Methods Notes: Eligible intervention studies include randomized controlled trials and non-randomized controlled studies of interventions. Interventions must primarily target mental health and include a mental health primary outcomes.

Summary of Results: We have identified and extracted data from two trials. In the first trial, the effect of progressive muscle relaxation (20-30 minutes for 5 consecutive days) on state anxiety was tested among patients in China hospitalized with COVID-19 compared to routine hospital care. The authors reported a standardized mean effect size of Hedges’ g = 1.09 (95% CI 0.50 to 1.67) for 25 intervention patients compared to 26 patients in the routine care group. Trial reporting, however, was poor, and there was serious risk of bias concerns (see table). Furthermore, the effect size reported is approximately double what others have found for relaxation techniques on anxiety [1,2]. The second trial compared a very brief exercise in which paid online volunteers were asked to write psychologically affirming text, and their stress and well-being were compared immediately after to participants who wrote text about a colour.

We have identified 46 registrations for trials that appear to be eligible once completed. All but four of the interventions where delivery method is reported provide videoconference or teleconference interactive interventions or self-help interventions delivered via websites or videos. One intervention provides in-person peer-support for hospital staff. Another intervention provides in-person psychological counseling to COVID-19 patients. A similar intervention provides Basic Body Awareness training for medical staff and COVID-19 patients at a hospital. An intervention in Belgium provides face to face counseling from community health workers. Of the 46 trial registrations, 17 target the general population, 11 target medical staff, 6 target patients infected with COVID-19, 4 target people with pre-existing mental health conditions, 7 target people with pre-existing medical conditions, and 1 targets university students. A trial in the Australian general population plans to enrol 9000 participants in online neighborhood engagement challenges. One trial, in Canadian medical staff, plans to enrol 1000 participants to test its peer-support model. One trial, in the German general population, plans to enrol 600 participants to test an online modular support program. One trial, in the Canadian general population plans to enrol 600 participants to test a virtual fitness intervention. One trial, in the Spanish general population plans to enrol 560 medical staff to test a psychoeducation app. One trial, in the Canadian general population, plans to enrol 490 participants to test a workout app intervention. One trial, in the American general population, plans to enrol 400 participants to test an meditation app; 19 trials will attempt to enrol between 100 and 300 participants; the rest plan to enrol fewer than 100 participants. Registered trials are from Australia (N = 4), Brazil (N=1), Belgium (N=2), Canada (N = 4), China (N = 4), Denmark (N = 1), France (N = 1), Germany (N = 2), Honduras (N=1), Hungary (N=1), Israel (N=1), Iran (N = 5), Malaysia (N=1), Oman (N=1), Spain (N = 5), Sweden (N = 2), Turkey (N = 3), and the United States (N=7). 

Most of the registered trials plan to recruit a small sample of participants and use standard interventions not adapted for COVID-19 (e.g., yoga, exercise). There are several feasibly implemented COVID-19 tailored interventions being tested that may be useful for rapid uptake, pending results. One trial, in Canadian medical staff, plans to enrol 1000 participants to test a peer-support model in a stepped-wedge trial (NCT04373382, results expected February 2022). Another trial, which plans to enrol 600 participants, is testing an online modular support program to encourage coping with COVID-19 in the German general population (NCT04324190, results expected June 2020). Another trial of interest is enrolling 195 participants with the rare autoimmune disease scleroderma and at least mild symptoms of anxiety and testing the effect of a videoconference-delivered group intervention on anxiety (NCT04335279, results expected June 2020).

Comment: Large trials with relevance to the general population and vulnerable groups are pending.



Effects of Interventions on Mental Health Symptoms


  1. Manzoni GM, Pagnini F, Castelnuovo G, et al. Relaxation training for anxiety: a ten-years systematic review with meta-analysis. BMC Psychiatry. 2008;8:41.

  2. Kim HS, Kim EJ. Effects of relaxation therapy on anxiety disorders: a systematic review and meta-analysis. Arch Psychiatr Nurs. 2018;32:278-284.





Summary of

Study Outcomes


Study Quality







*Abbreviations: BAI= Beck Anxiety Inventory; BBQ= Brunnsviken Brief Quality of Life Scale; BDI= Beck Depression Inventory; BRS= Brief Resilience Scale; BVK= Berner Verbitterungsinventar Kurzversion; CAS= Coronavirus Anxiety Scale; CD-RISC= Connor-Davidson Resilience Scale; CES-D= Center for Epidemiologic Studies Depression Scale; CRISIS= CoRonavIruS Health Impact Survey; DASS-21= Depression, Anxiety; DTS= Davidson Trauma Scale; Stress; EPDS= Edinburgh Postnatal Depression Scale; EQ-5D-5L= Quality of Life Measure; FFMQ= Five-Facet Mindfulness Questionnaire; FS= Psychological Flourishing Scale; GAD-7= Generalized Anxiety Disorder; GBQ= Goldard Burnout Questionnaire; GDS= Geriatric Depression Scale; GHQ= General Health Questionnaire; HADS= Hospital Anxiety and Depression Scale; HADS-A= Hospital Anxiety and Depression Scale-Anxiety; HADS-D= Hospital Anxiety and Depression Scale-Depression; HAM-A= Hamilton Anxiety Rating Scale; HAM-D= Hamilton Rating Scale for Depression; HAQ= Health Anxiety Questionnaire; K10= Kessler Psychological Distress Scale; LOT-R= Revised Life Orientation Test; MASQ= Mood and Anxiety Symptom Questionnaire; MBI= Maslach Burnout Inventory; NR= Not Reported; NRS= Non-Random controlled intervention Study; OASIS= Overall Anxiety Severity and Impairment Scale; ODSIS= Overall Depression Severity and Impairment Scale; OSI-R= Occupational Stress Inventory-Revised; PCL-5= PTSD Checklist; PDQ= Parkinson’s Disease Quality of Life Scale; PHQ-9= Patient Health Questionnaire; PROMIS= Patient-Reported Outcomes Measurement Information System; PS = Parenting Stress Scale; PSS/PSS-10= Perceived Stress Scale; PTSS= Post-Traumatic Stress Symptoms Subscale; QWL= Quality of Work Life; RCT= Randomized Controlled Trial; SBQ-R= Suicide Behavior Questionnaire Revised; SEK-27= Emotion regulation skills; SF-12= Short Form Health Survey; SF-36= Short-Form-36 (SF-36) Health Survey - Mental Health Component; STAI= State Anxiety Inventory; S-STS= Sheehan Suicidality Tracking Scale; SUD= Subjective Units of Distress Scale; SUPPH= Strategies Used by People to Promote Health Questionnaire; SWE= Skala zur Allgemeinen Selbstwirksamkeit; UCLA-8= University of California Los Angeles Loneliness Scale; ULS= UCLA Loneliness Scale (Döring & Bortz, 1993); WE-MWBS= Warwick Edinburgh Mental Wellbeing Scale; WHO-5= WHO (Five) Wellbeing Index; WHOQOL-BREF= World Health Organization Quality of Life Questionnaire Bref; ZSAS= Zung Self-rating Anxiety Scale; ZSDS= Zung Self-rating Depression Scale

*Pregnant women were coded under the pre-existing medical condition population; All hospital staff are coded as medical staff.

*Funding: Some registries do not discriminate between funder and sponsor. If only the researcher’s institution was listed as a sponsor, funder was coded as NR.




*Abbreviations: NR= Not Reported; PSS= Perceived Stress Scale; RCT = Randomized Controlled Trial; STAI= State Anxiety Inventory; WE-MWBS= Warwick Edinburgh Mental Wellbeing Scale

*A positive effect size indicates an increase in symptoms




Please contact us if you identify an error or if we have not included an eligible study