Post by Sarah Hill
What's the science?
Is the use of psychiatric or neurologic medications during pregnancy associated with higher (or lower) incidence of autism spectrum disorders (ASD) in children? This has been a difficult question to answer as pregnant women are often excluded from clinical trials. Further, separating out whether effects on children stem from exposure to medications or from increased risks associated with the mother's medical diagnosis has not been possible using prior study design methods. This week in JAMA Psychiatry, Janecka and colleagues use an innovative study design strategy to test whether prenatal exposure to drugs that act on neurotransmitter systems is associated with increased or decreased prevalence of ASD.
How did they do it?
The authors obtained data for 95,978 children born between 1997 and 2007 from a health maintenance organization in Meuhedet, Israel, and verified that maternal prescription rates and ASD prevalence in the sample matched the national rates. The sample population consisted of all children who had received an ASD diagnosis before 2015, randomly sampled ASD and non-ASD individuals, and all siblings of ASD and non-ASD individuals. They then grouped individuals based on the neurotransmitter system that was targeted by the mother's medication (55 non-mutually exclusive medication groups in total; for example, the serotonin system), in contrast to previous studies which grouped individuals by medication. They next calculated the hazard ratios, a measure of the relative risk of outcome in one group compared to another group, using a statistical method that simultaneously evaluates how several factors influence the rate of a particular event happening (in this case, an ASD diagnosis) at a particular timepoint [Cox proportional hazards regression]. Importantly, the models tested also took into account the mothers' psychiatric or neurological conditions (e.g. anxiety disorder), number of maternal medical diagnoses within 1 year prior to birth (e.g. acute bronchitis), birth year, maternal and paternal age, and socioeconomic status.
What did they find?
Due to the number of factors that may play a role, establishing a clear relationship between prenatal exposure to psychiatric drugs and altered risk of ASD is difficult. A link between neuronal acetylcholine receptor α antagonists, such as certain medications used to treat epilepsy, and increased rates of ASD was the only clear finding in this study that remained significant after accounting for all other factors, although there were very few women who took these medications during pregnancy. Certain medications used as anti-analgesic/anti-inflammatory agents were found to be associated with decreased rates of ASD when the number of maternal diagnoses was also taken into account, though the associations were not statistically significant in subsequent analyses. The authors found no other statistically significant links between medication group and ASD when all variables were considered, indicating that the use of most neurologic medications, including antidepressants and antipsychotics, during pregnancy is not associated with increased or decreased prevalence of ASD. Of all factors considered, the number of maternal medical diagnoses appeared to have the greatest effect on whether a relationship between medication group and ASD was found, suggesting it may be confounding previously observed associations.
What's the impact?
In contrast to previous reports, this study found no evidence that the use of drugs that act on the serotonin system (as well as most other neurotransmitter systems) during pregnancy increase a child's risk of ASD. Instead, it suggests that maternal general health can impact the risk of ASD, indicating that prior studies that found a link between prenatal medication use and ASD may need to be re-evaluated. Large-scale population-based studies such as this one are imperative for informing prescription practices of doctors treating pregnant patients.
Janecka et al. Association of Autism Spectrum Disorder with prenatal exposure to medication affecting neurotransmitter systems. JAMA Psychiatry (2018). Access the original scientific publication here.