Skip to main content

Rating Your Entrustability (EPA) with the AI Assistant

Dotan Asselmann

You can upload your program's EPA (Entrustable Professional Activity) rubric to the AI Assistant and ask it to rate your entrustability for a case. Attach the rubric PDF in chat, point the assistant at the case, and ask it to apply the rubric's scales — for example, to your intraoperative EPA. This article explains how to do that, what the assistant weighs, and where its limits are.

What This Workflow Does

You attach your institution's EPA rubric as a PDF, and the assistant reads it and applies the rubric's scales to the surgical data it can access for the case. It returns a rating with a short rationale, so you can use it as a structured self-assessment against the scales your program already uses.

The assistant reads the text of the PDF you upload — roughly the first 50,000 characters — so a standard one- or two-EPA rubric is read in full. For a very long rubric packet, split it into smaller files.

Before You Start

If you want the rating to reflect only the portion of the case you performed, mark it first with Claim Your Section. Claiming records which sections you performed, so the assistant focuses on your attributed portions of the case rather than the whole recording. See Claim Your Section.

Step by Step

  1. Open the case and open the AI Assistant.

  2. In the message box, click Attach file and choose your EPA rubric PDF, or drag the file onto the message box. Wait for the upload to finish.

  3. Type your request and send. For example:

    Using the scales in the PDF for the intraoperative EPA, rate my entrustability for this case, taking into account the percentage of the case I was primarily operating.

  4. Read the assistant's rating and rationale, and ask follow-up questions — for instance, ask it to look at a specific moment in the video to justify a score.

What the Assistant Examines

When it rates against the rubric, the assistant works from the surgical data in your Theator account: the attributed sections you claimed, operative times compared to your personal and peer averages, AI annotations and the case timeline, and any specific video moment you point it to. It reports only what it can clearly see and will not guess at findings it cannot confirm.

What It Can't Do — Read This Before You Rely on a Rating

  • It does not watch the whole recording to auto-grade you. The assistant looks at the moments you point it to; it will not scan an entire case end to end on its own. Point it at the parts of the video that matter for the rubric.
  • It reads Theator data, not free-text hospital records. It cannot pull scores from dictated notes or documents that are not connected to Theator.
  • A chat rating is a self-assessment aid, not an official submission. The assistant does not write, submit, or file evaluations, and uploading a rubric does not change or publish anything. It is not a rating submitted to the ABS or any accreditation body. Review its output yourself before using it.

Building an EPA Portfolio

To keep the cases you have rated together, add them to a collection — for example, a collection you use for EPA portfolio documentation. Ask the assistant to add a case to a collection, or build one yourself. See Finding Technique Clips and Building a Learning Collection with the AI Assistant.

What's Next?

Comments

0 comments

Please sign in to leave a comment.