The COVID epidemics is responsible for a huge number of death following COVID acute respiratory failure. First instance treatment includes oxygenotherapy up to 15L/min in spontaneous ventilation. However COVID infection can ultimately lead to an acute respiratory distress syndrome (ARDS) requiring mechanical ventilation in the intensive care unit (ICU). Guidelines on ARDS management are based on small ventilation volume (6 mL/kg), a pulmonary end expiratory pressure (PEEP) chosen to get the best pulmonary compliance, a plateau pressure lower than 30 cm of water and daily prone positioning when PaO2/FiO2 ratio is lower than 150. In ventilated ARDS patients, prone positioning has shown survival improvement. Though they applied this optimized management of ARDS patients, Chinese intensivists have recently reported mortality rate higher than 50% in ARDS COVID patients requiring intubation and mechanical ventilation. Before being intubated and admitted to ICU, COVID patients require increasing rate of oxygen delivery. From the start of the epidemics, we have observed that an oxygenotherapy rate higher than 3L/min at the initial phase of the disease was associated with a high risk of severe acute respiratory distress (30%)

The investigators hypothesize that prone positioning in patients in spontaneous ventilation (not tubed) from the stage of oxygenotherapy higher than 3L/min (to get an SpO2 of 95% or higher) would prevent respiratory worsening and the need for intubation. Prone positioning is easy to apply in patients in spontaneous ventilation since they can change position by themselves.

Patients admitted to the hospital with laboratory-confirmed SARS-CoV-2 infection as determined by PCR and/or CT scan showing typical radiological findings (ground glass abnormalities) and requiring O2 3L/min to get an SpO2 higher or equal to 95%.

Patients will be randomised for their position : a group with prone positioning at least 6 hours a day during hosital stay until oxygen weaning or intubation (interventional group) and a group with no instruction regarding positioning (control group). In the interventional group, patients will be suggested to spend 6 hours or more in prone position and will report themselves the time spent in prone position each day. In the control group, patients won't be given any instruction regarding positioning.

Patients will be clinically assessed as usual : monitoring 4 to 8 times a day in the ward and continuous monitoring in ICU.

The need for admission to ICU, high flow nasal oxygenation, non invasive ventilation or intubation or the occurrence of death will be recorded.

Inclusion Criteria:

• Patients admitted to the hospital
• Laboratory-confirmed SARS-CoV-2 infection as determined by PCR and/or CT scan showing typical radiological findings (ground glass abnormalities)
• Need for O2 3L/min to get an SpO2 higher or equal to 95%.
• Patient able to understand and to get in prone postion themself
• No therapeutic limitation

Exclusion Criteria:

• Age > 80 years
• Pregnancy
• Impossibility to get in prone position