Three Ways Into the Same Coding Engine: Progress Note, Voice, and Worklist

Every practice we talk to has the same worry about adding an AI layer to documentation and coding: whose workflow has to change? The physician who has no time between patients, the clinician who would rather talk than type, or the coding team that reviews encounters in batches at the end of the day?

Our answer is that none of them should have to move. Medical Copilot runs one engine, the same chart reading, gap detection, compliant queries, diagnosis and service coding, and write-back, and offers three ways in. Everything runs the same no matter how the data comes in.

Way one: from the Progress Note

The first way in is the Progress Note itself. The physician clicks the app shortcut and Medical Copilot runs right there, without ever leaving the chart. In around ten seconds it reads the notes, labs, orders, and services already in the record, runs everything in parallel, and gets the encounter ready to finalize: queries if the documentation has gaps, suggested diagnoses and service codes with their evidence if it does not.

This is live inside eClinicalWorks today, and athenahealth is coming very soon. There is no separate system to log into and no copy and paste; responses and confirmed codes are written back to the chart. For the physician, the entire experience is one click and, at most, a couple of one-question answers.

Way two: by voice

Voice is another way in. The clinician talks, Medical Copilot captures the encounter, and the conversation drops straight into that same engine. Clinicians stay hands-free, and the documentation is complete with a full Progress Note.

The important part is what happens after the capture. Ambient documentation on its own produces a note; ours produces a note that immediately flows through the same CDI and coding pass as every other encounter. If the spoken encounter leaves the CKD stage unstated or the heart failure type unspecified, the same single clarification question comes up, and the answer lands in the note the same way.

Way three: through the worklist

The other way in is the backend worklist. Once encounters are completed, Medical Copilot pulls them regularly from the EHR into a ready queue. Physicians can review in the time between patients or at the end of the day, and the results write back to the chart.

The worklist is also where the coding team works. Coders monitor and confirm the codes on behalf of the physicians: they confirm the clear cases themselves and refer only the queries, the questions that genuinely need clinical judgment, to the doctors. That division of labor is the point. Physicians see only what requires a physician, and everything else moves without them.

One engine, one standard

Because all three paths feed the same engine, the practice gets one standard of output instead of three:

  • The same coding rules and guidelines apply whether the encounter arrived by click, by voice, or by queue.
  • The same evidence discipline holds everywhere: every suggestion carries its source text, the guideline it applied, and its HCC mapping where relevant.
  • The same write-back keeps the chart authoritative: query answers and confirmed codes land in the Progress Note, ready for any insurance audit.
  • The same audit trail records who confirmed what, on every path.

The takeaway

You should not have to redesign your clinic around an AI tool. Medical Copilot meets each role where it already works: the physician in the Progress Note, the clinician who prefers to speak, and the coding team in a worklist, and runs every encounter through the same engine so the documentation, the diagnoses, and the codes prove the care that was delivered.

To see it on your own encounters, book a demo.