Key PointsQuestion
Which antidepressants used by pregnant women are associated with specific birth defects and do associations between antidepressants and specific birth defects remain after partially accounting for the underlying condition?
Findings
In this case-control study of 30 630 mothers of infants with birth defects and 11 478 control mothers, there were previous and new associations between individual selective serotonin reuptake inhibitors, venlafaxine, and bupropion and specific birth defects. Many selective serotonin reuptake inhibitor and birth defect (particularly heart defect) associations attenuated after partially accounting for the underlying condition; most venlafaxine associations remained.
Meaning
Venlafaxine was associated with more birth defects than other antidepressants, which needs confirmation; studies to assess birth defect risk among women taking antidepressants should account for the underlying condition.
Importance
Antidepressants are commonly used during pregnancy, but limited information is available about individual antidepressants and specific birth defect risks.
Objective
To examine associations between individual antidepressants and specific birth defects with and without attempts to partially account for potential confounding by underlying conditions.
Design, Setting, and Participants
The population-based, multicenter case-control National Birth Defects Prevention Study (October 1997–December 2011) included cases with selected birth defects who were identified from surveillance systems; controls were randomly sampled live-born infants without major birth defects. Mothers of cases and controls participated in an interview after the expected delivery date. The data were analyzed after the completion of the National Birth Defects Prevent Study’s data collection.
Exposures
Self-reported antidepressant exposure was coded to indicate monotherapy exposure to antidepressants.
Main Outcomes and Measures
We used multivariable logistic regression to calculate adjusted odds ratios (aORs) and 95% confidence intervals for associations between maternal antidepressant use and birth defects. We compared early pregnancy antidepressant-exposed women with those without antidepressant exposure and, to partially account for confounding by underlying maternal conditions, those exposed to antidepressants outside of the birth defect development critical period.
Results
This study included 30 630 case mothers of infants with birth defects and 11 478 control mothers (aged 12–53 years). Early pregnancy antidepressant use was reported by 1562 case mothers (5.1%) and 467 control mothers (4.1%), for whom elevated aORs were observed for individual selective serotonin reuptake inhibitors (SSRIs) and selected congenital heart defects (CHD) (eg, fluoxetine and anomalous pulmonary venous return: aOR, 2.56; 95% CI, 1.10-5.93; this association was attenuated after partially accounting for underlying conditions: aOR, 1.89; 95% CI, 0.56-6.42). This pattern was observed for many SSRI-CHD combinations. Associations between SSRIs and non-CHD birth defects often persisted or strengthened after partially accounting for underlying conditions (eg, citalopram and diaphragmatic hernia: aOR, 5.11; 95% CI, 1.29-20.24). Venlafaxine had elevated associations with multiple defects that persisted after partially accounting for underlying conditions (eg, anencephaly and craniorachischisis: aOR, 9.14; 95% CI, 1.91-43.83).
Conclusions and Relevance
We found some associations between maternal antidepressant use and specific birth defects. Venlafaxine was associated with the highest number of defects, which needs confirmation given the limited literature on venlafaxine use during pregnancy and risk for birth defects. Our results suggest confounding by underlying conditions should be considered when assessing risk. Fully informed treatment decision-making requires balancing the risks and benefits of proposed interventions against those of untreated depression or anxiety.
Depressive and anxiety disorders affect 7.1% and 19.1% , respectively, of the US adult population annually and disproportionately affect reproductive-aged women.1 Antidepressants and evidence-supported psychotherapies are first-line treatment options for most depressive and anxiety disorders.2-5 Approximately 10% to 15% of US reproductive-aged women are prescribed antidepressants annually.6,7 These disorders are prevalent among pregnant women,8 with 6% to 8% of US pregnant women reporting being prescribed or using an antidepressant.9-12
Managing these disorders during pregnancy and the postpartum period remains challenging,13,14 but effective management can maintain maternal and infant health,15 improve maternal prenatal health care practices,16 and improve maternal-infant attachment.17 However, concerns remain about the adverse associations of antidepressants with fetal and infant health,18,19 including birth defect risk after early pregnancy exposure.20,21 Research examining the associations between antidepressant use during pregnancy and infant health, while accounting for the underlying maternal condition, is needed to inform evidence-based guidance for clinical management before and during pregnancy.20,22,23
Some antidepressants may be associated with an increased risk for birth defects or congenital heart defects overall,20,24-27 although findings are occasionally mixed. Fewer studies have examined the associations between individual antidepressants and specific birth defects, which are critical as teratogens rarely increase risks for all defects.28,29 Most analyses have focused on selective serotonin reuptake inhibitors (SSRIs),30-36 with limited studies on other first-line antidepressants, such as serotonin norepinephrine reuptake inhibitors (SNRIs) and bupropion. The clinical usefulness of prior research is limited by the inability to account for associations between the underlying condition and birth defects, which may confound associations between early pregnancy antidepressant use and birth defect risk33,37 and possibly yield inaccurate risk estimates. Building on interim data from the National Birth Defects Prevention Study (NBDPS),38-42 we present final NBDPS data on associations between antidepressant use during early pregnancy and specific birth defects. Given increased sample size, we examine associations between more antidepressants and birth defects and account, at least partially, for potential confounding by underlying conditions and environmental factors.
Participants and Procedures
The NBDPS was a US population-based, multisite (Arkansas, California, Georgia, Iowa, Massachusetts, New Jersey, New York, North Carolina, Texas, and Utah) case-control study that examined risk factors for major structural birth defects.43 Cases with selected birth defects (a full list of eligible defects is available43) were identified from surveillance systems using standard case definitions and included live births (all sites), stillbirths (all sites but New York before 2000 and New Jersey), and terminations (all sites except Georgia before 1999, Massachusetts before 2011, New York before 2000, and New Jersey).43 Clinical data were abstracted from medical records and classified by clinician geneticists and other clinicians.43-45 Sites collected data on pregnancies ending on or after October 1, 1997, through those with an estimated date of delivery (EDD) on or before December 31, 2011. Infants with known single-gene disorders or chromosomal abnormalities were excluded. Controls were randomly sampled live-born infants without major birth defects identified via vital records or hospital birth logs from the same birth months and state- or county-level (depending on the site) geographic catchment area as cases.43 Mothers participated in a computer-assisted telephone interview 6 weeks to 24 months after the EDD, with a median time to interview of 11 months for case and 9 months for control mothers (67% case and 65% control mother participation, respectively). Institutional review board approval was provided by all sites and mothers gave oral interview informed consent.
Mothers were asked about the use of citalopram, fluoxetine, paroxetine, sertraline, venlafaxine, and bupropion during the 3 months before conception or during pregnancy; specific probing for citalopram and venlafaxine began with EDDs in 2006. Other antidepressants could be reported when women were asked to report any medications used, as well as any disease and subsequent medication treatment in the 3 months before or during pregnancy. Mothers were not asked specifically about depressive or anxiety diagnoses. Mothers were asked start and stop dates, duration, and frequency of medication use. Timing before or during pregnancy was calculated using pregnancy months (consecutive 30-day intervals from the estimated conception date to delivery/end of pregnancy). Medications were coded using the Slone Drug Dictionary (Boston University), which links medication products to corresponding active components. We identified antidepressants in the following classes: SSRI, SNRI, tricyclic antidepressants and other norepinephrine reuptake inhibitor, monoamine oxidase inhibitor, and other antidepressants (eTable 1 in the Supplement). Early pregnancy exposure was defined as maternal report of using 1 or more medication product(s) in these classes in any dose, duration, or frequency from the month before conception through the third pregnancy month; the month before conception was included in early pregnancy estimates to account for the imprecision of conception date estimates and medication exposure dates. Early pregnancy antidepressant use was coded to indicate component (eg, sertraline only) or class-level (eg, SSRIs only) monotherapy exposure; if multiple antidepressants were used, women were coded as such. Women were considered unexposed if they reported no antidepressant use during the 3 months before conception through their pregnancy’s end. Women were considered exposed only outside of early pregnancy if they reported exposure to any antidepressant, but solely during the 2 to 3 months before conception and/or pregnancy months 4 to 9.
The NBDPS included 32 200 case and 11 829 control mothers. Mothers were excluded from this analysis if they had incomplete medication history (797 [1.8%]), prepregnancy type 1 or 2 diabetes (830 [1.9%]), or used a known teratogenic medication during the 3 months before conception or during pregnancy (215 [0.5%]). Mothers were excluded if they, without prompting, reported having a depressive, bipolar, or anxiety disorder but did not report antidepressant use (79 [0.2%]). There were 30 630 case and 11 478 control mothers included in our analysis.
We estimated the prevalence of early pregnancy use of each antidepressant and broader medication class for case and control mothers separately. For antidepressants taken as monotherapy with a 0.2% or greater prevalence among control mothers in early pregnancy, we examined the time trends for each antidepressant in 2-year increments. We used χ2 tests to examine the distributions of characteristics among control mothers with any early pregnancy monotherapy or polytherapy antidepressant exposure compared with (1) unexposed control mothers and (2) control mothers whose only exposure to an antidepressant(s) occurred outside of early pregnancy. For specific antidepressants with a 0.2% or greater monotherapy exposure prevalence among control mothers during early pregnancy, we examined percentage distributions among these antidepressants used in monotherapy by selected maternal characteristics.
To identify associations between early pregnancy antidepressant use and specific birth defects, we used multivariable logistic regression to calculate adjusted odds ratios (aORs) and 95% confidence intervals for NBDPS-eligible birth defects with 4 or more exposed cases. We tested 2 model sets: (1) mothers exposed to each antidepressant during early pregnancy compared with mothers unexposed to an antidepressant before or during pregnancy and (2) mothers who were exposed to each antidepressant during early pregnancy compared with mothers only exposed to an antidepressant(s) outside of early pregnancy. The approach used in set 2 was intended to account partially for confounding by the underlying condition. That is, women exposed only outside of early pregnancy may have had similar conditions, as well as other factors associated with these conditions, during and around the time of pregnancy but did not take antidepressants during the birth defect critical period. In this analysis, the phrase underlying condition is intended to describe the condition necessitating treatment with an antidepressant as well as social and environmental factors associated with the condition. In set 2, one would expect to see elevated aORs for associations between the antidepressant and specific birth defects if the antidepressant was associated with the outcome, as the underlying condition and its corollaries would be taken at least partially into account given the analytic strategy. Covariates with known associations with birth defects were tested for inclusion in model sets 1 and 2 by assessing statistical differences for each covariate between the early pregnancy antidepressant-exposed group and both comparison groups in line with a common cause covariate selection approach.46 In final models, covariates were included if they resulted in a 10% or greater change in any point estimate for associations between the exposure and any defects tested.46,47 Covariates were consistent within a model set, but may have differed between sets based on the previously described criteria. We examined individual antidepressants and classes with 0.2% or greater early pregnancy antidepressant monotherapy prevalence among control mothers. Except for analyses examining any antidepressant or multiple medication class use, monotherapy use of individual antidepressants or of specific classes in early pregnancy was examined. We had substantial power to detect small to moderate effect sizes for many associations; the power to detect rarer birth defects was limited given smaller control-case ratios,48 especially for the confounding by underlying condition analyses. Given this, and to align with American Statistical Association guidelines to consider effect sizes when interpreting results instead of statistical significance,49 we noted associations as meaningfully elevated if aORs were 2.0 or greater and lower confidence interval bounds were 0.8 or greater. Analyses were conducted with SAS, version 9.4 (SAS Institute).
Early pregnancy antidepressant use was reported by 1562 cases (5.1%) and 467 control mothers (4.1%; Table 1); among control mothers, the most common (≥0.2% monotherapy prevalence) antidepressants were sertraline, fluoxetine, paroxetine, citalopram, escitalopram, venlafaxine, and bupropion. Sertraline, fluoxetine, citalopram, and escitalopram use during early pregnancy increased over the study years (Figure). Compared with unexposed control mothers, early pregnancy antidepressant-exposed control mothers were more likely to be older, non-Hispanic white, have higher levels of education, have a higher prepregnancy body mass index (calculated as weight in kilograms divided by height in meters squared), report early pregnancy smoking or alcohol use, periconceptional folic acid use, and have a previous live birth (Table 2). Compared with control mothers only exposed to an antidepressant outside of early pregnancy, early pregnancy antidepressant-exposed control mothers had higher levels of education. The characteristics of control mothers who used each antidepressant during early pregnancy are in Table 2.
In set 1 of the birth defect analyses, which compared early pregnancy antidepressant-exposed women with women unexposed to antidepressants before and during pregnancy, models were adjusted for maternal race/ethnicity, prepregnancy body mass index, education, and early pregnancy smoking and alcohol use. Mothers who used paroxetine or fluoxetine in early pregnancy had the highest proportion of elevated aORs with specific birth defects among the SSRIs examined, followed by citalopram and sertraline (eTable 2 and eFigures 1-5 in the Supplement). There were no elevated aORs between escitalopram and specific birth defects. Mothers who used venlafaxine had elevated aORs for most examined defects; some were relatively high (ie, aORs 3.34 [95% CI, 1.69-6.60]–5.26 [95% CI, 1.96-14.12]) (eTable 2 and eFigure 6 in the Supplement). We observed elevated aORs between bupropion and 3 defects (eTable 2 and eFigure 7 in the Supplement). The results for the associations between antidepressant classes and birth defects are in eTable 3 in the Supplement.
In set 2 of the birth defect analyses, which compared early pregnancy antidepressant-exposed women with women exposed only to an antidepressant outside of early pregnancy, models were adjusted for maternal education. Many of the previously elevated aORs for the 5 SSRIs and heart defects were attenuated (eTable 2 vs eTable 4 and eFigures 1-5 in the Supplement; for example, fluoxetine and anomalous pulmonary venous return [aOR, 2.56; 95% CI, 1.10-5.93; this association was attenuated after partially accounting for the underlying condition: aOR, 1.89; 95% CI, 0.56-6.42]). Although this pattern of attenuation after accounting for the underlying condition was observed for many SSRI and CHD combinations, there were 2 exceptions: fluoxetine with coarctation of the aorta (aOR, 2.06; 95% CI, 0.89-4.74) and citalopram with atrioventricular septal defect (aOR, 3.73; 95% CI, 0.86-16.27). After at least partially accounting for the underlying condition, we observed elevated aORs for specific SSRIs and nonheart defects, including sertraline and diaphragmatic hernia (aOR, 2.72; 95% CI, 0.84-8.81); fluoxetine and esophageal atresia (aOR, 2.61; 95% CI, 0.98-6.93); paroxetine and anencephaly and craniorachischisis (aOR, 3.43; 95% CI, 0.99-11.82); paroxetine and gastroschisis (aOR, 2.11; 95% CI, 0.97-4.59); and citalopram and diaphragmatic hernia (aOR, 5.11; 95% CI, 1.29-20.24) (eTable 4 and eFigures 1-4 in the Supplement). There were no elevated aORs between escitalopram and the examined defects. After accounting for the underlying condition, the elevated venlafaxine aORs persisted for most defects; in some instances, the association strength increased (anencephaly and craniorachischisis [aOR, 9.14; 95% CI, 1.91-43.83]; eTable 4 and eFigure 6 in the Supplement). To test the robustness of these findings, we split the NBDPS data into previously reported (1997–2007)41 and new (2008–2011). Venlafaxine was associated with many of the same defects across the samples (data not shown). After accounting for the underlying condition, we observed an elevated aOR for bupropion and diaphragmatic hernia (aOR, 6.50; 95% CI, 1.85-22.88) (eTable 4 and eFigure 7 in the Supplement). Antidepressant class and birth defect association results, after accounting for the underlying condition, are in eTable 5 in the Supplement. To assess whether attenuations or increases in the aORs were because of using a different set of covariates in the second set (vs set 1) of the birth defect analyses, we also ran a sensitivity analysis controlling for the same covariates used in model set 1. The observed associations between early pregnancy antidepressant use and specific birth defects, after accounting at least partially for the underlying condition, were largely similar in the main analysis (eTable 4 in the Supplement) and the sensitivity analysis (eTable 6 in the Supplement).
We found several previously reported associations between individual antidepressants and specific birth defects and extended prior research by identifying associations that, to our knowledge, have not been previously reported or examined. Comparing early pregnancy exposed mothers with mothers exposed to antidepressants only outside of early pregnancy (the birth defect development critical period) attenuated estimates for many specific SSRI and heart defect associations. This attenuation was not observed for many specific SSRI and nonheart birth defect associations, as well as associations between venlafaxine and heart or nonheart defects. Venlafaxine had the highest proportion of elevated aORs with specific birth defects; escitalopram had the lowest proportion of elevated aORs (none).
Our findings on antidepressant use prevalence among US pregnant women9,11,12 and factors associated with use12,35,50,51 align with previous research. Associations between sertraline,24,32,33,35,51 fluoxetine,31,32,34,36,51 paroxetine,30-36,51,52 and citalopram31-34,36,51,52 and specific birth defects have been reported in multiple studies, including NBDPS42; research on escitalopram is limited. Reports using interim NBDPS data have been published on SSRIs,38 venlafaxine,41 bupropion,40 and specific SSRIs (using bayesian methods to account for other published findings).42 In this analysis, we used the final NBDPS data, adding 613 SSRI, 71 venlafaxine, and 187 bupropion-exposed women in early pregnancy to previous reports, which allowed us to examine previously reported associations for individual antidepressants and specific defects with more precision and explore additional birth defect associations.
Most previous studies30-32,34,36,51,52 examining associations between individual antidepressants and specific birth defect risks have not accounted for the possible effects of the underlying condition. In our analysis, associations between the SSRIs and specific heart defects were largely attenuated when compared with women who took antidepressants only outside of early pregnancy. Our findings suggest that the elevated associations between heart defects and antidepressants may be confounded by the underlying condition; other work examining SSRI use and the risk of any heart defect has also concluded that associations between antidepressants and heart defects may be attributable to the underlying condition.37,51 For each SSRI (except escitalopram), we noted some degree of higher risk for some nonheart defects among women taking antidepressants who were exposed during early pregnancy compared with those exposed outside of early pregnancy.
Among the non-SSRI antidepressants examined, venlafaxine had meaningfully elevated associations with many specific birth defects. Previous studies on venlafaxine have almost exclusively examined associations with any birth defect or any heart defect using teratogen information service53,54 or administrative data.32,55,56 These studies did not find an increased risk for combined birth defect groups after early pregnancy exposure.22,32,53-56 An analysis of Nordic population data examined associations between venlafaxine and septal defects or hypospadias and concluded there was no association with either defect.32 An earlier NBDPS analysis using data from 1997 to 2007 that did not account at least partially for the underlying condition41 found that early pregnancy venlafaxine use was associated with many birth defects. Our analyses of the earlier and later (1997–200741 and 2008–2011, respectively) NBDPS data indicated results were similar across the 2 samples. After accounting at least partially for the underlying condition, bupropion was only associated with diaphragmatic hernia. While several studies have suggested bupropion exposure does not result in increased risk for all birth defects combined,57,58 other research has suggested early pregnancy bupropion use is associated with specific heart defects.21,59
Strengths and Limitations
The NBDPS is among the largest studies worldwide that examines risk factors for specific birth defects with systematic case verification, which allowed us to examine associations with specific defects more accurately than is possible with administrative data.60,61 We examined self-reported monotherapy use of specific antidepressants across several antidepressant classes and associations with a range of birth defects while accounting for many possible confounders. We accounted at least partially for confounding by the underlying condition, which, to our knowledge, previous analyses have not commonly done. This has clinical relevance because findings from such analyses can support work to identify medications with the highest and lowest birth defect risks independent of the underlying condition. However, the NBDPS interview did not systematically ascertain diagnoses associated with drug treatment, limiting our ability to account directly for mental health conditions in our analyses.12 Antidepressant selection is not random and is based on shared clinician-patient decision-making about safety and effectiveness; we were unable to examine how differential medication selection may have affected our results. There may also be differences in disease severity, relapse risk, or other clinical differences among women who select specific antidepressants, continue using antidepressants during pregnancy compared with those who discontinue treatment before pregnancy, or those who initiate treatment after the first trimester. This may have resulted in bias in our comparison group of women exposed only outside of early pregnancy, which we were unable to account for in our analysis. Further, we did not account for the differential half-life of antidepressants,62 which could have resulted in exposure during periods classified as unexposed. Recall bias because of the retrospective data collection and differing time to maternal interview between case and control mothers could have affected these results, potentially leading to differential misclassification of exposure and confounders. Because status as a case and control mother affected participation, results could also be affected by selection bias if antidepressant exposure was also associated with interview participation. As this analysis was hypothesis generating and not determining causality, we did not adjust for multiple testing in considering whether a medication and birth defect association met the criteria for an elevated association; each medication and birth defect association should be confirmed in other samples.
Depressive and anxiety disorder treatment during pregnancy remains clinically challenging, in part because of the competing reproductive safety risks of poorly treated mental health disorders and potential safety concerns regarding antidepressant treatment during pregnancy.63,64 Many studies on the fetal safety of antidepressants have grouped medications into broad classes65 that limit clinical usefulness. Our analysis, which focused on commonly prescribed individual antidepressants and specific birth defects, does not have this limitation. Our findings suggest varied risks for specific birth defects after early pregnancy use of individual SSRIs, venlafaxine, and bupropion, all of which are first-line antidepressants.66 Our results indicated venlafaxine had the highest proportion of elevated birth defect risks while escitalopram had the fewest. However, to our knowledge, ours is among the few studies to examine associations between individual antidepressants and specific birth defects while at least partially accounting for the underlying condition and its corollaries, and our results need replication. Our findings underscore the importance of research on the relative reproductive and fetal safety of all first-line antidepressants. For women who require antidepressants during pregnancy, relative differences in the safety of specific medications may be useful to consider in risk-benefit decision-making. Fully informed decision-making requires balancing the risks and benefits of any proposed intervention against the maternal and fetal risks of untreated depression or anxiety, mindful that with every pregnancy an underlying risk of a birth defect exists regardless of antidepressant treatment.67
Accepted for Publication: May 29, 2020.
Corresponding Author: Kayla N. Anderson, PhD, Centers for Disease Control and Prevention, 4770 Buford Hwy, MS S106-3, Atlanta, GA 30341 (kanderson5@cdc.gov).
Published Online: August 5, 2020. doi:10.1001/jamapsychiatry.2020.2453
Author Contributions: Drs Anderson and Reefhuis had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Anderson, Lind, Simeone, Mitchell, Polen, Reefhuis.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Anderson, Mitchell, Riehle-Colarusso, Polen, Reefhuis.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Anderson, Lind, Simeone, Polen.
Administrative, technical, or material support: Anderson, Reefhuis.
Supervision: Reefhuis.
Other - Development of study design: Mitchell.
Conflict of Interest Disclosures: Dr Bobo reported contributing chapters on bipolar disorder treatment for UpToDate outside the submitted work and research support from the National Institute of Mental Health, Agency for Healthcare Research and Quality, and Mayo Foundation for Medical Education and Research. Dr Mitchell reported grants from Massachusetts Department of Public Health during the conduct of the study and personal fees from Biogen outside the submitted work. No other disclosures were reported.
Funding/Support: This project was supported through the US Centers for Disease Control and Prevention (CDC) cooperative agreements under grants PA #96043, PA #02081, FOA #DD09-001, FOA #DD13-003, and NOFO #DD18-001 to the Centers for Birth Defects Research and Prevention participating in the National Birth Defects Prevention Study (NBDPS) and/or the Birth Defects Study to Evaluate Pregnancy exposureS (BD-STEPS).
Role of the Funder/Sponsor: Funding was used in the design and conduct of the study, and the collection, management, analysis, and interpretation of the data. This funding was not directly related to the design, conduct, or data presented in this analysis. Coding of drug information in the NBDPS used the Slone Drug Dictionary under license from the Slone Epidemiology Center of Boston University.
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC.
Meeting Presentation: Findings were presented at the 35th International Conference on Pharmacoepidemiology and Therapeutic Risk Management; August 24, 2019; Philadelphia, Pennsylvania.
5.National Institute for Health and Care Excellence. Clinical guideline [CG159]: social anxiety disorder: recognition, assessment, and treatment. Accessed August 20, 2019.
https://www.nice.org.uk/guidance/CG159 6.Dawson
AL, Ailes
EC, Gilboa
SM,
et al. Antidepressant prescription claims among reproductive-aged women with private employer-sponsored insurance—United States 2008–2013.
MMWR Morb Mortal Wkly Rep. 2016;65(3):41-46. doi:
10.15585/mmwr.mm6503a1PubMedGoogle ScholarCrossref 7.Pratt
LA, Brody
DJ, Gu
Q. Antidepressant use among persons aged 12 and over: United States, 2011–2014.
NCHS Data Brief. 2017;(283):1-8.
PubMedGoogle Scholar 9.Ailes
EC, Simeone
RM, Dawson
AL, Petersen
EE, Gilboa
SM. Using insurance claims data to identify and estimate critical periods in pregnancy: an application to antidepressants.
Birth Defects Res A Clin Mol Teratol. 2016;106(11):927-934. doi:
10.1002/bdra.23573PubMedGoogle ScholarCrossref 15.Bérard
A, Sheehy
O, Zhao
J-P,
et al; MotherToBaby Collaborative Research Committee. Impact of antidepressant use, discontinuation, and dosage modification on maternal depression during pregnancy.
Eur Neuropsychopharmacol. 2019;29(7):803-812. doi:
10.1016/j.euroneuro.2019.06.007PubMedGoogle ScholarCrossref 17.Flykt
M, Kanninen
K, Sinkkonen
J, Punamaki
R-L. Maternal depression and dyadic interaction: the role of maternal attachment style.
Inf Child Dev. 2010;19:530–550. doi:
10.1002/icd.679Google Scholar 19.Masarwa
R, Bar-Oz
B, Gorelik
E, Reif
S, Perlman
A, Matok
I. Prenatal exposure to selective serotonin reuptake inhibitors and serotonin norepinephrine reuptake inhibitors and risk for persistent pulmonary hypertension of the newborn: a systematic review, meta-analysis, and network meta-analysis.
Am J Obstet Gynecol. 2019;220(1):57.e1-57.e13. doi:
10.1016/j.ajog.2018.08.030PubMedGoogle ScholarCrossref 22.Lassen
D, Ennis
ZN, Damkier
P. First-trimester pregnancy exposure to venlafaxine or duloxetine and risk of major congenital malformations: a systematic review.
Basic Clin Pharmacol Toxicol. 2016;118(1):32-36. doi:
10.1111/bcpt.12497PubMedGoogle ScholarCrossref 23.Tak
CR, Job
KM, Schoen-Gentry
K,
et al. The impact of exposure to antidepressant medications during pregnancy on neonatal outcomes: a review of retrospective database cohort studies.
Eur J Clin Pharmacol. 2017;73(9):1055-1069. doi:
10.1007/s00228-017-2269-4PubMedGoogle ScholarCrossref 24.Bérard
A, Iessa
N, Chaabane
S, Muanda
FT, Boukhris
T, Zhao
J-P. The risk of major cardiac malformations associated with paroxetine use during the first trimester of pregnancy: a systematic review and meta-analysis.
Br J Clin Pharmacol. 2016;81(4):589-604. doi:
10.1111/bcp.12849PubMedGoogle ScholarCrossref 25.Gao
S-Y, Wu
Q-J, Sun
C,
et al. Selective serotonin reuptake inhibitor use during early pregnancy and congenital malformations: a systematic review and meta-analysis of cohort studies of more than 9 million births.
BMC Med. 2018;16(1):205-218. doi:
10.1186/s12916-018-1193-5PubMedGoogle ScholarCrossref 27.Shen
Z-Q, Gao
S-Y, Li
SX,
et al. Sertraline use in the first trimester and risk of congenital anomalies: a systemic review and meta-analysis of cohort studies.
Br J Clin Pharmacol. 2017;83(4):909-922. doi:
10.1111/bcp.13161PubMedGoogle ScholarCrossref 29.Mitchell
AA. Special considerations in studies of drug-induced birth defects. In: Strom
BL, ed. Pharmacoepidemiology (2nd Edition). John Wiley and Sons; 1994:595-608.
30.Bakker
MK, Kerstjens-Frederikse
WS, Buys
CHCM, de Walle
HEK, de Jong-van den Berg
LT. First-trimester use of paroxetine and congenital heart defects: a population-based case-control study.
Birth Defects Res A Clin Mol Teratol. 2010;88(2):94-100.
PubMedGoogle Scholar 31.Bérard
A, Zhao
J-P, Sheehy
O. Antidepressant use during pregnancy and the risk of major congenital malformations in a cohort of depressed pregnant women: an updated analysis of the Quebec Pregnancy Cohort.
BMJ Open. 2017;7(1):e013372-e013385. doi:
10.1136/bmjopen-2016-013372PubMedGoogle ScholarCrossref 32.Furu
K, Kieler
H, Haglund
B,
et al. Selective serotonin reuptake inhibitors and venlafaxine in early pregnancy and risk of birth defects: population based cohort study and sibling design.
BMJ. 2015;350:h1798. doi:
10.1136/bmj.h1798PubMedGoogle ScholarCrossref 36.Wemakor
A, Casson
K, Garne
E,
et al. Selective serotonin reuptake inhibitor antidepressant use in first trimester pregnancy and risk of specific congenital anomalies: a European register-based study.
Eur J Epidemiol. 2015;30(11):1187-1198. doi:
10.1007/s10654-015-0065-yPubMedGoogle ScholarCrossref 38.Alwan
S, Reefhuis
J, Rasmussen
SA, Olney
RS, Friedman
JM; National Birth Defects Prevention Study. Use of selective serotonin-reuptake inhibitors in pregnancy and the risk of birth defects.
N Engl J Med. 2007;356(26):2684-2692. doi:
10.1056/NEJMoa066584PubMedGoogle ScholarCrossref 39.Alwan
S, Reefhuis
J, Rasmussen
SA, Friedman
JM; National Birth Defects Prevention Study. Patterns of antidepressant medication use among pregnant women in a United States population.
J Clin Pharmacol. 2011;51(2):264-270. doi:
10.1177/0091270010373928PubMedGoogle ScholarCrossref 40.Alwan
S, Reefhuis
J, Botto
LD, Rasmussen
SA, Correa
A, Friedman
JM; National Birth Defects Prevention Study. Maternal use of bupropion and risk for congenital heart defects.
Am J Obstet Gynecol. 2010;203(1):52.e1-52.e6. doi:
10.1016/j.ajog.2010.02.015PubMedGoogle ScholarCrossref 41.Polen
KND, Rasmussen
SA, Riehle-Colarusso
T, Reefhuis
J; National Birth Defects Prevention Study. Association between reported venlafaxine use in early pregnancy and birth defects, national birth defects prevention study, 1997-2007.
Birth Defects Res A Clin Mol Teratol. 2013;97(1):28-35. doi:
10.1002/bdra.23096PubMedGoogle ScholarCrossref 42.Reefhuis
J, Devine
O, Friedman
JM, Louik
C, Honein
MA; National Birth Defects Prevention Study. Specific SSRIs and birth defects: bayesian analysis to interpret new data in the context of previous reports.
BMJ. 2015;351:h3190. doi:
10.1136/bmj.h3190PubMedGoogle ScholarCrossref 43.Reefhuis
J, Gilboa
SM, Anderka
M,
et al; National Birth Defects Prevention Study. The national birth defects prevention study: a review of the methods.
Birth Defects Res A Clin Mol Teratol. 2015;103(8):656-669. doi:
10.1002/bdra.23384PubMedGoogle ScholarCrossref 44.Botto
LD, Lin
AE, Riehle-Colarusso
T, Malik
S, Correa
A; National Birth Defects Prevention Study. Seeking causes: classifying and evaluating congenital heart defects in etiologic studies.
Birth Defects Res A Clin Mol Teratol. 2007;79(10):714-727. doi:
10.1002/bdra.20403PubMedGoogle ScholarCrossref 45.Rasmussen
SA, Olney
RS, Holmes
LB, Lin
AE, Keppler-Noreuil
KM, Moore
CA; National Birth Defects Prevention Study. Guidelines for case classification for the national birth defects prevention study.
Birth Defects Res A Clin Mol Teratol. 2003;67(3):193-201. doi:
10.1002/bdra.10012PubMedGoogle ScholarCrossref 48.Schlessman
JJ. Case-control Studies: Design, Conduct, and Analysis. Oxford Press; 1982.
50.Petersen
I, Evans
SJ, Gilbert
R, Marston
L, Nazareth
I. Selective serotonin reuptake inhibitors and congenital heart anomalies: comparative cohort studies of women treated before and during pregnancy and their children.
J Clin Psychiatry. 2016;77(1):e36-e42. doi:
10.4088/JCP.14m09241PubMedGoogle ScholarCrossref 51.Jimenez-Solem
E, Andersen
JT, Petersen
M,
et al. Exposure to selective serotonin reuptake inhibitors and the risk of congenital malformations: a nationwide cohort study.
BMJ Open. 2012;2(3):e001148. doi:
10.1136/bmjopen-2012-001148PubMedGoogle Scholar 52.Jordan
S, Morris
JK, Davies
GI,
et al. Selective serotonin reuptake inhibitor (SSRI) antidepressants in pregnancy and congenital anomalies: analysis of linked databases in Wales, Norway, and Funen, Denmark.
PLoS One. 2016;11(12):e0165122. doi:
10.1371/journal.pone.0165122PubMedGoogle Scholar 55.Oberlander
TF, Warburton
W, Misri
S, Riggs
W, Aghajanian
J, Hertzman
C. Major congenital malformations following prenatal exposure to serotonin reuptake inhibitors and benzodiazepines using population-based health data.
Birth Defects Res B Dev Reprod Toxicol. 2008;83(1):68-76. doi:
10.1002/bdrb.20144PubMedGoogle ScholarCrossref 58.Cole
JA, Modell
JG, Haight
BR, Cosmatos
IS, Stoler
JM, Walker
AM. Bupropion in pregnancy and the prevalence of congenital malformations.
Pharmacoepidemiol Drug Saf. 2007;16(5):474-484. doi:
10.1002/pds.1296PubMedGoogle ScholarCrossref 59.Thyagarajan
V, Robin Clifford
C, Wurst
KE, Ephross
SA, Seeger
JD. Bupropion therapy in pregnancy and the occurrence of cardiovascular malformations in infants.
Pharmacoepidemiol Drug Saf. 2012;21(11):1240-1242. doi:
10.1002/pds.3271PubMedGoogle ScholarCrossref 60.Grzeskowiak
LE, Gilbert
AL, Morrison
JL. Exposed or not exposed? Exploring exposure classification in studies using administrative data to investigate outcomes following medication use during pregnancy.
Eur J Clin Pharmacol. 2012;68(5):459-467. doi:
10.1007/s00228-011-1154-9PubMedGoogle ScholarCrossref 61.Cooper
WO, Hernandez-Diaz
S, Gideon
P,
et al. Positive predictive value of computerized records for major congenital malformations.
Pharmacoepidemiol Drug Saf. 2008;17(5):455-460. doi:
10.1002/pds.1534PubMedGoogle ScholarCrossref 63.Grigoriadis
S, VonderPorten
EH, Mamisashvili
L,
et al. The impact of maternal depression during pregnancy on perinatal outcomes: a systematic review and meta-analysis.
J Clin Psychiatry. 2013;74(4):e321-e341. doi:
10.4088/JCP.12r07968PubMedGoogle ScholarCrossref 65.Grigoriadis
S, VonderPorten
EH, Mamisashvili
L,
et al. Antidepressant exposure during pregnancy and congenital malformations: is there an association? A systematic review and meta-analysis of the best evidence.
J Clin Psychiatry. 2013;74(4):e293-e308. doi:
10.4088/JCP.12r07966PubMedGoogle ScholarCrossref