Labour induction in Nova Scotia: What has contributed to rising rates?

Abstract

Background: Labour induction, initiated to reduce perinatal harms, is increasing. Labour induction entails risks, so examining trends can inform practice. Therefore, the objectives of this study were to describe the incidence of induction at term gestation (≥37 weeks) in Nova Scotia, and assess the contribution of obstetric characteristics to temporal trends in two gestational periods (37-<39 and 39-<41 weeks). Methods: A population-based retrospective cohort study of singleton pregnancies ≥37 weeks from 2008 to 2022 was conducted with data from the Nova Scotia Atlee Perinatal Database. Analyses were stratified by epoch (2008-2012, 2013-2017, 2018-2019, and 2020-2022) and parity (nulliparous, multiparous). The cumulative incidence function of induction was calculated within each gestational week among individuals still pregnant at the beginning of each week (pregnancies-at-risk approach). Secondly, cause-specific time-to-event Cox models, with gestational age as the time scale variable, were used to estimate cause-specific hazard ratios (csHR) with 95% confidence intervals (CI) for the association between epoch and the rate of induction. Covariates (obstetric characteristics) were added to the models to assess if the association between epoch and induction was explained by these covariates. Results: 89,034 pregnancies were included. The incidence of induction increased over time. The increase was greater at 37-<39 weeks (8.0% in 2008-2012 to 15.6% in 2020-2022) than at 39-<41 weeks (21.1% in 2008-2012 to 28.4% in 2020-2022). Similar patterns were noted when stratified by parity, but induction at later gestational age was more frequent in nulliparous parents than parous parents. The association between epoch and induction remained largely unexplained after adding all covariates, particularly in 37-<39 week analyses and in the 2018-2019 and 2020-2022 epochs compared to 2008-2012 (e.g., 37-<39 weeks: Nulliparous 2020-2022: unadjusted csHR 2.04, 95% CI 1.88-2.20, vs. adjusted [Model 4] csHR 1.68, 95% CI 1.54-1.82). Of the covariates, medicated gestational diabetes, birth parent chronic conditions, and pre-eclampsia contributed most to trends. Conclusion: The incidence of induction has increased in Nova Scotia at both early term and later gestations. Induction trends were largely unexplained even after accounting for multiple birthing parent and fetal characteristics.