AI for Clinicians

Home / Gynaecology

Use caseBeginner5 min read

Writing Pre- and Post-Operative Patient Instructions

Using an AI tool to draft clear pre- and post-operative instructions for common gynaecological procedures, reducing patient anxiety and avoidable calls to the ward.

The problem

Patients who have clear written instructions before and after surgery have better outcomes and generate fewer avoidable contacts with your team. They know what to expect. They know what is normal. They know when to seek help.

Most gynaecology units have standard instruction sheets for common procedures. But they are often generic, outdated, or pitched at the wrong reading level. Patients ignore them or cannot follow them. Then they ring the ward at 11pm to ask whether it is normal to have light bleeding after a hysteroscopy.

Writing tailored instructions for individual procedures — or updating existing ones — is a task that falls to the bottom of the list.

How AI helps

An AI tool can draft a complete set of pre- and post-operative instructions for any procedure you name. You specify the procedure, the typical patient, the key information to include, and the reading level. The tool drafts. You check the clinical accuracy against your unit's protocols and adjust accordingly.

This takes far less time than writing from scratch — and less than editing an outdated existing sheet, if the draft is more accurate to begin with.

The tool is particularly good at producing instructions that are genuinely readable — short sentences, clear headings, numbered lists. It is not good at knowing your specific unit's protocols, your anaesthetic department's requirements, or your local post-operative care pathways. You supply those details.

A real example

Mr Jerome is a consultant gynaecologist who performs laparoscopic hysterectomy regularly. His unit's current patient instruction sheet for this procedure was written in 2019. It does not mention the same-day discharge pathway now used for most cases. Patients are arriving expecting an overnight stay and are anxious when they are told they are going home the same day.

He opens an AI tool and types:

Try it yourself
Write pre-operative and post-operative patient instructions for a laparoscopic 
hysterectomy performed as a day case (same-day discharge). The patient is an adult 
woman. Use plain English, short sentences, and numbered lists. Include sections for: 
what to bring to hospital, what to expect on the day, what to expect at home in the 
first week, when to seek medical advice, and who to contact. Do not include 
specific drug names or doses.

The tool produces a two-page draft. Jerome checks it against his unit's fasting guidelines (the tool had listed a general timeframe rather than his anaesthetic department's specific instruction) and corrects it. He adds the ward phone number. He reviews the "when to seek medical advice" section carefully — the tool had listed general warning signs, all clinically reasonable, but he adds one specific to his patient population. The updated sheet is ready in thirty minutes, replacing the 2019 version.

Try it yourself

Try it yourself
Write patient instructions for [[procedure — e.g. "outpatient hysteroscopy"]]. Use plain English and numbered lists. Include: how to prepare beforehand, what to expect during the procedure, what to expect in the [[timeframe — e.g. "24–48 hours"]] afterwards, normal symptoms versus symptoms that need medical attention, and when to contact the clinic. Do not include specific drug names or doses.

Things to watch for

The tool does not know your unit's protocols. Fasting times, pre-operative medications, discharge criteria, and follow-up pathways vary between institutions. Treat the draft as a framework you populate with your own unit's specifics.

It may be over-cautious or under-cautious. The tool tends toward inclusion — it would rather list too many warning symptoms than too few. That is usually the right instinct, but review for symptoms that may cause unnecessary alarm in your patient population.

Reading level requires checking. Ask the tool to write at a specific reading level. Even then, read the draft aloud to gauge whether a patient under post-operative stress could follow it. If in doubt, simplify further.

It cannot incorporate patient-specific factors. A patient with diabetes, anticoagulation, or significant anxiety needs instructions that differ from the standard template. The AI tool's draft is a starting point; you adapt it for the individual.

Remember: AI is a helpful assistant, not a clinician. You make the call.

Was this lesson helpful?

Related lessons