The role of pilot testing for a randomised control trial of a complex intervention in critical care

Abstract

Randomised controlled trials (RCTs) involving diverse population groups in complex health care settings can be difficult to successfully undertake and pose practical and methodological challenges. For this reason, undertaking a pilot study is recommended by the Medical Research Council prior to conducting a definitive trial. This paper reports a pragmatic review of the challenges encountered by the research team during a single centre pilot RCT and discusses how feasible it was to deliver and evaluate a complex intervention in practice. The psycho-social intervention designed for this trial was a ‘user-centred critical care discharge information pack’. The pack was designed to support patients and their families during and after their discharge from a critical care unit. It consisted of two written booklets, one for the patient and one for the family member. In total, 221 patients met our inclusion criteria, of whom 158 (71%) were recruited. The pilot RCT identified important lessons for the design of future trials. Challenges included those associated with the recruitment of potential participants, assessing capacity and obtaining informed consent. Problems with attrition, intervention delivery and the choice of data collection tools and time-points were also apparent. Our findings demonstrate that the evaluation of complex interventions is feasible in clinical practice. The importance of adhering to a robust research protocol, maintaining efficient and effective communication between researchers and clinical staff is emphasised. Findings further support the importance of conducting a pilot study prior to embarking on a definitive RCT.