A longitudinal analysis of the predictors and consequences of prenatal antidepressant use among women requiring these medications before pregnancy

Women with chronic conditions who become pregnant have a difficult choice to consider: continue pharmaceutical treatment, though it may have teratogenic risks for the fetus, or stop treatment, though the condition itself may harm both mother and baby. Unfortunately, very little is known about prescription medication use in pregnancy among women requiring treatment for chronic conditions prior to pregnancy to help guide their decisions. This is due, partly, to the scarcity of data from population-based studies assessing the consequences of medication use or discontinuation on pregnancy outcomes. It is also due to problems of confounding that complicate efforts to untangle the roles of medication and disease in pregnancy outcomes.In this study, we examined a series of questions to address these issues: Are pregnant women more likely to discontinue antidepressant use than are non-pregnant women, i.e. is pregnancy a major determinant of medication discontinuation? What are the maternal characteristics associated with antidepressant discontinuation in pregnancy? Finally, does maternal antidepressant use and discontinuation have consequences on maternal health? The answers may help us broaden our knowledge of an understudied area, as well as shape clinical guidelines.Our data derive from a large, population-based cohort of women identified through Quebec’s health administrative databases (RAMQ). We compared medication use in pregnancy among women using antidepressants before pregnancy to medication use in matched non-pregnant women, and determined the predictors of antidepressants discontinuation. We then assessed the risk of preeclampsia in women continuing use of antidepressants in pregnancy compared to (a) women who stopped all use in pregnancy; (b) women with a depression diagnosis and no antidepressant use; and (c) women with neither a depression diagnosis nor antidepressant use. Finally, we assessed the risk of miscarriage in women taking antidepressants in the first trimester compared to depressed and non-depressed unexposed women. To account for the risk of induced abortions, which may be high among antidepressant users, and may bias the miscarriage risk estimates, we employed an appropriate correction factor. We found that pregnant women are significantly more likely to discontinue antidepressants compared to non-pregnant women, with discontinuation rates differing within classes of antidepressants. The main predictors of continuing use in pregnancy were factors related to disease severity and overall health (e.g. duration of pre-pregnancy antidepressant use, being on welfare and older age). The risk of preeclampsia among women who continued antidepressants in the first 20 weeks of pregnancy was significantly higher than those who stopped use before pregnancy; discontinuers and depressed, unexposed women did not have a significantly elevated risk compared to non-depressed unexposed women. Women using antidepressants in the first trimester had an increased risk of miscarriage compared to either depressed or non-depressed unexposed women, and these findings persist even after accounting for induced abortions.The findings of this thesis research suggest that pre-pregnancy antidepressant users are likely to discontinue use in pregnancy, and the likelihood of discontinuation depends on disease severity and medication class. Our results support an association between antidepressant use itself and an increased risk of miscarriage and preeclampsia because of the persistent elevated findings in antidepressant users when compared to depressed women, and the higher risks associated with continuers compared to stoppers. While residual confounding by factors related to disease severity cannot be ruled out, our findings are nevertheless relevant to the clinical management of pregnant women requiring the use of antidepressants, and should be considered in physician-patient discussions and decision-making.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.