Main Findings
We have developed a novel and user-friendly tool wherein comprehensiveness of outcome reporting in obstetric clinical trials can be graphically represented as a heatmap. This tool will enable trialists, systematic reviewers, and clinicians to critically assess outcome selection in obstetric trials and use this information to contextualize findings of studies and meta-analyses.
Ensuring comprehensiveness of outcome reporting is vital to designing a clinical trial, and its assessment should be integral to conducting a thorough systematic review of clinical trials, in any area of medicine. In obstetrics, where the implications of decisions made during pregnancy can have life-long consequences to both mother and fetus, in regard to life, life-expectancy, and resource-use, it is particularly important that conclusions pertaining to the benefits of an intervention are based on a holistic set of outcomes. Although obstetrical interventions can impact both individuals, researchers have drawn attention to a lack in reporting of fetal-neonatal outcomes,23 and more recently, to the importance of considering long-term maternal outcomes, such as cardiovascular mortality and morbidity from prolonging pregnancies complicated with hypertensive disorders of pregnancy, to improve neonatal outcomes.24 Drawing conclusions based exclusively on outcomes related to one individual may neglect important negative outcomes in the other.
The World Health Organization mandates that “choosing the most important outcome is critical to producing a useful guideline”, acknowledging that “different groups value outcomes differently”.25 A priority-setting exercise that involved 48 clinical trial directors, identified “choosing appropriate outcomes to measure” among their top three priorities, while designing trials.26 The focus of contemporary medicine goes beyond avoiding mortality, to preventing morbidity and improving the quality-of-life.27, 28 Comprehensive selection of outcomes related to these goals is challenging with current financial pressures and practical constraints. Yet, it should not be acceptable for trialists to draw conclusions on effectiveness or efficacy based on a limited set of outcomes that do not ‘improve health’ or ‘prevent infirmity’, in the holistic sense of these terms, especially when reporting these outcomes is feasible. Studying outcomes that are not relevant to patient care contributes to significant, and avoidable, waste of health research dollars and continues to force healthcare professionals to make decisions based on literature that is less than ideal in regard to transparency.
Published trials seem to place greater emphasis on physiological/clinical outcomes, in favour over other often neglected aspects of health such as emotional well-being, patient satisfaction, and resource costs. A systematic review that analyzed 16525 outcomes from 3515 different Cochrane reviews, showed that 83% of reported outcomes were categorized as physiological, while less than a quarter (24%) measured functional impact and quality of life.3 This is also true of obstetric conditions such as pregnancy in obesity and cardiac disease, wherein published studies are found to be lacking in their inclusion of important patient-centric outcomes.5-9 This barrier to translation is further magnified when trying to interpret evidence across studies, due to the vast variation in outcome selection and measurement.8, 9, 29
Other criticisms with the choice of outcomes include the inappropriate use intermediate outcomes as surrogate outcomes,30 the lack of patient perspective when choosing outcomes, under-reporting of adverse events,11 and the use of poorly selected components within composite endpoints.31 Ultimately, many of these gaps originate from a lack of comprehensive outcome selection, the very issue our COR Tool has been designed to mitigate.