Association of Maternal Antipsychotic Prescription During Pregnancy With Standardized Test Scores of Schoolchildren in Denmark

Abstract

ImportanceAn increasing number of individuals fill antipsychotic prescriptions during pregnancy, and concerns have been raised about prenatal antipsychotic exposure on neurodevelopmental outcomes.ObjectiveTo examine whether maternal prescription fill for antipsychotics during pregnancy was associated with performance in standardized tests among schoolchildren.Design, Setting, and ParticipantsThis register-based cohort study included 667 517 children born in Denmark from January 1, 1997, to December 31, 2009, and who were attending public primary and lower secondary school. All children had completed at least 1 language (Danish) or mathematics test as part of the Danish National School Test Program between 2010 and 2018. Data were analyzed from November 1, 2021, to March 31, 2022.ExposuresAntipsychotic prescriptions filled by pregnant individuals were obtained from the Danish National Prescription Register.Main Outcomes and MeasuresDifferences in standardized test scores (range, 1-100; higher scores indicate better test results) in language and mathematics between children of mothers with and without antipsychotic prescription fills during pregnancy were estimated using linear regression models. Seven sensitivity analyses, including a sibling-controlled analysis, were performed.ResultsOf the 667 517 children included (51.2% males), 1442 (0.2%) children were born to mothers filling an antipsychotic prescription during pregnancy. The mean (SD) age of children at the time of testing spanned from 8.9 (0.4) years in grade 2 to 14.9 (0.4) years in grade 8. Maternal prescription fill for antipsychotics was not associated with performance in language (crude mean test score: 50.0 [95% CI, 49.1-50.9] for the exposed children vs 55.4 [95% CI, 55.4-55.5] for the unexposed children; adjusted difference, 0.5 [95% CI, −0.8 to 1.7]) or in mathematics (crude mean test score: 48.1 [95% CI, 47.0-49.3] for the exposed children vs 56.1 [95% CI, 56.1-56.2] for the unexposed children; adjusted difference, 0.4 [95% CI, −1.0 to 1.8]). There was no evidence that results were modified by the timing of filling prescriptions, classes (first-generation and second-generation) of antipsychotics, or the most commonly prescribed antipsychotic monotherapies, including chlorprotixene, flupentixol, olanzapine, zuclopenthixol, quetiapine, perphenazine, and methotrimeprazine. The results remained robust across sensitivity analyses, including sibling-controlled analyses, negative control exposures analyses, and probabilistic bias analyses.Conclusions and RelevanceIn this register-based cohort study, maternal prescription fill for antipsychotics during pregnancy did not appear to be associated with standardized test scores in the offspring. The findings provide further reassuring data on offspring neurodevelopmental outcomes associated with antipsychotic treatment during pregnancy.

Prenatal Antipsychotic Exposure and Performance in Neurodevelopmental Outcomes Among Schoolchildren—Reply

“…In Reply We appreciate the opportunity to respond to the letter about our study on maternal prescriptions filled for antipsychotics during pregnancy and offsprings’ school performance. Dr Jingjing and colleagues suggest the implementation of propensity score matching over multivariable regression to balance the covariables.…”

“…In this scenario, the propensity score approach would yield similar or slightly weaker associations to the traditional multivariable regression and would reduce precision and statistical power due to nonmatched observations being discarded. Furthermore, our conclusions across all 7 sensitivity analyses were consistent, including analyses of better-balanced populations (eg, sibling comparisons and children of mothers with antipsychotic prescription fills before vs during pregnancy) …”

“…Therefore, prepregnancy BMI could potentially be associated with the likelihood of receiving antipsychotic treatment during pregnancy. Our study was based on the linkages of Danish nationwide registers, and information on prepregnancy BMI was not included in the Danish Medical Birth Registry until 2003 . We acknowledged potential residual confounding from any unmeasured factors in our discussion .…”

“…These results completely align with our findings in the primary analyses in the full population (children born in 1997–2009, adjusted difference, 0.5 [95% CI, −0.8 to 1.7] for language and 0.4; [95% CI, −1.0 to 1.8] for mathematics). Therefore, further adjustment for prepregnancy BMI did not change the overall conclusion of the study; that is, maternal antipsychotic prescription during pregnancy did not appear to be associated with offspring standardized test scores …”

Prenatal Antipsychotic Exposure and Performance in Neurodevelopmental Outcomes Among Schoolchildren

“…First, the group of children exposed to prenatal antipsychotics (n = 1442) was significantly smaller than the group of unexposed children (n = 666 075) in this cohort study, and there were important differences in the baseline characteristics of the 2 groups, such as maternal age at delivery and maternal smoking during pregnancy. To make the characteristics comparable, we suggest implementing the propensity score matching method instead of traditional regression analysis …”