The Effects of Social Media Restriction on Social Anxiety, Fear of Missing Out, and Health-related Quality of Life: a Randomised Controlled Trial (RESET)

Background With more than 5.4 billion users worldwide in 2025, social media has become an integral part of daily life. Young adults aged 18 to 24 are common users of social media platforms. Social media offers several benefits, such as providing a sense of social support, entertainment, an outlet for creativity, and learning opportunities. However, it is also associated with negative experiences such as psychological distress, depression, loneliness, anxiety, the fear of missing out (FoMO), and problematic social media use (PSMU).

Interest in reducing social media use is increasing. Some studies suggest that restricting social media use can have a positive effect on anxiety and FoMO. One randomised controlled trial (RCT) found that limiting social media use to one hour per day for three weeks reduced symptoms of anxiety, depression and FoMO, and improved sleep in young people aged 17-25 with symptoms of generalised anxiety and depression (Davis & Goldfield, 2025). However, restrictions may also have counterproductive effects, such as continued use despite the restriction, anxiety when separated from one's smartphone, or poorer sleep (OECD, 2024).

Our trial aims to test whether reducing social media use improves social anxiety, FoMO,, health-related quality of life (HRQoL) and safety. It addresses current knowledge gaps by including a non-clinical sample of students and by assessing potential adverse consequences, such as an rebound effect, with an increased use of social media after the intervention.

Objectives The primary objective is to evaluate the impact of restricting social media usage on social anxiety at the end of the one-week intervention. The secondary objectives are to evaluate the impact on 1) anxiety three weeks post-baseline, 2) FoMO, and 3) HRQoL, and 4) to compare these impacts between participants with and without PSMU. A further fifth secondary safety objective is to evaluate whether restricting social media use leads to an increase in use after the intervention (the 'rebound effect'), or whether it results in sustained reduced use.

Design This is a two-arm, parallel-group, open-label, pragmatic randomised controlled superiority trial, with an allocation ratio of 1:1. Data will be collected via self-reporting questionnaires at three time points: baseline (T0), the end of the one-week intervention (T1) and three weeks after the intervention begins (T2).

Participants will be allocated using block randomisation (ratio 1:1) and will be stratified by PSMU status. PSMU will be assessed using the Bergen Social Media Addiction Scale (BSMAS), with a cut-off score of 24 (Luo et al., 2021). The randomisation list will be generated using LimeSurvey, and participants will be assigned to a group when they complete the baseline questionnaire.

Setting The study will take place at the University of Lausanne in Switzerland. First-year undergraduate psychology students will be invited to participate during a lecture. Screening and data collection will be conducted online using LimeSurvey.

Participants First-year Bachelor's degree students of psychology will be invited to participate. They will be required to meet the following criteria: be aged 18 or over; have a good level of French (minimum B2); use the social media platforms included in the study for an average of at least two hours per day, based on cumulative smartphone screen time over one week; and provide informed consent. The two-hour threshold was based on Davis and Goldfield (2025), who defined regular social media use as an average of at least two hours per day.

Participants will be excluded if they use social media mainly for professional purposes (e.g., as a community manager, influencer or content creator), have voluntarily reduced their social media use or undertaken a digital detox (lasting at least two weeks) in the previous 12 months, or are unable to provide smartphone screen-time data (e.g., due to an incompatible smartphone or refusal to share these data).

Intervention and comparator Participants in the intervention group will be instructed to restrict their total daily use of the social media platforms to one hour for one week. This limit applies to the total time spent on all included platforms, not to each platform separately. This limit of one hour per day is based on Davis and Goldfield's (2025) RCT, in which restricting use of seven social media platforms to one hour per day for three weeks was found to significantly reduce symptoms of anxiety, depression, and FoMO. We expect that our short intervention would be more feasible and acceptable for students, thus improving adherence. It will provide additional evidence of the potential benefits of short-term interventions.

Participants in the control group will be instructed to continue their usual social media use (treatment as usual). This comparator was chosen to reflect real-world social media use, in line with the pragmatic nature of the trial. It is also consistent with the control condition used by Davis and Goldfield (2025), facilitating comparison with their findings.

Outcomes Liebowitz Social Anxiety Scale - Self-Report (LSAS-SR). The French version of the LSAS-SR will be used as the primary outcome to assess social anxiety (Yao et al., 1999; Heeren et al., 2012). This scale includes 24 social situations, each rated for fear and avoidance on a four-point Likert scale scored from 0 to 3. A total score will be calculated (range 0-144), with a higher score indicating higher social anxiety. The primary outcome will be assessed at T1. Assessment at T2 will be considered as a secondary outcome.

Fear of Missing Out Scale (FoMOS). The validated eight-item French version of the FoMOS will be used as a secondary outcome (Andraos et al., 2025). Items are rated on a five-point Likert scale scored from 1 to 5. A sum score will be calculated (range 8-40), with a higher score indicating higher FoMO. The FoMOS will be assessed at T1 and T2.

Child Health Utility 9 Dimensions (CHU9D). The CHU9D will be used as a secondary outcome to assess HRQoL. It comprises nine dimensions: worried, sad, pain, tired, annoyed, schoolwork/homework, sleep, daily routine, and able to join in activities, each with five severity levels (Stevens, 2012; Ratcliffe et al., 2016). The schoolwork/homework item was reworded as "academic work/studies" to suit a student sample. Responses will be analysed as a continuous score and can be converted into a utility score anchored at 0 (dead) and 1 (full health). The CHU9D will be assessed at T1 and T2.

Social media use. Social media use will be assessed using the Red Box, Green Box approach, a self-report behavioural frequency measure that has been developed for research into behavioural addictions (Stevens et al., 2026). Changes in social media use will be used as a secondary safety outcome to assess the potential for a rebound effect. Participants will be asked to estimate the total number of hours they spend on social media. The green box assesses leisure use, while the red box assesses use that displaces other activities. Higher scores indicate more frequent use. Social media use will be assessed at T1 and T2.

Other variables Bergen Social Media Addiction Scale (BSMAS). The French version of the BSMAS will be used to assess PSMU (Rouleau et al., 2023). It includes six items on a five-point Likert scale scored from 1 to 5 (range 6-30), with a higher score indicating higher PSMU. It will be administered at all three time points. The baseline score will be used to stratify participants (cut-off of 24, Luo et al., 2024).

We will assess sociodemographic and lifestyle variables at baseline. The LSAS-SR, FoMOS, CHU9D, and social media use will also be assessed at baseline.

Sample size We performed a power analysis using G*Power 3.1. We used a two-tailed independent t-test with α=.05, 1-β=.80, and a moderate effect size (Cohen's d=0.50, close to the effect of 0.42 reported by Davis and Goldfield (2025) for anxiety symptoms). A total of 128 participants (64 per group) are needed. To account for dropouts and missing data, we will recruit 160 participants.

Statistical analyses The main analyses will follow the intention-to-treat principle. A per-protocol analysis will be performed as a secondary analysis, including only participants who adhered to the intervention instructions. We will test intervention effects using regression models for the primary outcome (assessed at T1). Then, mixed ANOVAs with a 2 (group) × 2 (PSMU status) × 3 (time) design will be used for secondary outcomes. These models will include group, time, PSMU status (the stratification factor) and the relevant pre-defined covariates. PSMU moderation effects will also be tested. No adjustment will be made for variables measured after randomisation. A statistical analysis plan will be written before the end of data collection.

Expected results We expect participants in the intervention group to report lower levels of social anxiety and FoMO, and higher HRQoL than participants in the control group. We also anticipate that the benefits of the intervention will be more pronounced among individuals with PSMU. These findings could enhance our understanding of the impact of restricting social media use on mental health and well-being. They may also inform the development of interventions aimed at promoting healthier social media use and preventing mental health issues, particularly among individuals with PSMU or social anxiety.

References Andraos, E., Graziani, P., & Del Monte, J. (2025). Validation of a French Fear of Missing Out Scale. L'Encéphale, 51(6), 609-614. https://doi.org/10.1016/j.encep.2024.12.003 Davis, C. G., & Goldfield, G. S. (2025). Limiting social media use decreases depression, anxiety, and fear of missing out in youth with emotional distress: A randomized controlled trial. Psychology of Popular Media, 14(1), 1-11. https://doi.org/10.1037/ppm0000536 Heeren, A., Maurage, P., Rossignol, M., Vanhaelen, M., Peschard, V., Eeckhout, C., & Philippot, P. (2012). Self-report version of the Liebowitz Social Anxiety Scale: Psychometric properties of the French version. Canadian Journal of Behavioural Science / Revue canadienne des sciences du comportement, 44(2), 99-107. https://doi.org/10.1037/a0026249 Luo, T., Qin, L., Cheng, L., Wang, S., Zhu, Z., Xu, J., Chen, H., Liu, Q., Hu, M., Tong, J., Hao, W., Wei, B., & Liao, Y. (2021). Determination the cut-off point for the Bergen social media addiction (BSMAS): Diagnostic contribution of the six criteria of the components model of addiction for social media disorder. Journal of Behavioral Addictions, 10(2), 281-290. https://doi.org/10.1556/2006.2021.00025 OECD. (2024). Élèves et écrans : Performance académique et bien-être. OECD Publishing. https://doi.org/10.1787/b3c4552d-fr Ratcliffe, J., Huynh, E., Stevens, K., Brazier, J., Sawyer, M., & Flynn, T. (2016). Nothing about us without us? A comparison of adolescent and adult health-state values for the Child Health Utility-9D using profile case best-worst scaling. Health Economics, 25(4), 486-496. https://doi.org/10.1002/hec.3165 Rouleau, R. D., Beauregard, C., & Beaudry, V. (2023). A rise in social media use in adolescents during the COVID-19 pandemic: The French validation of the Bergen Social Media Addiction Scale in a Canadian cohort. BMC Psychology, 11(1), 92. https://doi.org/10.1186/s40359-023-01141-2 Stevens, K. (2012). Valuation of the Child Health Utility 9D Index. PharmacoEconomics, 30(8), 729-747. https://doi.org/10.2165/11599120-000000000-00000 Stevens, M. W. R., Radunz, M., Galanis, C., Quinney, B., Zajac, I., Billieux, J., Delfabbro, P. H., & King, D. L. (2026). Red Box, Green Box: Psychometric evaluation of a self-report behavioral frequency measurement approach for behavioral addictions research. Addiction, 121(2), 429-439. https://doi.org/10.1111/add.70192 Yao, S. N., Note, I., Fanget, F., Albuisson, E., Bouvard, M., Jalenques, I., & Cottraux, J. (1999). Social anxiety in social phobics: Validation of Liebowitz's social anxiety scale - French version. L'Encéphale, 25(5), 429-435.