Objective: Medical letters contain detailed information, but most of the time they are difficult to understand for patients. The use of artificial intelligence (AI) in healthcare is still in its early stages, but it has the potential to become a key pillar of the modern healthcare system. Our objective was to evaluate whether ChatGPT-5 can help both patients and healthcare professionals to better transmit and understand medical communication regarding their clinical condition and hypertensive status. Design and method: We selected 100 consecutive medical letters belonging to hypertensive patients, which were anonymized. We then used ChatGPT to generate an easy-to-understand version of each letter. These responses were subsequently evaluated by specialized medical staff as well as by patients. Results: Our results highlight ChatGPT's potential to generate medically correct output (92%, p = 0.00138), depending on the complexity of the cases (p = 0.001), as well as to provide relevant recommendations for both healthcare professionals and patients (96%, p = 0.0222; 98%, p = 0.00039). The impact of ChatGPT was more expressed especially in older patients suffering from hypertension, in those with blood pressure values above 150/100 mm Hg, in those with multiple comorbidities and in those with target organ damage. Conclusions: Simplifying medical letters significantly improves patient understanding and engagement, highlighting ChatGPT's potential as a tool for enhancing patient-centered communication. Although promising, future research should focus on ensuring accuracy in order to support the equitable and effective integration of artificial intelligence into healthcare communication.
Mihăilă et al. (Fri,) studied this question.