Graduate student Vanderbilt University Nashville, United States
Background: The study aims to understand the impact of antidepressant use during pregnancy on neonatal outcomes, providing important information for informed treatment decisions and guidelines.
Objectives: To quantify the effect of prenatal exposure to antidepressants on adverse neonatal outcomes.
Methods: Data source: We will use data from Vanderbilt University's Mother/Baby Dyad Database, a database of 55k mother-baby linked records with corresponding demographic data, drug exposures, and clinical outcomes.
Causal framework: A causal framework introduced by Hernán et al. will be used, which applies causal inference to emulate a randomized trial using observational data.
Eligibility criteria: pregnant women with health records in Vanderbilt who have at least 1 year of records prior to conception and were exposed to antidepressants during that year.
Exposure of interest: Initiation of an antidepressant ingredient or drug class during pregnancy.
Outcomes: Neonatal outcomes such as NICU admission, gestational age, birth weight, congenital malformations, etc.
Statistical analysis: Statistical models used will depend on the type of outcome, and may include linear or logistic regression models. We will adjust for baseline confounders (depression severity, socioeconomic status, biological sex, race, maternal age) using inverse probability weighting.
Results: We validated our study design using positive (benzodiazepines) and negative control (SSRIs) exposures. The outcomes were NICU admission, gestational age, birth weight (adjusted for gestational age), and heart defects. We employed a simple linear regression model for gestational age and birth weight and logistic regression for NICU admission and heart defects (adjusted for sex, race, and maternal age).
benzodiazepine ----------------------- beta CI p-value NICU admission 0.78 [0.39, 1.17] 7.5e-05 gestational age -5.41 [-7.79, -3.04] 8.43e-06 birth weight -0.12 [-0.17, -0.067] 1.2e-05 heart defect 0.63 [0.028, 1.23] 4.03e-02
SSRI ------- beta CI p-value NICU admission 0.12 [-0.2, 0.46] 0.46 gestational age -1.85 [-3.28, -0.43] 0.01 birth weight -0.04 [-0.08, -0.004] 0.03 heart defect 0.15 [-0.3, 0.6] 0.53
Conclusions: This study aimed to investigate the effect of prenatal antidepressant exposure on adverse neonatal outcomes. We used positive and negative control exposures to validate our data and study design. The results showed that the positive control exposure was associated with an increased risk of NICU admission, lower gestational age, lower birth weight, and a higher risk of heart defects in newborns compared to the negative control exposure. The next step would be to further refine the study design, define and validate more neonatal phenotypes, and test other antidepressant ingredients.
