Table 3.
Study, year | Outcomes | Measures | Comparison measurement | Postintervention-reported values (unless labeled) | Effect size (Cohen d) and interpretation | Main finding | Direction of effect | |
Interventions involving full abstinence from social media | ||||||||
|
Brown and Kuss [36], 2020 |
|
|
Mean difference between before and after the intervention (SD) |
|
|
Significant improvements after the intervention | ↑g |
|
Fioravanti et al [41], 2020 |
|
|
Calculated mean difference between IGj and CGk for women and men |
|
|
Significantly improved life satisfaction and positive affect for women but not for men in the IG compared with those in the CG | ↕o |
|
Hall et al [42], 2019 |
|
|
Mean (SD) |
|
|
No significant difference for the IG compared with the CG across all measures | ↔q |
|
Hanley et al [43], 2019 |
|
|
Standardized coefficients from multiple regression |
|
|
No significant difference for the IG compared with the CG across all measures | ↔ |
|
Lambert et al [44], 2022 |
|
|
Mean (SD) |
|
|
Significant improvements for the IG compared with the CG across all measures | ↑ |
|
Mitev et al [45], 2021 | Mental well-being | Daily satisfaction question, self-esteem scale, and positive and negative affect scales were combined to create an overall composite score of participants’ well-being | F value and partial eta–squared value | F2,416=0.11; P=.89; ηp2=.001 | 0.06 (X) | No significant difference for the IG compared with the CG | ↔ |
|
Mosquera et al [46], 2020 |
|
|
Mean difference between before and after the intervention (SD) |
|
—t | Significant improvements for the IG compared with the CG for depression but not for life satisfaction | ↕ |
|
Przybylski et al [47], 2021 |
|
|
Mean (SD) |
|
|
No significant difference for the IG compared with the CG | ↔ |
|
Tromholt [48], 2016 |
|
|
Mean (SD) |
|
|
Significant improvements for the IG compared with the CG across all measures | ↑ |
|
Turel et al [49], 2018 |
|
|
Marginal means (95% CI) |
|
|
Significantly reduced absolute stress in the IG but not relative stress compared with the CG | ↕ |
|
Vally and D’Souza [50], 2019 |
|
|
Mean (SD) |
|
|
Significantly reduced life satisfaction and increased negative feelings and loneliness for the IG compared with the CG but no difference in positive affect or stress | ↕ |
|
Vanman et al [51], 2018 |
|
|
Mean difference between before and after the intervention (SD) |
|
0.54 (M) | The IG had significantly reduced life satisfaction compared with the CG but no other outcomes were significantly different | ↕ |
Therapy-based interventions | ||||||||
|
Chen et al [22], 2022 |
|
|
Mean (SD) |
|
|
Significant improvements for the IG compared with the CG across all measures | ↑ |
|
Hou et al [40], 2019 |
|
|
Mean (SD) |
|
|
Significant improvements for the IG compared with the CG across all measures | ↑ |
|
O’Connell [35], 2020 |
|
|
Mean difference between before and after the intervention (N/Rac) |
|
|
Significant reduction in mindfulness after the intervention but no difference in FOMO, well-being, depression, or anxiety | ↕ |
|
Esmaeili Rad and Ahmadi [38], 2018 |
|
|
Mean rankae (N/R) |
|
|
Significant improvements within the IG across all measures | ↑ |
|
Throuvala et al [39], 2020 |
|
|
Mean (SD) |
|
|
Significant improvements for the IG compared with the CG across all measures | ↑ |
|
Zhou et al [37], 2021 | Life satisfaction | SWLS | Mean (SD) |
|
0.50 (M) | Significant improvements for the IG compared with the CG | ↑ |
Interventions involving limited social media use | ||||||||
|
Brailovskaia et al [52], 2020 |
|
|
Mean difference (95% CI) between the groups |
|
|
No significant improvement in the IG compared with the CG across both measures | ↔ |
|
Graham et al [53], 2021 | Well-being | WEMWBS | Mean (SD) |
|
0.01 (X) | No significant improvement in the IG compared with the CG | ↔ |
|
Hunt et al [54], 2018 |
|
|
Mean |
|
— | Significant improvements in depression and loneliness for the IG compared with the CG; no significance for other outcomes | ↕ |
|
Hunt et al [55], 2021 | Depression | BDI | Mean (SD) |
|
— | Significant improvements for IG participants who were highly depressed compared with CG and AG participants | ↑ |
|
Thai et al [56], 2021 |
|
|
Mean (SD) |
|
|
No significant improvement in the IG compared with the CG across both measures | ↔ |
aFOMO: fear of missing out.
bFoMOs: Fear of Missing Out Scale.
cWEMWBS: Warwick-Edinburgh Mental Well-Being Scale.
dP<.001.
eM: medium.
fS: small.
gBeneficial effect.
hSWLS: Satisfaction With Life Scale.
iPANAS: Positive and Negative Affect Schedule.
jIG: intervention group.
kCG: control group.
lP<.05.
mL: Large.
nX: negligible.
oMixed effects.
pSF-36: 36-item Short Form Health Survey.
qNo effect.
rGAD-7: General Anxiety Disorder-7.
sOECD: Organisation for Economic Co-operation and Development.
tNot available (number in each group was not specified to calculate effect size).
uAverage mean and SD reported across the 3 countries because the relationship among variables was the same across the countries. Estimates based on unadjusted means because the adjusted means were not provided.
vCES-D: Center for Epidemiological Studies Depression Scale.
wPSS: Perceived Stress Scale.
xCES-D-20: Center for Epidemiological Studies Depression Scale, 20-item version.
yULS-8: University of California Los Angeles Loneliness Scale.
zGHQ-30: General Health Questionnaire-30.
aaMAAS: Mindful Attention Awareness Scale.
abPWB: Psychological Well-Being Scale.
acN/R: not reported.
adBDI: Beck Depression Inventory.
aeComparing the IG between before the intervention and after. The effect of the intervention between the IG and the CG was not reported and could not be calculated.
afDASS-21: Depression Anxiety Stress Scales 21.
agAG: active group (limited use of social media at 30 minutes per day plus 1 action every 3 minutes, eg, posting and replying).