test interventions targeted to specific PC needs.

References: 1. Schoenherr L, Goto Y, Sharpless J, O'Riordan D, Pantilat S. A Validated Electronic Medical Record-Based Algorithm to Identify Hospitalized Patients with Serious Illness. J Palliat Med. 2025 Feb; 28(2): 201-206. PMID: 39607739.

2. Sharpless J, Goto Y, O'Riordan D, Pantilat S, Schoenherr L. A Validated EMR-Based Algorithm to Detect and Quantify Palliative Care Needs in Seriously Ill Hospitalized Patients (Abstract Only). J Pain Symptom Manage. 2024 May; 67(5): e628-e629.

3. Sharpless J, Goto Y, Schoenherr L, O'Riordan D, Pantilat S. A Survey to Characterize the Palliative Care Needs of All Hospitalized Patients in an Academic Medical Center (Abstract Only). J Pain Symptom Manage. 2024 May; 67(5): e670-e671.