AI Meets Aesthetic Surgery: Enhancing Tummy Tuck Patient Information Sheets to Empower Patient Decisions

Goals/Purpose: In aesthetic plastic surgery, clear communication is essential to support patients in making well-informed choices tailored to their preferences. However, many patient information sheets exceed the sixth-grade reading level recommended by the American Medical Association, creating a gap that can hinder understanding. This disconnect may prevent patients from fully understanding important details that would further empower their healthcare decisions. The recent rise of Artificial Intelligence (AI) tools like ChatGPT-4.0 offers unprecedented potential to enhance patient education by simplifying complex medical information. Building on prior efforts to adjust cosmetic surgery information sheets to a sixth-grade reading level using ChatGPT-4.0, this study evaluates the effectiveness of AI-optimized American Society of Plastic Surgeons (ASPS) Tummy Tuck informational sheets in the general population. By comparing participant responses to both original and AI-enhanced versions, this study aims to evaluate the impact of AI on readability, patient preference, and comprehension. Ultimately, this research seeks to establish whether AI-optimized materials can enhance patient engagement and accessibility, paving the way for broader applications in patient-centered communication for aesthetic plastic surgery.

Methods/Technique: Tummy Tuck, the third most popular procedure based on the 2023 Aesthetic Plastic Surgery Statistics, was selected as the focus for this study. We sourced the Tummy Tuck patient information sheet directly from the ASPS website. To create the ChatGPT-4.0-optimized version, we used a standardized prompt: “Generate a patient information sheet on [surgery] tailored to a 6th-grade reading level for the average U.S. patient," followed by inputting the original ASPS Tummy Tuck sheet content. A comprehensive, 37-item survey was administered to assess participant demographics, understandability, comprehension confidence, and preference between the original ASPS and AI-optimized sheets. The survey utilized a modified Patient Education Materials Assessment Tool (PEMAT) to measure understandability and confidence levels. Participants were recruited through SurveyMonkey and were blinded to the source of each information sheet to eliminate potential bias. Statistical analysis included paired two-tailed t-tests for understandability and confidence and a binomial test for preference, with p<0.05 for significance.

Results/Complications: We received 147 completed responses. The majority of participants identified as male (66.7%), white (79.9%), aged between 45-54 (38.4%), holding a master’s degree (32.1%), and reporting an annual income of over $100,000 (57.2%). Analysis revealed that the AI-optimized sheet scored an average understandability of 96.78% ± 8.02%, comparable to the original ASPS sheet’s score of 94.37% ± 13.30%, with no statistically significant difference between them (p = 0.061). No significant differences were found in confidence regarding procedure understanding (p=0.591), comprehension of risks/benefits (p=0.591), or informed decision-making (p=0.712). For patient readability and preference, 78.91% of participants (n = 116) found the AI-optimized sheet easier to read, while 68.03% (n = 100) preferred it for clinical use; both results were statistically significant (p < 0.001).

Conclusion: Our findings indicate that AI-optimized and ASPS sheets had similarly high understandability scores, with no significant difference between them. However, participants showed a statistically significant preference for the AI-optimized sheet in clinical settings and its reading ease, likely due to its straightforward language and minimized medical terms. The preference for simplified language points to a broader role for AI in enhancing medical communication and bridging health literacy gaps without compromising accuracy. Although promising, limitations of this study include a primarily high-income, educated sample, which may not fully represent the general patient population. In addition, given that Tummy Tuck surgery is more common amongst women, but most survey participants in this study were male, future research should ensure gender balance to capture potential gender-specific preferences. The next step of this field involves integrating AI-enhanced materials within clinical settings to monitor their real-world influence on patient engagement, satisfaction, and long-term educational outcomes, ultimately supporting AI’s role in advancing patient-centered care in cosmetic surgery.