NewNew: The enterprise guide to Agentic AI — 24 min read.

Read →
Case study · Healthcare Providers · Ambient AI

72 minutes returned per clinician per day at an academic medical center — ambient AI scribe

Clinician documentation was the top-cited driver of burnout at an academic medical center. pronix.ai rolled out an ambient AI scribe integrated with Epic across 2,400 clinicians — 72 minutes returned per clinician per day and a 38% drop in reported burnout scores in six months.

Client
Academic medical center, 2,400 clinicians
Industry
Healthcare Providers
Platform
Epic · Azure OpenAI · Nuance DAX-class ambient scribe
By pronix.ai CX Engineering8 min readQ1 2026
72 min
Returned per clinician per day
-38%
Burnout score (Maslach)
+11%
E/M level-appropriate coding
83%
Six-month clinician adoption

*Representative outcome; results vary by client, scope and platform configuration.

The challenge

Clinicians spent 2 hours after clinic closing charts, and 41% self-reported burnout on the Maslach index. Prior scribe pilots stalled because notes weren't Epic-ready and coders had to rework 30% of encounters.

Our approach

Step 01

Specialty-tuned templates

Note templates tuned per specialty — primary care, cardiology, orthopedics — matched to E/M coding rules and Epic SmartText slots.

Step 02

Epic-native ingestion

Notes landed in the encounter draft with structured problem-list and med-rec updates — clinician signed, didn't retyped.

Step 03

Coder-in-the-loop for E/M

Coders reviewed the top 5% highest-value encounters; feedback re-trained the E/M classifier monthly.

Step 04

Consent and PHI boundary

Explicit patient consent captured in the workflow; ambient audio never left the session — transcript only.

Step 05

Adoption as a change program

Peer champions in every service line, weekly office hours, and clinician-visible metrics on time saved.

For the first time in a decade, I finish clinic when clinic ends.

Chief of Primary Care
For CMIOFor CMOFor CIO

Illustrative case study. Scenarios, metrics, quotes and client details are representative composites based on Pronix engagements and industry benchmarks unless a named client is shown with written consent. Outcomes vary by client, scope, data quality and platform configuration. Nothing on this page is a guarantee, warranty or professional advice. See our Terms of Use for the full disclaimer.

Trademarks.

AWS, Salesforce, Microsoft, NICE, Genesys, Kore.ai, Google, ServiceNow, Amazon Connect and logos referenced on this site are trademarks of their respective owners. References are for descriptive purposes only and do not imply endorsement, sponsorship or partnership beyond stated partner relationships. See our Disclosures.

Bring this to your program

Talk to the team that shipped it.

Book a working session with a pronix.ai lead on this practice — we'll map the approach to your platform, industry and constraints.