Patient Access Agent

Books, reschedules and confirms appointments directly in the EHR. The agent verifies the patient, reasons over provider availability, visit types and scheduling rules, and writes the booking straight into the EHR — including reschedules, cancellations and confirmations. It runs proactive confirmation and waitlist outreach, so open slots are refilled instead of lost. The moment a clinical question or a symptom appears, the call goes to patient-access staff with the full context.

It works across EHR scheduling, Patient identity, CCaaS, Referral queue. The named integration surface is EHR scheduling (Epic, Cerner or equivalent); Patient identity and verification; CCaaS voice and digital channels; Referral and waitlist queues.

We deliver it on the platforms you already own — Amazon Connect, Amazon Lex, Kore.ai. Pronix.ai is a systems integrator, not a software vendor, so the code, prompts and evaluation sets belong to you.

Autonomous autonomy. No clinical advice, triage or symptom interpretation — those intents escalate on detection. Scheduling rules, visit types and provider constraints are enforced from the EHR, not learned by the model. Every action outside policy stops at a reviewer queue with the agent's reasoning, evidence and proposed change attached. Identity verification must pass before any record is read or written.

Scheduling API access to the EHR with a non-production environment for build. Documented scheduling rules per specialty and visit type. An agreed identity-verification standard for phone and digital.

HIPAA-aligned handling with a signed BAA and PHI confined to your tenancy and region. Customer and employee data stays inside your tenancy and region; no training on your data by default. PII is redacted before it reaches a model, and prompts, responses and tool calls are retained under your retention policy. Every tool call, input, decision and system write is logged and replayable for audit and model-risk review.

Delivery runs in 4 phases. Weeks 1–2 · Scope: Specialty and visit-type selection, scheduling-rule capture, identity standard agreed. Quarter 2+ · Scale & run: Specialty-by-specialty expansion with waitlist and proactive confirmation outreach.

50–75% of scheduling demand handled without staff. Expected ranges from comparable production engagements: Scheduling demand resolved without staff 50–75%; Call abandonment at peak 30–50% lower; Slot utilisation from waitlist refill 3–8 points higher; Booking accuracy against EHR rules 98%+ sustained. Your baseline is agreed before build starts and reported on the same measures after go-live.

Production pilot: One service line booking, rescheduling and confirming live in the EHR, measured against the access baseline. Scale and run: Rolled out across specialties with waitlist refill, reminders and multilingual coverage under an access SLA. Fixed-price pilot with milestone billing, then a retained pod with quarterly outcome reviews.

Price on request. Every agent is scoped to your integration surface, data readiness and compliance posture, so we send a written estimate after a 30-minute scoping call rather than publish a band that would not apply to you.