P-746 Recruitment in randomised controlled trials: a conundrum

Abstract Study question What factors are associated with recruitment failure within the pre-planned study period? Summary answer Trials with a no-treatment arm, over four inclusion criteria, without a previous pilot and with low funding, are more likely to experience recruitment failure. What is known already Randomised controlled trials are considered to be the best strategy in evaluating the effectiveness of medical interventions, but they are often hampered by failure to recruit enough patients with the pre-planned study period. This represents a form of research waste. Little is known on variables related to recruitment failure within the pre-planned study period. Study design, size, duration This study was designed as a nationwide cohort study and included 71 randomized controlled trials supported by the Dutch Consortium for Women’s Health Research. We included trials initiated between March 1st 2003 and March 1st 2022. We extracted study data and information on predefined barriers and facilitators. Participants/materials, setting, methods Primary outcome was recruitment target not achieved within 6 months after the pre-planned recruitment period. Secondary outcomes included recruitment target not achieved within an extension of at least 12 months and premature termination of the trial. We used logistic regression to evaluate the association between potential indicators of recruitment failure and expressed these as odds ratios (OR) with corresponding 95% confidence intervals (CI). Main results and the role of chance Of the 71 trials included in our cohort, recruitment was not achieved within the pre-planned study period with a maximal extension period of 6 months or 12 months in 38 (54%) and 33 trials (46%), respectively. Indicators associated with higher chances on recruitment failure were a no treatment arm (OR 8.53 (95% CI 1.01 to 72.36)), a compensation fee of less than 200 euros per included patient (OR 3.16 (95% CI 1.18 to 8.45)) and a funding of less than 350.000 euros (OR 4.30 (95% CI 1.56 to 11.64)). Indicators associated with lower chances on recruitment failure were less than 4 inclusion criteria (OR 0.31 (95% CI 0.10 to 0.97)) and a preceding pilot study (OR 0.53 (95% CI 0.15 to 1.85)). Limitations, reasons for caution If we could have accessed an even larger cohort of studies we might have been able to identify more potential indicators for recruitment failure. Wider implications of the findings Investigators and grantproviders should consider our findings before the start of their studies, to optimize recruitmentcircumstances. If a trial is destined to have a suspected long recruitment, it seems wise to consider the relevance and urgency of the research question, and then accept a longer recruitment period with its consequences. Trial registration number not applicable