The update of the Cochrane systematic review of trial recruitment interventions has just been published. The review is led by HSRU's Prof Shaun Treweek and also involves Heidi Gardner and Cynthia Fraser. It's a sizeable update, increasing the number of included studies from 45 to 68 but in terms of help to trialists, the evidence base to support trial recruitment decisions is still rather underwhelming.
‘There are more studies but not that much more useful information to support decisions around what to include in a trial recruitment strategy. There is still only high certainty evidence for three things (see figure). We need a bit more focus.’ said Prof Treweek. ‘We’ve introduced a few changes into the review to try and help with focus, and this fits nicely with what we are trying to do with Trial Forge (https://www.trialforge.org) regarding improving trial efficiency generally.’
Those changes include:
- Not talking about high risk of bias studies in the Results and Discussion sections.
- Using structured results presentations in the Results section, along with GRADE, to avoid lots of narrative text. The wording of the structured sentences is consistent and based on the GRADE level of certainty and the importance of the outcome.
- Giving clear guidance in the Implication for methodological research section as to which interventions should be evaluated rather than a vague ‘more research is required’ statement.
- Providing evaluation protocols for the three top priority recruitment interventions in need of evaluation to reduce current uncertainty.
The latter point in particular is an innovation that we hope other reviews might start to pick up: not only saying what should be evaluated but providing a protocol for that evaluation. The three prioritised evaluations are a design called Studies Within A Trial (SWATs) and all three (SWATs 59-61) are publicly registered on the SWAT Repository. The intention is that by making protocols available, more trial groups will be able to build SWAT evaluations into their trials and that by using a common protocol, the results from these evaluations can be pooled.
The review is one of a trio of systematic reviews on recruitment that HSRU is working on as part of a Trial Forge initiative called a Trinity Package (https://www.trialforge.org/about/), which aims to link both qualitative and quantitative systematic review evidence to inform trial process decision-making and research. Heidi Gardner leads the review of non-randomised evaluations of trial recruitment interventions as part of her PhD. The final review looks at factors that influence recruitment and is led by one of our Trial Forge collaborators, Catherine Houghton at the National University of Ireland Galway. The Cochrane protocol for this review is at http://onlinelibrary.wiley.com/doi/10.1002/14651858.MR000045/full.
Prof Treweek thinks things are improving. ‘There are some chinks of light in our review linked to the benefits of coordination and collaboration to speed things up. By doing both reviews and initiating studies, especially SWATs, to fill gaps in evidence, we hope that our work in HSRU will help trial teams move towards a more evidence-informed approach to trial recruitment.’
For more information contact Shaun Treweek (streweek@mac.com).