Umeå universitets logga

Background: Two out of three randomised controlled trials (RCTs) fail to meet their recruitment goals. Recruitment to Efficacy oF Fluoxetine - a randomisEd Controlled Trial in Stroke (EFFECTS), fluoxetine for stroke recovery was slower than anticipated. We aimed to evaluate an intervention to improve recruitment to EFFECTS.

Methods: This stepped wedge, cluster randomised study investigated whether a teleconference with the study personnel and the head of department could enhance recruitment in the ongoing EFFECTS. We included 20 low- and medium recruiting active centres. We excluded high recruiting centres. All centres started as controls and were followed by 60 days of observation. We used block randomisation. The primary outcome was a 20% increase of recruitment within 60 days post intervention compared within 60 days pre intervention. Secondary outcomes were comparing recruitment between different types of centres, that is small versus large or experienced versus non-experienced centres, and university versus non-university hospitals. In exploratory analyses, recruitment within 30 days post versus 30 days pre intervention was compared.

Results: The recruitment increased by 10% at 60 days. We noticed a short-lived increase of 23% the first month. The increased recruitment was most pronounced in low-recruiting, small and non-university hospitals. The recruitment of patients increased after the first contact with the centres where we announced that there would be a conference.

Conclusion: A teleconference with the study personnel and the head of department increased the recruitment by 23% within 30 days and by 10%, 60 days post intervention in this embedded RCT. This implies that this structured intervention aimed at increased recruitment was short-lived and would need frequent repetitions in order to be effective.