MedReadCtrl: Personalizing medical text generation with readability-controlled instruction learning

TL;DR

MedReadCtrl is a novel framework that enables large language models to generate medical texts tailored to specific readability levels, improving patient comprehension without losing medical accuracy.

Contribution

It introduces a readability-controlled instruction tuning method for LLMs, enhancing medical text accessibility while maintaining content integrity.

Findings

Significantly reduced readability instruction errors compared to GPT-4

Substantial improvements on unseen clinical tasks in ROUGE-L and SARI scores

Experts preferred MedReadCtrl outputs, especially for low-literacy users

Abstract

Generative AI has demonstrated strong potential in healthcare, from clinical decision support to patient-facing chatbots that improve outcomes. A critical challenge for deployment is effective human-AI communication, where content must be both personalized and understandable. We introduce MedReadCtrl, a readability-controlled instruction tuning framework that enables LLMs to adjust output complexity without compromising meaning. Evaluations of nine datasets and three tasks across medical and general domains show that MedReadCtrl achieves significantly lower readability instruction-following errors than GPT-4 (e.g., 1.39 vs. 1.59 on ReadMe, p<0.001) and delivers substantial gains on unseen clinical tasks (e.g., +14.7 ROUGE-L, +6.18 SARI on MTSamples). Experts consistently preferred MedReadCtrl (71.7% vs. 23.3%), especially at low literacy levels. These gains reflect MedReadCtrl's ability to restructure clinical content into accessible, readability-aligned language while preserving medical intent, offering a scalable solution to support patient education and expand equitable access to AI-enabled care.