489: Rate of preterm birth in a prematurity prevention clinic after adoption of progestin supplementation

branes. STUDY DESIGN: We performed a decision analysis of six test strategies predicting early preterm delivery, based on data from the APOSTEL1 cohort, a nationwide cohort study in 660 women with threatened early preterm labor (submitted as abstract no 1478706). The strategies included (1) treat if fFN positive, (2) treat if CL 25 mm, (3) treat if both fFN positive and CL 25 mm, (4) treat if fFN positive or CL 25 mm, (5) measure CL and only test fFN if CL is 15-30 mm, and treat if CL 15 mm or if both fFN positive and CL15-30 mm, treat if CL 15 mm or if both fFN positive and CL15-30 mm and (6) treat all patients. Strict protocol adherence was assumed. Costs were either measured in our study or retrieved from the literature. Strategies were evaluated from a healthcare perspective in terms of averted neonatal mortality and morbidity. Effects of uncertainty were evaluated in probabilistic sensitivity analyses. We estimated cost-acceptability curves. RESULTS: Additional fFN testing in case of a CL between 15-30 mm (5) had better outcomes at lower costs than either fFN testing or CL measurement alone. This strategy generated average costs until neonatal discharge €18.814 per patient, as compared to €21.657 per patient for a treat all strategy, while neonatal outcome was comparable (table). Combining fFN with CL is dominant in all acceptable willingness-topay ranges. In the Netherlands, a country with 180.000 deliveries and 25.000 cases of threatened preterm birth annually, CL measurement is used for predicting preterm birth. Here additional fFN testing could save up to €11 million/year. CONCLUSION: In women undergoing a work-up for threatened preterm birth, fFN testing in case of an inconclusive CL measurement is able to decrease health care costs considerably, without major effects on neonatal outcome.