Staff training to improve participant recruitment into surgical randomised controlled trials: A feasibility study within a trial (SWAT) across four host trials simultaneously
Adwoa Parker, Catherine Arundel, Nicola Mills, Leila Rooshenas, Marcus Jepson, Jenny Donovan, Jane Blazeby, Elizabeth Coleman, Laura Clark, Laura Doherty, Catherine Hewitt, Prasanna Partha Sarathy, David Beard, Peter Bower, Stephen Brealey, Paul Brocklehurst, Cindy Cooper, Julie Croft, Lucy Culliford, Joseph Dias, Declan Devane, Sandra Eldridge, Richard Emsley, Sandra Galvin, Elke Gemperlé Mannion, David Jayne, Andrew Metcalfe, Alan Montgomery, Amar Rangan, Chris Sutton, Puvan Tharmanathan, Shaun Treweek, David Torgerson
Abstract
Objective To test the feasibility of undertaking a simultaneous Study Within A Trial (SWAT) to train staff who recruit participants into surgical randomised controlled trials (RCTs), by assessing key uncertainties around recruitment, randomisation, intervention delivery and data collection. Study design and setting Twelve surgical RCTs were eligible. Interested sites (clusters) were randomised 1:1, with recruiting staff (surgeons and nurses) offered training or no training. The primary outcome was the feasibility of recruiting sites across multiple surgical trials simultaneously. Secondary outcomes included numbers/types of staff enrolled, attendance at training, training acceptability, confidence in recruiting and participant recruitment rates six months later. Results Four RCTs (33%) comprising 91 sites participated. Of these, 29 sites agreed to participate (32%) and were randomised to intervention (15 sites, 29 staff) or control (14 sites, 29 staff). Research nurses attended and found the training to be acceptable; no surgeons attended. In the intervention group, there was evidence of increased confidence when pre- and post- training scores were compared (mean difference in change 1.42; 95% CI 0.56, 2.27; p = 0.002). There was no effect on recruitment rate. Conclusion It was feasible to randomise sites across four surgical RCTs in a simultaneous SWAT design. However, as small numbers of trials and sites participated, and no surgeons attended training, strategies to improve these aspects are needed for future evaluations.