Using AI to Make Patient Education Clear, Kind, and Actionable

Clinicians work hard to explain complex information, but many patients still leave visits confused about what to do next. Studies show that many discharge instructions and education materials are written above the reading level of the average adult, which can increase the risk of errors and poor outcomes. Plain language and structured communication training can close this gap and support safer care. AI tools, when guided with thoughtful prompts, can help clinicians create clearer messages, practice empathy, and reach patients where they are.

This article walks through four practical prompt patterns you can use today: low‑literacy translation, empathetic bad‑news simulations, FAQ generation, and friendly SMS reminders.

1. Turning “Med‑Speak” into Plain Language

Many patient materials still rely on dense language and jargon, even though guidelines recommend using clear, concise, logically organized information that people can understand the first time they read or hear it. For post‑surgical instructions and discharge summaries, targeting about a fifth–sixth grade reading level can make a real difference, especially for patients with limited health literacy.

A simple way to use AI here is to treat it like a plain‑language editor. You paste in your existing instructions and give explicit rules about reading level, structure, and words to avoid.

Example prompt

“Rewrite these post‑surgical instructions for a gallbladder removal at a 5th‑grade reading level. Use short sentences and bullet points. Avoid medical jargon like ‘ambulation’ or ‘analgesic,’ and use everyday words instead (for example, ‘walk around a few times a day’ instead of ‘begin ambulation’). Here are the original instructions: [paste text].”

You are looking for output that:

• Uses concrete actions with clear timing, such as “walk every few hours” instead of “ambulate as tolerated.”

• Replaces terms like “incision,” “NSAID,” or “analgesic” with common alternatives and brief explanations.​

• Groups ideas under clear mini‑headings like “Pain Control,” “Eating and Drinking,” and “When to Call the Doctor,” which supports navigation and recall.

After you get a draft back, you can add a quick self‑check prompt: “Now summarize these instructions in three key points a patient must remember.” This highlights whether the core message is truly simple and actionable.

2. Practicing Empathy When Delivering Bad News

Breaking bad news is one of the most emotionally demanding tasks in medicine. Effective communication requires both verbal and nonverbal skills, and simulation is now widely used to help clinicians practice in a safe environment. Frameworks and workshops emphasize privacy, adequate time, clarity, honesty, and an empathetic attitude.

AI can function as a communication coach. Residents can paste in their “script” for delivering a new diagnosis, such as Type 2 Diabetes, and ask for targeted feedback on empathy, clarity, and next steps.

Example prompt

“I am a resident practicing how to tell a patient they have Type 2 Diabetes. Here is what I plan to say: [paste script]. You are an attending and communication coach. Please critique my delivery for empathy, clarity, and actionable next steps. Point out where I did well, suggest exact sentences to sound more supportive, and list 3–5 follow‑up questions I should ask to check understanding and feelings.”

High‑quality feedback will typically:

• Highlight strengths, such as non‑blaming language or clear explanation of the condition.

• Suggest specific phrasing to invite emotion (“How does this land with you?”) and check permission (“Is it okay if we talk about your test results now?”), which align with structured breaking‑bad‑news approaches.

• Emphasize small, achievable steps for the first days or weeks after diagnosis, so patients leave with a sense of direction rather than only fear.

You can integrate this into teaching sessions by having residents draft their script, run it through the AI coach, and then discuss the feedback in group debrief.

3. Turning Dense PDFs into Patient FAQs

Medication handouts and clinical monographs often overwhelm patients with detail, especially around dosing instructions and side effects. Yet, in practice, patients tend to ask a small, predictable set of questions: how to take the drug, what to expect, and what to do when things go wrong.

AI can help you convert a long PDF—for example, about metformin—into a short list of frequently asked questions written in plain language.

Example prompt

“Using the attached PDF about ‘Metformin,’ generate a list of 5 frequently asked questions patients have about common side effects, when and how to take each dose, and what to do if they miss a dose. Write each question in plain language so it fits on a clinic handout.”

The resulting FAQs might include:

• “What should I do if I forget to take my Metformin?”

• “Can Metformin upset my stomach, and how can I make this better?”

• “Is it okay to drink alcohol while I am taking Metformin?”

• “Should I take Metformin before or after I eat?”

• “Can Metformin make my blood sugar go too low?”

Once you have the questions, you can ask for one‑sentence answers and then refine them against your local protocols. This gives you a patient‑friendly FAQ that can live in print, on your portal, or inside automated chat flows.

4. Writing Friendly SMS Reminders That Patients Actually Read

Short message service (SMS) reminders can improve appointment adherence and screening completion, especially for procedures that require preparation. For colonoscopy, reminder texts that are clear and supportive can help patients follow the pre‑procedure diet and bowel prep instructions, which in turn improves exam quality.

AI can help you design SMS templates that fit within character limits, avoid jargon, and gently explain why prep matters.

Example prompt

“Draft a friendly SMS reminder for a colonoscopy appointment. Use the patient’s first name placeholder [Name]. Emphasize the importance of the pre‑procedure diet in simple words. Keep the tone warm and encouraging, not clinical. Stay under 320 characters and avoid jargon like ‘bowel prep’ or ‘mucosa.’ Include one short sentence about why following the diet helps the test go smoothly.”

You can then refine the tone (more formal or more conversational) and add your clinic’s name or callback details. Over time, you can A/B test slightly different messages to see which versions lead to better prep quality and fewer no‑shows.

Bringing It All Together in Your Workflow

Plain‑language communication is not about “dumbing down” medicine; it is about making information accurate, clear, and actionable for the people who need it most. When you combine those principles with AI, you can:

• Standardize discharge instructions at an appropriate reading level, reducing confusion and errors.

• Give trainees a low‑risk space to practice difficult conversations and receive structured feedback.

• Turn dense reference materials into concise FAQs that mirror real patient concerns.

• Deliver timely, human‑sounding reminders that support preparation and follow‑through.

You do not need to overhaul your entire system at once. Start with a single use case—such as post‑operative instructions or colonoscopy reminders—build a small library of prompts and examples, and invite both patients and clinicians to react to the new language. Their feedback will show you where the AI is helping, where it still sounds too technical, and how you can keep improving the way you communicate care.