Background: It is well established that singletons born of assisted reproductive technology are at higher risk of preterm birth and other adverse outcomes. What remains unclear is whether the increased risk is attributable to the effects of the treatment alone or whether the underlying causes of infertility also play a role. The aim of this study was to examine whether any of the six categories of causes of infertility were associated with a direct effect on preterm birth using causal mediation analysis.
Methods: We assembled a hospital-based cohort of births delivered at a large tertiary care hospital in Montreal, Canada between 2001 and 2007. Causes of infertility were ascertained through a clinical database and medical chart abstraction. We employed marginal structural models (MSM) to estimate the controlled direct effect of each cause of infertility on preterm birth compared with couples without the cause under examination.
Results: The final study cohort comprised 18,598 singleton and twin pregnancies, including 1689 in couples with ascertained infertility. MSM results suggested no significant direct effect for any of the six categories of causes. However, power was limited in smaller subgroup analyses, and a possible direct effect for uterine abnormalities (e.g. fibroids and malformations) could not be ruled out.
Conclusion: In this cohort, most of the increased risk of preterm birth appeared to be explained by maternal characteristics (such as age, body mass index, and education) and by assisted reproduction. If these findings are corroborated, physicians should consider these risks when counselling patients.
Keywords: causal mediation analysis; causes of infertility; infertility; marginal structural models; preterm birth.
© 2015 John Wiley & Sons Ltd.