Antidepressants in Pregnancy: No Link to Autism, ADHD

Batya Swift Yasgur, MA, LSW

April 21, 2017

Use of antidepressants before and during pregnancy does not cause autism, or attention-deficit/hyperactivity disorder (ADHD) new research shows.

Three studies demonstrate that antidepressant use in pregnant women is likely not responsible for autistic spectrum disorders (ASDs) in children and that the association found in previous studies was likely due to confounding factors.

A team of researchers headed by Simone N. Vigod, MD, of Women's College Hospital, Toronto, Canada, evaluated the association between gestational serotonergic antidepressant exposure and childhood ASD in 35,906 births.

In the 2837 pregnancies (7.9%) involving exposure to antidepressants, 2% of children were diagnosed with ASD, a percantage that was higher than in the children who had not been exposed. However, once the researchers adjusted for confounding factors, the difference was no longer significant. Nor was it significant when exposed children were compared with unexposed siblings.

"We will never be able to say for certain that a medication has zero risks," Dr Vigod told Medscape Medical News. "But these data, together with the findings of the other studies published this week, are reassuring and suggest that antidepressants during pregnancy are reasonably low risk."

In another study, Brian M D'Onofrio, PhD, of Indiana University, Bloomington, and colleagues focused on the confounding factors between first-trimester antidepressant exposure and birth and neurodevelopmental problems.

The researchers analyzed a database of more than 1.5 million Swedish offspring born between 1992 and 2012 for whom follow-up was available through 2013. Of these, 1.4% (n = 22,544) were born to mothers who self-reported antidepressant use during their first trimester.

The researchers used sophisticated statistical methods to account for a variety of confounding factors. After controlling for these factors, they found that first-trimester antidepressant exposure, as compared to nonexposure, was associated with a "small increased risk of preterm birth," but not with increased risk for having small size for gestational age, ASD, or ADHD.

"Our study suggests that it is not the medication use during pregnancy per se that explains why children of mothers who use these medications have more problems than children whose mothers do not," Dr D'Onofrio told Medscape Medical News.

The studies were published in the April 18 issue of JAMA.

The third article is a meta-analysis conducted by Florence Gressier, MD, PhD, of the Bicêtre University Hospital, Le Kremlin-Bicêtre, France, and colleagues. The researchers analyzed 10 studies with "inconsistent results" regarding the association of ASD with fetal exposure to antidepressants for each trimester and during the preconception period.

Although the researchers found a significant association between increased ASD risk and maternal use of antidepressants during pregnancy, the association was more consistent during the preconception period than during each trimester. Moreover, the association was weaker when the researchers controlled for past maternal mental illness.

"Previously reported differences between exposed and unexposed children may come from statistical fluctuations, different criteria of evaluation, and different modelings," Dr Gressier told Medscape Medical News.

This study was published in the April 17 issue of JAMA Pediatrics.

Because it is not feasible to conduct randomized controlled trials of antidepressant safety during pregnancy, researchers have been forced to rely upon observational studies, which have found an association between the use of serotonergic antidepressants and ASD.

However, the association may be attributable not to causal serotonergic mechanisms but rather to the indication for antidepressant treatment, genetic contributions, and poor health practices during pregnancy, according to Dr D'Onofrio and colleagues.

The presence of these confounders renders the findings of the previous studies "inconclusive," Dr Gressier and coauthors note, and necessitate further clarification.

Using varying study designs and populations, three three studies sought to investigate the association between serotonin use during pregnancy and ASD and the potential role of confounders in this association.

No Increased Autism Risk

Dr Vigod's team used a large administrative health dataset to conduct a retrospective cohort study of children born to mothers (aged 16 to 50 years) for whom public prescription drug coverage was available during pregnancy in Ontario, Canada, between 2002 and 2010 (n = 42,274 deliveries among 35,906 women).

The researchers used a process called inverse probability of treating weighting, which is based on a high-dimensional propensity score (HDPS). This approach makes it possible to identify confounding factors in administrative data in order to "balance exposure group difference."

The final cohort consisted of 35,906 singleton births; the mean duration of follow-up was 4.95 years. Of these, 2837 pregnancies (7.9%) involved exposure to serotonergic antidepressants during pregnancy.

Users of serotonergic antidepressants differed in baseline characteristics from nonusers, with users having a more severe psychiatric history, including psychiatric diagnoses and psychiatric emergency department visits.

In the overall cohort, 1.1% of children (95% confidence interval [CI], 1.0% - 1.2%) were diagnosed with ASD by the end of follow-up, compared to 2.0% (95% CI, 1.6% - 2.6%) of children with ASD born to women exposed to serotonergic antidepressants.

By the end of follow-up, the risk for ASD was 4.51 per 1000 person-years among exposed children, as compared to 2.03 per 1000 person-years among unexposed children (between-group difference, 2.48 [95% CI, 2.33 - 2.62] per 1000 person-years; hazard ratio [HR], 2.16 [95% CI, 1.64 - 2.86]; adjusted HR, 1.59 [95% CI, 1.17 - 2.17]).

However, the association was no longer significant after inverse probability of treatment weighting, based on the HDPS (HR, 1.61 [95% CI, 0.997 - 2.59]). The association was also not significant when exposed children were compared with unexposed siblings in the crude analysis or after adjustment for maternal age, parity, and calendar year of delivery.

"We tried to use creative methods to improve upon previous studies by using HDPS, a special method that looks at many variables in healthcare claims that together can act as proxy for confounders we may not be able to measure," Dr Vigod said.

"When we did this, we no longer found an increased risk of ASD attributable to antidepressant drugs," she reported. "And when we compared children of the same mother, where one sibling was and the other was not exposed —meaning that genetics factors were accounted for — we also found no increased risk for autism."

Role of Confounders

Like Dr Vigod and his team, Dr D'Onofrio and colleagues set out to investigate the role of confounding factors, focusing on first-trimester antidepressant exposure and birth and neurodevelopmental problems.

Along with analyzing measured variables, the researchers utilized three additional approaches to analyze data from a retrospective cohort of Swedish offspring born between 1995 and 2012 (n = 1,580,629) who were followed up through 2013.

The researchers used complex sensitivity analyses that controlled for pregnancy and maternal and paternal covariates and also included sibling comparisons, timing of exposure comparisons, and paternal comparisons. Their analysis incorporated both women's self-report of medication use and data regarding medications that were dispensed.

The main outcomes were preterm births (<37gestational weeks), small size for gestational age (birth weight <2 SDs below the mean for gestational age), and first inpatient or outpatient clinical diagnosis of ASD or ADHD.

Offspring exposed to maternal first-trimester antidepressants (n = 22,544 offspring [1.4%]) were born to 943,776 mothers. Of these, 82% (18,470) were exposed to selective serotonin reuptake inhibitors.

Of the exposed offspring, 6.98% were preterm, 2.54% were small for gestational age, 5.28% were diagnosed with ASD by age 15 years, and 12.63% were diagnosed with ADHD by age 15 years, compared to 4.78%, 2.19%, 2.14%, and 5.46% of unexposed offspring, respectively.

At the population level, first-trimester exposure to antidepressants was associated with all outcomes, compared with nonexposure.

But when the researchers adjusted for confounders such as pregnancy and maternal and paternal traits in models that compared siblings, first-trimester antidepressant exposure was associated only with preterm birth (OR, 1.34 [95% CI, 1.18 - 1.52]), but not with being small for gestational age (OR, 1.01 [95% CI, 0.81 - 1.25]), ASD (HR, 0.83 [95% CI, 0.62 - 1.13]), or ADHD (HR, 0.99 [95% CI, 0.79 - 1.25]).

Medication Not the Culprit

Dr D'Onofrio elaborated on the unusual "advanced designs used to pull apart the hypothesis of a causal effect of antidepressants on autism from other factors, such as depression, stress, genetic factors, or poor prenatal care.

"We used a very large population sample, we measured women's self-report as well as medication dispensed from a pharmacy — most studies rely on only one of these. We compared differentially exposed siblings, and we explored medication use both during and before pregnancy. We also looked at the father's use of antidepressants during pregnancy as an indicator of factors that influence either parent to have depression," he said.

The study "suggests that it is not medication use during pregnancy that explains why children with gestational antidepressant exposure have more problems, but that there are an array of other genetic and environmental factors involved," he said.

Dr Gressier and colleagues investigated the association of prenatal antidepressant exposure and ASD by conducting a meta-analysis of 10 studies encompassing not only pregnancy as a whole but also each individual trimester and the time preceding pregnancy.

For prenatal exposure, the six case-controlled studies (n = 117,737 participants) demonstrated a positive association between antidepressant exposure and ASDs (OR, 1.81; 95% CI, 1.49 - 2.20). However, when the researchers controlled for past maternal mental illness, the association weakened (OR, 1.52; 95% CI, 1.09 - 2.12). A similar pattern was found when each trimester of exposure was considered separately.

Preconception exposure to antidepressants, which was examined in four studies, was found to be significantly associated with an increased risk for ASDs. But when the researchers pooled the findings of two cohort studies (n = 772,331 patients) for the whole pregnancy, no association was found (HR, 1.26; 95% CI, 0.91 - 1.74).

"As we also observed an association with preconception exposure, the increased risk for ASDs in children exposed in utero could well be related to maternal depression or other psychiatric disorders rather than to antidepressant medications," Dr Gressier said.

Hitchhiker Drugs

Commenting on the findings for Medscape Medical News, Tim F. Oberlander, MD, of the Department of Pediatrics, University of British Columbia, Vancouver, Canada, called the studies "sophisticated" and "thoughtful."

"I think the three papers shed an incredibly important light on this very complicated research methodology of how to address confounding factors by indication — especially the mother's depression, which is what led her to use an antidepressant," said Dr Oberlander, who is author of an accompanying editorial. He was not involved with any of the studies.

"The strength of all these studies is the size of the data and the sophistication of the techniques that try to control for unmeasured variables in the large data," he said.

Taken together, the studies help "tease apart" the association between gestational exposure to an antidepressant and the development of an ASD, he said. He noted that there is likely a "shared genetic risk" between autism and depression.

"The drug may be a hitchhiker on the back of the genetic propensity for the mother's depression that may be associated with the outcome in the child," he suggested.

"This topic is incredibly important, and many women rest heavily on getting the science right," he added. "The mother's mental health during pregnancy has downstream implications long after the baby is born."

The study investigators also emphasized the importance of treating depression during pregnancy.

"Previous studies reporting an association between fetal antidepressant exposure and ASDs are not sufficient to warrant a discontinuation or avoidance of antidepressants during pregnancy," Dr Gressier warned.

"Each prescription should be evaluated individually, and clinicians should always balance the risks of maternal depression against the potential neurodevelopmental risks of antidepressant exposure for the fetus," she said.

She added that other treatment approaches, including emotional support and psychosocial interventions, may be helpful for some women. "It seems advisable to reserve psychotropic treatment for women with severe major depression."

Dr Vigod agreed, advising a "stepped-care model that delivers the most noninvasive intervention first, so psychotherapy would be tried before medication."

Ultimately, the decision is highly individual and should be based on a risk-benefit analysis, she said.

"The results of these studies tell us that if we weigh the risks and need to treat a pregnant woman with antidepressants, we should not be afraid of using them."

Funding for the studies and authors' relationships with industry are described in the original articles.

JAMA. Published online April 18, 2017. Vigod et al, Abstract; D'Onofrio et al, Abstract

JAMA Pediatr. Published online April 17, 2017. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.