ANNOTATION (Jane Phelps-Tschang)

The Finding: The results from this systemic review and meta-analysis broadly support a significant association between prenatal SSRI exposure and an increased risk of ASDs. Seven observational studies (cohort and case controlled) are selected for inclusion in the analysis, with each found to be of adequate study design quality. The qualitative review of the two cohort studies, both of which were conducted using the same population dataset, reveal one significant and one non-significant association between prenatal SSRI exposure and ASD in children. Based on the quantitative meta-analysis, data pooled from four case-control studies, both pooled OR (2.13, 95% CI 1.66-2.73) and pooled AOR (1.81, 95% CI 1.47-2.24) demonstrate significant association between SSRI use during pregnancy and ASD risk.

Strength and Weaknesses: This is the first and only systemic review with rigorous meta-analysis of prenatal SSRI exposure and ASD risk in children. The meta-analysis employed a validated statistical model to control for heterogeneity of the studies and appropriate weighing of results according to study size. As noted by the authors, the small number of available studies included in the meta-analysis may contribute to overestimation of significance due to publication bias. Similar to the strength of the Gentile article (Pub #2 below), the studies reviewed in this paper are based on a relatively large population dataset. In fact, six of the studies included in the Man, et a,l review are also included in the Gentile review. Similar to other review studies of this type, the result of this systemic review and meta-analysis may be limited by the variabilities across study designs and potential contribution of confounding factors, including methods or consistency of ADS diagnosis and maternal baseline covariates (lifestyle factors, psychiatric diagnosis, severity of psychopathology, genetic contribution).

Relevance: There has been considerable increase in recent antidepressant use during pregnancy, and SSRIs account for the majority (84%) of the prescribed antenatal antidepressants. Given that untreated maternal depression is associated with poor maternal and child health outcomes, physicians need to thoughtfully weigh the potential benefits of treatment with SSRIs against the potential future neurodevelopmental risks in children. While this qualitative and quantitative review suggests a significant association between prenatal SSRI use and ASDs, this systemic review and meta-analysis further emphasize the need for high-quality epidemiological research assessing the potential confounder contributions to risk of ASDs.

ANNOTATION (Jane Phelps-Tschang)

The Finding: Current available evidence remains limited regarding the use of SSRIs during pregnancy and the risk of ASDs in children. A number of studies suggest a potential association between prenatal exposure to SSRIs and increased risk of ASDs. Of the eight selected articles in this systemic review, six studies (including case-control, cohort, and prospective studies) show a link between prenatal SSRI exposure and ASDs, with AOR or OR ranging from 1.6 to 3.5; the other two large cohort studies demonstrate no statistically significant association between maternal use of SSRIs during pregnancy and ASDs.

Strength and Weaknesses: This is one of only two existing systemic reviews of prenatal SSRI exposure and ASDs, with both reviews published in 2015. The studies reviewed in this paper all have results drawn from a relatively large healthcare database, contributing more likely to broad applicability of the study results. However, in this systemic review, there exists limited method description of inclusion and exclusion criteria used in the process of article selection, which restricts the quality assessment of the articles reviewed. The author appropriately notes potential confounders in the various studies, such as baseline maternal covariates and lifestyle factors, severity of maternal depression, inability to control for genetic factors, and recall bias or lack of clarity in actual maternal use of the prescribed SSRIs during pregnancy. Additional confounders to be considered may include methods or consistency of ADSs diagnosis, particularly as diagnostic criteria of ASDs has broadened in conjunction with increased use of SSRIs in this time.

Relevance: An estimated 1 out of every 68 children (1.5%) in the US is diagnosed with an ASD. Approximately 4.5% of women reported using antidepressants three months prior to or during pregnancy between 1998 and 2005 with data indicating up to a threefold increase in antidepressant use in this time period. The most commonly prescribed antidepressants in pregnancy are SSRIs (84%), and their increased use during pregnancy requires physicians to consider carefully the neurodevelopmental outcomes of prenatal SSRI exposure on children. As medication discussions in pregnancy require a thoughtful risk-benefit consideration of the effects of both medications and untreated maternal psychiatric illness in neonatal outcomes, this systematic review has highlighted the urgent need for high-quality research in this area and especially the role of depressive illness on ASDs. Addressing the confounders not investigated in this review will allow physicians to appropriately advise women in understanding the role of depression, prenatal SSRIs, and other maternal factors in the long-term development of children.