Quality, Reliability, and Dissemination of In Vitro Fertilization-Related Videos on Chinese Social Media: Cross-Sectional Analysis of 300 Short Videos.

BACKGROUND

Patients increasingly rely on short-video platforms for information regarding in vitro fertilization (IVF), yet the relationship between the scientific quality of this content and its algorithmic dissemination remains unclear.

OBJECTIVE

This study aimed to assess the quality, reliability, and key drivers of dissemination of IVF-related short videos on major Chinese social media platforms.

METHODS

A cross-sectional content analysis was conducted on 300 popular IVF-related videos (the top 100 results from each platform) retrieved from Douyin, Bilibili, and Xiaohongshu between January 10 and 15, 2025. Video quality and reliability were evaluated using the Global Quality Score and a modified DISCERN instrument. Predictors of video dissemination were identified using an Extreme Gradient Boosting machine learning model, with the number of "likes" serving as the primary outcome variable.

RESULTS

Content produced by medical professionals demonstrated significantly higher quality and reliability (median mDISCERN 11.0, IQR 9.0-15.0) compared to non-medical sources (median mDISCERN 8.0, IQR 5.0-13.0; P< .001). However, the Extreme Gradient Boosting analysis identified the uploader's follower count as the most powerful predictor of video "likes." In contrast, quality metrics (Global Quality Score and modified DISCERN scores) had a negligible impact on dissemination.

CONCLUSIONS

In the current Chinese social media landscape, the dissemination of IVF-related videos is strongly associated with creator influence rather than scientific merit. This disconnect between engagement and quality poses a potential risk of misinformation, highlighting the need for medical professionals to adopt platform-native communication strategies to ensure that high-quality information reaches patients.