Successful recruitment to trials: findings from the SCIMITAR+ Trial

BACKGROUND: Randomised controlled trials (RCT) can struggle to recruit to target on time. This is especially the case with hard to reach populations such as those with severe mental ill health. The SCIMITAR+ trial, a trial of a bespoke smoking cessation intervention for people with severe mental ill health achieved their recruitment ahead of time and target. This article reports strategies that helped us to achieve this with the aim of aiding others recruiting from similar populations. METHODS: SCIMITAR+ is a multi-centre pragmatic two-arm parallel-group RCT, which aimed to recruit 400 participants with severe mental ill health who smoke and would like to cut down or quit. The study recruited primarily in secondary care through community mental health teams and psychiatrists with a smaller number of participants recruited through primary care. Recruitment opened in October 2015 and closed in December 2016, by which point 526 participants had been recruited. We gathered information from recruiting sites on strategies which led to the successful recruitment in SCIMITAR+ and in this article present our approach to trial management along with the strategies employed by the recruiting sites. RESULTS: Alongside having a dedicated trial manager and trial management team, we identified three main themes that led to successful recruitment. These were: clinicians with a positive attitude to research; researchers and clinicians working together; and the use of NHS targets. The overriding theme was the importance of relationships between both the researchers and the recruiting clinicians and the recruiting clinicians and the participants. CONCLUSIONS: This study makes a significant contribution to the limited evidence base of real-world cases of successful recruitment to RCTs and offers practical guidance to those planning and conducting trials. Building positive relationships between clinicians, researchers and participants is crucial to successful recruitment.