Debunking Myths and Patient Education About Complex Regional Pain Syndrome Using Popular Generative Artificial...

Area of Science:

• Medical Informatics
• Pain Management
• Artificial Intelligence in Healthcare

Background:

• Complex Regional Pain Syndrome (CRPS) significantly impacts quality of life, necessitating effective patient education.
• Current patient education materials often lack accessibility and readability, perpetuating myths about CRPS.
• This study compares AI-generated patient education materials (PEMs) with traditional online resources for CRPS.

Purpose of the Study:

• To evaluate the accuracy, readability, completeness, emotional tone, and myth-debunking capabilities of AI-generated PEMs.
• To compare AI chatbots (ChatGPT, Gemini, Llama) against established sources (Cleveland Clinic, Mayo Clinic).
• To assess the potential of AI in improving patient education for chronic pain conditions.

Main Methods:

• Comparative analysis of PEMs from three AI chatbots and two reputable medical institutions.
• Readability assessed using Flesch Reading Ease, Gunning Fog Index, Flesch-Kincaid Grade Level, and SMOG Index.
• Accuracy and completeness evaluated by pain experts; myth-debunking assessed using sensitivity and specificity metrics; sentiment analysis for emotional tone.

Main Results:

• Cleveland Clinic materials showed highest accuracy (4.8/5) and completeness (4.8/5), outperforming AI chatbots.
• Gemini AI offered superior readability (Flesch score 62) but lower myth-debunking specificity (66.67%).
• ChatGPT achieved 100% specificity in debunking myths; Gemini had the most positive sentiment (83).

Conclusions:

• AI-generated PEMs offer potential for accurate and readable CRPS patient education.
• Traditional resources from leading institutions provide more comprehensive information.
• AI tools show promise for complementing existing patient education but require further validation for clinical use.