Obstetric technologies: What determines clinical acceptance or rejection of results of randomized controlled trials?

Objective: The acceptance or rejection of obstetric technologies in clinical practice is not always based on the results of randomized controlled trials (RCTs). In this study, possible reasons for this discordance are examined. Study Design: Eleven technologies introduced over the past 30 years were examined. RCT results were determined from the published literature, with emphasis on meta-analyses and the Cochrane reviews. Results were graded for quality of evidence (I-III) and strength of recommendations (A-E) on the basis of criteria adopted by the United States Preventive Services Task Force. Clinical use was based on data from the literature, or when this was lacking, on an estimate, and graded + to +++ or – to – – –, based on degree of acceptance or rejection. Results: There was concordance of results with midtrimester ultrasound (US) imaging, home uterine activity monitoring (HUAM), Doppler velocimetry in high-risk pregnancy, short-term tocolysis, single-course antenatal corticosteroids (BMZ), thyroid-releasing hormone (TRH), and fetal fibronectin (fFN). There was discordance of results with fetal heart rate monitoring (FHRM), fetal blood sampling (FBS), US imaging in late pregnancy, Doppler velocimetry in low-risk pregnancy, and tocolysis beyond 7 days. In several cases there are published RCTs, but clinical acceptance is still evolving (multiple course BMZ, pulse oximetry, and electrocardiogram ST-segment analysis). Conclusion: These observations suggest that technologies which are simple to apply, and have a single endpoint, show concordance between RCT results and clinical use (HUAM, short-term tocolysis, single-course BMZ, TRH, and fFN). Those that are complex to apply (FBS), or interpret (FHRM), and/or do not have a single clear end point (FHRM, Doppler), or have results that are of great personal interest to the patient (FHRM, US imaging, long-term tocolysis), are more likely to show discordance. In several cases despite published RCTs, clinical acceptance or rejection is still evolving (multiple course BMZ, pulse oximetry, and electrocardiogram ST-segment analysis). These inconsistent relationships suggest that the decision-making processes of obstetric providers and patients would be a fruitful area of investigation. (Am J Obstet Gynecol 2003;188:1622-8.)