Background: The FIRST-ABC trial comprises of two pragmatic, multicentre, parallel groups, non-inferiority randomised clinical trials designed to evaluate the clinical non-inferiority of first-line use of high flow nasal cannula (HFNC) to continuous positive airway pressure (CPAP) in critically ill children who require non-invasive respiratory support (NRS). Objectives: To describe the pre-specified statistical and health economic analysis for the FIRST-ABC trial before completion of patient recruitment and data collection. Methods: The statistical analysis plan was designed by the chief investigators and statisticians. We define the primary and secondary outcomes, summarise methods for data collection and safety monitoring, and present a detailed description of the planned statistical and health economic analysis. Results: The primary clinical outcome is time to liberation from respiratory support. The primary effect estimate will be the adjusted hazard ratio, reported with a 95% confidence interval. As a sensitivity analysis, the primary analysis will be repeated using time to start weaning of NRS. Subgroup analyses will be performed to test for interactions between the effect of allocated treatment group and pre-specified baseline covariates. The health economic analysis will follow the intention-to-treat principle and report the mean (95% confidence interval) incremental costs, quality-adjusted life years (QALYs) and cost-effectiveness up to 6 months. All analyses will be performed separately for each of the two trials, and any results will not be combined. Conclusion: The FIRST-ABC trial will assess the non-inferiority of HFNC compared to CPAP in two parallel trials with shared infrastructure (step-up RCT and step-down RCT). We have developed a pre-specified statistical and health economics analysis plan for the FIRST-ABC study before trial completion to minimise analytical bias. Trial registration: ISRCTN ISRCTN60048867. Registered on 19 June 2019.
- Intensive care
- Oxygen delivery
- Respiratory medicine
- Statistics, epidemiology and research design