Objective: It is important to evaluate how a randomized controlled trial (RCT) sample was assembled from the general patient population in order to determine whether a patient differs from those who participated in the trial in a meaningful way. The aim of this study is to assess the adequacy of reporting of the recruitment process of rheumatoid arthritis (RA) patients participating in RCTs with biologic agents.
Methods: We searched PubMed for all reports of RCTs involving etanercept, infliximab, adalimumab, golimumab, certolizumab, abatacept, tocilizumab, and rituximab in RA patients. Data recorded were eligibility fraction, enrollment fraction, recruitment fraction, and number of patients needed to be screened (NNS) in order to randomize 1 participant.
Results: Of the 66 trials included in the analysis, 23 (35%) reported the number of individuals assessed by investigators for eligibility, and 18 (27%) reported the number eligible for participation. Of the studies that reported quantitative recruitment information, the median eligibility fraction was 80.6% (interquartile range [IQR] 71.8-91.1%) and the median enrollment fraction was 100% (IQR 88.4-100%). The median NNS was found to be 1.28 (IQR 1.18-1.43).
Conclusion: A substantial majority of RCTs conducted in RA patients with biologic agents did not provide sufficient information about the patient recruitment process, which makes assessments of external validity difficult. The rate of reporting of the recruitment process in this study was found to be lower as compared to similar studies conducted in different specialties.
Copyright © 2012 by the American College of Rheumatology.