| Using secondary data on more than 40,000 hospital-based births in five countries, this thesis examines the role of select clinical and demographic variables in postpartum blood loss. Each chapter builds on prior work in the field and addresses outstanding questions with important implications for clinical practice.;In chapter one I assess systematically the independent and combined effectiveness of three interventions that are recommended for prevention of postpartum hemorrhage: oxytocin prophylaxis, controlled traction of the umbilical cord, and uterine massage. I also examine for the first time the effect of route of oxytocin administration (intravenous versus intramuscular). Results affirm the importance of oxytocin and controlled cord traction, but suggest that their relative importance depends upon the combination of interventions used. Findings also indicate that route of administration is important only when oxytocin is given alone, and in such cases intravenous administration is superior to intramuscular administration.;In chapter two I examine the relationship between duration of the third stage of labor and postpartum blood loss among women with and without oxytocin prophylaxis. While delays in placental delivery are known to increase postpartum hemorrhage risk, the precise trajectory of risk over the duration of the third stage has not been established. In addition, the potential effect of oxytocin prophylaxis on this relationship is not known. Results indicate that hemorrhage odds increase with third stage duration, peaking at durations of 25 minutes or more. Findings also suggest that oxytocin prophylaxis has a large and consistent effect on absolute hemorrhage odds at all third stage durations, but does not obviate the incremental change in hemorrhage odds over time.;Chapter three examines the role of low and high maternal hemoglobin on postpartum blood loss. While anemia is widely considered a risk factor for postpartum hemorrhage, the role of elevated hemoglobin has not been previously explored. Results suggest that anemia is an important correlate of hemorrhage risk, irrespective of the receipt of oxytocin prophylaxis. However there was considerable country-level heterogeneity in the magnitude of effect, which raises questions about the underlying nature of this relationship. Findings also suggest that elevated hemoglobin may increase hemorrhage odds among those with no oxytocin prophylaxis. |
