Co-occurrence between social media use and eating disorders

Exposure to pro-eating disorder (ED) content on social media (SM) is an indicator of high risk for ED.¹ Various studies have confirmed an association between posting edited photographs and body dissatisfaction and ED symptomatology.^2,3 However, most existing research has focused on the general population rather than on clinical populations. Our objective was to explore the co-occurrence of SM use and EDs in a clinical population of individuals with EDs. Co-occurrence refers to a situation in which two or more phenomena appear or occur together in the same context.

The RESTCA study was a descriptive, cross-sectional, multicenter study of individuals diagnosed with an ED and receiving outpatient care at a specialist facility. It was conducted from January 2024 to January 2025 in France. Participants were asked to complete a semistructured questionnaire including questions about EDs, sociodemographic characteristics, SM use, and weight. They also completed the 26-item Eating Attitudes Test (EAT-26).^4,5 Mean and SD, or count (n) and percentage (%), were calculated to describe variables. We used the Wilcoxon test and the point-biserial correlation coefficient (rpb) to test associations. Analyses were performed using R version 4.4.1.⁶

Of the 53 patients included in the study, 49 (92.4%) were women, with a mean age of 20.0 years (SD, 3.9). Almost half of the patients (n = 22) were adolescents. In the sample, 22 patients (41.5%) were diagnosed with anorexia nervosa, 10 (18.9%) with bulimia nervosa, and 21 (39.6%) with eating disorder not otherwise specified (EDNOS).

All participants used SM. As shown in Fig. 1, the main SM platforms used were Instagram (n = 49), Snapchat (n = 48), and TikTok (n = 43).

Fig. 1.

Word cloud of social media platforms used by participants. The size of the words and the intensity of the black color in the text correlate with usage frequency within the sample. Only the 20 most frequently used SM platforms are represented. Instagram, TikTok, Snapchat, and Facebook are the most widely used platforms.

Among participants, 49 patients (92.4%) reported that they had already viewed content related to EDs, and 37 (75.5%) had viewed content promoting weight loss. A significant proportion of patients (n = 36, 67.9%) had followed an influencer specializing in EDs on social media, and 2 (3.8%) had used a coach to lose weight. By contrast, 5 patients (9.4%) had used a coach to help them overcome their ED, and 11 (20.7%) had made purchases to help them overcome their ED. More details are provided in Table 1.

Table 1.

Description of social media use (N = 53).

	Mean or %	SD or n
Initiation
Age at start of SM use	12.7	2.1
Age at first exposure to ED-related social media content	16.9	3.3
ED-related content
Viewing ED-related content on SM, lifetime	92.4	49
Types of ED-related content viewed on SM, lifetime
Type 1: focused on weight loss	18.4	9
Type 2: focused on recovery	24.5	12
Type 3: both types	57.1	28
Viewing ED-related content on SM, past month	69.2	36
Types of content viewed on SM, past month
Type 1: focused on weight loss	8.3	3
Type 2: focused on recovery	58.3	21
Type 3: both types	33.3	12
Following influencers and coaches
Following ED-related influencers on social media	67.9	36
Following ED-related influencers in the past month	56.6	30
Consulting a professional coach for ED recovery	9.4	5
Consulting a professional coach for weight loss	3.8	2
Purchasing products or services for recovery/improvement	20.7	11
Purchasing products or services for weight loss	11.3	6

ED, eating disorder; n, count; N, sample size; SD, standard deviation; SM, social media.

The mean EAT-26 score was high (mean, 34.2; SD, 17.8). We found a significant positive association between a high EAT-26 score and viewing ED-related social media content in the month preceding the interview (rpb = 0.43; P < .003).

Although the level of ED symptoms was associated with recent viewing of content dealing with EDs, it is important to note that most patients used SM to find support, feel less alone in coping with their illness, or find advice on overcoming their ED. This aspect should therefore be considered carefully to avoid drawing premature conclusions about the effects of SM. Of note, the role of algorithms in the content viewed by users. In a study focusing specifically on the TikTok platform, 64% of patients reported finding pro-ana content without deliberately searching for it.⁷ Finally, only 20% of healthcare professionals asked their patients about their SM use.⁸

Some limitations should be considered. Because of the cross-sectional nature of the study, it is impossible to establish a causal relationship between the variables studied. Our study was conducted in a small sample, and inferences should therefore be made with caution. The results should be considered preliminary.

The results highlight the role of social media and the important place it can occupy in the lives of patients with EDs. There is currently no scientific evidence establishing a framework for the relationship between EDs and SM. For EDs, SM could be a risk factor, a maintaining factor, an obstacle to treatment, a coping strategy, or a combination of these possibilities.

Larger-scale studies would be useful to further explore these findings and examine the impact of SM on patients with EDs. The high co-occurrence of EDs and SM use among patients with EDs suggests that discussions about SM use should be initiated with patients during treatment. Collecting this type of information could help us understand the influence of SM on patients’ psychopathology.

Ethical approval statement

The RESTCA study protocol was submitted to the ethics committee in accordance with French procedures and was assigned to the South-East I Human Protection Committee (CPP) under No. 2023-A02380-45. The RESTCA study is a type 3 observational study according to the Jardé law. A declaration of compliance with the CNIL (Commission Nationale de l’Informatique et des Libertés) was made and registered under No. YPU33345426.

Funding statement

The RESTCA project did not receive funding.

Conflicts of interest

None declared.

Acknowledgments

We would like to thank the participants for their commitment to this project, as well as the clinical research department of the CentreHospitalier de Calais.