Comparing Optimized Models of Primary And Specialist Services for Palliative Care (COMPASS-PC)

Palliative care (PC) seeks to reduce suffering and improve quality of life for patients with serious illnesses and their families. National guidelines recommend that clinicians either provide palliative care themselves (generalist PC) or consult experts (specialist PC) as a standard part of serious illness care. This pragmatic clinical trial will be conducted with 48 hospitals at two large U.S. health systems and enroll more than 78,000 seriously ill hospitalized patients. Eligibility is determined by a mortality prediction score where enrolled patients have at least a 70% risk of dying within 1 year. Enrollment assessment occurs as close as possible to 36 hours post admission. The 48 hospitals will be randomized to 3 arms: (1) standardized usual care, (2) trained generalist PC, or (3) specialist PC. Generalist clinicians are trained using the Center to Advance Palliative Care (CAPC) online trainings. This pragmatic, hybrid effectiveness-implementation parallel-cluster RCT will assess the comparative effectiveness of triggering generalist PC and specialist PC on several patient-centered outcome measures, and follows a pilot feasibility study. We will collect Patient-Reported Outcomes (PROs) surveys from a random subset of enrolled patients.

Study Overview

• Status: Enrolling by invitation
• Conditions: Seriously Ill Hospitalized Patients
• Intervention / Treatment: Behavioral: Default Order; Behavioral: Accountable Justification; Behavioral: Standardized Usual Care

• Study Type: Interventional
• Enrollment (Estimated): 78302
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • California
    • Pasadena, California, United States, 91101
      • Kaiser Permanente Southern California
  • Michigan
    • Livonia, Michigan, United States, 48152
      • Trinity Health

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Age 18 years of age or older; AND Predicted 1-year mortality risk of 70% or greater; AND Admitted to a study hospital.

Exclusion Criteria:

• Patients who die or have an active or completed discharge order prior to enrollment time OR
• Readmission within 182 days of an eligible encounter OR
• Ineligible service line, with current admission status labeled as: hospice, acute rehabilitation, skilled nursing facility, long-term acute care, psychiatry, obstetrics

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Health Services Research
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

3

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Standardized Usual Care; Active control group, where high-risk patients (i.e., with a 1-year mortality risk between 70% and 94%) will receive usual care. A specialist PC consult is ordered by default for the very high-risk patients (i.e., 1-year mortality risk ≥ 95%), unless clinicians cancel the order.	Behavioral: Standardized Usual Care; High-risk patients (i.e., with a 1-year mortality risk between 70% and 94%) will receive usual care. For very high-risk patients (i.e., with a 1-year mortality risk of ≥ 95%), an EHR-based Our Practice Advisory (OPA) alert on Open Chart informs clinicians when the default order will become active, and how to cancel an order within 24 hours if they elect to do so.
Experimental: Trained Generalist Palliative Care; Generalist clinicians trained in PC domains receive an EHR-based alert to document whether or not they have addressed PC domains for high-risk patients ('accountable justification intervention'). A specialist PC consult is ordered by default for the very high risk patients (i.e., 1-year mortality risk ≥ 95%) unless clinicians cancel the order.	Behavioral: Default Order; A specialist PC consult is automatically ordered for patients meeting a certain threshold of 1-year mortality risk (dependent on arm). An EHR-based Our Practice Advisory (OPA) alert on Open Chart informs clinicians when the default order will become active, and how to cancel an order within 24 hours if they elect to do so.; Behavioral: Accountable Justification; An EHR-based Our Practice Advisory alert asks generalist clinicians to self-report whether they have provided primary PC by clicking which of 4 key PC domains they have addressed or to provide a brief justification as to why not.
Experimental: Specialist Palliative Care; A specialist PC consult is ordered by default for all patients with a ≥ 70% 1-year mortality risk ('default order intervention'), unless clinicians cancel the order.	Behavioral: Default Order; A specialist PC consult is automatically ordered for patients meeting a certain threshold of 1-year mortality risk (dependent on arm). An EHR-based Our Practice Advisory (OPA) alert on Open Chart informs clinicians when the default order will become active, and how to cancel an order within 24 hours if they elect to do so.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Hospital free days	Count of days from enrollment spent alive outside of an acute care hospital through 182 days.	Enrollment- 182 days.

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Patient quality of life	Patient or caregiver report of the patient's quality of life using the 15-item McGill Quality of Life survey instrument.	1-month, 3-months, and 6-months post-enrollment
Clinician communication	Patient or caregiver report of how much patients feel heard and understood, using the 4-item CMS-MACRA survey instrument	1-month post-enrollment
Pain management	Patient or caregiver report of whether patients receive desired help for pain, using the 3-item CMS-MACRA survey instrument.	1-month post-enrollment
Social interaction	Patient or caregiver report of patient's social support using the 4-item Duke Social Support Index's Social Interaction Sub-scale	1-month post-enrollment
Loneliness	Patient or caregiver report of patient's loneliness using the 3-item Duke UCLA loneliness scale.	1-month post-enrollment
Hospital free days at 3 months	Count of days from enrollment spent alive and not in an acute care hospital through 3 months.	Enrollment - 3-months post-enrollment
Institution free days at 3 months	Count of days from enrollment spent alive and not in any care facility through 3 months.	Enrollment - 3-months post-enrollment
Institution free days at 6 months	Count of days from enrollment spent alive and not in any care facility through 6 months.	Enrollment - 6-months post-enrollment
30-day hospital readmissions	Count of readmissions	Enrollment - 30 days post-enrollment
90-day hospital readmissions	Count of readmissions.	Enrollment - 90 days post-enrollment
Community-based palliative care use	Binary variable indicating any use (use/no use) captured through EHR and claims data and count of days of PC visits.	Enrollment - 6-months post-enrollment
Hospice use	Binary variable indicating any use (use/no use) captured through EHR and claims data and count of days of hospice.	Enrollment - 6-months post-enrollment
Change in code status	Binary indicator of whether code status changed from enrollment.	Enrollment- Discharge from hospital (an average of 7 days)
Goal concordant care	Patient or caregiver report of the patient's perception of whether their treatment matched what they wanted.	1-month post-enrollment
Time to palliative care consult	Time from enrollment until receipt of the first documented inpatient palliative care consultation note.	Enrollment- Discharge from hospital (an average of 7 days)

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
End-of-Life Care	Caregiver report of the patient's end-of-life experience using the 14-item Family Assessment of Treatment at the End of Life Short Form (FATE-S) survey instrument and two additional questions regarding place of death and whether it was their preferred location.	3 months post-patient death

Collaborators and Investigators

• Sponsor: University of Pennsylvania
• Collaborators: Duke Clinical Research Institute
• Investigators: Principal Investigator: Scott Halpern, MD, PhD, University of Pennsylvania; Principal Investigator: Katherine Courtright, MD, MS, University of Pennsylvania

Study record dates

Study Major Dates

• Study Start (Actual): February 16, 2026
• Primary Completion (Estimated): November 1, 2029
• Study Completion (Estimated): July 31, 2030

Study Registration Dates

• First Submitted: October 24, 2025
• First Submitted That Met QC Criteria: October 31, 2025
• First Posted (Actual): November 4, 2025

Study Record Updates

• Last Update Posted (Actual): March 12, 2026
• Last Update Submitted That Met QC Criteria: March 10, 2026
• Last Verified: January 1, 2026

More Information

Terms related to this study

• Keywords: Palliative
• Other Study ID Numbers: 855378; PLACER-2022C3-30553 (Other Grant/Funding Number: Patient-Centered Outcomes Research Institute (PCORI))

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: De-identified data (the Analyzable Data Set, Full Protocol, metadata, data dictionary, full statistical analysis plan (including all amendments and all documentation for additional work processes)), and analytic code from a PCORI-funded research project from this PCORI-funded clinical trial will be made available on the Patient-Centered Outcomes Data Repository.
• IPD Sharing Time Frame: The IPD and supporting information will be available between 3 and 6 months after the study completion date, and will be available on the Patient Centered Outcomes Data Repository in line with PCORI's policies for data retention.
• IPD Sharing Access Criteria: Applicants will follow the guidelines as set out on the Patient Centered Outcomes Data Repository for accessing this data set.
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP; ANALYTIC_CODE

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No