Abstract
To assess studies which compare or recommend perinatal methods of treatment or management, all such studies (86) published in 1979 in J. Pediatrics, Pediatrics, Am. J. Ob.Gyn, and Ob.Gyn.were reviewed independently by one of two neonatologists and one of two biostatisticians. A 5 page evaluation form was completed by each reviewer. Items noted by both reviewers included prospective design (48%), predetermined criteria for study completion (3%), adequate number of subjects (15%), adequate description of subjects (39%), clear indications for treatment (47%), use of blinding when feasible (36%), evaluation of treatment hazards (45%), and use of a randomized trial to support treatment recommendations (22%). Both reviewers considered the treatment recommendations fully justified in 10% of articles and partially justified in 69%. An overall score was calculated based on the 34 items considered most important to a well executed study. Despite disagreement on individual items, the total score for important items was highly correlated (r=.99). The mean score for all studies was 16.9 for biostatisticians (SD=6.5) and 17.2 for neonatologists (SD=6.5). The range was 3 to 32. Despite tragic past errors in the use of diethylstilbestrol, oxygen, sulfonamide, and chloramphenicol, poorly designed and executed studies are often the basis for therapeutic recommendations in perinatal medicine.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Tyson, J., Furzan, J., Reisch, J. et al. 696 QUALITY OF PERINATAL THERAPEUTIC STUDIES. Pediatr Res 15 (Suppl 4), 558 (1981). https://doi.org/10.1203/00006450-198104001-00719
Issue Date:
DOI: https://doi.org/10.1203/00006450-198104001-00719