Propensity score method for analyzing the effect of labor induction in prolonged pregnancy

摘要

Abstract Introduction There is an ongoing debate on the optimal time of labor induction to reduce the risks associated with prolonged pregnancy. Material and methods Registry‐based study of 212?716 term, singleton cephalic deliveries between 2006 and 2012 in Finland comparing the outcomes of labor induction with those of expectant management in five, three‐day gestational age periods between 40 and 42?weeks (group 1: 40 +0 –40 +2 ; group 2: 40 +3 –40 +5 ; group 3: 40 +6 –41 +1 ; group 4: 41 +2 –41 +4 ; group 5: 41 +5 –42 +0 ). Using Poisson regression, induced deliveries in each of the gestational age periods were compared with all ongoing pregnancies. Propensity score matching was applied to reduce confounding by indication. Results In the gestational age groups 1 and 2, labor induction significantly decreased the risk of meconium aspiration syndrome [relative risk (RR) 0.40, 95% confidence interval (CI) 0.18–0.91 (group 1), RR 0.44, 95% CI 0.21–0.91 (group 2)] but increased the risk for prolonged hospitalization of a neonate [RR 1.30, 95% CI 1.10‐1.54 (group 1) and RR 1.23, 95% CI 1.03–1.47 (group 2)]. In groups 3 and 4, labor induction significantly increased the risk for emergency cesarean section [RR 1.17, 95% CI 1.06–1.28 (group 3) and RR 1.19, 95% CI 1.09–1.29 (group 4)] but still reduced the risk for meconium aspiration syndrome. In group 5, labor induction did not affect the risk for any of the studied outcomes (operative delivery, obstetric trauma, neonatal mortality, respirator treatment, Apgar 7). Conclusions Propensity score matching is a novel approach to studying the effect of labor induction. It highlighted the conflicting maternal and neonatal risks and benefits of the intervention, and supported expectant management as a valid option, at least until close to 42?weeks.