Share this article on:

Does Antidepressant Use Attenuate the Risk of a Major Depressive Episode in Pregnancy?

Yonkers, Kimberly A.a,b; Gotman, Nathana; Smith, Megan V.a,c; Forray, Ariadnaa; Belanger, Kathleend; Brunetto, Wendy L.a; Lin, Haiqund; Burkman, Ronald T.e; Zelop, Carolyn M.f; Lockwood, Charles J.b

doi: 10.1097/EDE.0b013e3182306847

Background: Many women become pregnant while undergoing antidepressant treatment and are concerned about continuing antidepressant medication. However, antidepressant discontinuation may increase the risk of a new episode of major depressive disorder. We sought to estimate differences in the risk of developing a new major depressive episode among pregnant and postpartum women with recurrent illness who either did or did not use antidepressants.

Methods: Participants were recruited from obstetrical settings; we analyzed a subgroup of 778 women with a history of a depressive disorder. Diagnoses were determined by the Composite International Diagnostic Interview administered twice in pregnancy and once after delivery. We used Cox Regression to model onset of a major depressive episode with a time-dependent predictor of antidepressant use.

Results: There was no clear difference in risk of a major depressive episode between women who took antidepressants and women who did not (hazard ratio [HR] = 0.88; 95% CI = 0.51-1.50). After accounting for antidepressant use, clearly hazardous factors included 4 or more depressive episodes before pregnancy (HR = 1.97; 95% CI = 1.09-3.57), black race (HR = 3.69; 95% CI = 2.16-6.30), and Hispanic ethnicity (HR = 2.33; 95% CI = 1.47-3.69).

Conclusions: Failure to use or discontinuation of antidepressants in pregnancy did not have a strong effect on the development of a major depressive episode. Women with 4 or more episodes before pregnancy were at high risk of a major depressive episode, independent of antidepressant use. Black and Hispanic women also were at high risk of a major depressive episode, but it is possible that this effect is attributable to unmeasured factors.

From the Departments of aPsychiatry, bObstetrics, Gynecology, and Reproductive Sciences, cChild Study, dEpidemiology and Public Health, Yale University School of Medicine, New Haven, CT; eDepartment of Obstetrics and Gynecology, Tufts University School of Medicine, Springfield, MA; and fBeth Israel Deaconess Medical Center, Harvard University School of Medicine, Boston, MA.

Submitted 29 March 2011; accepted 4 July 2011; posted 7 September 2011.

Supported by a grant awarded to Dr. Yonkers by the National Institute of Child Health and Development (5-R01-HD045735) entitled “Effects of Perinatal Depression on PTD and LBW.”

Kimberly A. Yonkers is the lead author and received royalties from Up To Date, support for an investigator-initiated study from Eli Lilly, and study medication for an NIMH trial from Pfizer. Dr. Burkman acknowledges consultation to the Veritech corporation. The remaining authors have no disclosures.

Correspondence: Kimberly Yonkers, Departments of Psychiatry and Obstetrics, Gynecology and Reproductive Sciences, Yale University School of Medicine, 142 Temple Street, Suite 301, New Haven, CT 06510. E-mail:

Up to 5% of women who conceive are undergoing treatment with antidepressant agents.1 Concerns about possible adverse fetal effects lead many of those women to discontinue medication treatment.2 Other women with recurrent depression stop antidepressant medication before conception to avoid any use of medication in pregnancy. However, patients and their doctors must also consider the risk of relapse into a new major depressive episode if they stop (or decide not to reinitiate) antidepressant medication.3 A recent prospective cohort study found that women with a history of recurrent major depressive episodes who stopped taking antidepressants in pregnancy or just before conception, had a 5-fold increased risk of another episode as compared with women who continued medication.4 Although findings from this study have great clinical importance, they have not been replicated. Moreover, specific requirements in that study for study participation and substantial recruitment from psychiatric settings may have led to selection of an “antidepressant-requiring” group.4

We assessed onset of a major depressive episode in a cohort of women who were recruited from community and hospital-based prenatal care sites. The data were collected as part of a prospective cohort study designed to evaluate the possible association of major depressive episodes or antidepressant treatment in pregnancy with adverse birth outcomes. Detailed monthly data on major depressive symptoms and use of antidepressants allowed us to assess the relationship between the two. We hypothesized that among participants with a history of depressive episodes, those who continue antidepressants would have lower risk for onset of a major depressive episode in pregnancy and after delivery.

Back to Top | Article Outline


Inclusion/Exclusion Criteria

To be enrolled in the study, women had to be at least 18 years of age and not yet completed their 17th week of pregnancy. Women were ineligible if they had a known multifetal pregnancy, had insulin-dependent diabetes, did not speak English or Spanish, did not have access to a telephone, had plans to relocate, or intended to terminate their pregnancy.

Back to Top | Article Outline

Recruitment and Assessment Procedures

We recruited women from 137 obstetrical practices and hospital-based clinics in Connecticut and western Massachusetts. Obstetrical providers or a project screener gave patients a form on which patients could indicate interest in participation. The form also included the estimated date of delivery. Providers faxed forms to the central data collection site where research staff obtained consent by telephone and completed a screening interview. The structured screening questionnaire included questions about pregnancy dates, mood, psychotherapy, antidepressant use, medical conditions, and plans to relocate or terminate the pregnancy. The interviewer confirmed eligibility criteria and administered the gateway questions for a depressive disorder from the World Mental Health Composite International Diagnostic Interview v2.1 (WMH-CIDI)5): (1) felt sad, empty, or depressed; (2) felt discouraged; or (3) lost interest in most things they enjoy for most of the day during any 2-week interval in the current pregnancy. The screener also asked respondents about past episodes or treatment for a depressive disorder. We used these results to enrich the cohort with women who used antidepressants or were at risk for a major depressive episode. To do this, we offered participation to women who were undergoing antidepressant treatment or had a current or prior history of a depressive disorder. We randomly selected and invited participation from one of every 3 women who neither provided a positive response to screening questions for a depressive disorder nor reported treatment for a depressive disorder in the last 5 years.

We obtained written consent during a face-to-face interview before completion of 17 weeks' gestation. We reinterviewed participants by telephone at 28 (±2) weeks' gestation (“monitoring” telephone interview) and again at 8 (±4) weeks postpartum (“postpartum” telephone interview). Pregnancies in the cohort occurred between March 2005 and May 2009; follow-up continued until September 2009. Institutional Review Boards at Yale University School of Medicine and affiliated hospitals provided approval for the study.

At each assessment point, we administered the depressive disorder, post-traumatic stress disorder (PTSD), generalized anxiety disorder, and panic disorder modules of the WMH-CIDI5 to the participants. The Composite International Diagnostic Interview is a valid and reliable, fully structured lay interview instrument,6 and has been administered to >150,000 persons from 28 countries. Although the interview has not been specifically validated for use in pregnant women, such women have been well represented among those interviewed. In a validity study of a mixture of participants, the interview had high concordance with a semistructured clinical psychiatric interview for a 12-month period prevalence.7 The area under the receiver operating curve between the semi-structured clinical interview and the Composite International Diagnostic Interview was between 0.8 and 0.9 for any depressive or anxiety disorder. The specificity for any depressive disorder in the prior 12 months was 97% (SE = 0.9) and the sensitivity was 69% (SE = 11.8). The Diagnostic Interview is similarly reliable when administered over the telephone.8

At the initial face-to-face interview, we obtained a lifetime history of previous depressive episodes and the self-reported number of lifetime episodes by repeating the aforementioned gateway questions. We asked for greater detail regarding depressive episodes in the 6 months before and during pregnancy. The latter information included questions about all criteria for a major depressive episode according to the Diagnostic and Statistical Manual of Mental Disorders-4th Edition (DSM-IV).9 At this interview, we asked women about symptoms in pregnancy months 1-3; at the monitoring telephone interview, we asked about symptoms in months 4-7; and at the postpartum telephone interview, we asked about symptoms during pregnancy months 8-10 and the first 2 months after delivery. We gave women specific dates to aid recall. We applied the algorithm for a major depressive episode from the Composite International Diagnostic Interview to determine, on a monthly basis, whether a participant was in an episode of illness. A diagram of the interview schedule and time period queried is shown in Figure 1.



For information on antidepressant use or psychotherapy, we relied primarily on participant interviews. Interviewers collected information about antidepressant use for periods corresponding to those for the depressive episodes. We asked participants to show us pill bottles, if available, at the home interview. We also showed participants pictures of various antidepressant pills to obtain more accurate information. Although we attempted to collect records from outpatient behavioral health clinicians, many clinicians would not provide this information. Because of missing data, we deemed data from this source inadequate for analysis. Information about covariates, such as drug and alcohol use, race, ethnicity, and socioeconomic status was also collected by interview.

Back to Top | Article Outline

Interviewers and Quality Control

Interviewers received a minimum of 4 days of didactic training followed by at least 6 practice interviews and a minimum of 4 supervised interviews before becoming eligible to conduct independent interviews. We audiotaped interviews with the permission of participants. We randomly selected participants for quality-control assessment. For 5% of interviews, supervising staff subsequently called the participant and confirmed demographic and other information. For an additional 5%, the entire interview tape was reviewed for quality of data collection. Finally, all interviews were reviewed by second- and third-level coders. Any inconsistencies, unresolved questions, or missing information triggered review of the audiotape or a call-back to the participant. We used the same reliability procedures for telephone interviews.

Back to Top | Article Outline

Study Recruitment and Analytic Group

Study recruitment is illustrated in Figure 2. We enrolled 9525 women, of whom 3087 (32%) were ineligible, declined to enroll, or were not screened. Another 1905 (20%) screened positive for a depressive episode in the last 5 years or were currently undergoing antidepressant treatment. The remaining 4533 women (48%) screened negative for a depressive episode or antidepressant treatment. We invited all women who screened positive to participate in the study, and we randomly selected 1612 (36%) of the women who screened negative to participate. Of these 3517 women, we interviewed 2793 (79%). We excluded 1928 of these women because they were not relevant for the analysis. These included participants who were already experiencing a major depressive episode in month 1, and women with no history of a depressive disorder (who would have less reason to consider antidepressant use in pregnancy). In addition, we excluded participants with one of the following conditions that may appear similar to a major depressive episode and could have led to misclassification errors (n = 87): HIV (n = 4), hypothyroidism (n = 31), sickle cell anemia (n = 4), pancreatitis (n = 1), substance abuse (n = 12), or alcohol abuse (n = 7). Substance abuse was defined as use of an illegal drug 4 or more days per week in month 3 or later; alcohol abuse was defined as use of alcohol 4 days or more per week or 5 or more drinks per occasion. Finally, we excluded participants who used antipsychotic (n = 24) or anticonvulsant (n = 19) agents that would indicate a history of bipolar or other psychotic disorder. Some participants in our sample met more than one exclusion criteria.



The remaining group of 778 women constituted our analytic sample (Fig. 2). Of this group, 21 women (3%) miscarried, 617 (79%) successfully completed both subsequent interviews, and 689 (89%) completed at least one of the 2 remaining interviews. When a monitoring telephone interview was missed, we did not collect data for pregnancy months 4-7, and when a postpartum telephone interview was missed, we did not collect information for pregnancy months 9 and 10 and postpartum months 1 and 2.

Back to Top | Article Outline

Statistical Procedures

We used Cox regression to model risk for a major depressive episode. An event required full DSM-IV criteria for at least 2 weeks. The risk period began in the second month of pregnancy and ended 2 months after delivery. Women with missing data for a major depressive episode were treated as censored. In a sensitivity analysis, we considered censorship to be an event. For each model, we used the exact method of computation to account for a large number of ties in times to event.

The primary exposure of interest was antidepressant treatment, a time-dependent variable. We examined models comparing treatment with onset of a major depressive episode during the corresponding months. We also created models for time-lagged antidepressant indicators of 1, 2, and 3 months to resolve issues of causal ordering. In the end, we chose a lag of 2 months, so as to relate risk of episodes with antidepressant use 2 months earlier. The longer lag was chosen to avoid artificially inflating the hazard for antidepressant users by including antidepressants taken following onset of a major depressive episode. Lags of 2 and 3 months produced similar hazard ratios for antidepressant use but a lag of 2 months was more sensitive to changes in use.

We preselected covariates that might confound the relationship between antidepressant use and onset of a major depressive episode. A history of prior depressive episodes (categorized as 1, 2, 3, or 4 or more lifetime episodes), illness onset before the age of 14 years, and psychiatric hospitalization are correlates of severe depressive illness.10 We included age, socioeconomic status, and race/ethnicity in our adjusted models because perinatal depressive episodes may be higher in poor, ethnic-minority women,11 and in adolescents.12 We included factors related to physical health: prepregnancy body mass index (BMI) and binary indicators for doctor visits related to diabetes and hypertension in the 12 months before pregnancy. Finally, we included an indicator for a major depressive episode in the 6 months before pregnancy because women with more recent illness appear to have greater risk after medication discontinuation.13 All two-way interactions between antidepressant use and other factors, including the antidepressant-time interaction, were not statistically significant. We asked about psychotherapy during pregnancy but only by trimester, making it difficult to model this factor to account for causal ordering.

For each covariate, we computed an unadjusted hazard ratio (HR) for a major depressive episode with the associated 95% confidence interval (CI). We also computed an adjusted HR that assessed the jointly modeled effects of each factor. Missing data were uncommon due to extensive data quality checks; age of initial depressive illness onset and BMI were the only predictors with missing values, and this was due to participant recall. Missing values for these factors were included as a separate category. Missing values for antidepressant use always corresponded to missed interviews and censored data. Analysis was performed using SAS version 9.2 (SAS Institute Inc., Cary, NC).

Back to Top | Article Outline


Participant Characteristics

Characteristics of the analytic sample are illustrated in Table 1. Over half of the cohort was between the ages of 25 and 34 years, well-educated, and white non-Hispanic. It was relatively rare to have had a depressive disorder onset before age 14 years or to have been hospitalized for a depressive episode before pregnancy. A sizable minority of participants had a major depressive episode in the 6 months before pregnancy or 4 or more depressive episodes before pregnancy.



Back to Top | Article Outline

Antidepressant Treatment and Psychiatric Illness

Antidepressant use decreased early in pregnancy. Twenty-one percent of participants took an antidepressant in the month before pregnancy, decreasing to 18% in trimester 1, and to 14% in trimesters 2 and 3. Use of antidepressants returned to 21% after pregnancy. In all, 123 women (16%) developed a major depressive episode during pregnancy or postpartum. For those who became depressed, most had onset in the first trimester (n = 93; 76%). Regarding other psychiatric disorders during pregnancy, 5% of the cohort met criteria for post-traumatic stress disorder, 15% for generalized anxiety disorder, and 6% for panic disorder. Eight percent of the sample had more than one of these conditions during pregnancy.

Back to Top | Article Outline

Onset of a Major Depressive Episode

Table 2 provides hazard ratios for onset of a major depressive episode associated with various demographic and clinical factors. There was little evidence of an association between antidepressant use and reduced occurrence of depression (unadjusted HR = 0.85 [95% CI = 0.52-1.40]); adjusted HR = 0.88 [ 0.51-1.50]). A major depressive episode in the 6 months before pregnancy and a history of 4 or more previous depressive episodes were risk factors for a major depressive episode. A depressive episode before 14 years of age or unknown age of onset appeared to be associated with increased risk, although adjustments substantially reduced those associations. Black and Hispanic women also appeared to be at increased risk compared with white women. These clinical and racial ethnic associations were consistent in unadjusted and adjusted models that accounted for antidepressant use.



We performed a sensitivity analysis to estimate the potential effect of attrition on hazard ratio estimates for antidepressant use. When we assumed that all women lost to follow-up experienced a major depressive episode, the estimated hazard ratio for antidepressant use remained neutral (HR = 1.00 [95% CI = 0.69-1.45]), reflecting similar attrition between antidepressant users and nonusers.

The mean dose of antidepressants taken by participants is shown in Table 3. Of women who took antidepressants during the study period, 112 took one type, 16 took 2 types, and 2 took 3 types. These medications were not necessarily taken concurrently. In general, average monthly doses were consistent with the manufacturer's recommended treatment guidelines. For the most common drugs (those with at least 10 users), 71%-90% of the monthly use met recommended levels.



Back to Top | Article Outline


In this prospective study of pregnant women with recurrent depressive illness who were followed through pregnancy and postnatally, risk for onset of a major depressive episode was similar whether or not women took antidepressant medications. Controlling for antidepressant use, the highest risks of a major depressive episode were in women with an episode in the 6 months before conception, among women with 4 or more depressive episodes before pregnancy, and among black or Hispanic women. Unknown or young age at the first depressive episode were also possible risk factors.

Our findings differ from a retrospective study14 and another prospective cohort study.4 In the only previously published prospective study, Cohen and colleagues4 found a markedly increased risk of a major depressive episode among participants who discontinued medication (HR = 5.0 [95% CI = 2.8-9.1]). In our study, the hazard ratio for antidepressant users compared with nonusers was 0.88 (95% CI = 0.51-1.5). These differences may be due to differences between the 2 study populations. Many subjects from the study by Cohen et al4 were recruited from psychiatric treatment centers and had characteristics suggestive of severe depressive illness. In that study, 43% of participants developed a major depressive episode during the observation period and 53% suffered from a concurrent psychiatric condition. In our study, only 16% had the onset of a major depressive episode and 8% had another psychiatric disorder. The 2 cohorts also differed in the percentage of women who had a depressive episode before age 14 (23% in the study by Cohen et al4 and 13% in our group). Of the women in our cohort who had ≥4 depressive episodes before pregnancy, 26% experienced a depressive episode, similar to that of the cohort in the study by Cohen et al.4

Differences in the study populations may not entirely account for the disparate findings. Cohen et al used the clinician-administered Structured Clinical Interview for DSM-IV15 rather than the Composite International Diagnostic Interview (CIDI). It is possible that our use of lay interviewers led to under-detection of depressive episodes, although interviews were taped and checked by secondary coders. Recall bias may have been an issue in our study as we had several months rather than a single month between assessments. If there were under-detection or recall bias, this may have diminished estimated group differences through nondifferential misclassification bias. Other limitations of our study included our reliance on self-report rather than blood tests to document antidepressant use; participants may have under-reported their use of medication. Additionally, given our data-collection methods, we had limited ability to account for the effect of psychotherapy.

Our study was observational and cannot attribute causality. Of particular concern is the issue of confounding by indication. It is likely that women who chose to continue antidepressant treatment had a higher risk for a major depressive episode. We attempted to account for this bias by including measures of depressive disorder before pregnancy in the analysis. However, unobserved or unquantifiable differences beyond these factors may have remained. A clinical trial that randomly assigns treatment condition to pregnant women would better control this type of bias, but many would find such a design ethically objectionable.

Synthesizing our findings with that of Cohen et al,4 it is possible that some women are able to discontinue medication in pregnancy without a relapse, while others are not. Numerous parameters may indicate greater illness severity and thus risk for a major depressive episode. We found that 4 or more prepregnancy illness episodes and an episode of illness within the 6 months before conception were the strongest predictors. These characteristics may be less common among women typically attending prenatal care, from which we recruited the majority of our subjects. Although racial and ethnic minority women had elevated risk, it is not clear whether other psychosocial or environmental factors might better explain the association. This deserves further exploration.

In summary, the present cohort provides data on a large group of women with a history of a depressive episode or recent antidepressant treatment who were prospectively followed through pregnancy and into the postpartum period. Our data provide no evidence that failure to use antidepressants increased the risk for onset of a depressive episode in pregnancy. This could be reassuring to women who wish to discontinue treatment in pregnancy. However, our findings are limited by the possibility that women who most strongly required antidepressant treatment are the ones who continued therapy in pregnancy. Clinicians and patients should consider the women's psychiatric history because this appears to greatly contribute to outcomes. Patients and their physicians should carefully discuss the risks and benefits of antidepressant treatment in pregnancy.

Back to Top | Article Outline


1.Andrade SE, Raebel MA, Brown J, et al. Use of antidepressant medications during pregnancy: a multisite study. Am J Obstet Gynecol. 2008;198:e1-194.
2.Einarson A, Selby P, Koren G. Abrupt discontinuation of psychotropic drugs during pregnancy: fear of teratogenic risk and impact of counselling. J Psychiatry Neurosci. 2001;26:44-49.
3.Montgomery SA, Dufour H, Brion S, et al. The prophylactic efficacy of fluoxetine in unipolar depression. Br J Psychiatry. 1988;134(suppl 3): 69-76.
4.Cohen L, Altshuler L, Harlow B, et al. Relapse of major depressive during pregnancy in women who maintain or discontinue antidepressant treatment. JAMA. 2006;295:499-507.
5.WHO. Composite International Diagnostic Interview (CIDI, Version 2.1). Version 2.1 ed. Geneva, Switzerland: World Health Organization; 1997.
6.Wittchen HU. Reliability and validity studies of the WHO Composite International Diagnostic Interview (CIDI): a critical review. J Psychiatr Res. 1994;28:57-84.
7.Haro JM, Arbabzadeh-Bouchez S, Brugha TS, et al. Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health surveys. Int J Methods Psychiatr Res. 2006;15:167-180.
8.Kessler RC, Avenevoli S, Costello EJ, et al. National Comorbidity Survey Replication Adolescent supplement (NCS-A): II. Overview and design. J Am Acad Child Adolesc Psychiatry. 2009;48:380-385.
9.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders-DSM-IV. 4th ed. Washington, D.C.: American Psychiatric Association; 1994.
10.Rush AJ, Golden WE, Hall GW, et al. Depression in Primary Care: Vol. 1. Detection and Diagnosis. Clinical Practice Guideline Number 5. Rockville, MD: Agency for Health Care and Policy Research; 1993.
11.Hobfoll SE, Ritter C, Lavin J, Hulsizer MR, Cameron RP. Depression prevalence and incidence among inner-city pregnant and postpartum women. J Consult Clin Psychol. 1995;63:445-453.
12.Troutman BR, Cutrona CE. Nonpsychotic postpartum depression among adolescent mothers. J Abnorm Psychol. 1990;99:69-78.
13.Yonkers K, Wisner K, Stewart D, et al. The management of depression during pregnancy: a report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. Gen Hosp Psychiatry. 2009;31:403-413.
14.Cohen L, Altshuler L, Stowe Z, Faraone S. Reintroduction of antidepressant therapy across pregnancy in women who previously discontinued treatment. A preliminary retrospective study. Psychother Psychosom. 2004;73:255-258.
15.First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for Axis I DSM-IV Disorders-Patient Edition. New York: New York State Psychiatric Institute, Biometrics Research Department; 1994.
© 2011 Lippincott Williams & Wilkins, Inc.