← The Health AI Toolkit

Review a patient message before updating the record

Turn a patient message into a draft note you can check. Keep the original message and approve any change to the record yourself.

Starter · About 15 min to try; setup approval may take longer.

Bring: A fictional message and record excerpt; Maya’s S4 and S5 work for this exercise.

Make: A proposed change beside the original record.

Suggested starting tools: ChatGPT for Clinicians, Claude (personal / Team). Use fictional inputs for this exercise; check the exact account and data path before real records.

Jump to the copyable prompt ↓

Guide sources and teaching inputs checked . Vendor directory claims retain their own review dates.

Draft a change instead of silently replacing a fact

Give the assistant a fictional message and the relevant fictional record excerpt. Ask for the source and date, a concise patient-reported summary, what conflicts with the existing record and what requires confirmation. Show the proposed change beside the existing text.

For Maya, “I stopped the cholesterol pill” is a patient report. It should trigger reconciliation rather than silently removing a medication or recording a confirmed drug adverse effect. Preserve the original message and the uncertainty.

Keep approval at the clinical boundary

A clinician confirms the facts and decides which change belongs in the record. A staff member may prepare the draft within their role; the AI does not choose treatment, approve a medication change or send a patient reply on its own.

For a practical pilot, begin with drafts that a person manually reviews and enters in the approved clinical system. This keeps the review step explicit while you learn whether the draft helps.

An integration needs its own verification

Before enabling automated record writes, test the correct patient and encounter, access controls, duplicate messages, failed writes, retry behavior and the audit history. Confirm the saved result, not just a success message. Make a failed or overdue review visible to a named owner.

The presentation illustrates this workflow; it does not demonstrate a deployed clinical integration. Permission to read a message is separate from permission to edit a chart or send a reply.

Copy a prompt to try

Proposed update from a fictional message

Documentation

Use this fictional message: {{fictional_message}}.
Compare it with this fictional record excerpt: {{fictional_record}}.
Return the source/date, patient-reported summary, conflicts, missing information
and proposed record change beside the original text. Preserve uncertainty.
Do not overwrite the record, approve a medication change or send a reply.
Finish with the questions a clinician must confirm before approving an update.

Practice with fictional inputs. Real messages need an approved data path and clinician review; connecting an EHR requires separate testing and authorization.

Example of a reviewable result

This is an authored teaching example, not a captured AI output or a clinical plan.

Existing record: Atorvastatin 10 mg daily. [S4]
Patient report: Stopped the cholesterol pill; has not spoken with the prescriber. [S5]
Proposed draft: Patient reports stopping the medication; medication reconciliation needed.
Requires review: Confirm the medication, current use and prescriber plan. Do not remove a medication or record a confirmed adverse effect from this message alone.

Check before you use the result

  • Keep the original message and its source.
  • Show the proposed change beside the current text.
  • Keep patient reports separate from verified changes.
  • Confirm the clinician’s decision before saving a record change or sending a reply.
Next: Set up a weekly research brief →Explore the practical workflows →