Screening for bacterial vaginosis prior to delivery: a cost-effectiveness study.

Abstract

The objective of this study was to compare the cost and effectiveness of three strategies for screening and/or treating bacterial vaginosis (BV) during pregnancy prior to delivery: (1) the current standard of care, neither test nor treat for BV (Treat None); (2) test all patients for BV at 36 weeks gestation; treat if positive (Test Treat); and (3) treat all patients undergoing cesarean delivery with intravenous metronidazole at time of surgery (Treat All Cesarean). Effectiveness was defined as avoidance of postpartum surgical site infection. A decision analytic cost-effectiveness model was designed from a third-party payer perspective using clinical and cost estimates obtained from the literature, American College of Surgeons National Surgical Quality Improvement Program participant use file, National Vital Statistics, Medicare costs, and wholesale drug costs. Cost estimates were inflated to 2020 United States dollars. For this study, effectiveness was defined as postpartum surgical site infections (SSI) avoided. In the base case analysis, the current standard of care, Treat None, was the most expensive and least effective strategy, with a mean cost of $59.16 and infection rate of 3.71%. Treat All Cesarean was the most effective and least expensive strategy, with a mean cost of $53.50 and an infection rate of 2.75%. Test Treat was also relatively inexpensive and effective, with an infection rate of 2.94% and mean cost of $57.05. Compared to Treat None, we would expect the Treat All Cesarean strategy to reduce the infection rate by 26%. These findings suggest that treating pregnant patients with intravenous metronidazole at time of cesarean delivery could be an effective and cost-saving strategy. Testing and treating for BV could also be considered a reasonable strategy, as it has the added benefit of preserving antibiotic stewardship. In no analysis was the standard of care the preferred strategy.