Design and implementation of a large and complex trial in emergency medical services

BACKGROUND: The research study titled “Cluster randomised trial of the clinical and cost effectiveness of the i-gel supraglottic airway device versus tracheal intubation in the initial airway management of out-of-hospital cardiac arrest (AIRWAYS-2)” is a large-scale study being run in the English emergency medical (ambulance) services (EMS). It compares two airway management strategies (tracheal intubation and the i-gel) in out-of-hospital cardiac arrest. We describe the methods used to minimise bias and the challenges associated with the set-up, enrolment, and follow-up that were addressed. METHODS: AIRWAYS-2 enrols adults without capacity when there is no opportunity to seek prior consent and when the intervention must be delivered immediately. We therefore adopted a cluster randomised design where the unit of randomisation is the individual EMS provider (paramedic). However, because paramedics could not be blinded to the intervention, it was necessary to automatically enrol all eligible patients in the study to avoid bias. Effective implementation required engagement with four large EMS and 95 receiving hospitals. Very high levels of data capture were required to ensure study integrity, and this necessitated collaborative working across multiple organisations. We sought to manage these processes by using a large and comprehensive electronic study database, implementing efficient trial procedures and comprehensive training. RESULTS: Successful implementation of the study design was facilitated by the approaches used. The necessary regulatory and ethical approvals to conduct the study were secured, and benefited from strong patient and public involvement. Early and continued consultation with decision makers within the four participating EMS resulted in a coordinated approach to study set-up. All receiving hospitals gave approval and agreed to collect data. A comprehensive database and programme of training and support were implemented. More than 1500 paramedics have been recruited to the study, and patient enrolment and follow-up has proceeded as planned. CONCLUSION: Care provided by EMS needs to be based on evidence. Although participants may be experiencing life-threatening emergencies, high-quality pre-hospital research is possible in well-designed and well-managed studies. The approaches described here can be used to support successful research that will lead to improved treatment and outcomes in similar patient groups. TRIAL REGISTRATION: ISRCTN08256118. Registered on 22 July 2014. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s13063-019-3203-0) contains supplementary material, which is available to authorized users.