A new study examining the relationship between social media and eating behaviors suggests that the connection may be more complex than often assumed. The research indicates that college-aged women with existing disordered eating symptoms are more likely to increase their engagement with TikTok content promoting restrictive diets, rather than the content itself directly causing those symptoms to develop. The study, published in the journal Eating Behaviors, also found that a university-wide ban on the social media platform did not reduce students’ consumption of this content.
The link between social media use and body dissatisfaction has been a subject of extensive scientific inquiry. Platforms that are highly visual in nature, such as TikTok, frequently feature content centered on diet, fitness, and idealized body types. This has generated concern among researchers and clinicians about the potential for such content to contribute to the development or worsening of eating disorders, particularly among young people who are the heaviest users of these platforms.
A team of researchers sought to clarify the direction of this relationship over time. Their work was guided by a central question: Does viewing restrictive eating content on TikTok lead to disordered eating, or do individuals with pre-existing vulnerabilities to disordered eating actively seek out this type of content?
The research team included Samantha R. Strickland from the University of North Texas, along with Alejandra Medina Fernandez and Pamela K. Keel from Florida State University. They designed a study to track changes in both social media habits and eating behaviors over a period of several weeks.
The project gained an additional layer when, partway through the data collection process, the university where the study was being conducted implemented a ban on TikTok. This event provided the researchers with a natural experiment, allowing them to examine whether restricting access to the platform on a systemic level had any effect on students’ engagement with potentially harmful content.
To conduct their investigation, the researchers recruited 252 female undergraduate students between the ages of 18 and 24. At the beginning of the study, which served as the baseline, each participant completed a series of standardized questionnaires. These surveys were designed to measure the presence and severity of eating pathology, which includes a range of thoughts and behaviors such as concerns about weight and shape, food restriction, and binge eating.
Participants also reported on their specific engagement with TikTok content related to restrictive eating, such as “what I eat in a day” videos, diet challenges, and other posts promoting weight loss through food limitation.
Nine weeks after the initial assessment, the same students were asked to complete the questionnaires a second time. This follow-up allowed the researchers to observe changes in both disordered eating symptoms and TikTok engagement over the study period. The university’s TikTok ban was put in place after some participants had already completed the study but before others had started their nine-week follow-up.
This split the participants into two groups: those who completed the entire study before the ban and those who completed the follow-up assessment after the ban was in effect. This setup enabled a comparison to see if the policy change influenced behavior.
The analysis of the collected data yielded several key observations. The researchers found a predictive relationship in one specific direction. Higher levels of disordered eating symptoms reported by students at the beginning of the study were associated with greater engagement with restrictive eating content on TikTok at the nine-week follow-up. This suggests that individuals already experiencing symptoms related to eating disorders may be more drawn to this type of online content over time.
However, the reverse relationship was not observed. The amount of time students spent engaging with restrictive eating content at the study’s outset did not predict an increase in their disordered eating symptoms nine weeks later. This finding indicates that, within this timeframe, exposure to such content did not appear to be the initial driver of eating pathology. Instead, it seems to be something that individuals with existing concerns gravitate toward, potentially creating a feedback loop where their preoccupations are reinforced by what they see online.
Another significant result concerned the campus-wide TikTok ban. The researchers found that the policy had no measurable impact on students’ engagement with restrictive eating content. Students who completed the study after the ban was implemented reported consuming this type of content at similar rates to those who participated before the ban.
This suggests that the institutional restriction, which applied to university Wi-Fi networks and devices, was not effective in preventing access, as students likely used personal cellular data or other means to continue using the platform. In addition, across the entire group of participants, both disordered eating symptoms and engagement with restrictive eating content on TikTok showed a slight increase over the nine-week period.
The study has certain limitations that invite further exploration. The findings are based on self-reported data, which relies on the accuracy of participants’ memories and their willingness to report honestly on sensitive topics like eating behaviors and social media use.
The nine-week duration of the study also provides a relatively short window into what are often long-term, complex behaviors. The results are specific to a sample of female college students at a single university and may not be generalizable to other populations, such as men, older adults, or individuals who are not in college.
Future research could build on these findings by employing more objective measures of social media use, such as data gathered directly from smartphones, to bypass the potential inaccuracies of self-reporting. Longer-term studies would also be helpful to understand how these dynamics between media consumption and personal well-being evolve over months or even years.
The authors propose that interventions aimed at mitigating harm may be more effective if they are integrated directly into social media platforms. For example, apps could be designed to identify users who are engaging heavily with potentially harmful content and proactively offer them resources for mental health support. Such an approach may prove more successful than broad institutional bans, which appear to be easily circumvented.
The study, “TikTok and disordered eating: Delineating temporal associations and effects of a ban,” was authored by Samantha R. Strickland, Alejandra Medina Fernandez, and Pamela K. Keel.
