UC35: After-Hours Triage (Urgent Care)

The Problem

After-hours calls overwhelm on-call providers. A patient calls at 2 AM with a sore throat; the on-call physician must spend 15 minutes assessing symptoms and deciding whether the patient needs an ER visit or can wait for morning urgent care. Multiply this by 30 to 50 calls per night, and on-call providers are disrupted constantly. Many calls are non-urgent (routine cold symptoms, medication refills) and could be handled by triage protocols without waking a physician.

What the Agent Does

Receives after-hours patient call (phone or text-based chat); uses IVR or conversational agent to initiate triage
Asks symptom-focused questions based on chief complaint using Schmitt-Thompson triage protocols: red-flag symptoms (chest pain, difficulty breathing, severe bleeding) trigger immediate ER referral
Categorises into three pathways: (1) Emergent , go to ER immediately, (2) Urgent , schedule morning appointment or urgent care visit, (3) Self-Care , OTC remedies, call back if worse
For non-emergent calls, offers guidance: OTC medication recommendations, self-care instructions, hydration/rest advice
Documents call in EHR for continuity: provides summary to morning provider team
Tracks outcomes: call volume, triage accuracy (compare AI triage vs. physician triage for validation)

Data Requirements

Data Sources:

Schmitt-Thompson Triage Protocols , Evidence-based triage logic by chief complaint (sore throat, fever, abdominal pain, headache, etc.)
Patient Contact Info , Phone number, preferred language
Patient History Summary , Recent visits, allergies, medications (from EHR, provided at call initiation)
Urgent Care / ER Availability , Real-time appointment availability or ER wait times
On-Call Schedule , Physician on-call coverage, escalation procedures if patient insists on physician

Data Classification:

Triage protocols: Internal, Structured
Patient contact: Sensitive, Internal, Structured
Patient history: Sensitive, Internal, Structured
Appointment availability: Internal, Structured
On-call schedule: Internal, Structured

Data Quality Requirements:

Triage protocols: reviewed annually by medical leadership (align with Schmitt-Thompson updates)
Patient contact info: verified at registration; phone validated
Patient history: current as of last visit (medications, allergies accurate)
Appointment availability: real-time updates as slots fill
On-call schedule: current and confirmed by each on-call physician

Integration Complexity: Medium

IVR or conversational agent platform (Twilio IVR, Amazon Connect, or similar)
EHR patient summary retrieval API (Epic FHIR, Cerner REST)
Urgent care appointment system integration (if available; may default to manual scheduling)
Call recording and EHR documentation: call summary must be logged in patient chart
Escalation workflow: transfer to on-call physician if patient insists or if red flags detected

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%	5	0.75
Ease of Implementation	10%	3	0.30
Fallback Available	10%	3	0.30
Audience (Int/Ext)	10%	3	0.30
Composite	100%		3.55

Why It Scores Well

After-hours triage is a physician workload and access issue: reducing non-urgent calls by 50% saves significant on-call burden while improving access for truly urgent cases. The data is rule-based (Schmitt-Thompson protocols); triage logic is deterministic. Outcomes are measurable (call volume reduction, ER utilisation, patient safety).

Regulatory Alignment

Medical Practice Acts: State laws vary on scope of AI-driven triage; physician must have oversight; AI cannot independently advise emergency action
EMTALA: Telehealth-based triage must not inappropriately divert patients from ER (patients with true emergencies cannot be diverted)
Telehealth Licensure: Physician providing after-hours triage must be licensed in patient's state
Malpractice: Triage accuracy is critical; AI recommendations must align with standard protocols; liability insurance must cover AI-assisted triage

Sprint Factory Fit

Sprint 0 (2 weeks) + 3 build sprints (6 weeks)

After-hours triage is event-driven: triggered by patient call during off-hours. The initial 2-week sprint focuses on IVR implementation and Schmitt-Thompson protocol integration; subsequent sprints add conversational AI, EHR integration, and urgent care appointment routing. This is a medium-complexity use case suitable for clinical operations and telephony teams.

Comparable Implementations

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

Real-Time Clinical Operations

Bed Management "Traffic Controller"

Score: 4.15/5.0

Real-Time Clinical Operations

Supply Chain & Surgical Kit Inventory

Score: 3.7/5.0

Real-Time Clinical Operations

Intelligent Patient Triage (Urgent Care/ED)

Score: 3.6/5.0