Male Self-Reported Eating Disorders on Chinese Social Media: A Computational Analysis of Xiaohongshu Posts and Comments

Background Males with eating disorders (EDs) remain overlooked, particularly due to gendered stereotypes that characterize EDs as primarily affecting females. Social media provides a naturalistic context in which content creators disclose illness experiences and receive public feedback. This study used computational text analysis to examine how male ED experiences are discussed on Xiaohongshu (Little Red Note), a major Chinese social media platform. Methods We collected 502 posts from eight male self-reported ED content creators and 20,844 associated public comments posted between October 2021 and June 2025. In posts and comments, sentiment analysis was used to quantify emotional polarity (positive, neutral, negative), and latent Dirichlet Allocation topic modeling was used to identify semantic patterns. Results The posts were predominantly positive (57.0%), reflecting recovery motivation, while 22.9% expressed negative affect, and 20.1% were neutral. Three associated thematic categories were revealed:(1) Emotional Struggles and Meaning-Making around EDs , (2) Male Binge–Purge Episodes and Self-Rescue Attempts , and (3) Journaling Treatment, Control, and Recovery Journeys . Comments also showed mixed sentiment, with a majority of positive responses (48.1%), followed by negative (34.5%) and neutral (17.3%). Three themes appeared from the comments: (1) Body Image- and Eating-related Communication , (2) Empathic Care and Psychological Support , and (3) Male Body Ideals and Appearance Judgments . Conclusion Male self-reported ED discourse on social media presents a mixed emotional landscape, yet positive sentiments reflect the majority of posts and comments. Computational analysis of naturalistic digital traces provides scalable insights into male self-reported EDs, underscoring the need for sex/gender-sensitive communication and platform-level strategies to support interventions.