Study Challenges Link Between Social Media Use and Body Image

Total time spent on social media may not have an adverse impact on how adolescents think about their weight and dieting, together referred to as body image–related factors, according to new research published online today in JAMA Pediatrics.¹

Highlighting mixed results from previous research, including that social media may be a risk factor for developing an eating disorder² and that research methods have thus far been limited in their ability to accurately measure this relationship,³ the present study authors went directly to the source of cell phone social media use: the adolescents themselves. Using online ads that promoted participation in data gathering, the authors reached out to Android users between May 2021 and June 2022. All were adolescents aged 13 to 17 years and the reported sole users of the phone. They also used both within-person and between-person levels of modeling; measured the potential impact of gender, age, race and ethnicity, parental educational level, body mass index, and perceived weight changes as moderating factors; and evaluated associations between weight concerns and dieting frequency and the use of Instagram, Snapchat, Facebook, Discord, Twitter (which became known as X in July 2023), and TikTok.

Participation was close to equal between genders, with 74 female participants (47.7%) and 68 male participants (43.9%). In addition, 7.1% (n = 11) did not specify their gender identity, and 1.3% (n = 2) left this response blank. For self-identified race and ethnicity, most (54.2%) were White (n = 84), followed by Black or African American (14.2%; n = 22), multiracial (12.9%; n = 20), Asian (8.4%; n = 13), Hispanic or Latino (7.7%; n = 12), American Indian or Alaska Native (1.9%; n = 3), and unreported (0.6%; n = 1).

Following consent to participate, data for analysis were collected through the Stanford Screenomics app. Most parents of the participants had a master’s or doctoral degree (39.3%; n = 61) or a bachelor’s degree (29.0%; n = 45).

Overall, the adolescent participants provided 217,612 total hours of smartphone data and 769 reports of weight concerns and dieting. Their median (IQR) daily hours of reported social media use were 1.35 (0.40-2.94), and these were broken down as follows, with most median time (in minutes) spent on Instagram and least time spent on Twitter:

• Instagram: 13.29 (0.03-52.73) (range, 0.00-373.26)
• Snapchat: 1.15 (0.00-15.16) (range, 0.00-306.42)
• Discord: 0.68 (0.00-9.49) (range, 0.00-237.19)
• TikTok: 0.65 (0.00-76.05) (range, 0.00-316.52)
• Facebook: 0.00 (0.00-0.95) (range, 0.00-83.71)
• Twitter: 0.00 (0.00-0.13) (range, 0.00-39.70)

Total social media use came in at 80.97 minutes (24.20-176.65) (range, 0.00-520.52).

Weight concerns and dieting frequency were both scored on a scale of 1 to 3, with higher total points indicating more instances of both. The study authors did not see any significant relationships for their within-person and between-person analyses:

1. Social media use overall:
   1. Within: weight concerns, P = .65, and dieting frequency, P = .68
   2. Between: P = .25 and P = .39, respectively
2. Male gender:
   1. Within: P = .28 and P = .92
   2. Between: P = .46 and P = .61
3. Baseline age:
   1. Within: P = .86 and P = .75
   2. Between: P = .22 and P = .30
4. White race/ethnicity:
   1. Within: P = .88 and P = .35
   2. Between: P = .31 and P = .80
5. Parental educational level:
   1. Within: P = .75 and P = .81
   2. Between: P = .95 and P = .56
6. Body mass index:
   1. Within: P = .91 and P = .99
   2. Between: P = .56 and P = .40
7. Weight changes:
   1. Within: P = .83 and P = .99
   2. Between: P = .76 and P = .91

However, suggested relationships were seen for between-person associations with weight concerns and Snapchat (γ, 0.29; 95% CI, 0.01-0.58), Twitter (γ, 0.49; 95% CI, 0.01-0.96), and TikTok (γ, 0.49; 95% CI, 0.01-0.96) following exploratory analyses.

There are potential limitations to these findings, including that the only participants were adolescents who used an Android device, use of other smart devices was not considered, and the total sample size (N = 155) may not have been sufficient to identify heterogeneous associations among certain subgroups.

“Future studies should use larger and more nationally representative samples,” the authors concluded. “We recommend that future studies obtain objective measures of individual app use and specific content experienced by adolescents to understand the role of digital media environments in body image–related issues among adolescents.”

References

1. Sun X Molaib K, Xu T, et al. Objectively measured social media use and weight concerns and dieting in adolescents. JAMA Pediatr. Published online December 1, 2025. doi:10.1001/jamapediatrics.2025.4744
2. Dane A, Bhatia K. The social media diet: a scoping review to investigate the association between social media, body image and eating disorders amongst young people. PLOS Glob Public Health. 2023;3(3):e0001091. doi:10.1371/journal.pgph.0001091
3. Jarman HK, McLean SA, Griffiths S, et al. Critical measurement issues in the assessment of social media influence on body image. Body Image. 2022;40:225-236. doi:10.1016/j.bodyim.2021.12.007