JAMA Pediatrics published Antidepressant Use During Pregnancy and the Risk of Autism Spectrum Disorder in Children by Boukhris et al. on December 14th, 2015. The numbers quoted in the abstract seem very scary, and the media went wild with inaccurate data interpretations. Women who are or have been on antidepressants during pregnancy panicked. The Reproductive Psychiatry community through Postpartum Support International worked together to write response pieces to clarify the findings, highlight the study’s limitations, and reduce unwarranted fear.
As the Research Chair on the Executive Board of Postpartum Support International, I am proud that we were able to respond quickly and accurately, and I hope that our response can help women everywhere understand the facts and minimize undue distress.
We write on behalf of Postpartum Support International (PSI), the leading organization dedicated to helping women suffering from perinatal mood disorders, and its clinical member group, the Reproductive Psychiatry Group. As a group, we are deeply concerned by the harmful repercussions of Boukhris et al.’s study ‘Antidepressant Use During Pregnancy and the Risk of Autism Spectrum Disorder in Children’, just published in JAMA Pediatrics. This study and the resulting media fervor may have harmful repercussions on women who may previously, currently or in the future need antidepressants during pregnancy for their mental health and well being.
We hope to highlight serious methodological errors in this study. First, epidemiological studies such as this one fall short of identifying unique risks related to antidepressants during pregnancy. There is already strong evidence that mothers with untreated depression are more likely to have a child with autism. The genetic inheritability of autism was found to be more than 90%, and the genes found to be involved in depression overlap with those playing a role in autism. Despite these clear genetic links, the authors failed to control for family history of both parents when designing this recent study.
In terms of results, the findings of an association between prenatal exposure to antidepressants and offspring with autism were no longer significant once autism diagnosis was made only by a neurologist or by a psychiatrist. Furthermore, the risk of autism was significantly decreased after controlling for maternal history of depression.
Finally, the authors frame the main finding as an ‘87% increase’ in offspring of mothers who took antidepressants during pregnancy, while in fact the absolute risk increase was less than 1%.
Many studies have explored this topic, yet unfortunately the media focuses on papers with positive findings, regardless of the veracity of the work. Two larger, well-controlled epidemiological studies in 2013 (one including nearly 630 000, and the other nearly 670,000 Danish children) failed to find an association between prenatal exposure to antidepressant medication and autism spectrum disorders. We implore the mainstream media to present the negative studies to help women feel more secure, not just the positive studies that feed into their greatest fears.
A risk versus benefit analysis occurs daily among women, their partners, and clinicians involved in reproductive psychiatry. Alarmist headlines, such as those we are seeing worldwide in response to this study, cause many women with a history of depression to stop their antidepressant medication, increasing their chance of relapsing to more than 50%. Both antenatal and postnatal depression have known and quantifiable risks to mother and child, and every treatment decision for a pregnant woman must balance those known risks against any risks of medication. As a society, we must be very cautious about overly and falsely advertising the risk associated with psychiatric treatment. PSI, the Reproductive Psychiatry Group, and its members support well-designed research, accurate use of statistics, and thoughtful publication to help families and providers make well-informed decisions regarding the use of psychotropic medication during pregnancy.