- To systematically gather, extract and synthesise recruitment data from stroke rehabilitation trials over a 10 year period and to examine their recruitment efficiency and reporting standards.
- To explore the challenges encountered by stroke rehabilitation researchers when planning recruitment and implementing recruitment methods in randomised controlled trials (RCTs).
- To explore the adequacy of participant sample sizes for (RCTs) of stroke rehabilitation interventions in relation to statistical accuracy and effect size interpretation
Half of the clinical trials experience difficulty reaching their recruitment targets, resulting in underpowered studies with less reliable findings. Only 31% of stroke rehabilitation trials (funded by two UK funders) achieved their original recruitment targets, half were awarded monetary or time extensions, and recruitment problems were identified in 63%. While extensions to trial recruitment periods are possible there are considerable costs associated with such a step. There appeared to be a complex link between piloting trials and recruitment successes. Recruitment improved when changes were made to written materials, exclusion criteria, recruitment targets, recruitment strategy, and numbers of sites.
Working with a comprehensive list of stroke trials (published 2005 to 2015) we applied the following inclusion criteria: evaluation of a stroke rehabilitation intervention, patient participants only, non-pharmacological experimental intervention. We employed no language limitations. Information on recruitment efficiency was extracted; (i) Rate: the percentage of people screened for eligibility that was randomised, (ii) Speed: monthly recruitment numbers by site (iii) and Dropout rates. Information was also extracted on; the number of recruitment sites, number of recruiters, the setting, profession of the recruiter, funding support, ethical approval, type of intervention, targeted impairment, control condition, and country of recruitment. Two independent reviewers screened titles, abstracts, full texts, and data extracted from included RCTs. Discrepancies were resolved by a third reviewer.
To date, we have reviewed 12,939 titles, 1,270 abstracts, 788 full texts. A total of 515 trials were included in our review. We found recruitment inefficiencies in stroke rehabilitation RCTs with 39% of stroke survivors screened being randomised. Subgroup analysis revealed that recruitment efficiency was significantly affected by the type of intervention, stage of stroke survivor rehabilitation, targeted impairment, control condition, and recruitment setting. The second and third phases of our study are currently underway and results are pending.