Type of study: Cohort Study

The finding: Children of mothers using an SSRI in late pregnancy (> 29 weeks gestational age) had a greater risk of developing anxious or depressed behaviors by age 5 relative to those who were unexposed (95% CI: 0.04-0.96); however this is confounded significantly by indication, illness severity, and postnatal environment. The increased risk appeared to increase with age, reaching a significant effect at age 5 years old or greater. The observed risk was of medium magnitude corresponding to an odds ratio of 2.5 (8 children would be expected to have this behavioral problem for every 100 women treated with an SSRI in late gestation). Such risk was not observed in children exposed to SSRIs in early (0-16 weeks) or mid (17-28 weeks) pregnancy. There was no significant effect on externalizing behaviors and no increased risk for more problematic temperament in terms of emotionality, sociability, activity, or shyness.

Strengths:

• Data was obtained from the prospective population-based MoBa study (the Norwegian Mother and Child Cohort Study).
• Used a widely-used, validated measure to assess child development at 1.5, 3, and 5 years of age.
• Controlled for confounders including: maternal BMI, parity, maternal education, gross yearly income, marital status, folic acid use, smoking and alcohol use in pregnancy, illicit substance use and paternal education; co medication with analgesics, anxiolytics and sedatives, antipsychotics and non SSRI antidepressants.
• Used a negative control (“SSRI Discontinuers” or women exposed to SSRIs in the six month period before pregnancy but not during pregnancy).
• The study only included pregnancies wherein the women reported depressive/anxiety disorders prior to and/or during pregnancy.

Weakness:

• Maternal depressive/anxiety disorders and medication use were self-reported.
• Outcome measures on child development were parent-reported.
• The effect of early pregnancy SSRI exposure could not be estimated in the weighted analysis.
• The MoBa study has a low response rate (41%), potentially selecting for the healthiest women.
• Women lost to follow-up were more likely to not be in a relationship, to have low education and more severe mental illness.
• The instrument used to measure maternal depressive/anxiety disorders, while validated in general populations, was not designed to measure perinatal mood/anxiety specifically.

Relevance: SSRI use in pregnancy is common, reaching 1-4% in Europe and up to 8% in the United States. Unfortunately, there is inconsistent evidence pertaining to the effects of prenatal SSRI exposure on behavioral outcomes in children. Taking into account the impact that childhood social, emotional and behavioral problems can have on psychiatric disorders later in life, it is important to better understand the role of potential risk factors, such as prenatal SSRI exposure, though these studies are significantly confounded by maternal illness state and severity. This is the first study to examine the time-dependent effect of prenatal SSRI exposure on behavioral, emotional, and social outcomes in preschool-age children. While the observed association must be confirmed by future research, this study may provide some insight into potentially important periods of fetal vulnerability to SSRI exposure and, in turn, may prove clinically useful in assisting physicians when evaluating the risk of treatment with an SSRI during pregnancy. Confounding by indication, severity, and genetics makes research in these areas challenging and conclusions must be drawn with caution.