Table 2 Summary of included studies

Al-Alawi, 2021

(RCT) [35]

Oman; Community.

Adults in the general population living in Oman during COVID-19 pandemic (N = 60) but data only available for N = 46 (n = 22 intervention group; n = 24 control group); of which: 78% female; mean age 28.51 years (Standard Deviation (SD): 8.70 years).

Internet-based (email-delivered) therapist-guided online therapy focusing on symptoms of anxiety and depression. One session per week for 6 weeks.

Follow-up: 6 weeks.

Control group received an automatic weekly newsletter via email containing self-help information and tips to cope with psychological distress associated with COVID-19.

Outcome measures used: Patient Health Questionnaire-9 (PHQ-9) and General Anxiety Disorder-7 (GAD-7) scale.

Analysis of covariance indicated a significant reduction in the GAD-7 scores (F1,43 = 7.307; P = .01) between the two groups after adjusting for baseline scores. GAD-7 scores of participants in the intervention group were considerably more reduced than those of participants in the control group (β = − 3.27; P = .01). A greater reduction in mean PHQ-9 scores was observed among participants in the intervention group (F1,43 = 8.298; P = .006) than those in the control group (β = − 4.311; P = .006). Although the levels of anxiety and depression reduced in both study groups, the reduction was higher in the intervention group (P = .049) than in the control group (P = .02).

Carbone, 2021

(RCT) [36]

Italy; Community.

Adults in the general population living in Italy during the COVID-19 pandemic (N = 54) (n = 26 intervention group; n = 28 control group): 77% female (N = 41); mean age: 34.34 years (SD: 8.27 years).

Online counselling session focusing on reducing clinical symptoms and increasing wellbeing during the first COVID-19 Italian lockdown. One session (only) lasting 60 min.

Follow-up: none.

Control group were on a waiting list.

Outcome measures used: Self-report questionnaires: BSI Inventory; Positive and Negative Affect Schedule (PANAS); State-Trait Inventory of Cognitive and Somatic Anxiety (STICSA); The Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS)

Compared to the control group, the intervention group showed a significant reduction in anxiety (M ± SD 36.65 ± 8.35 vs 48.04 ± 11.51; p < 0.001).

Dincer, 2021

(RCT) [37]

Turkey; Hospital.

Hospital-based nurses (N = 72 completed the intervention) (n = 35 intervention group; n = 37 control group):

88.8% female; mean age: 33.45 years (SD: 9.63 years).

Brief online form of Emotional Freedom Techniques (EFT) aimed at prevention of stress and anxiety in nurses involved in treatment of COVID-19 patients. One session (only) lasting 20 min.

Follow-up: none.

Control group received no intervention.

Outcome measures used: Subjective Units of Distress (SUD) scale, State-Trait Anxiety Inventory (STAI-1), burnout scale via online survey

• The mean anxiety score reduction on the post-test for the intervention group was highly significant (p < 0.001).

• The mean post-test anxiety score for the control group was not statistically significantly different.

Fiol-DeRoque, 2021

(Parallel RCT) [38]

Spain; Hospital and community.

Healthcare workers (N = 482) (n = 248 intervention group; n = 234 control group): 83.2% female; median age: 42 years; 33.4% nurses; 31.7% physicians;30.5% nurse assistants; 81.3% worked in a hospital setting; 12.7% worked in primary care; 6% worked in home-care settings.

A psychoeducational, mindfulness-based mHealth intervention entitled PsyCovidApp.

Two-week duration.

Follow-up: by telephone between 24 h and 10 days post-intervention using the same questionnaire used at baseline.

Control app comprising brief written information about the mental healthcare of healthcare workers during the COVID-19 pandemic (adapted from a set of materials developed by the Spanish Society of Psychiatry).

Outcome measures used: A composite of depression, anxiety, and stress: Depression Anxiety and Stress Scale (DASS21).

• The Intervention Group presented significantly lower overall scores (suggesting improved mental health) at 2 weeks than the Control Group (adjusted standardized mean difference − 0.29; 95% CI: − 0.48 to − 0.09; p = 0.004).

• The Intervention significantly improved symptoms of:

• Anxiety (adjusted standardised mean differences) (− 0.26; 95% CI-0.45 to − 0.08; p = .004);

• Stress (adjusted standardised mean differences)

(− 0.30; 95% CI − 0.50 to − 0.09; p = .003);

• PTSD (adjusted standardised mean differences)

(− 0.20; 95% CI-0.37 to − 0.03; p = 0.01)

• No significant differences were observed for symptoms of depression.

Gharaati Sotoudeh, 2020

(RCT) [39]

Iran; Hospital

Patients with diagnosis of COVID-19 (N = 30) (n = 16 intervention group; n = 14 control group): 53.3% female; age range: 20 to 70 years; mean age Intervention Group: 41.92 years (SD: 12.2 years); mean age Control Group: 44.7 years (SD: 14.2 years).

Brief Crisis Intervention Package.

Duration: four 60-min sessions for 1 month.

Weekly sessions comprised: (1) greetings and introduction to the package; (2) adjustments skills; (3) responsibility and factualism; (4) spirituality.

Follow-up: 1 month.

Routine care.

Outcome measures used: DASS21, symptom checklist (SCL-25), Quality of Life Assessment developed by the World Health Organisation (WHO-QOL)

• The t-test results showed that the average score of depression, anxiety and stress after the intervention was statistically significant compared to the pre-test (p < 0.05).

• The results of the ANCOVA showed statistically significant differences as a result of the intervention vs usual care in depression, anxiety, stress, mental health, and quality of life (p < 0.05).

Kong, 2020

(RCT) [40]

China; Hospital.

Patients with COVID-19 (N = 144) (n = 13 intervention group; n = 13 control group): 51.4% female; 51.4% aged over 50 years.

Psychological-Behavioural Intervention which included breathing exercises and psychosocial support.

Duration: 10 days.

Breathing exercises performed daily for 20 mins in the morning. Psychosocial support lasted approx. 15 mins.

Delivered by two trained medical staff.

Follow-up: after 10 days treatment.

Usual care.

Outcome measures used: Hospital Anxiety and Depression Scales for Anxiety and Depression (HADS-A and HADS-D); Perceived Social Support Scale (PSSS) self-report.

After a 10-day intervention, the Hospital Anxiety and Depression Scale-Anxiety (HADS-A) score (Mean 6.15 +/−  3.579) and the Hospital Anxiety and Depression Scale-Depression (HADS-D) score (5.92 +/−  3.730) were significantly reduced in the Intervention Group (both p < 0.001).

Li, 2020

(RCT) [41]

China; Hospital.

Patients with COVID-19 diagnosis (N = 93) (n = 47 intervention group; n = 46 control group): 64.5% female; mean age: 48 years; 20.4% had chronic diseases.

Cognitive Behavioural Therapy (CBT).

Comprised: cognitive intervention, relaxation techniques training, problem-solving training, and social support strategy.

Performed once a day in the morning, taking 30 mins to complete. Recorded by nurses. Delivered face-to-face and adjusted to suit individual patient’s needs.

Follow-up: not reported.

Routine care.

Outcome measure used: Chinese version of the Depression, Anxiety and Stress Scale (DASS-21)

• A significant decrease in the means for scales of depression, anxiety and total DASS-21 (DASS-12 is the Chinese version of the Depression Anxiety and Stress Scale-21) were found in both intervention (p < 0.001) and control groups (p = 0.001). Participants in the Intervention Group had a bigger reduction on means for scales of depression and anxiety.

• After the intervention, more participants in the Intervention Group had no depression or anxiety symptoms than in the Control Group, but no statistical differences were found (p > 0.05).

• Compared with participants with chronic disease, participants with no chronic disease had a statistically significantly larger reduction of total DASS-21 scale (mean difference on the DASS-21 scale = − 4.74, 95% CI: − 9.31 to − 0.17; p = 0.04).

• The length of hospital stay was statistically significantly associated with a greater increase in anxiety in the Intervention Group (p = 0.005), whilst no statistically significant association was found in the Control Group (p = 0.29).

Liu, 2020

(RCT) [42]

China; Hospital.

Patients with confirmed COVID-19 (N = 51) (n = 25 intervention group; n = 26 control group): 56% male; mean age 50.41 years (SD: 13.04 years).

Progressive muscle relaxation.

Performed for between 20–30 min per day over a period of five consecutive days.

Follow-up: none.

Routine care.

Outcome measures used: State-Trait Anxiety Scale (STAI).

• The t-test results showed that the average score of anxiety before intervention was not statistically significant [between the two groups] (p = 0.713), and the average score of anxiety after the intervention was statistically significant (Intervention group: Mean 44.96+/−(SD)12.68; Control group: Mean 57.15+/−(SD)9.24; p < 0.001).

Liu, 2021(a)

(RCT) [43]

China; Hospital.

Patients with COVID-19 hospitalised infections (N = 140) (n = 70 intervention group; n = 70 control group); (59% female; 54% aged 36 years and over.

‘WeChat’ intervention group.

Promoted during ward rounds and through daily broadcasts. Included daily broadcasts which provided knowledge about COVID-19 – including prevention, treatment, and recovery measures. Participants encouraged, through a WeChat app platform, to conduct self-introductions, make friends, share experiences, help each other build confidence, satisfy spiritual issues, and soothe stress.

Follow-up: none.

Routine care.

Outcome measures used: State Anxiety Inventory (SAI).

• Results showed that the average State Anxiety Inventory (SAI) score of the trial group was 38.5 ± 13.2, and it was 15.9% lower than the control group (45.8 ± 10.4) resulting in a statistically significant difference (p < 0.001).

• Females, young, well-educated, and those without underlying diseases were more willing to involved in, and more vigorous to fulfil intervention activities and rehabilitation exercises; moreover, they found it easier to understand COVID-19 prevention methods and countermeasures.

Liu, 2021(b)

(RCT) [44]

China; Hospital.

Patients with COVID-19 from five sites (N = 273) (n = 137 intervention group; n = 136 control group): 59% male; mean age intervention group: 43.76 years (SD: 14.31 years); mean age control group: 41.52 years (SD: 11.51 years).

Computerised Cognitive Behavioural Therapy (cCBT)

A self-help intervention delivered through 10 min of self-directed individual therapy per day for 1 week at each trial centre. The cCBT intervention was installed on an iPad only available to research therapists. Therapists first show participants how to use the system before the participants can use the self-help intervention.

Follow-up: one-month post-intervention.

Treatment as usual.

Outcome measures used: Hamilton Depression Rating Scale-17 (HAMD); Hamilton Anxiety Scale (HAMA), Self-Rating Depression Scale, Self-Rating Anxiety Scale.

A mixed-effects repeated measures model revealed statistically significant improvement in depression (p < 0.001), anxiety (p < 0.001), during the post-intervention and follow-up periods in the intervention group compared to the control group.

Malboeuf-Hurtubise, 2021

(Cluster RCT) [45]

Canada; Schools.

Schoolchildren (N = 22): (n = 14 intervention group; n = 8 control group): mean age: 11.3 years; 50% female.

Emotion-based directed drawing intervention. The intervention was group-based, delivered online and remotely. Duration: 5 weeks (1 session per week with each session lasting approximately 45 mins). Content involved: story of a virus (Comic strip); drawing how you feel; drawing viruses with funny names; drawing what you are afraid of and putting it in a bottle and throwing this in the bin; drawing what makes you anxious and where you feel it in your body; drawing your COVID-19 cure; and forecasting how your heart feels.

Follow-up: none reported.

Mandala drawing intervention. Group-based, delivered online and remotely.

Outcome measure used: Behaviour Assessment Scale for Children-3rd edition (BASC 111)

• No statistically significant impact on levels of anxiety or depression in either the intervention or control group as measured by ANCOVA (p = 0.26 for anxiety; p = 0.68 for depression).

• For anxiety: Intervention group had means (SD) pre-test: 3.71(1.48) and post-test: 3.5(1.70); Control group had means (SD) pre-test: 3.25(2.05) and post-test: 2.87(0.83).

• For depression Intervention group had means (SD) pre-test: 2.46(1.71) and post-test: 2.07(1.49); Control group had means (SD) pre-test: 2.62(1.84) and post-test: 2.62(1.50).

Ng, 2006

(RCT) [46]

Hong Kong; Community.

Community Rehabilitation Network for participants with chronic disease (N = 51) (n = 25 intervention group; n = 26 control group): 74.5% female; mean age of intervention group: 53.9 years.

Strength-Focused

and Meaning-Oriented Approach for Resilience and Transformation (SMART) debriefing intervention for people exposed to SARS. One-day psycho-educational intervention.

Follow-up: one-month post-intervention.

No intervention

Outcome measure used: Brief Symptom Inventory (BSI)

Paired t-tests were conducted for changes between baseline (T₀) and immediately at the end of the session (T₁) only for the intervention group:

▪ The Depression score dropped significantly (p < 0.01).

▪ The Anxiety, Somatization and Hostility scores did not.

Repeated-measure ANOVA was conducted for changes between T₀ and 1 month after the intervention (T₂), comparing the intervention group (N = 21 valid responses) and the control group (N = 13 valid responses):

▪ Group effects were found in Personal-Positive (p < 0.01) and Social-Negative scores (p < 0.05);

▪ Depression was the only subscale in BSI which had statistically significant group effect (p < 0.05).

Ozlu, 2021

(RCT) [47]

Turkey; Hospital.

Patients receiving treatment in hospital for COVID-19 (N = 73 – data available for N = 67) (n = 33 intervention group; n = 34 control group): 44.77% female.

Progressive muscle relaxation exercises.

Compact Disk (CD) provided to the intervention group (N = 33) and exercises performed twice a day for 5 days over a period of 3 months. Exercises took approximately 20-30 min to perform.

Follow-up: none reported.

Routine care

Outcome measures used: State-Trait Anxiety Inventory (STAI) (Turkish Version).

• No statistically significant differences were found between the state and trait anxiety levels of the groups, which were determined to be homogeneous (p > 0.05).

• The experimental group’s mean ± SD post-test score on the State-trait Anxiety Inventory Scale (SAS) was 44.67 ± 5.41, and the control group’s mean ± SD post-test score on the SAS was 61.29 ± 7.95.

• A statistically significant difference was found between their mean post-test SAS scores. The mean post-test SAS score of the control group was higher than that of the experimental group (p < 0.05).

• A statistically significant difference was found between the experimental group’s mean pre-test and post-test SAS scores.

• No significant differences were found between the mean pre-test and post-test SAS scores of the control group.

Parizad, 2021

(RCT) [48]

Iran; Community.

Patients with COVID-19 (N = 110) (n = 55 intervention group; n = 55 control group): 54.4% male in the intervention group; Mean age intervention group: 43.14 years (SD: 12.22 years).

Guided imagery under a psychiatrist’s supervision. Ten sessions for five consecutive days, twice a day for an hour and a half. Delivered by audio track via headphones, administered by a nurse. Instructional guided imagery tracks lasted approximately 25 mins. Five different audio tracks. During each session, the patient closed their eyes, took deep breaths, and relaxed their muscles. Then they moved towards relevant imagery using mind and imagination.

Follow-up: none reported.

Routine care

Outcome measure used: Spielberger State-Trait Anxiety Inventory (STAI).

• The results of the paired-samples t-test revealed that the mean scores of the state (p < 0.001) and the trait anxiety (p < 0.001) statistically significantly differed in the intervention group (pre- vs post-intervention).

• The mean scores of the state (p = 0.214) and the trait anxiety (p = 0.629) did not have a statistically significant difference in the control group (pre- vs post-intervention).

• The difference in the mean score of the state (Cohen’s d = 1.10) and trait anxiety (Cohen’s d = 1.07) in the guided imagery group after the intervention compared to before the intervention was large.

• The difference in the mean score of the state (Cohen’s d = 0.16) and trait anxiety (Cohen’s d = 0.06) in the control group was small.

Perri, 2021

(RCT) [49]

Italy; Hospital.

Healthcare professionals requiring psychological support to manage ongoing trauma associated with work on COVID-19 hospital wards (N = 38) (n = 19 intervention group; n = 19 control group): 71.1% female.

Eye Movement Desensitization and Reprocessing (EMDR).

Dispensed online by 8 experienced psychotherapists.

Seven-sessions therapy: two sessions per week for a duration of approximately 3 weeks.

Follow-up: 1 month.

Comparison between EMDR and Trauma-focused cognitive-behavioural therapy (TF-CBT).

Dispensed online by six experienced psychotherapists.

Seven-sessions therapy: two sessions per week for a duration of approximately 3 weeks.

Outcome measures used: Post Traumatic Syndrome Disorder (PTSD) Checklist for DSM-V (PCL-5) – a self-report checklist to assess the presence and severity of PTSD symptoms; State Trait Anxiety Inventory (STAI-Y1); Beck Depression Inventory (BDI-II)

• No intervention was found to be superior to the other. The RM-ANOVAs effect of therapy and therapy time interaction did not reach statistical significance (all p-values > 0.05).

• State anxiety decreased by approximately 30% in both intervention groups after the seven-session treatment.

• Traumatic and depressive symptoms reduced by approximately 55% after the seven-sessions in both interventions.

• RM-ANOVAs yielded a significant main effect of time for the PCL-5 (p < 0.0001), STAI Y-1 (p < 0.0001, 2p = 0.53) and BDI-II scores (p < 0.0001).

• Post-hoc comparisons revealed similar values of EMDR and TF-CBT for all the considered measures, and a significant score decrease from pre- to post-treatment, and from pre-treatment to follow-up for both groups (all Bonferroni-corrected p-values< 0.0001).

• Results were confirmed at a one-month follow-up.

Procaccia, 2021

(RCT) [50]

Italy; Hospital.

Healthcare workers (caring for patients with COVID-19) from two hospitals (N = 55) (n = 30 intervention group; n = 25 control group):54.4% nurses; 27.27% physicians; 18.18% allied healthcare workers; 74.54% female.

Expressive Writing (EW) Intervention.

Participants asked to write for three consecutive days at home for 20mins each describing their thoughts, feelings, and moods.

Duration of study: 3 months.

Follow-up: none reported.

Neutral writing (NW) task.

Participants asked to write at home for three consecutive days for 20 mins each time, but they were asked to describe their experiences objectively.

Outcome measures used: Beck Depression Inventory (BDI-II), Los Angeles Symptom Checklist – a self-report instrument with 43 items which measures overall distress related to trauma exposure, overall PTSD symptomology severity and PTSD symptoms sub-scales.

• For the EW group - Statistically significant interaction effects were found for PTSD symptoms, depression symptoms, and Global Severity Index.

• No effects for social support and resilience were found.

• Plots showed that: (1) PTSD symptoms reduced significatively only in EW group (PTSD writing condition: p = 0.002) (2) depression symptoms reduced in EW group while it increased in NW group (depression writing condition: p = 0.02).

Shabahang, 2020

(RCT) [51]

Iran; College.

College students (N = 150) (n = 75 intervention group; n = 75 control group): age range:18 to 32 years.

Cognitive Behavioural Therapy (CBT) (10 × 90 min sessions, 5-day week). Duration unclear. Delivered by two CBT experts.

Follow-up: none.

Unclear.

Outcome measures used: Short Health Anxiety Inventory, adapted to coronavirus disease anxiety, Somatosensory Amplification Scale, and Beck Depression Inventory–Second Edition.

• The cognitive-behavioural intervention was effective for health anxiety and depression of healthy individuals with high levels of coronavirus anxiety.

- F statistic for health anxiety) was significant at 0.01 level.

- F statistic for depression was significant at 0.01 level.

- Small effect sizes were obtained for anxiety and depression.

Thombs, 2021

(RCT) [52]

Australia, Canada, France, Mexico, Spain, UK and USA; Community.

Patients with self-reported systemic sclerosis diagnosis (N = 172) (n = 86 intervention group, n = 86 control group): mean age: 55 years (SD: 11 years); 94% female; 79% identified as White.

Scleroderma Patient centered Intervention Network COVID-19 Home-isolation Activities Together (SPIN-CHAT) (10 weeks: 4 weeks intervention comprising 3 × 90 min sessions per week, and 6 weeks follow-up). Video-conferencing group intervention providing education and practice with mental health coping strategies and social support to reduce isolation.

Follow-up: 6 weeks post-intervention.

Waiting list participants received reminders to complete trial measures only. (They received the SPIN-CHAT intervention following the six-week post-intervention outcome assessment).

Outcome measures used: Patient-Reported Outcomes Measurement Information System (PROMIS). The primary outcome analysed in the intention-to treat population was anxiety symptoms (PROMIS Anxiety 4a version 1.0) immediately post-intervention.

• The intervention did not significantly reduce anxiety symptoms post-intervention (score difference: − 1.57 points; 95% CI: − 3.59 to 0.45; standardised mean difference [SMD]: − 0.22 points) but reduced symptoms 6 weeks later (score difference: − 2.36 points; 95% CI: − 4.56 to − 0.16; SMD − 0.31 points).

• Depression symptoms were significantly lower 6 weeks post-intervention (score difference: − 1.64 points, 95% CI: − 2·91 to − 0.37; SMD: − 0.31 points).

Vanden Bossche, 2021

(RCT) [53]

Belgium; Community (city of Ghent).

Patients living in Ghent and known by a family physician working in the same urban area (N = 135) (n = 67 intervention group; n = 68 control group): mean age: 60.4 years (range: 19 to 93 years); 62.2% female; 32.6% migrants.

Community Health Worker (CHW) intervention: N = 50 CHWs matched with N = 67 pairs of patients. 8 weeks of tailored psychosocial support for the intervention group.

Follow-up: none after the 8 weeks (end of intervention).

Usual care.

Outcome measures used: Patient-Reported Outcomes Measurement Information System (PROMIS).

• For anxiety and fear of COVID-19, independent-samples t-test results revealed a statistically significant difference (p = 0.049 and p = 0.017, respectively), despite random attribution to intervention and control group. The patients of the intervention group had higher scores (meaning higher levels of anxiety and fear of COVID-19) than the control group patients at baseline.

• The 95% confidence intervals fell entirely between the margins of meaningful change of [5, + 5%], except for anxiety where the mean decrease in anxiety in the control group might be larger than in the intervention group.

• For anxiety: baseline-adjusted mean difference between groups post-intervention: 2.32 (95% CI: − 1.89 to 6.52; p = 0.278).

Vukcevic Markovic, 2020

(RCT) [54]

Most probably Serbia, but not clearly specified; Community.

General population (N = 135) (n = 67 intervention group; n = 68 control group): 70% female in the control group; 77% female in the intervention group; mean age control group: 32.67 years; mean age intervention group: 31.79 years.

Expressive writing (EW) intervention.

Five EW sessions, each lasting 20mins, set 3 days apart, over a 2 weeks period.

In EW the participant is asked to disclose their deepest thoughts and feelings surrounding a stressful life event. The notion is that EW can decrease negative feelings, and improve physical and mental health, by engaging in deep and meaningful writing about a traumatic or difficult event.

Follow-up: 1 month.

Usual care.

Outcome measures used: Depression Anxiety Stress Scale (DASS); WHO well-being index; Manchester Short Assessment of Quality of Life.

• A main effect of group on post-test Depression and Anxiety, after controlling for baseline scores, was not found: p = 0.082, pη2 = 0.030 and p = 0.735, pη2 = 0.001, respectively.

• The study found no evidence that five sessions of remote EW generate benefits in lowering depression, anxiety, and stress, and increasing overall well-being.

• On the contrary, the results showed that engaging in EW during the pandemic elevates the stress level of participants from the intervention group. The same results were obtained when controlling for the baseline results.

Wahlund, 2021

(RCT) [55]

Sweden; Community.

Adults in Swedish population reporting daily uncontrollable worry about COVID-19 and its possible consequences (N = 670) (n = 335 intervention group; n = 335 control group): 42% with previous psychiatric diagnosis; 4% had confirmed COVID-19 diagnosis; 81.64% female.

Brief self-guided online psychological intervention. Three-week duration completely self-guided via encrypted website and organised into five brief modules.

Follow-up: 1 month.

Waiting list of equal duration. Participants randomised to control group were crossed over to the intervention after the controlled study period (week 3).

Outcome measure used: COVID-19 adapted version of the Generalised Anxiety Disorder 7-item scale.

• The primary intention-to-treat mixed effects regression model showed that both groups improved significantly over time (ß = 0.74–1.89; Z = 9.36–19.84; p < 0.001) but the intervention group had a larger reduction in COVID-19-related worry than the control group (ß = 1.14; Z = 9.27; p < 0.001).

• The intervention was associated with significant improvements including depressive symptoms: GAD-7 from post-treatment to the 1-month follow-up (ß = 1.78; Z = 8.06; p < 0.001).

Zheng, 2021

(Cluster RCT) [56]

China; Schools (Grade 7) in the Duanzhou district of Zhaoqing City, Guangdong Province, Southern China.

Children. (N = 954) (n = 485 intervention group; n = 469 control group); in n = 6 schools: mean age: 13.5 years (SD: 0.5 years): 52.3% male.

Digital behaviour change intervention which included: health education information promoting exercise and ocular relaxation, access to digital behaviour change intervention, with live streaming and peer sharing of promoted activities. Duration and follow-up unclear – presumed to be 2 weeks (the only follow-up timepoint).

Health education information only.

Outcome measures used: Self-reported Spence Children’s Anxiety Scale (SCAS) and a parent questionnaire.

The assigned interventions were completed by 896 children (intervention: N = 467, 96.3%; control: N = 429, 91.5%).

• The 2-week change in square-root–transformed self-reported anxiety score was greater in the intervention (− 0.23;, 95% CI: − 0.27 to − 0.20) vs control group (0.12; 95% CI: 0.09 to 0.16); unadjusted difference: − 0.36, 95% CI: − 0.63 to − 0.08; p = 0.02).

• In linear regression models, randomisation to receive the peer-to-peer live-streaming intervention was associated with a statistically significant reduction in self-reported anxiety compared to the controls (ß = − 0.36; 95% CI: − 0.63 to − 0.08; p = 0.02), after adjusting for sex and household income.