Comparative Safety of commonly prescribed Psychotropic Drugs in pregnant Women

点击翻译按钮获取中文摘要

ABSTRACT Mental illnesses complicate over 15% of all pregnancies. Many of these conditions require treatment with psychotropic drugs. However, multiple studies have suggested an association between prenatal exposure to psychotropic drugs and an increased risk of major malformations, growth restriction, neonatal withdrawal syndrome and respiratory problems in the newborn, as well as gestational diabetes, preeclampsia and cesarean section in the mother. On the other hand, pregnant women with mental and/or neurological conditions who discontinue therapy may experience relapse of symptoms that can be dangerous for them and their offspring. Because many of these conditions require pharmacologic treatment and many pregnancies are unplanned, the clinically relevant questions are 1) which specific medication to recommend for women of childbearing age and 2) whether and when to discontinue treatment during pregnancy. Despite the difficult decisions faced by patients and physicians, sufficient information to guide therapeutic recommendations for the use of psychotropic drugs during pregnancy is lacking. Even when these medications may be safe, the lack of knowledge may instill unnecessary anxiety in pregnant women. Our primary objective is to provide direct evidence on the comparative safety of specific strategies used to manage mood disorders, psychosis, and attention deficit hyperactivity disorders (ADHD) in a population of pregnant women, which includes vulnerable indigent young women. We hypothesize that not all drugs are equal and that the risk-benefit trade- off of continuing treatment during pregnancy will differ by drug. We propose to quantify the risk of maternal and fetal adverse events associated with continuing and discontinuing specific psychotropic drugs during pregnancy. Within both the Medicaid Analytic eXtract (MAX) and the UnitedHealth Research database, two large population-based claims databases linkable to additional obstetrical, birth, and psychiatric records, we will identify cohorts of over 18,500 women on mood stabilizers (antiepileptics), 6,500 on antipsychotics, and 3,000 on psychostimulants at conception who delivered between 2000 and 2012. Relative risks and 95% confidence intervals will be calculated using regression models that account for the possible correlation induced by multiple gestations. We will apply an innovative combination of high dimensional propensity score techniques, instrumental variables and external adjustment to minimize the potential for confounding, and Bayesian approaches to account for prior information and support patient subgroup analyses. We will obtain information from medical records to validate specific outcomes and from the National Health and Nutrition Examination Survey to obtain information on potential confounders such as smoking, alcohol and obesity. The current proposal is an interdisciplinary collaboration between the Brigham and Women's Hospital and the Department of Epidemiology at Harvard School of Public Health.