3219 A case of myasthenia gravis lambert-eaton overlap syndrome

MethodsWe utilised an OpenAI GPT-4 model to adapt generic migraine information into four distinct communication styles: Analytical, Intuitive, Functional, and Personal.Patients at a single site were then presented online with both the generic information and the version personalized to their preferred communication style.Preferences were quantified based on their overall preference and engagement with the material (-2 generic to +2 personalized), using a structured questionnaire.Statistical analysis was conducted to evaluate the significance of patient preferences.Results Analysis of the questionnaire responses from 33 patients revealed a statistically significant preference for personalized information sheets, both in terms of overall preference (Mean Preference = 0.576, P-Value = 0.028) and engagement (Mean Preference = 0.606, P-Value = 0.025) using one sample T test.Conclusion This pilot study underscores the potential importance of personalizing patient education materials to match individual communication preferences.Personalised information is shown to enhance patient engagement and is preferred over generic information.This highlights the potential for AI-driven tools to significantly improve the effectiveness of patient communication and education in healthcare settings.Future research should explore the implementation of such personalised approaches in a broader range of medical information and its impact on patient outcomes.