Statistical Analysis
This meta-analysis will be carried out on studies which are conducted in multiple countries, which increases effect size estimates across different conditions as well as the external validity of the results. We plan a prospective meta-analysis of individual patient data. Common variables from all datasets will be combined to conduct the analysis. Analyses will follow the intention-to-treat principle, considering all participants in the treatment group to which they were assigned.
We will use a Bayesian approach for analysis of the pooled individual patient data and for frequent monitoring of the accumulating data for efficacy and safety. We also plan to undertake additional subgroup analyses of the following variables: protocol individual studies (which includes dose and nebuliser system), severity of COVID-19 (according to the PaO2/FiO2 or SpO2/FiO2 ratio and the modified ordinal scale), duration of intervention, time from admission to start of intervention, time from development of symptoms to start of intervention, influence of other therapies, age of the patients. We describe our detailed statistical analysis plan in a separate manuscript in this journal. [reference to SAP]