The aim of the study was to evaluate various methods of chest compressions in patients with suspected/confirmed SARS-CoV-2 infection conducted by medical students wearing full personal protective equipment (PPE) for aerosol generating procedures (AGP). This was prospective, randomized, multicenter, single-blinded, crossover simulation trial. Thirty-five medical students after an advanced cardiovascular life support course, which included performing 2-min continuous chest compression scenarios using three methods: (A) manual chest compression (CC), (B) compression with CPRMeter, (C) compression with LifeLine ARM device. During resuscitation they are wearing full personal protective equipment for aerosol generating procedures. The median chest compression depth using manual CC, CPRMeter and LifeLine ARM varied and amounted to 40 (38-45) vs. 45 (40-50) vs. 51 (50-52) mm, respectively (p = 0.002). The median chest compression rate was 109 (IQR; 102-131) compressions per minute (CPM) for manual CC, 107 (105-127) CPM for CPRMeter, and 102 (101-102) CPM for LifeLine ARM (p = 0.027). The percentage of correct chest recoil was the highest for LifeLine ARM - 100% (95-100), 80% (60-90) in CPRMeter group, and the lowest for manual CC - 29% (26-48). According to the results of this simulation trial, automated chest compression devices (ACCD) should be used for chest compression of patients with suspected/confirmed COVID-19. In the absence of ACCD, it seems reasonable to change the cardiopulmonary resuscitation algorithm (in the context of patients with suspected/confirmed COVID-19) by reducing the duration of the cardiopulmonary resuscitation cycle from the current 2-min to 1-min cycles due to a statistically significant reduction in the quality of chest compressions among rescuers wearing PPE AGP.