Mode of conception and risk of spontaneous vs. provider-initiated preterm birth: population-based cohort study

Abstract

To study the association between mode of conception and risk of preterm birth, including, spontaneous and provider-initiated subtypes. Population-based retrospective cohort study. Not applicable. All singleton livebirths and stillbirth in Ontario, Canada, 2006-2014. The main exposure was mode of conception, namely unassisted conception, infertility without fertility treatment (i.e., known infertility but conceived without assistance), ovulation induction (OI) or intrauterine insemination (IUI), and invitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI). Modified Poisson regression generated risk ratios (RRs) and 95% confidence intervals for the association between exposure categories and preterm birth adjusted for clinically relevant covariates using a propensity score. The primary outcome was preterm birth <37 weeks, further categorized as spontaneous or provider-initiated subtypes. The secondary outcome was preterm birth <34 weeks. We included 732,810 singleton births born to 649,918 mothers, of which 646,926 (88.3%) were from an unassisted conception, 68,822 (9.4%) with infertility but no fertility treatment, 9,024 (1.2%) following OI/IUI, and 8,038 (1.1%) following IVF/ICSI. Preterm birth <37 weeks occurred among 6.0% of births by unassisted conception, 7.7% with infertility without fertility treatment, 8.0% with OI/IUI, and 10.8% following IVF/ICSI. Relative to unassisted conception, the unadjusted RR of provider-initiated preterm birth was 1.30 (1.26-1.33) in women with infertility without fertility treatment, 1.36 (1.26-1.45) after OI/IUI, and 1.82 (1.70-1.93) after IVF/ICSI. The corresponding adjusted RRs (aRR) were 1.23 (1.16-1.31), 1.48 (1.29-1.69), and 2.35 (2.09-2.64). The unadjusted RR of spontaneous preterm birth was 1.22 (1.18-1.27) in women with infertility without fertility treatment, 1.22 (1.12-1.34) after OI/IUI, and 1.47 (1.35-1.60) after IVF/ICSI. The corresponding aRR were 1.15 (1.10-1.19), 1.19 (1.09-1.31), and 1.40 (1.27-1.53). For preterm birth <34 weeks, the RRs followed a similar pattern as for preterm birth <37 weeks, with the exception of women with infertility without fertility treatment (aRR 1.08; confidence interval, 0.95-1.23). Infertility and receipt of fertility treatment are each associated with a higher risk of preterm birth, spontaneous and provider-initiated subtypes, even in singleton pregnancies. Strategies are needed to reduce the risk for preterm birth in these women.