Recent research published in the International Journal of Eating Disorders suggests a connection between the use of TikTok, a popular video-based social media platform, and eating disorder symptoms. The study found that users who actively engage with content related to appearance and eating are more likely to exhibit symptoms of eating disorders, potentially due to the nature of the platform’s content recommendation algorithms.
Social media platforms, particularly those that are photo-based, have been consistently linked to symptoms of eating disorders. These symptoms are often attributed to the exposure to idealized body images and the ensuing comparisons users make between themselves and others.
However, most previous studies have not taken into account the dynamic and user-responsive nature of modern social media environments, particularly those driven by machine-learning algorithms. TikTok, known for its engaging short videos and personalized content suggestions, has become increasingly popular among youth. Researchers aimed to investigate whether the way TikTok’s algorithm suggests content could contribute to the development or exacerbation of eating disorder symptoms.
The researchers recruited 230 participants from the University of Western Australia, ensuring all had been using TikTok for at least four weeks. The sample consisted predominantly of women (179), with a smaller number of men (45), non-binary individuals (3), and participants who preferred not to disclose their gender (3). Participants were asked to complete several questionnaires and report their TikTok activity.
Eating Disorder Symptoms: The Eating Disorder Examination Questionnaire assessed participants’ eating disorder symptoms over the past 28 days, focusing on dietary restraint, eating concerns, weight concerns, and shape concerns. Higher scores indicated more severe symptoms.
Upwards Social Media Appearance Comparisons: This was measured using an adapted version of the Upward Appearance Comparison Scale, which assessed how often participants compared themselves to others on social media who they perceived to look better.
TikTok Activity: Participants accessed their TikTok data to report their screen time and details of their last 20 recommended videos. They noted whether these videos contained themes related to appearance and eating, had hashtags indicating such themes, or featured bodies perceived to be attractive. They also recorded any active engagement with these videos, such as liking, commenting, or sharing.
The researchers used statistical models to test the relationships between engagement with appearance/eating-related content, the proportion of such content in recommendations, upwards appearance comparisons, and eating disorder symptoms. They employed a serial mediation model to explore these connections.
The study found that higher engagement with appearance and eating-related content on TikTok was associated with higher proportions of similar content being recommended. This, in turn, was linked to greater frequency of upwards social media appearance comparisons, which were associated with more severe eating disorder symptoms. Specifically, the results showed significant indirect effects when the content was tagged with appearance/eating-related hashtags or featured bodies perceived as attractive by the user.
However, not all forms of content engagement showed these effects. For example, general screen time and engagement with recommended content that did not have these specific themes were not associated with eating disorder symptoms or appearance comparisons. This suggests that it is the nature of the content, rather than overall usage, that is critical in these associations.
While the findings provide valuable insights, the study has several limitations. First, it focused solely on active engagement with TikTok content and did not consider other factors, such as video watch time, that also influence the platform’s recommendations.
Additionally, the study’s cross-sectional design means it cannot establish causality or the directionality of the observed relationships. Future research should use experimental and longitudinal designs to explore these associations over time and determine whether engagement with specific types of content leads to increased eating disorder symptoms or vice versa.
“In conclusion, these findings make an important and novel contribution to the extant empirical work documenting the relationships between social media use and [eating disorder] symptoms. Findings provide preliminary support for the ways in which users’ engagement with TikTok content is related to [eating disorder] symptoms. Future work in this area is warranted, and social media platforms should be urged to create policies and practices that protect youth from these risks,” the researchers wrote.
The study, “Association between engagement with appearance and eating related TikTok content and eating disorder symptoms via recommended content and appearance comparisons,” was authored by Laura Dondzilo, Rachel F. Rodgers, and Fanny Alexandra Dietel.
