Drafting Patient Education for Newly Diagnosed Type 2 Diabetes

The Problem

A patient leaves your clinic with a new diagnosis of type 2 diabetes. They have questions. Their family has questions. You have given them your best ten-minute explanation, but research consistently shows patients retain very little of what they hear in clinic when they are anxious or overwhelmed.

You want to send them home with something written. Something clear, not frightening, and actually useful. But writing that from scratch — or adapting a generic hospital leaflet — takes time you rarely have between patients.

How AI Helps

An AI writing assistant can produce a first draft of a patient education document in under a minute. You describe what you want: the audience, the key topics, the tone. The AI writes it. You read it, adjust it, remove anything that does not match your clinical approach, and add your name or clinic details.

The AI is not generating medical advice. It is generating words. You are the clinician reviewing those words. Think of it like having a very fast, tireless typist who has read widely about diabetes education.

You remain in full control. If the draft mentions something you do not want included — a specific monitoring frequency, a food list that does not match your local dietitian's guidance — you simply delete or rewrite it.

A Real Example

Dr Nair sees a 54-year-old patient, "Mr P", who has just been diagnosed with type 2 diabetes. He is a Tamil-speaking patient who also speaks English, and he runs a small catering business — so his diet is both complex and central to his life.

Dr Nair types the following into an AI tool:

"Write a plain-English patient education summary for a 54-year-old man newly diagnosed with type 2 diabetes. He works in catering and has a South Asian diet. Cover: what type 2 diabetes is, why blood sugar matters, three key lifestyle changes, and what to expect at his next appointment. Aim for a reading age of around 12. Use short paragraphs."

The AI produces a two-page draft in about thirty seconds. Dr Nair reads it, adjusts the section on carbohydrate portions to reflect what the clinic dietitian actually advises, removes a sentence she finds too alarming, and adds a line about the next HbA1c (glycated haemoglobin) check. She saves it as a template.

Next time a similar patient comes through, she uses the same template with small edits. She has saved roughly fifteen minutes per patient.

Try It Yourself

Copy and paste this prompt into ChatGPT, Claude, or a similar AI tool. Replace the items in brackets with your own details. Do not include any real patient names, dates of birth, or NHS or hospital numbers.

Write a patient education summary for a {{patient_age}}-year-old patient newly diagnosed with type 2 diabetes. Their main dietary pattern is {{dietary_pattern}}. Cover the following topics in plain English, using short paragraphs and no jargon:
1. What type 2 diabetes is and why it develops
2. Why keeping blood sugar stable matters
3. Three practical lifestyle changes (diet, activity, sleep)
4. What the patient can expect at their next review

Keep the reading level simple. Do not include specific drug names or doses.

Things to Watch For

Do not paste identifiable patient information. This includes names, dates of birth, addresses, hospital numbers, or any detail that could identify a real person. Most AI tools send your text to external servers. Use placeholders like "a 54-year-old patient" instead.

Review every draft carefully. AI tools can produce plausible-sounding but inaccurate statements. Check all clinical content before sharing anything with patients.

Your local guidelines take precedence. If the AI draft mentions monitoring targets or dietary advice that differs from what your clinic or trust recommends, update accordingly.

Plain language is not the same as no nuance. Sometimes simplifying a concept too far changes its meaning. Read the draft as if you are the patient.

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