Abstract
Background:
Antidepressant use later in pregnancy has been associated with preeclampsia and postpartum hemorrhage (PPH) in some studies.
Objective:
To evaluate the association between patterns of prenatal antidepressant dose across gestation and risk of precclampsia and PPH.
Methods:
We utilized OptumLabs® Data Warehouse (2012–2016) administrative health care claims, identifying 226,932 singleton live-born deliveries for this retrospective cohort study. Antidepressant dispensing doses were converted to fluoxetine equivalents. Using k-means longitudinal, we identified women with similar patterns of antidepressant exposure, i.e., trajectory groups, during the first 20 and 35 gestational weeks. We estimated risk ratios (RR) and 95% confidence intervals (CI) for the association between trajectory groups and preeclampisa (20-week groups) and PPH (35-week groups), adjusting for demographics, comorbidities, and other psychotropic medications. Linear trend tests assessing increasing risk of the outcomes across groups were performed.
Results:
Among 15,041 (6.6%) pregnancies exposed to an antidepressant, the following trajectory groups were identified: A-low exposure, starting pregnancy at ~10mg/day, with 1st trimester reduction/discontinuation, B-low sustained exposure of ~20 mg/day, C-moderate exposure (~40mg/day) with 1st trimester reduction/discontinuation, D-moderate sustained exposure of ~40 mg/day, and E-high sustained exposure of ~75mg/day. In the low exposure with reduction/discontinuation trajectory, risks were 8.2% for preeclampsia and 2.7% for PPH. Compared with this group, low, moderate, and high sustained trajectories were associated with preeclampsia [adjusted (a)RR: 1.17 (95% CI: 1.01, 1.34), aRR: 1.31 (95% CI: 1.12, 1.54), aRR: 1.41 (95% CI: 1.05, 1.90), respectively] and PPH [aRR: 1.32 (1.05, 1.66), aRR: 1.35 (95% CI: 1.03,1.78), aRR: 2.51 (95% CI: 1.69, 3.71), respectively]; p<0.01 for linear trend tests for both outcomes. There was no increased risk for either outcome for moderate exposure with reduction/discontinuation (trajectory C).
Conclusions:
Women with sustained antidepressant exposure, especially at higher doses, were at increased risk for preeclampsia and PPH, but underlying depression and anxiety may contribute to the increased risk.
Keywords: Antidepressants, anxiety, depression, pregnancy, postpartum hemorrhage, preeclampsia
Background
Untreated prenatal depression is associated with poor self-care, unhealthy behaviors, sucide attempts, and adverse offspring outcomes.1–4 In the US, 7–13% of pregnant women use antidepressants during pregnancy.5–7 Some studies have found antidepressant exposure during pregnancy to be associated with a range of adverse perinatal outcomes including preeclampsia, especially for exposures into the second trimester, and postpartum hemorrhage (PPH), for late-pregnancy exposures.8–19 Other studies,generally those with smaller numbers of exposed cases, conclude no association between antidepressants and preeclampsia and PPH.20–23
Recently, studies have used longitudinal cluster analysis to classify medication use during pregnancy instead of a dichotomous “any use” versus “no use” approach.24–28 These methods classify individuals with similar patterns of use, allowing for comparisons across groups with less within-group heterogeneity in dose and gestational timing of use that may be important for understanding risks for preeclampsia and PPH.
To better understand potential risks associated with antidepressant use during pregnancy, we identified antidepressant exposure patterns with respect to dose and gestational timing in a large US cohort of commercially insured women. The objective of this study was to estimate the risk of preeclampsia and PPH associated with patterns of prenatal antidepressant exposure.
Methods
Data source and study cohort
We utilized OptumLabs® Data Warehouse (OLDW), an administrative health care datasource of commercially insured enrollees with comprehensive insurance coverage for physician, hospital and prescription drug services. Specifically, we used a de-identified claims database that contains longitudinal health information on over 200 million unique lives, including a mixture of ethnicities and US geographical regions.29 Analyses were performed using R. Per OptumLabs policy, counts less than 11 and corresponding crude risk ratios were suppressed. The research was determined to be exempt from oversight by the University of California San Diego Institutional Review Board.
International Classification of Disease (ICD) 9 or 10 diagnosis or procedure codes and Current Procedural Terminology (CPT) procedure codes used to define deliveries, outcomes, and covariates, and medications of interest are listed in the supplement (eTable 1).
We identified 585,902 live-born deliveries between 2012 and 2016 among females ages 12–49 using delivery-related diagnosis and procedure codes. Women could have more than one delivery identified because of more than one birth during the study period or because of multiple delivery-related claims for the same delivery. Therefore, we linked mothers and newborns to eliminate multiple delivery dates for the same delivery by matching on Family ID and by matching the newborn’s earliest insurance coverage date with the date associated with the woman’s delivery-related code. Because gestational age at delivery was not available, we set the date of the last menstrual period (LMP) to 35 weeks before delivery for those with diagnosis codes for preterm birth and 39 weeks otherwise, consistent with previous research.30 Then, to help ensure complete claims information during the study period, we excluded deliveries from women who did not have continuous enrollment with medical and pharmacy benefits 3 months before LMP through 3 months post-delivery. We excluded deliveries with newborns who did not have continuous benefits during the first 3 months after birth. There were 226,932 deliveries from 208,271 women after implementing linkage, applying enrollment criteria, and restricting to singleton births.
Outcomes
Preeclampsia was identified from diagnosis codes present from 140 gestational days through 14 days after delivery.9,10 PPH was based on diagnosis codes within 14 days after delivery.
Antidepressant exposure
We used antidepressant pharmacy dispensing information from 3 months before the LMP through 35 gestational weeks, i.e., consistent with the assigned gestational length for preterm births. Dispensing date plus the number of days supply of antidepressant that was dispensed was used to determine days with antidepressant exposure between LMP and 35 gestational weeks. Dose on each day was determined by multiplying the antidepressant’s strength by the number of drug units per day and then converting to fluoxetine-equivalent dose.31 The doses for multiple antidepressant agents per day were summed. However, multiple overlapping dispensings for the same antidepressant agent per day were counted only once, retaining exposure from only the highest dose. Doses were expressed as average daily dose (mg/day) per week.
Antidepressant trajectories
We used k-means longitudinal to cluster women into groups with similar individual trajectories of antidepressant exposure.32 We implemented the method using R statistical software package ‘kml’33 allowing for k=2 to k=8 clusters. We selected the number of clusters for further analysis based on: 1) optimization of 3 statistical quality criterion,32 2) clinical relevance of the trajectories, and 3) at least 200 women per cluster. To assess the association with preeclampsia, we identified antidepressant trajectories from LMP through 20 gestational weeks. Although the pathophysiology of preeclampsia may initiate earlier, preeclampsia diagnosis occurs at 20 weeks or later.34 For PPH, we identified antidepressant trajectories from LMP through 35 gestational weeks. We plotted the mean dose for each cluster, i.e., trajectory group, for each gestational week during the period of interest. Five clusters best described both periods (Figure 1).
Figure 1.
Average daily flouxeine equivalent dose per week by gestational week for A) 20-week and B) 35-week antidepressant (AD) trajectory groups.
Covariates
Depression, anxiety, other mental health disorders, pain disorders and sleep disorders, i.e., possible antidepressant indications, were identified from diagnosis codes from LMP through 20 gestional weeks for preeclampsia and through 35 gestational weeks for PPH, and antipsychotic and benzodiazepine exposures were identified from pharmacy dispensings with supply overlapping the same time frames. Diagnosis of maternal preexisting hypertension was assessed in the 3 months before LMP. Finally, we obtained information on maternal age at delivery, race, and education closest to the LMP from enrollment files. We did not have data on parity, although we do not consider it to be a confounder in this study as we do not expect it to be associated with antidepressant exposure patterns independent of age.
Analyses
We estimated risk ratios (RR) and 95% confidence intervals (CI) for the association between exposure groups and the outcomes using log-linear regression with a Poisson distribution and robust standard errors to account for correlations within women with more than one delivery identified during the study period.35 For the primary analysis, we a priori restricted to antidepressant-exposed pregnancies and the lowest trajectory group was the referent group. We used an active comparator as the reference in the primary analysis to reduce confounding by disease severity because we expected pregnant women with antidepressant exposure to be more similar with respect to unmeasured confounders than pregnant women with depression or anxiety who are treated versus not treated with antidepressants. We adjusted for potential confounders including maternal age, race/ethnicity, education, depression, anxiety, pain, other mental health disorders, antipsychotic medications and benzodiazepines. Even with the large data set, some trajectories had 50 or fewer outcomes. Therefore, we aimed for parsimonious models and did not adjust for maternal preexisting hypertension or sleep disorders because they did not vary greatly by exposure status. Moreover, we did not adjust for delivery year because doing so did not materially change our estimates. For each model, we created a categorical variable for trajectory group (one level per group) and included it as a continuous variable to assess linear trend of increasing trajectory group with increasing outcome risk, and included continuous and quadratic terms to assess quadratic trend.
We conducted a multiply adjusted probabistic bias analysis of the crude associations comparing highest versus lowest trajectory groups, adjusting for differential exposure misclassification due to incorrect estimation of the LMP, adjusting for unmeasured confounding due to obesity and smoking, and accounting for random error.36,37 We report 95% simulation limits and median simulation RR estimates. Additional methodologic details are in the supplement (eMethods). Separately, we calculated E-values from adjusted RRs comparing the highest versus lowest trajectory groups (Tables 3–4) to identify the minimum association needed between an unmeasured confounder and the exposure and the outcome, conditional on the adjusted covariates, to nullify the RRs.37,38
Table 3.
Associations between antidepressant (AD) 20-week trajectory groups and risk of preeclampsia with women in trajectory group A as the primary comparison and women with depression and no AD and women with anxiety and no AD as secondary comparisons.
| Exposure Group | Preeclampsia | Trajectory A as reference | Depression and no AD as reference | Anxiety and no AD as reference | |||
|---|---|---|---|---|---|---|---|
| n (%) | Crude RR (95% CI) | Adjusted RR (95% CI) a | Crude RR (95% CI) | Adjusted RR (95% CI) a | Crude RR (95% CI) | Adjusted RR (95% CI) a | |
| Trajectory A-Low AD exposure with reduction/discontinuation | 606 (8.2) | 1.00 (Reference) | 1.00 (Reference) | 0.97 (0.85, 1.09) | 0.88 (0.77, 1.02) | 0.99 (0.89, 1.11) | 0.88 (0.79, 1.00) |
| Trajectory B-Low sustained AD exposure | 295 (9.5) | 1.16 (1.01, 1.33) | 1.17 (1.01, 1.34) | 1.12 (0.97, 1.30) | 1.03 (0.88, 1.21) | 1.16 (1.01, 1.32) | 1.05 (0.91, 1.21) |
| Trajectory C- Moderate AD exposure with reduction/discontinuation | 178 (8.4) | 1.03 (0.87, 1.22) | 1.00 (0.84, 1.18) | 0.99 (0.83, 1.18) | 0.89 (0.74, 1.07) | 1.02 (0.87, 1.20) | 0.89 (0.75, 1.06) |
| Trajectory D-Moderate sustained AD exposure | 221 (11.0) | 1.35 (1.15, 1.57) | 1.31 (1.12, 1.54) | 1.30 (1.10, 1.53) | 1.17 (0.99, 1.40) | 1.34 (1.15, 1.55) | 1.18 (1.01, 1.39) |
| Trajectory E-High sustained AD exposure | 49 (12.5) | 1.53 (1.15, 2.05) | 1.41 (1.05, 1.90) | 1.48 (1.10, 1.99) | 1.28 (0.95, 1.74) | 1.52 (1.14, 2.04) | 1.29 (0.96, 1.74) |
| Depression no AD | 419 (8.5) | NA | NA | 1.00 (Reference) | 1.00 (Reference) | NA | NA |
| Anxiety no AD | 773 (8.2) | NA | NA | NA | NA | 1.00 (Reference) | 1.00 (Reference) |
Adjusted for maternal age, race/ethnicity, education, diagnosis of depression, anxiety, other mental health disorders, and pain disorder within first 20 gestational weeks, and benzodiazepines and antipsychotics dispensings overlapping with the first 20 gestational weeks.
Table 4.
Associations between antidepressant (AD) 35-week trajectory groups and risk of postpartum hemorhage (PPH) with women in trajectory group A as the primary comparisons and women with depression and no AD and women with anxiety and no AD as secondary comparisons.
| Exposure Group | PPH | Trajectory A as reference | Depression and no AD as reference | Anxiety and no AD as reference | |||
|---|---|---|---|---|---|---|---|
| n (%) | Crude RR (95% CI) | Adjusted RR (95% CI) a | Crude RR (95% CI) | Adjusted RR (95% CI) a | Crude RR (95% CI) | Adjusted RR (95% CI) a | |
| Trajectory A-Low AD exposure with reduction/discontinuation | 204 (2.7) | 1.00 (Reference) | 1.00 (Reference) | 0.78 (0.63, 0.96) | 0.79 (0.65, 0.98) | 0.99 (0.82, 1.19) | 1.01 (0.84, 1.22) |
| Trajectory B- Low sustained AD exposure | 116 (3.7) | 1.36 (1.09, 1.72) | 1.32 (1.05, 1.66) | 1.07 (0.84, 1.35) | 1.06 (0.83, 1.34) | 1.36 (1.09, 1.69) | 1.34 (1.07, 1.67) |
| Trajectory C- Moderate AD exposure with reduction/discontinuation | 50 (2.6) | 0.98 (0.72, 1.33) | 0.95 (0.69, 1.30) | 0.76 (0.56, 1.05) | 0.77 (0.56, 1.06) | 0.97 (0.72, 1.31) | 0.98 (0.72, 1.33) |
| Trajectory D-Moderate sustained AD exposure | 73 (3.8) | 1.42 (1.09, 1.86) | 1.35 (1.03, 1.78) | 1.11 (0.84, 1.46) | 1.10 (0.83, 1.45) | 1.41 (1.09, 1.83) | 1.39 (1.06, 1.81) |
| Trajectory E-High sustained AD exposure | 31 (7.3) | 2.73 (1.87, 3.99) | 2.51 (1.69, 3.71) | 2.13 (1.45, 3.12) | 2.09 (1.41, 3.08) | 2.71 (1.86, 3.93) | 2.62 (1.78, 3.85) |
| Depression no AD | 170 (3.4) | NA | NA | 1.00 (Reference) | 1.00 (Reference) | NA | NA |
| Anxiety no AD | 254 (2.7) | NA | NA | NA | NA | 1.00 (Reference) | 1.00 (Reference) |
NA, not applicable.
Adjusted for maternal age, race/ethnicity, education, diagnosis of depression, anxiety, other mental health disorders, and pain disorder during first 35 gestational weeks, and benzodiazepine and antipsychotic dispensings overlapping the first 35 gestational weeks.
In secondary analyses, we included women with a depression diagnosis or an anxiety diagnosis (during the relevant exposure period) without antidepressant dispensings as the reference groups, separately. We made this comparison to help inform the risk-benefit tradeoff of antidepressant treatment that women and their providers may consider when making decisions about treatment of depression and anxiety during pregnancy. Previous studies have reported differing risks of preeclampsia and PPH according to antidepressant class, e.g., serotonin-norepinephrine reuptake inhibitors (SNRI) and tricyclic antidepressants were associated with stronger risks for preeclampsia than selective serotonin reuptake inhibitors (SSRI),9,10 and SNRIs were associated with a stronger risk for PPH than SSRIs.18 In a sensitivity analysis, we restricted to pregnancies exposed to SSRIs only; the study size was insufficient to compare non-SSRIs.
Results
Cohort characteristics
Between LMP and 35 gestational weeks, there were 15,041 (6.6%) pregnancies exposed to an antidepressant, 4,949 (2.2%) with a depression diagnosis not exposed to an antidepressant, and 9,406 (4.4%) with an anxiety diagnosis not exposed to an antidepressant. Of the latter two groups, 1,775 had both a diagnosis of depression and anxiety. Overall, the 27,621 women with antidepressants, or depression or anxiety disorders were most often between the ages of 25–34 years old, White, and had more than a high school degree and less than a Bachelor’s degree (Table 1). Women with depression and anxiety diagnoses without antidepressants during the first 35 gestational weeks were more often non-White compared with those with antidepressants. Among women with antidepressant exposure, 69.9% had SSRI monotherapy, 6.6% had SNRI monotherapy, and 8.9% had bupropion monotherapy.
Table 1.
Maternal characteristics according to antidepressant trajectory group, depression with no antidepressants, or anxiety with no antidepressants during the first 35 weeks of gestation (n=27,621).
| Characteristics n, (%) | Trajectory A Low AD exposure with reduction/discontinuation | Trajectory B Low/moderate sustained AD exposure | Trajectory C Moderate AD exposure with reduction/discontinuation | Trajectory D Moderate sustained AD exposure | Trajectory E High sustained AD exposure | Depression no ADa | Anxiety no ADa |
|---|---|---|---|---|---|---|---|
| n=7,622 | n=3,170 | n=1,907 | n=1,918 | n=424 | n=4,949 | n=9,406 | |
| Maternal age | |||||||
| <25 years | 269 (3.5) | 69 (2.2) | 44 (2.3) | 18 (0.9) | <11 (<2.6) | 162 (3.3) | 254 (2.7) |
| 25–34 years | 4942 (64.8) | 1936 (61.1) | 1146 (60.1) | 1101 (57.4) | >224 (>52.8) | 3102 (62.7) | 6023 (64.0) |
| >34 years | 2411 (31.6) | 1165 (36.8) | 717 (37.6) | 799 (41.7) | 189 (44.6) | 1685 (34.0) | 3129 (33.3) |
| Number of births in study period | |||||||
| 1 | 7065 (92.7) | 2915 (92.0) | 1797 (94.2) | 1761 (91.8) | 384 (90.6) | 4343 (87.8) | 8334 (88.6) |
| ≥2 | 557 (7.3) | 255 (8.0) | 110 (5.8) | 157 (8.2) | 40 (9.4) | 606 (12.2) | 1072 (11.4) |
| Maternal race | |||||||
| White | 6148 (80.7) | 2738 (86.4) | 1585 (83.1) | 1676 (87.4) | 365 (86.1) | 3741 (75.6) | 7161 (76.1) |
| Black | 503 (6.6) | 140 (4.4) | 124 (6.5) | 80 (4.2) | 23 (5.4) | 378 (7.6) | 598 (6.4) |
| Hispanic | 643 (8.4) | 155 (4.9) | 119 (6.2) | 93 (4.9) | 24 (5.7) | 454 (9.2) | 936 (10.0) |
| Other or unknown | 328 (4.3) | 137 (4.3) | 79 (4.1) | 69 (3.6) | 12 (2.8) | 376 (7.6) | 711 (7.6) |
| Education | |||||||
| Less than or equal to high school degree or unknown | 1400 (18.4) | 427 (13.5) | 358 (18.8) | 236 (12.2) | 46 (10.9) | 725 (14.6) | 1233 (13.1) |
| Less than Bachelor’s degree | 4401 (57.7) | 1825 (57.6) | 1062 (55.7) | 1091 (56.9) | 244 (57.6) | 2756 (55.7) | 4939 (52.5) |
| Bachelor’s degree or higher | 1821 (23.9) | 918 (29.0) | 487 (25.5) | 591 (30.8) | 134 (31.6) | 1468 (29.7) | 3234 (34.4) |
| Comorbidities and co-exposuresb | |||||||
| Depression | 2270 (29.8) | 1279 (40.4) | 812 (42.6) | 932 (48.6) | 282 (66.5) | 4949 (100.0) | 1775 (18.9) |
| Anxiety | 2838 (37.2) | 1406 (44.4) | 867 (45.5) | 924 (48.2) | 258 (60.8) | 1775 (35.9) | 9406 (100.0) |
| Other mental health diagnosis | 1360 (17.8) | 754 (23.8) | 429 (22.5) | 506 (26.4) | 157 (37.0) | 1334 (27.0) | 1909 (20.3) |
| Pain | 1349 (17.7) | 514 (16.2) | 355 (18.6) | 345 (18.0) | 104 (24.5) | 855 (17.3) | 1825 (19.4) |
| Sleep disorder | 427 (5.6) | 145 (4.6) | 133 (7.0) | 109 (5.7) | 33 (7.8) | 220 (4.5) | 499 (5.3) |
| Hypertension | 210 (2.8) | 73 (2.3) | 62 (3.3) | 63 (3.3) | 13 (3.1) | 128 (2.6) | 268 (2.8) |
| Antipsychotics | 55 (0.7) | 44 (1.4) | 52 (2.7) | 74 (3.9) | 40 (9.4) | 50 (1.0) | 52 (0.6) |
| Benzodiazepines | 657 (8.6) | 292 (9.2) | 280 (14.7) | 269 (14.0) | 112 (26.4) | 220 (4.4) | 731 (7.8) |
AD=antidepressant; cell sizes <11 are suppressed per requirements of OptumLabs
1,775 women have depression and anxiety diagnoses and are in both the depression and no AD group and the anxiety and no AD group.
Diagnosis or pharmacy dispensing through 35 gestational weeks.
Trajectory groups
In the k-means longitudinal analysis, 5 trajectories best described antidepressant exposure through 20 and 35 gestational weeks (Figure 1, Table 2). For both of these timeframes, trajectory groups were characterized by (A) low exposure of approximately 10mg/day with 1st trimester reduction or discontinuation, (B) low sustained exposure of approximately 20 mg/day, (C) moderate exposure of approximately 40mg/day with 1st trimester reduction or discontinuation, (D) moderate sustained exposure of approximately 40 mg/day, and (E) high sustained exposure of approximately 75mg/day. As illustrated by the median average daily fluoxetine equivalent dose of 0 mg/day in trajectories A and C, discontinuation was common in these groups. For the 35-week reduction/discontinuation trajectory groups (A, C), the percentages of women that had no antidepressant exposure beyond week 13 were 60% and 61%, respectively, and the percentages with no antidepressant exposure after week 27 were 71% and 82%, respectively.
Table 2.
Mean, standard deviation (SD), and median, interquartile range (IQR) of average daily fluoxetine equivalent dose (mg/day) per week for 20- and 35-week trajectory groups.
| 20-week trajectories fluoxetine equivalent dose mg/day | 35-week trajectories fluoxetine equivalent dose mg/day | |||||
|---|---|---|---|---|---|---|
| Trajecory Group | N | Mean (SD) | Median (IQR) | N | Mean (SD) | Median (IQR) |
| Trajectory A-Low AD exposure with reduction/discontinuation | 7414 | 4.2 (8.3) | 0.0 (0.0) | 7622 | 3.7 (8.2) | 0.0 (0.0) |
| Trajectory B- Low/moderate sustained AD exposure | 3108 | 18.6 (12.9) | 20.3 (17.9) | 3170 | 18.1 (13.4) | 20.3 (15.9) |
| Trajectory C- Low/moderate AD exposure with reduction/discontinuation | 2118 | 16.5 (22.0) | 0.0 (20.3) | 1907 | 12.5 (20.5) | 0 (22.2) |
| Trajectory D-Moderate sustained AD exposure | 2010 | 40.2 (18.8) | 40.6 (14.4) | 1918 | 37.9 (18.9) | 40.6 (11.1) |
| Trajectory E-High sustained AD exposure | 391 | 76.7 (30.9) | 80.1 (21.2) | 424 | 73.1 (30.4) | 80 (20.9) |
AD, antidepressant
Maternal characteristics differed by 35-week trajectory groups (Table 1). Women with 1st trimester antidepressant reduction or discontinuation, either from low or moderate doses (Trajectories A and C), more often were Black or Hispanic and had a high school diploma or less education compared with those that had sustained low, moderate or high doses. Women who reduced or discontinued from low doses (Trajectory A) tended to be younger with fewer mental health diagnoses and less psychotropic medication exposure than women in other trajectories, however these patterns did not hold for women who reduced or discontinued from moderate doses (Trajectory C). Women in the high sustained group (Trajectory E) more often had psychiatric diagnoses and antipsychotic or benzodiazepine exposure between 3 months before the LMP and 35 gestational weeks than women in the other trajectories.
Preeclampsia
The risk for preeclampsia in the lowest trajectory group was 8.2% (Table 3). Compared with this group, low, moderate, or high sustained antidepressant exposure (Trajectories B, D, and E) were associated with increased preeclampsia risk [adjusted (a)RR: 1.17 (95% CI: 1.01, 1.34), aRR: 1.31 (95% CI: 1.12, 1.54), aRR: 1.41 (95% CI: 1.05, 1.90), respectively], and p <0.01 for linear trend. Moderate exposure with 1st trimester reduction/discontinuation (Trajectory C) was not associated with an increased risk of preeclampsia. In secondary analysis, the risk for preeclampsia among women with depression and no antidepressant exposure was 8.5% and it was 8.2% among women with anxiety and no antidepressant exposure (Table 3). All trajectory groups included 1 in the confidence interval when compared with a depression diagnosis without antidepressants, although the moderate and high sustained exposure groups (Trajectories D and E) suggested an increased risk of preeclampsia. When compared with an anxiety diagnosis without antidepressants, associations with preeclampsia appeared to be elevated with moderate sustained (Trajectory D) and high sustained (Trajectory E) exposures.
PPH
The risk for PPH in the lowest trajectory group was 2.7% (Table 4). Compared with this group, low, moderate or high sustained antidepressant exposure (Trajectories B, D, and E) was associated with increased risk of PPH [aRR: 1.32 (1.05, 1.66), aRR: 1.35 (95% CI: 1.03,1.78), aRR: 2.51 (95% CI: 1.69, 3.71), respectively], and p< 0.01 for linear trend. Moderate dose with reduction or discontinuation early in pregnancy (Trajectory C) was not associated with an increased risk of PPH (Table 4). In secondary analysis, the risk for PPH among women with depression and no antidepressant exposure was 3.4% and it was 2.7% among women with anxiety and no antidepressant exposure (Table 4). Findings generally attenuated when compared with women with depression and no antidepressants, although the risk for PPH remained elevated among women with high sustained exposure (Trajectory E; aRR: 2.09 (95% CI: 1.41, 3.08)). The associations indicating increased PPH risk for low, moderate and high sustained antidepressant exposure (Trajectories B, D, and E) were similar between the primary comparison group and when the reference group was women with anxiety and no antidepressans [Table 4; aRR: 1.34 (95% CI: 1.07, 1.67), aRR: 1.39 (95% CI: 1.06, 1.81), aRR: 2.62 (95% CI: 1.78, 3.85), respectively].
Bias analyses
The crude RR comparing the highest versus the lowest trajectory groups attenuated for preeclampsia (median simulation RR: 1.33, (95% simulation limits: 0.99, 1.76)) and did not change materially for PPH (median simulation RR: 2.67, (95% simulation limits: 1.87, 3.89)) when adjusting for differential exposure misclassification due to incorrect estimatimation of the LMP and confounding by obesity and smoking. The E-value indicating the minimum association between an unmeasured confounder and the exposure and the outcome needed to explain the aRR comparing the highest versus lowest trajectory group and preeclampsia was 2.17, and it 4.46 was for PPH.
SSRI only analyses
After restricting to pregnancies with SSRI monotherapy, the risk of preeclampsia and PPH tended to be lower compared with examining all antidepressants, particularly at higher doses. These lower risks resulted in slightly attenuated risk ratios for the highest dose trajectory group (Table 5).
Table 5.
Associations between antidepressant (AD) trajectory groups and risks of preeclampsia and postpartum hemorrhage (PPH) with women in trajectory group A as the comparison, restricting to selective serotonin reuptake inhibitor antidepressants (SSRIs).
| 20-Week trajectory groups | N | Preeclampsia n (%) | Crude RR (95% CI) | Adjusted RR (95% CI)a |
|---|---|---|---|---|
| Trajectory A-Low SSRI exposure with reduction/discontinuation | 5546 | 424 (7.7) | 1.00 (Reference) | 1.00 (Reference) |
| Trajectory B- Low sustained SSRI exposure | 2395 | 209 (8.7) | 1.14 (0.97, 1.34) | 1.14 (0.96, 1.33) |
| Trajectory C- Moderate SSRI exposure with reduction/discontinuation | 1167 | 98 (8.4) | 1.10 (0.89, 1.36) | 1.06 (0.87, 1.31) |
| Trajectory D-Moderate sustained SSRI exposure | 1241 | 127 (10.2) | 1.34 (1.11, 1.62) | 1.30 (1.07, 1.57) |
| Trajectory E-High sustained SSRI exposure | 168 | 18 (10.7) | 1.40 (0.90, 2.19) | 1.28 (0.81, 2.01) |
| 35-Week trajectory groups | N | PPH n (%) | Crude RR (95% CI) | Adjusted RR (95% CI)b |
| Trajectory A-Low SSRI exposure with reduction/discontinuation | 5636 | 137 (2.4) | 1.00 (Reference) | 1.00 (Reference) |
| Trajectory B- Low sustained SSRI exposure | 2430 | 105 (4.3) | 1.78 (1.38, 2.28) | 1.69 (1.31, 2.18) |
| Trajectory C- Moderate SSRI exposure with reduction/discontinuation | 1057 | 21 (2.0) | 0.82 (0.52, 1.29) | 0.78 (0.50, 1.23) |
| Trajectory D-Moderate sustained SSRI exposure | 1206 | 44 (3.7) | 1.50 (1.07, 2.10) | 1.38 (0.99, 1.93) |
| Trajectory E-High sustained SSRI exposure | 188 | <11 (<5.9) | suppressed | 1.50 (0.73, 3.08) |
Adjusted for maternal age, race/ethnicity, education, diagnosis of depression, anxiety, other mental health disorders, and pain disorder during the first 20 weeks of gestation, and benzodiazepine and antipsychotic dispensings overlapping with the first 20 gestational weeks
and the first 35 gestational weeks.
Cell sizes <11 are suppressed per requirements of OptumLabs.
Comment
Principal findings
Women with sustained antidepressant exposure, compared with women who started pregnancy with low antidepressant exposure and reduced dose or discontinued during the first trimester, were at increased risk for preeclampsia and PPH. There was a trend of higher dose exposures being associated with the highest risk of these outcomes.
Furthermore, we observed that after adjusting for potential measured confounders, the risk for preeclampsia and PPH was somewhat similar between women with depression without antidepressant exposure and most antidepressant exposure groups, with the exception of the high sustained exposure groups. This finding informs risk-benefit tradeoff of antidepressant use during pregnancy, suggesting that many patterns of antidepressant exposure are no worse for preeclampsia and PPH than depression not treated with antidepressants, or factors associated with untreated depression. In contrast, risk for PPH was higher in the sustained antidepressant exposure groups than anxiety not treated with antidepressants. Preeclampsia risk was higher across all groups (8%−13%) than the expected 5%,39 whereas PPH risk was similar (3%−7%) to previous reports of 3–6%.40,41
Interpretation
We observed a group of women who reduced or discontinued antidepressants in the first trimester, which has been reported in previous studies.5,8,24 These women tended to have lower risks for preeclampsia and PPH than women with sustained antidepressant exposure. However, results may reflect confounding by underlying disease severity associated with sustained antidepressant exposure as we did not have clinical measures of depression and anxiety severity.42 We attempted to mitigate this by adjusting for co-exposure to other psychotropic medications and central nervous system comorbidities.
Similar to our study, use of antidepressants into the second trimester has been associated with an increased risk of preeclampsia in some previous studies.8–13 Studies have differed in whether SSRI antidepressants in particular are associated with preeclampsia or not. In our study, moderate and high sustained SSRI exposure appeared to be associated with at least a borderline increased risk of preeclampsia.
Many studies have reported an increased13,14,16–19 or borderline increased risk43 for PPH associated with antidepressants late in pregnancy. The two studies that have not reported this association have had 18 or fewer exposed cases.22,23 Our results are in line with previous large studies, each with more than 55 exposed cases, that found antidepressant use late in pregnancy was associated with an increased risk of PPH.13,14,16–19 Related to our finding that higher sustained doses of antidepressants were associated with higher PPH risk, one previous study among low income women reported a dose response for serotonin reuptake inhibitors (SRI), but not non-SRIs.14 Proposed mechanisms for the effect include interference by SRI antidepressants with serotonin-mediated platelet aggregation and uterine myometrial contraction after delivery.14,44,45 Here, the cluster based analysis approach was of interest, as it allowed for the isolation of late exposure (Trajectories B, D and E) as opposed to early moderate exposure with reduction/discontinuation (Trajectory C). Our results of increased risk in the sustained exposure trajectories versus early reduction/dicontinuation support targeted investigations into mechansisms in the later gestational weeks.
Limitations
In addition to being unable to account for depression and anxiety severity linked with higher antidepressant doses and sustained exposures, limitations of this research include that we had to make assumptions about antidepressant dose and gestational timing of use based on pharmacy dispensing data without maternal confirmation and gestational length from diagnosis codes for preterm birth instead of gestational age at delivery. The trajectory shapes and women’s membership within the groups may have differed from what they would have been without exposure misclassification. Bias analysis adjusting for exposure misclassification due to LMP estimation indicated that this source of systematic error did not impact our point estimates greatly. Furthermore, a limitation of k-means longitudinal is that each trajectory group contains some individual heterogeneity in antidepressant exposure and only common patterns of antidepressant exposure emerge as separate trajectories. For example, we expect some women to initiate antidepressant use during pregnancy and some women in the high sustained exposure group to have dose reductions. However k-means longitudinal did not identify trajectory groups with such patterns and these women would have been included in other trajectory groups. Nevertheless, this approach allows for more well-defined exposure groups than the traditional any exposure vs none approach that is often used when studying antidepressants in pregnancy.
Additionally, we included live birth deliveries only and selection bias could impact the results if one of the exposure patterns were associated with pregnancy loss or stillbirth. Also, women in this study had commercial insurance who were enrolled before pregnancy and findings may not generalize to women with public insurance. Finally, variables such as smoking, alcohol use and obesity are not captured well by health care claims data. Bias analysis indicated that overall, adjusting for smoking and obesity would likely have attenuated the results for preeclampsia but would not have strongly impacted the results for PPH. E-values indicated that very strong umeasured confounding would be needed to explain the association between high sustained antidepressant exposure versus low exposure with reduction/discontinuation and PPH.
Conclusions
Compared with the lowest antidepressant exposure pattern, sustained antidepressant exposure during gestation was associated with an increased risk of preeclampsia and PPH, especially for the highest antidepressant doses. Although in most cases, the risk for the outcomes associated with sustained antidepressant exposure was no higher or only modestly higher than the risk associated with depression not treated with antidepressants, suggesting the risks for preeclampsia and postpartum hemorrhage associated with untreated depression. Regardless of confounding by underlying depression and anxiety severity, providers should be aware of the increased risk of preeclampsia and PPH among women with sustained antidepressant use, especially use at higher doses, during pregnancy.
Clinical guidance supports pharmacologic or non-pharmacologic treatments of depression and anxiety disorders during pregnancy given the effects of untreated disorders.1–4,46,47 Moreover, although the data are limited,48–50 evidence suggests that women who discontinue antidepressant use during or shortly before pregnancy are at increased risk for depression relapse compared with women who continue to use antidepressants. Our study focuses on two perinatal outcomes and should be considered in context with data on the spectrum of maternal and child health outcomes to inform appropriate medication treatment decisions during pregnancy.
Supplementary Material
Synopsis.
Study question:
What is the association between patterns of prenatal antidepressant exposure, in terms of dose and gestational timing, and risk of precclampsia and postpartum hemorrhage?
What’s already known:
Antidepressant exposures into the second trimester have been associated with preeclampsia and antidepressant exposures late in pregnancy have been associated with postpartum hemorrhage (PPH) in some studies. Other studies, particularly those with smaller numbers of exposed cases, conclude no association between antidepressants and these outcomes.
What this study adds:
Compared with first trimester reduction/discontinuation of antidepressant dose, sustatined antidepressant exposure was associated with increased preeclampsia and postpartum hemorrhage risks, especially for higher doses. However, our data suggested that many patterns of antidepressant exposure are no worse for preeclampsia and postpartum hemorrhage risk than untreated depression.
Acknowledgements
We received access to the OptumLabs data without a fee through a Univesity of California and OptumLabs research credit. The authors would like to acknowledge Sarah Thayer, PhD and Nina Veeravalli, MPH for their contributions to the project.
Funding
Kristin Palmsten is supported by a career development award from the Eunice Kennedy Shriver National Institute of Child Health & Human Development, National Institutes of Health (R00HD082412). The OTIS Collaborative Research Group (PI: Chambers) receives research funding unrelated to this study from the following industry sponsors: AbbVie; Amgen Inc.; Apotex; Barr Laboratories, Inc.; Bristol-Myers Squibb; Celgene; Janssen Pharmaceuticals; Kali Laboratories, Inc.; Pfizer, Inc.; Hoffman La Roche-Genentech; Sandoz Pharmaceuticals; Genzyme Sanofi-Aventis; Takeda Pharmaceutical Company Limited; Teva Pharmaceutical Industries Ltd.; UCB, USA; Seqirus; Regeneron; Glaxo Smith Kline; and Astra Zeneca Medimmune.
Footnotes
Social media quote: Compared with first trimester antidepressant dose reduction/discontinuation, sustatined doses were associated with increased preeclampsia and postpartum hemorrhage risks. Many antidepressant exposure patterns were no worse for outcome risks than untreated depression.
The figure can be posted with the quote.
References
- 1.Bonari L, Pinto N, Ahn E, Einarson A, Steiner M, Koren G. Perinatal risks of untreated depression during pregnancy. Can J Psychiatry. 2004;49:726–735. [DOI] [PubMed] [Google Scholar]
- 2.Chisolm MS, Payne JL. Management of psychotropic drugs during pregnancy. BMJ. 2016;532:h5918. [DOI] [PubMed] [Google Scholar]
- 3.Jarde A, Morais M, Kingston D, Giallo R, MacQueen GM, Giglia L, et al. Neonatal Outcomes in Women With Untreated Antenatal Depression Compared With Women Without Depression: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2016;73:826–837. [DOI] [PubMed] [Google Scholar]
- 4.Gentile S Untreated depression during pregnancy: Short- and long-term effects in offspring. A systematic review. Neuroscience. 2017;342:154–166. [DOI] [PubMed] [Google Scholar]
- 5.Huybrechts KF, Palmsten K, Mogun H, Kowal M, Avorn J, Setoguchi-Iwata S, et al. National trends in antidepressant medication treatment among publicly insured pregnant women. Gen Hosp Psychiatry. 2013;35:265–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Andrade SE, Raebel MA, Brown J, Lane K, Livingston J, Boudreau D, et al. Use of antidepressant medications during pregnancy: a multisite study. Am J Obstet Gynecol. 2008;198:194 e191–195. [DOI] [PubMed] [Google Scholar]
- 7.Cooper WO, Willy ME, Pont SJ, Ray WA. Increasing use of antidepressants in pregnancy. Am J Obstet Gynecol. 2007;196:544 e541–545. [DOI] [PubMed] [Google Scholar]
- 8.Toh S, Mitchell AA, Louik C, Werler MM, Chambers CD, Hernandez-Diaz S. Selective serotonin reuptake inhibitor use and risk of gestational hypertension. Am J Psychiatry. 2009;166:320–328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Palmsten K, Setoguchi S, Margulis AV, Patrick AR, Hernandez-Diaz S. Elevated risk of preeclampsia in pregnant women with depression: depression or antidepressants? Am J Epidemiol. 2012;175:988–997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Palmsten K, Huybrechts KF, Michels KB, Williams PL, Mogun H, Setoguchi S, et al. Antidepressant use and risk for preeclampsia. Epidemiology. 2013;24:682–691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Avalos LA, Chen H, Li DK. Antidepressant medication use, depression, and the risk of preeclampsia. CNS Spectr. 2015;20:39–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bernard N, Forest JC, Tarabulsy GM, Bujold E, Bouvier D, Giguere Y. Use of antidepressants and anxiolytics in early pregnancy and the risk of preeclampsia and gestational hypertension: a prospective study. BMC Pregnancy Childbirth. 2019;19:146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Reis M, Kallen B. Delivery outcome after maternal use of antidepressant drugs in pregnancy: an update using Swedish data. Psychol Med. 2010;40:1723–1733. [DOI] [PubMed] [Google Scholar]
- 14.Palmsten K, Hernandez-Diaz S, Huybrechts KF, Williams PL, Michels KB, Achtyes ED, et al. Use of antidepressants near delivery and risk of postpartum hemorrhage: cohort study of low income women in the United States. BMJ. 2013;347:f4877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lindqvist PG, Nasiell J, Gustafsson LL, Nordstrom L. Selective serotonin reuptake inhibitor use during pregnancy increases the risk of postpartum hemorrhage and anemia: a hospital-based cohort study. J Thromb Haemost. 2014;12:1986–1992. [DOI] [PubMed] [Google Scholar]
- 16.Joseph KS, Sheehy O, Mehrabadi A, Urquia ML, Hutcheon JA, Kramer M, et al. Can drug effects explain the recent temporal increase in atonic postpartum haemorrhage? Paediatr Perinat Epidemiol. 2015;29:220–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Grzeskowiak LE, McBain R, Dekker GA, Clifton VL. Antidepressant use in late gestation and risk of postpartum haemorrhage: a retrospective cohort study. BJOG. 2016;123:1929–1936. [DOI] [PubMed] [Google Scholar]
- 18.Hanley GE, Smolina K, Mintzes B, Oberlander TF, Morgan SG. Postpartum Hemorrhage and Use of Serotonin Reuptake Inhibitor Antidepressants in Pregnancy. Obstet Gynecol. 2016;127:553–561. [DOI] [PubMed] [Google Scholar]
- 19.Heller HM, Ravelli ACJ, Bruning AHL, de Groot CJM, Scheele F, van Pampus MG, et al. Increased postpartum haemorrhage, the possible relation with serotonergic and other psychopharmacological drugs: a matched cohort study. BMC Pregnancy Childbirth. 2017;17:166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lupattelli A, Wood M, Lapane K, Spigset O, Nordeng H. Risk of preeclampsia after gestational exposure to selective serotonin reuptake inhibitors and other antidepressants: A study from The Norwegian Mother and Child Cohort Study. Pharmacoepidemiol Drug Saf. 2017;26:1266–1276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.De Ocampo MPG, Araneta MRG, Macera CA, Alcaraz JE, Moore TR, Chambers CD. Risk of gestational hypertension and preeclampsia in women who discontinued or continued antidepressant medication use during pregnancy. Arch Womens Ment Health. 2016;19:1051–1061. [DOI] [PubMed] [Google Scholar]
- 22.Lupattelli A, Spigset O, Koren G, Nordeng H. Risk of vaginal bleeding and postpartum hemorrhage after use of antidepressants in pregnancy: a study from the Norwegian Mother and Child Cohort Study. J Clin Psychopharmacol. 2014;34:143–148. [DOI] [PubMed] [Google Scholar]
- 23.Kim DR, Pinheiro E, Luther JF, Eng HF, Dills JL, Wisniewski SR, et al. Is third trimester serotonin reuptake inhibitor use associated with postpartum hemorrhage? J Psychiatr Res. 2016;73:79–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bandoli G, Kuo GM, Sugathan R, Chambers CD, Rolland M, Palmsten K. Longitudinal trajectories of antidepressant use in pregnancy and the postnatal period. Arch Womens Ment Health. 2018;21:411–419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hurault-Delarue C, Chouquet C, Savy N, Lacroix I, Beau AB, Montastruc JL, et al. Interest of the trajectory method for the evaluation of outcomes after in utero drug exposure: example of anxiolytics and hypnotics. Pharmacoepidemiol Drug Saf. 2017;26:561–569. [DOI] [PubMed] [Google Scholar]
- 26.Hurault-Delarue C, Chouquet C, Savy N, Lacroix I, Beau AB, Montastruc JL, et al. How to take into account exposure to drugs over time in pharmacoepidemiology studies of pregnant women? Pharmacoepidemiol Drug Saf. 2016;25:770–777. [DOI] [PubMed] [Google Scholar]
- 27.Palmsten K, Rolland M, Hebert MF, Clowse MEB, Schatz M, Xu R, et al. Patterns of prednisone use during pregnancy in women with rheumatoid arthritis: Daily and cumulative dose. Pharmacoepidemiol Drug Saf. 2018;27:430–438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Palmsten K, Bandoli G, Vazquez-Benitez G, Xi M, Johnson DL, Xu R, et al. Oral corticosteroid use during pregnancy and risk of preterm birth. Rheumatology (Oxford). 2020;59:1262–1271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wallace P, Shah N, Dennen T, Bleicher P, Crown W. Optum Labs: Building A Novel Node In The Learning Health Care System. Health Affairs. 2014;33:1187–1194. [DOI] [PubMed] [Google Scholar]
- 30.Margulis AV, Setoguchi S, Mittleman MA, Glynn RJ, Dormuth CR, Hernandez-Diaz S. Algorithms to estimate the beginning of pregnancy in administrative databases. Pharmacoepidemiol Drug Saf. 2013;22:16–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hayasaka Y, Purgato M, Magni LR, Ogawa Y, Takeshima N, Cipriani A, et al. Dose equivalents of antidepressants: Evidence-based recommendations from randomized controlled trials. J Affect Disord. 2015;180:179–184. [DOI] [PubMed] [Google Scholar]
- 32.Genolini C, Falissard B. KmL: a package to cluster longitudinal data. Comput Methods Programs Biomed. 2011;104:e112–121. [DOI] [PubMed] [Google Scholar]
- 33.Genolini C, Falissard B (2016) Package “kml.” https://cran.r-project.org/web/packages/kml/kml.pdf. [Accessed May 24, 2018].
- 34.Rana S, Lemoine E, Granger JP, Karumanchi SA. Preeclampsia: Pathophysiology, Challenges, and Perspectives. Circ Res. 2019;124:1094–1112. [DOI] [PubMed] [Google Scholar]
- 35.Zou GY, Donner A. Extension of the modified Poisson regression model to prospective studies with correlated binary data. Stat Methods Med Res. 2013;22:661–670. [DOI] [PubMed] [Google Scholar]
- 36.Lash TL, Fox MP, Fink AK. Applying Quantitative Bias Analysis to Epidemiologic Data, Springer, 2009. pp.109–150. [Google Scholar]
- 37.Package Haine D. ‘episensr.’ 2019. https://cran.r-project.org/web/packages/episensr/episensr.pdf. [Accessed December 17, 2019].
- 38.VanderWeele TJ, Ding P. Sensitivity Analysis in Observational Research: Introducing the E-Value. Ann Intern Med. 2017;167:268–274. [DOI] [PubMed] [Google Scholar]
- 39.Abalos E, Cuesta C, Grosso AL, Chou D, Say L. Global and regional estimates of preeclampsia and eclampsia: a systematic review. Eur J Obstet Gynecol Reprod Biol. 2013;170:1–7. [DOI] [PubMed] [Google Scholar]
- 40.Callaghan WM, Kuklina EV, Berg CJ. Trends in postpartum hemorrhage: United States, 1994–2006. Am J Obstet Gynecol. 2010;202:353 e351–356. [DOI] [PubMed] [Google Scholar]
- 41.Mehrabadi A, Liu S, Bartholomew S, Hutcheon JA, Kramer MS, Liston RM, et al. Temporal trends in postpartum hemorrhage and severe postpartum hemorrhage in Canada from 2003 to 2010. J Obstet Gynaecol Can. 2014;36:21–33. [DOI] [PubMed] [Google Scholar]
- 42.Palmsten K, Hernandez-Diaz S. Can nonrandomized studies on the safety of antidepressants during pregnancy convincingly beat confounding, chance, and prior beliefs? Epidemiology. 2012;23:686–688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Salkeld E, Ferris LE, Juurlink DN. The risk of postpartum hemorrhage with selective serotonin reuptake inhibitors and other antidepressants. J Clin Psychopharmacol. 2008;28:230–234. [DOI] [PubMed] [Google Scholar]
- 44.de Abajo FJ, Rodriguez LA, Montero D. Association between selective serotonin reuptake inhibitors and upper gastrointestinal bleeding: population based case-control study. BMJ. 1999;319:1106–1109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Cordeaux Y, Pasupathy D, Bacon J, Charnock-Jones DS, Smith GC. Characterization of serotonin receptors in pregnant human myometrium. J Pharmacol Exp Ther. 2009;328:682–691. [DOI] [PubMed] [Google Scholar]
- 46.Yonkers KA, Vigod S, Ross LE. Diagnosis, pathophysiology, and management of mood disorders in pregnant and postpartum women. Obstet Gynecol. 2011;117:961–977. [DOI] [PubMed] [Google Scholar]
- 47.Misri S, Abizadeh J, Sanders S, Swift E. Perinatal Generalized Anxiety Disorder: Assessment and Treatment. J Womens Health (Larchmt). 2015;24:762–770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Cohen LS, Altshuler LL, Harlow BL, Nonacs R, Newport DJ, Viguera AC, et al. Relapse of major depression during pregnancy in women who maintain or discontinue antidepressant treatment. JAMA. 2006;295:499–507. [DOI] [PubMed] [Google Scholar]
- 49.Roca A, Imaz ML, Torres A, Plaza A, Subira S, Valdes M, et al. Unplanned pregnancy and discontinuation of SSRIs in pregnant women with previously treated affective disorder. J Affect Disord. 2013;150:807–813. [DOI] [PubMed] [Google Scholar]
- 50.Yonkers KA, Gotman N, Smith MV, Forray A, Belanger K, Brunetto WL, et al. Does antidepressant use attenuate the risk of a major depressive episode in pregnancy? Epidemiology. 2011;22:848–854. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.