GPT-4 improves sex-specificity in cardiovascular patient education but may perpetuate gender biases: A mixed-methods audit
- 1School of Global Public Health, New York University, New York, New York, United States of America
- 2Herbert Wertheim College of Medicine, Florida International University, Miami, Florida, United States of America
Abstract
Patient education materials for cardiovascular disease (CVD) prevention frequently omit clinically important differences in disease manifestation, risk, and prevention between sexes. Socially constructed gender norms further shape how health information is communicated and received. Large Language Models (LLMs) like GPT-4 offer a potential tool for personalizing health communication, but their capacity to incorporate evidence-based sex differences without introducing or perpetuating gender biases is unknown. We identified seven publicly available English-language CVD prevention handouts from major health organizations. Using the GPT-4 API in August 2025, we generated sex-specific revisions for a 55-year-old male and female audience via standardized prompts. We provided structured prompts instructing the model to include evidence-based, sex-specific risk factors and symptoms. Original and revised materials were evaluated using Flesch-Kincaid Reading Ease, a novel 10-point sex-inclusivity checklist, and qualitative thematic analysis using a reflexive approach informed by thematic analysis principles. GPT-4 revisions substantially improved sex-inclusivity scores (Original median: 3.0/10, IQR = 0.0–3.0; Male-tailored median: 7.0, IQR = 7.0–8.5; Female-tailored median: 10.0, IQR = 10.0–10.0). Readability was maintained. However, qualitative analysis identified instances where outputs introduced content inconsistent with an equity-oriented framework: female-tailored content occasionally framed female symptoms as deviations from a male norm, male-tailored content frequently omitted erectile dysfunction as a CVD risk marker despite explicit prompting, and both versions at times reflected gendered social stereotypes (e.g., “bottling up emotions”) rather than clinical evidence. LLMs can rapidly improve sex-specificity in patient education materials but may also reproduce gender stereotypes and linguistically biased framings present in their training data. Their integration into clinical content development requires critical human oversight to support, rather than undermine, health equity goals. Why was this study done?: What did the researchers do and find?: What do these findings mean?:
Related
- Simulation-based education improves postpartum hemorrhage management knowledge and skills: a randomized controlled trial
- Bringing Cardiovascular Prevention into the Community: Protocol for an Observational Study of a Mosque-Based Health Initiative Combining Education and Cardiovascular Screening
- A long-term organotypic slice culture model of patient high-grade glioma preserving tumor heterogeneity and architecture, and specificity of CAR T cell targeting
- Interactive Nutrition Education Using Animated Videos and Puzzles Improves Fruit and Vegetable Nutrition Knowledge Among Elementary School Children
- Missed nursing care in postoperative patient education: a qualitative study from the perspective of surgical nurses
BibTeX
@article{khan2026gpt4,
title = {{GPT-4 improves sex-specificity in cardiovascular patient education but may perpetuate gender biases: A mixed-methods audit}},
author = {Samah Khan and Georgeta Vaidean},
journal = {PLOS Digital Health},
year = {2026},
month = sep,
doi = {10.1371/journal.pdig.0001122},
url = {https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001122},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - GPT-4 improves sex-specificity in cardiovascular patient education but may perpetuate gender biases: A mixed-methods audit
AU - Samah Khan
AU - Georgeta Vaidean
PY - 2026
DA - 2026/09/10
JO - PLOS Digital Health
DO - 10.1371/journal.pdig.0001122
UR - https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001122
AB - Patient education materials for cardiovascular disease (CVD) prevention frequently omit clinically important differences in disease manifestation, risk, and prevention between sexes. Socially constructed gender norms further shape how health information is communicated and received. Large Language Models (LLMs) like GPT-4 offer a potential tool for personalizing health communication, but their capacity to incorporate evidence-based sex differences without introducing or perpetuating gender biases is unknown. We identified seven publicly available English-language CVD prevention handouts from major health organizations. Using the GPT-4 API in August 2025, we generated sex-specific revisions for a 55-year-old male and female audience via standardized prompts. We provided structured prompts instructing the model to include evidence-based, sex-specific risk factors and symptoms. Original and revised materials were evaluated using Flesch-Kincaid Reading Ease, a novel 10-point sex-inclusivity checklist, and qualitative thematic analysis using a reflexive approach informed by thematic analysis principles. GPT-4 revisions substantially improved sex-inclusivity scores (Original median: 3.0/10, IQR = 0.0–3.0; Male-tailored median: 7.0, IQR = 7.0–8.5; Female-tailored median: 10.0, IQR = 10.0–10.0). Readability was maintained. However, qualitative analysis identified instances where outputs introduced content inconsistent with an equity-oriented framework: female-tailored content occasionally framed female symptoms as deviations from a male norm, male-tailored content frequently omitted erectile dysfunction as a CVD risk marker despite explicit prompting, and both versions at times reflected gendered social stereotypes (e.g., “bottling up emotions”) rather than clinical evidence. LLMs can rapidly improve sex-specificity in patient education materials but may also reproduce gender stereotypes and linguistically biased framings present in their training data. Their integration into clinical content development requires critical human oversight to support, rather than undermine, health equity goals. Why was this study done?: What did the researchers do and find?: What do these findings mean?:
SN - 2767-3170
ER -