COVID-19 is an infectious disease caused by a virus called SARS-CoV-2. The spread rate and mortality rate of the virus have had a significant impact on global health and economic systems. Comprehensive treatment and supportive care are required to help COVID-19 patients recover. During the treatment of COVID-19, the respiratory system of patients may be affected, and measures need to be taken to support respiratory function. Currently, awake prone positioning(APP) is an effective method for treating respiratory failure when mechanical ventilation is not feasible or unavailable.

In the prone position, the patient's face is facing the ground, the chest is sinking, and the abdomen is rising. This posture can improve alveolar ventilation and blood flow distribution, reduce pulmonary edema and pulmonary vascular resistance. In addition, the prone position can also improve the coordination of respiratory muscles, promote pulmonary expectoration, and increase functional residual capacity. Although the prone position has been widely used, there is still some controversy over when to use it. Some studies have shown that using the APP for early-stage COVID-19 patients can reduce mortality rates and respiratory dependence time, while others believe that using the APP in the later stages may increase the risk of lung injury.

Therefore, it is necessary to further study when to use the APP and its impact on treatment effectiveness and safety. Such research may need to consider multiple factors, such as the patient's clinical symptoms, the severity of lung lesions, lung function indicators, and the duration of using the prone position. Determining the optimal time to use the APP will help guide clinical physicians to better apply this treatment strategy in the treatment of COVID-19 patients. In addition, the prone position can also be used in the treatment of other respiratory system diseases, so relevant research can also provide references and inspiration for the treatment of other diseases.

The goal of this observational study is to learn about the details which might optimize the effect of APP in COVID-19 patients. The main questions it aims to answer are:

• The optimal time to start in the prone position
• the optimal duration of APP treatment
• the strategies to improve the tolerance of the awake prone position

1. Setting and design

   Retrospectively collect relevant data from the electronic medical records of COVID-19 patients who were admitted to the research center from December 2022 to February 2023.

2. Study Participants

   Convenient sampling of COVID-19 patients admitted to the research center from December 2022 to February 2023.

   Inclusion criteria:

   • age 18 years or older;
   • laboratory-confirmed COVID-19 patients;
   • radiographic evidence of pulmonary inflammation;
   • clear consciousness and ability to communicate verbally;
   • patients who have undergone prone position ventilation at least once during hospitalization;

   Exclusion criteria:

   • patients who received mechanical ventilation and oral intubation immediately after admission or within 2 hours of admission.

3. Sample Size Determination Sample size of the multivariable Cox regression model for patients' overall survival is calculated by using the previously reported method. Based on the generally accepted rule of thumb of 10 events per variable and the final Cox model containing 4 variables, the field size was expected to be 40 events. A sample size of at least 440 patients are warranted based on an estimated 10% event rate and a 10% loss to-review rate among the participants.

4. Data collection

   4.1 Baseline data: Gender, age, comorbidities, body mass index (BMI), smoking status, use of anti-Covid-19 drugs (Paxlovid and Azvudine), use of corticosteroids, and radiological results, etc., as baseline and demographic data.

   4.2 Data related to awake prone positioning: Time from admission to first use of awake prone positioning, oxygen delivery method and oxygen concentration at first use of awake prone positioning, peripheral oxygen saturation during use of awake prone positioning, systolic blood pressure and respiratory rate at first use of awake prone positioning, duration of daily use of awake prone positioning, total number of days using awake prone positioning, achievement of awake prone positioning goals, acceptance of maintenance measures for awake prone positioning, changes in oxygen delivery method and oxygen concentration after first use of awake prone positioning, and adverse events caused by awake prone positioning (dislodged catheter, skin injury, and gastrointestinal reactions).

   4.3 primary outcome: Mortality rate.

   4.4 Secondary outcomes: Intubation, upgrading of oxygen supply equipment, deterioration of respiratory failure, and length of hospital stay.

5. Data collection and management Data is collected through the hospital's electronic medical record management system and recorded in the original registration log. All data is recorded clearly to ensure accurate interpretation and traceability. All data and original documents are kept for a minimum of 3 years, and publication and data sharing agreements are made.

6. Statistical Analysis Quantitative variables will be described using means (standard deviations) or medians [interquartile ranges], while categorical variables will be described using counts (proportions). We will use t-tests to compare normally distributed variables and chi-square tests to compare proportions. For non-normally distributed variables, we will use the Mann-Whitney U test. We will examine the predictive ability of SPO2:FIO2 for various respiratory outcomes and mortality rates by calculating the area under the receiver operating characteristic curve and the Youden index and explore the optimal discriminating threshold. We will then divide study subjects into low and high SPO2:FIO2 groups based on the threshold and compare their characteristics in univariate analysis. In addition, we will compare characteristics of patients with and without adverse respiratory outcomes in univariate analysis. We will further investigate the relationship between adverse respiratory outcomes occurring within 28 days and variables with P values <0.10 in univariate analysis using Cox proportional hazards regression models to analyze the relationship between the time to occurrence of adverse respiratory outcomes (up to 28 days) and these variables. The same methods will be used to explore the relationship between these variables and 28-day hospital mortality rate. We will also use Kaplan-Meier curves to plot the percentage of Covid-19 patients who experience adverse respiratory outcomes or die within 28 hospital days based on the SpO2/FiO2 ratio at first application of awake prone positioning, controlling for these covariates through Cox regression. A significance level of 0.05 will be used for all two-tailed tests. Data will be analyzed using SPSS 26.0.

7. Quality Control

7.1 Establish a unified case information collection register. Before collecting data, all data collectors will receive unified training to clarify the concept of data collection and screening standards, ensuring the consistency of data collection.

7.2 Data collection will be carried out simultaneously by two individuals and entered into the electronic medical record information collection register to ensure accuracy.

7.3 If any unresolved or uncertain issues arise during the data collection process, timely communication with the principal investigator will be conducted to ensure consistency of the information.

Inclusion Criteria:

• age 18 years or older;
• laboratory-confirmed COVID-19 patients;
• radiographic evidence of pulmonary inflammation;
• clear consciousness and ability to communicate verbally;
• patients who have undergone prone position ventilation at least once during hospitalization;

Exclusion Criteria:

• Patients who were mechanically ventilated with oral intubation immediately after admission or 2 h after admission.

COVID-19 patients admitted to two tertiary hospitals in Shanghai, China from December 2022 to February 2023