UC53: Palliative Care Identification

The Problem

Patients who would benefit from palliative care (symptom relief, goals-of-care conversations, advance planning) are identified too late, often only in the final days of life. Early palliative care integration improves symptom management, patient satisfaction, and quality of life. Systematic screening identifies eligible patients for timely consultation.

What the Agent Does

Batch runs weekly: queries inpatient population
Agent identifies patients meeting palliative care criteria: advanced cancer, Stage 4 disease, >2 hospitalizations in 30 days, functional decline (ADL dependence), documented goals-of-care conversations absent, frailty indicators
Calculates palliative care readiness score: triggers consult if score exceeds threshold
Generates referral list for palliative care team, prioritised by urgency
Provides clinical context: "72M with metastatic lung cancer, 3 ED visits in 30 days, no goals-of-care documented; recommend PC consult"
Tracks consultation outcomes: was consult offered, accepted, completed; documented goals of care
Generates population health report: % eligible patients receiving PC; goals-of-care completion rate

Data Requirements

Primary Data Sources: Problem list (cancer diagnosis, functional status), hospitalization history (frequency, LOS), ICD codes (advanced disease), advance directive status, medications (opioids, symptom management), functional decline indicators (mobility, ADL dependence)
Classification: PHI/PII (diagnosis, functional status, advance directives)
Quality Metrics: Palliative care identification sensitivity >85%; specificity >70%; clinician satisfaction with referrals >4.0/5
Complexity: Medium (EHR data aggregation, palliative care criteria logic)
Systems: Epic EHR, palliative care consult module

Score Breakdown

Criterion	Weight	Score (1-5)	Weighted
Time Recaptured	15%	5	0.75
Error Reduction	10%	3	0.30
Cost Avoidance	10%	3	0.30
Strategic Leverage	5%	3	0.15
Data Availability	15%	4	0.60
Process Clarity	15%	4	0.60
Ease of Implementation	10%	3	0.30
Fallback Available	10%	3	0.30
Audience (Int/Ext)	10%	4	0.40
Composite	100%		3.50

Why It Scores Well

Patient Centered Care: Early PC integration improves symptom management and quality of life
Evidence Base: Studies show improved outcomes with early PC (better symptom control, goals-of-care alignment, family satisfaction)
Operational Benefit: PC team triaged to highest-need patients
Scalability: Applies to entire inpatient population
Regulatory Fit: No novel autonomy concerns; consult is recommendation, not autonomous decision

Regulatory Alignment

Joint Commission: Goals-of-care conversations and advance planning are care standards
CMS: Advance Care Planning (CPT 99497 to 99499) is billable service; incentivizes uptake
State Advance Directive Laws: Vary by state; agent must respect state-specific requirements
HIPAA: Advance directive status is PHI; standard access controls apply
POLST/MOLST Programs: Portable orders for life-sustaining treatment; agent may trigger POLST completion

Sprint Factory Fit

2 sprints (EHR data aggregation, palliative care criteria logic, consult workflow, outcomes tracking)

Sprints Required: 2 sprints (EHR data aggregation, palliative care criteria logic, consult workflow, outcomes tracking)
Timeline: 3 to 4 weeks from requirements to pilot
Governance: Palliative care medical director approval of identification criteria; monitor referral acceptance rate and outcomes
Success Metrics: Identification sensitivity >85%; referral acceptance rate >70%; goals-of-care documentation increase >20%; patient satisfaction with PC >4.2/5

Comparable Implementations

Palliative Care referral tools (health system-specific)
CMS Advance Care Planning: Billing and Coding
Joint Commission PC Standards

Deploy This Use Case with the Sprint Factory

From zero to a governed, production agent in 6 weeks.

Sprint Factory Schedule a Briefing

Related Use Cases

Batch Clinical

Organ Transplant Waitlist Monitor

Score: 3.35/5.0

On-Demand Clinical

AI Clinical Scribe & Note Generator

Score: 4.3/5.0

Real-Time Clinical

Sepsis Early Warning Agent

Score: 3.85/5.0