AI Healthcare Agent for an Urgent Care Network: 40% Faster Patient Intake

Challenge

A 12-location urgent care network serving a mid-Atlantic metro area processed an average of 3,200 patient visits per week. The intake process—registration, insurance verification, medical history review, and initial triage—averaged 18 minutes per patient, with insurance verification alone accounting for 8 of those minutes. During peak hours (Monday mornings, weekend afternoons, and flu season surges), lobby wait times exceeded 45 minutes and walk-away rates hit 14%. Front desk staff at each location juggled intake paperwork, phone calls, insurance calls, and patient questions simultaneously. The network had invested in a patient portal for pre-registration, but adoption was only 22% because the portal required patients to create an account, remember credentials, and manually enter insurance details—steps most people skipped when they were feeling sick and just wanted to walk in. Insurance verification was the single largest bottleneck: staff called payers or used a legacy eligibility tool that returned errors 30% of the time, forcing manual follow-up. Incorrect eligibility information led to $420K in annual claim denials that required rework.

Solution

The network deployed an AI healthcare agent across the intake workflow, integrated with their athenahealth EHR and a multi-payer eligibility clearinghouse.

Digital pre-registration. Instead of requiring portal accounts, the agent sent patients a text-based intake link when they checked in online or called ahead. The link opened a conversational interface—no app download, no account creation—that walked patients through demographics, insurance card photo capture (the agent extracted plan details via OCR), current symptoms, and medication list in under 4 minutes. For returning patients, the agent pre-populated known information and asked only for updates.

Real-time insurance eligibility verification. The agent ran eligibility checks through the clearinghouse the moment insurance information was captured—before the patient arrived or while they were still in the lobby. It verified coverage, co-pay amounts, deductible status, and prior authorization requirements, flagging issues (expired coverage, out-of-network plan) immediately so staff could address them before the visit rather than after.

Symptom-based triage prioritization. Based on the symptoms reported during pre-registration, the agent applied evidence-based triage protocols to assign an acuity score. Patients presenting with chest pain, difficulty breathing, or high-risk indicators were flagged for immediate attention, while lower-acuity visits (sore throat, minor sprain) were queued normally. This didn't replace clinical triage by nursing staff but gave the care team advance visibility into who was waiting and why.

Results

• Average intake time: Reduced from 18 minutes to 11 minutes—a 40% improvement
• Insurance verification time: From 8 minutes average to 45 seconds per patient
• Patient satisfaction scores: Up 28% on post-visit surveys, with "wait time" and "check-in experience" showing the largest gains
• Pre-registration adoption: Jumped from 22% (portal) to 67% (text-based AI intake)
• Claim denial rate from eligibility errors: Reduced by 62%, recovering an estimated $260K annually
• Front desk staff reallocation: Staff spent 40% less time on administrative intake tasks, redirecting that capacity to patient communication, care coordination, and in-person assistance
• Peak-hour walk-away rate: Decreased from 14% to 6%

Takeaway

The biggest insight was that the technology barrier to pre-registration adoption wasn't patient willingness—it was friction. When the network replaced a portal that required account creation, login credentials, and manual data entry with a text link and conversational interface, adoption tripled. The insurance verification speed improvement had a cascading effect: faster eligibility checks meant fewer surprises at checkout, fewer denied claims, and less time spent on post-visit billing corrections. For the clinical staff, the advance triage visibility was unexpectedly valuable—providers reviewed the symptom summary and acuity score before entering the room, shaving additional time off the encounter itself. For healthcare organizations evaluating similar workflows, see AI Healthcare Agent. To compare platforms and explore implementation paths, visit Solutions.