Remote Exercise Training Intervention During the COVID-19Pandemic:Randomized Controlled Trial

Background: Societal measures to contain the spread of COVID-19 (eg, lockdown and contact restrictions) have been associatedwith decreased health and well-being. A multitude of prepandemic studies identified the beneficial effects of physical exerciseon both physical and mental health. Objective: We report on the feasibility of a remote physical exercise intervention and its stress-buffering potential in 2 untrainedcohorts: a pre-COVID-19 cohort that completed the intervention in 2019 and a lockdown cohort that started the interventionshortly before pandemic-related restrictions were implemented.Methods: In a randomized controlled trial, participants were assigned to either an intervention group (IG; pre-COVID-19cohort: n=7 and lockdown cohort: n=9) or a control group (CG; pre-COVID-19 cohort: n=6 and lockdown cohort: n=6). IGparticipants received weekly individualized training recommendations delivered via web-based support. The intervention periodwas initially planned for 8 weeks, which was adhered to in the pre-COVID-19 cohort (mean 8.3, SD 0.5 weeks) but was extendedto an average of 17.7 (SD 2.0) weeks in the lockdown cohort. Participants'health parameters were assessed before and after theintervention: aerobic capacity was measured as peak oxygen uptake (VO2peak) via cardiopulmonary exercise testing. Depressivesymptoms were scored via the depression subscale of the Brief Symptom Inventory-18. Results: Dropout rates were low in both cohorts in the IG (pre-COVID-19 cohort: n=0, 0% and lockdown cohort: n=2, 16.7%)and the CG (pre-COVID-19 cohort: n=0, 0% and lockdown cohort: n=2, 20%). The mean adherence to the training sessions ofthe IG for both cohorts was 84% (pre-COVID-19 cohort: SD 5.5% and lockdown cohort: SD 11.6%). Aligned rank transformANOVAs in the lockdown cohort indicated deterioration of VO2peakand depressive symptoms from before to after the interventionin the CG but no longitudinal changes in the IG. Analyses in the pre-COVID-19 cohort revealed significant increases in VO2peakfor the IG compared to the CG (P=.04) but no intervention effects on depressive symptoms Conclusions: With low dropout rates and high adherence, the remote intervention was feasible for healthy adults under regularconditions and in the face of pandemic-related stressors. Moreover, our results hint at a stress-buffering effect as well as a bufferingof a lockdown-induced deconditioning of remote physical exercise interventions in the pandemic scenario, which can be used infuture studies to overcome equally stressful periods of life. However, due to limited statistical power, these findings should bereplicated in similar scenarios