Factors Influencing the Enrollment of Eligible Individuals in Orthopedic Randomized Controlled Trials

Abstract

Purpose: Low rates of subject enrollment are a threat to the external validity of clinical trials, which are necessary to confirm or contradict basic assumptions about clinical management. Our goal was to examine the association of subject enrollment rates in orthopedic randomized controlled trials (RCTs) with characteristics of the interventions being studied, the investigators of the studies, and the publications in which the RCTs are reported. Methods: We performed a search in PubMed/MEDLINE for RCTs involving an orthopedic surgical procedure, comparing different intraoperative interventions, published in English in a peer-reviewed journal during 2003 to 2014, and reporting both the numbers of enrolled and eligible subjects. The primary outcome variable was the enrollment rate, calculated as the number of enrolled subjects divided by the number of eligible subjects. We collected and analyzed data from papers meeting inclusion criteria. Results: The average enrollment rate across all 393 studies meeting inclusion criteria was 84.5% (standard deviation (SD) 16.6%). Trials in the United States and Canada had significantly lower enrollment rates when compared to trials in the rest of the world (72.9% vs. 87.6%, p<0.0001), and trials comparing an operative arm to a non-operative arm had significantly lower enrollment rates than trials comparing two different operative arms (73.1% vs. 86.3%, p<0.0001). The national differences were observed primarily in trials comparing operative and non-operative interventions, in which the average North American enrollment rate was 47.9% (SD 25.9%) and the average enrollment rate elsewhere was 81.1% (SD 15.8%). Conclusions: Trials may have variable rates of success recruiting subjects depending on their location and the difference between the interventions being studied, with North American trials and trials comparing operative and non-operative interventions having lower enrollment rates.