Cluster Randomized Trial of an Active, Multifaceted Educational Intervention Based on the WHO Reproductive Health Library to Improve Obstetric Practices

Abstract

The hypothesis underlying this study was that hospital obstetrical practices can be improved by interactive workshops that promote the use of evidence-based measures based on the World Health Organization's Reproductive Health Library (RHL). The RHL is an annually updated electronic publication featuring Cochrane reviews in the area of reproductive health along with expert commentary, practical guidance documents, and various aids such as educational videos. Participating in the trial were 22 hospitals in Mexico City and 18 in northeast Thailand, whose maternity units conducted more than a thousand deliveries each year, and which had no direct affiliation with a university or other academic or research facility. Three workshops were held during a 6-month period: the first provided information about the project and principles of evidence-based decision-making; the second focused on RHL content; and the third concerned how best to implement desired changes. A stratified cluster randomized design took country, type of hospital, and number of births per year into account. The chief outcome measures were changes in 10 clinical practices as recommended in the RHL, implemented within 4 to 6 months after the third workshop. The intervention failed to influence targeted practices in any consistent or substantive way. Three of the 10 practices (iron/folate supplementation, postnatal uterotonic use, breast feeding on demand) already were part of the practice regimen, and it proved impractical to measure external cephalic version. Selective rather than routine episiotomy was performed more often in the intervention hospitals than in control hospitals. In Mexico, antibiotics tended to be used more at the time of cesarean delivery. No significant differences were documented in companionship during labor, the use of magnesium sulfate to treat eclampsia, corticosteroid treatment of women delivering before 34 weeks' gestation, or the use of vacuum extraction. Awareness of the RHL increased substantially following the intervention at both study sites, as did its use. Despite the efficacy of active educational measures in expanding access to the RHL and promoting its use, this study failed to demonstrate lasting or substantial changes in reproductive medical practices. Within a 10- to 12-month time frame, at least, access to current knowledge of such practices apparently does not assure that they will become part of standard obstetrical care.