19-year HK data: No causal link between maternal psychotropic exposure and obstetric complications

Broad-class psychotropic exposure during pregnancy is not associated with an increased risk of obstetric or neonatal complications, although certain subclasses may carry higher risks, according to a 19-year population-based study in Hong Kong.

“Although current prescribing guidelines suggest continuation of psychotropic drugs during pregnancy, population-based evidence supporting their safety remains limited,” wrote the researchers from the University of Hong Kong (HKU). They therefore conducted a population-based study examining all births by Hong Kong residents aged ≥18 years in public hospitals. [Br J Psychiatry 2025;doi:10.1192/bjp.2025.10340]

“Our study is among the first in the Chinese population to investigate the safety of psychotropic drug use [ie, antipsychotics, antidepressants, anticonvulsants, benzodiazepines] and to consider a comprehensive list of obstetric complications,” highlighted the researchers.  “It is also the first to examine outcomes across all major psychotropic classes, which allowed realistic modelling of polypharmacy in a psychiatric population.”

A total of 587,419 newborns from 577,656 maternal episodes in 2004–2022 were identified for analysis. Among these, 7,182 episodes had maternal psychotropic exposure (cases), while 2,637 maternal episodes had discontinued psychotropic use within 1 year prior to conception (negative controls).

The significant associations observed in the broad drug class analysis were also observed in the negative control analysis, suggesting that these associations were probably driven by unmeasured confounders, since the negative control group should have had no maternal exposure.

“When considering psychotropic drug exposure in broad classes, no causal links to obstetric or neonatal complications were found at the population level,” pointed out the researchers.

These results were generally consistent with those of an umbrella review published in 2024, which found no convincing or highly suggestive level of evidence of adverse health outcomes associated with psychotropic medication use in pregnant patients with mental disorders. [Mol Psychiatry 2025;30:327-335]

Nonetheless, some psychotropic subclasses were associated with distinct risks of obstetric complications. Specifically, atypical antipsychotics were associated with an increased risk of genitourinary infection (odds ratio [OR], 2.70; 95 percent confidence interval [CI], 1.46–4.83), while valproate was linked to a higher risk of low birth weight (OR, 1.68; 95 percent CI, 1.16–2.37). [Br J Psychiatry 2025;doi:10.1192/bjp.2025.10340]

“Given the small sample of negative controls [particularly serotonin-norepinephrine reuptake inhibitors (SNRIs)], our subclass negative control analysis may be prone to false negative findings, Therefore, we validated the results using E-values,” added the researchers.

Of note, E-values for the associations between atypical antipsychotics and genitourinary infection (E=4.84; 95 percent CI, 2.56–7.12) and between valproate and low birth weight (E=2.75; 95 percent CI,1.16–4.34) were markedly higher than 1. This indicates that the observed associations are robust and unlikely due to unmeasured confounding.

After controlling for the effects of psychotropics, maternal diagnoses of schizophrenia and depression were independently associated with an increased risk of obstetric complications. “These results suggest that severe mental illnesses and obstetric complications may share an overlapped aetiology, independent of psychotropic drug use,” added the researchers.