RESUMO
BACKGROUND: The use of randomized clinical trials (RCTs) is predicated on the idea that the study group is representative of the overall clinical population; however, recent studies have suggested that this may not be the case in obstetrics.1,2 The generalizability of several seminal obstetrical RCTs, including the A Randomized Trial of Induction Versus Expectant Management trial, has been questioned because of the high ratio of assessed-to-enrolled individuals.3 Although difficulty in recruitment for obstetrical RCTs is known,4 it has been postulated that this high ratio increases the likelihood of the study population not being truly representative. Our primary objective was to analyze the assessed-to-enrolled ratio in contemporary obstetrical RCTs. STUDY DESIGN: During a 4-year period (January 2017 to December 2020), we identified all obstetrical RCTs published in 6 journals (The New England Journal of Medicine, Journal of the American Medical Association, Lancet, American Journal of Obstetrics & Gynecology, British Journal of Obstetrics and Gynaecology, and Obstetrics & Gynecology). The journals were manually reviewed by 2 coauthors to ensure all obstetrical RCTs were identified and abstracted accurately. Reported patient recruitment data, per Consolidated Standards of Reporting Trials recommendations,5 were abstracted from each RCT. Based on the primary outcome of the study, the purpose of the trial was categorized as preventative or treatment. Medians were compared using Wilcoxon rank-sum testing, and P<.05 was considered significant. RESULTS: Of 240 obstetrical RCTs that were identified, 193 (80%) reported data on the number of individuals screened and recruited. The median number of individuals screened per trial was 850 (interquartile range [IQR], 323-2582), of which 348 were eligible (IQR, 201-1347) and 300 (IQR, 133-894) were randomized. There was no difference in trials based internationally or in the United States (P=.26). Compared with preventative RCTs, the trials that had treatment as the primary outcome were significantly more likely to screen a larger number of individuals for eligibility (P<.01) and subsequently randomize (P<.02). There was no difference in the median number of individuals who declined enrollment between trials with prevention or treatment as their objective (P=.12). More individuals were lost to follow-up in trials with prevention as the primary outcome (P<.01), although the median number lost was low at 3 (IQR, 0-21). CONCLUSION: The low assessed-to-enrolled ratio found in this study suggested that researchers are casting a large net for patient recruitment. The relatively high eligible-to-randomized ratio suggested that most individuals are willing to participate in obstetrical RCTs. This finding was at odds with previous critiques of obstetrical RCTs that had called in to question study results secondary to high eligible-to-randomized ratios. Further research into how patients are recruited and counseled is indicated, to explore this variation. The ratio of patients assessed for study eligibility to patients enrolled in preventative vs treatment RCTs was not substantially different. Based on this result, patients seemed to be similarly interested in being enrolled to prevent a complication from pregnancy as in treating one. These results supported the concept that most individuals in the obstetrical RCTs represent the eligible population.
