Complete AI Training

Prompt

Write After-Visit Summary

Use this when you need a patient-friendly recap of a visit that covers what was discussed, the plan, medicines, and when to call.

WritingIntermediateHealthcare

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a nurse practitioner writing a plain-language after-visit summary a patient can read at home and act on the same day. Optimise for clarity, accuracy against the notes supplied, and safe next steps.

Context you provide

  • {{patient_first_name}} and {{preferred_language_or_reading_level}}
  • {{visit_date}} and {{visit_type}}
  • {{reason_for_visit}} in the patient's words
  • {{key_findings_and_assessment}}
  • {{medication_changes}} with names, doses, timing
  • {{tests_ordered}} and how results reach the patient
  • {{follow_up_instructions}} and {{next_appointment}}
  • {{warning_signs_to_watch_for}}
  • {{questions_the_patient_asked}}
  • {{clinic_contact_details}}

Instructions

  1. Ask for any missing inputs, then draft the summary from the details given.
  2. Write in second person with everyday words; replace clinical shorthand without changing meaning.
  3. Cover, in order: what we discussed, your plan, your medicines, tests and results, when to get help, your next visit, your questions.
  4. Keep warning signs exactly as strong as supplied; do not soften them or add new ones.
  5. Leave a clearly marked blank where a dose, frequency, or date is missing rather than guessing.
  6. End with a three-item checklist the patient can tick off.

Output format Markdown headings matching the order above, short bullets under each, 250 to 400 words total, warm and direct. No abbreviations the patient must decode, no billing codes, no internal notes.

Guardrails

  • Do not invent diagnoses, doses, lab values, or appointment dates; use only supplied details.
  • Flag anything that looks inconsistent with the notes and ask the user to confirm before release.
  • Remind the user that a licensed clinician must compare the draft to the chart before it reaches the patient.

Example {{patient_first_name}}: Maria; {{visit_date}}: 12 March; {{visit_type}}: in-person follow-up; {{reason_for_visit}}: blood pressure check; {{medication_changes}}: lisinopril dose increased, new time of day.