Anxiety disorders in pregnancy have been linked in prior literature to adverse obstetric and neonatal outcomes. This retrospective cohort study aimed to determine the contributions of diagnosed anxiety disorder and exposure to anxiolytic pharmacotherapy on perinatal outcomes, and to describe prescribing trends across a single tertiary medical center between 2013 and 2024.
The study included 142,028 singleton deliveries at one tertiary center from 2013 to 2024. Women were categorized into three mutually exclusive groups: no anxiety, unmedicated anxiety, and medicated anxiety. Reported group sizes were 136,563 (96.2%) with no anxiety, 942 (0.7%) with unmedicated anxiety, and 4,523 (3.2%) with medicated anxiety.
The predefined primary outcome was a composite measure comprising any of the following: preterm birth, Apgar score below 7 at 5 minutes, or neonatal intensive care unit (NICU) admission. Secondary outcomes included the incidence of preeclampsia and prolonged maternal hospital stay. The selection of endpoints reflects clinically relevant neonatal and obstetric events.
Multivariable logistic regression models were used to estimate adjusted odds ratios (aORs) for outcomes, controlling for pre-specified maternal and obstetric confounders. The abstract reports adjusted effect estimates and 95% confidence intervals; the exact covariates included in adjustment are described in the source article but are not restated in the abstract.
Anxiolytic prescribing during pregnancy increased substantially over the study period. The proportion of deliveries with recorded anxiolytic prescription rose from 2.25% in 2013 to 4.13% in 2024. This temporal trend was statistically significant (Cochran-Armitage Z = 13.287, P < 0.001), indicating nearly a doubling in prescribing rates at the reporting center.
Compared with women without anxiety, both anxiety groups demonstrated higher adjusted odds of the composite adverse neonatal outcome. Specifically, women with unmedicated anxiety had an aOR of 1.615 (95% CI 1.227–2.088), and women with medicated anxiety had an aOR of 1.685 (95% CI 1.491–1.897). These findings indicate that a diagnosis of anxiety during pregnancy was associated with increased odds of the composite endpoint regardless of documented pharmacotherapy.
When medicated and unmedicated anxiety groups were compared directly, the study found no significant differences in the composite primary outcome or in most secondary outcomes. In other words, treated and untreated women with anxiety had comparable risks for the reported perinatal endpoints, with one notable exception (see next section).
Among secondary outcomes, preeclampsia was the only endpoint where a difference emerged: medicated women demonstrated significantly lower odds of preeclampsia compared with unmedicated women (aOR 0.674, 95% CI 0.471–0.981). No other secondary outcome differences between medicated and unmedicated women were reported as significant in the abstract.
The principal interpretation is that maternal anxiety disorder during pregnancy was associated with adverse perinatal outcomes regardless of whether women received anxiolytic pharmacotherapy. The lack of observed differences between medicated and unmedicated women does not imply that pharmacotherapy is without effect or risk; the authors explicitly caution against interpreting results as evidence for safety or causality of pharmacological treatment.
The authors note important limitations arising from unavailable data in the dataset: information on anxiety severity, the timing and duration of pharmacological treatment, and medication adherence were not available. Because these variables can influence both medication exposure and outcomes, their absence limits causal inference. The abstract also does not report medication-specific effect estimates beyond keywords; classes mentioned in the keywords include selective serotonin reuptake inhibitors and benzodiazepines, but detailed breakdowns by drug class, dose, or trimester of exposure are not provided in the abstract.
In this large single-center retrospective cohort (142,028 singleton deliveries), anxiety disorder during pregnancy was associated with increased odds of a composite adverse neonatal outcome irrespective of recorded pharmacological treatment. Anxiolytic prescribing approximately doubled over the study interval. Given the absence of data on severity, treatment timing, duration, and adherence, these observational findings should not be used to draw definitive conclusions about the safety or causal effects of anxiolytic pharmacotherapy in pregnancy. Clinicians should continue individualized risk–benefit assessment when managing anxiety in pregnancy, and interpret these results within the study’s design and data limitations.
Note: This summary reflects results and statements reported in the source abstract. Detailed methods, covariate lists, and full subgroup or medication-class analyses, if present, are reported in the full article but are not reproduced here because they were not detailed in the abstract.