An evaluation of the impact and costs of three strategies used to recruit acutely unwell young children to a randomised controlled trial in primary care

Abstract

Background Recruitment to primary care trials, particularly those involving young children, is known to be difficult. There are limited data available to inform researchers about the effectiveness of different trial recruitment strategies and their associated costs. Purpose To describe, evaluate, and investigate the costs of three strategies for recruiting febrile children to a community-based randomised trial of antipyretics. Methods The three recruitment strategies used in the trial were termed as follows: (1) ‘local’, where paediatric research nurses stationed in primary care sites invited parents of children to participate; (2) ‘remote’, where clinicians at primary care sites faxed details of potentially eligible children to the trial office; and (3) ‘community’, where parents, responding to trial publicity, directly contacted the trial office when their child was unwell. Results Recruitment rates increased in response to the sequential introduction of three recruitment strategies, which were supplemented by additional recruiting staff, flexible staff work patterns, and improved clinician reimbursement schemes. The three strategies yielded different randomisation rates. They also appeared to be interdependent and highly effective together. Strategy-specific costs varied from £297 to £857 per randomised participant and represented approximately 10% of the total trial budget. Limitations Because the recruitment strategies were implemented sequentially, it was difficult to measure their independent effects. The cost analysis was performed retrospectively. Conclusions Trial recruiter expertise and deployment of several interdependent, illness-specific strategies were key factors in achieving rapid recruitment of young children to a community-based randomised controlled trial (RCT). The ‘remote’ recruitment strategy was shown to be more cost-effective compared to ‘community’ and ‘local’ strategies in the context of this trial. Future trialists should report recruitment costs to facilitate a transparent evaluation of recruitment strategy cost-effectiveness.