Women’s Mental Health

Journal Article Annotations
2016, 4th Quarter

Women’s Mental Health

Annotations by Shalini Dave, DO and Shelly Kucherer, MD
January 2017

  1. Postpartum depressive symptoms following consecutive pregnancies: Stability, change and mechanisms
  2. Risk of mortality with concomitant use of tamoxifen and selective serotonin reuptake inhibitors: Multi-database cohort study

PUBLICATION #1 — Women’s Mental Health
Postpartum depressive symptoms following consecutive pregnancies: Stability, change and mechanisms
Schetter CD, Saxbe D, Cheadle A, Guardino C
Clin Psychol Sci 2016; 4(5):909-918
Annotation  (Shalini Dave)

The finding: This 2016 prospective study utilizes data from the Community and Child Health Network (CCHN) and follows a cohort of 2510 mothers from 1 month following a birth for 2 years to characterize postpartum depression (PPD) across pregnancies. Specifically, the authors looked at patterns of stability and change in postpartum depressive symptoms among consecutive pregnancies, as well as life stress mediators (e.g., perceived stress, parenting stress, number of stressful life events over the preceding year, and interpersonal violence/aggression over the preceding year) as an explanation for the association between depressive symptoms following consecutive postpartum periods. The study found that while 76% of the women had a negligible change in depressive symptoms between subsequent postpartum periods (specifically, fewer than 5 points difference on the EPDS), there was considerable variation as to whether postpartum depressive symptoms increased or decreased. 36% of women showed a decrease in depressive symptoms over two postpartum periods, while 20% reported identical depressive symptoms, and 44% of the sample reported increased depressive symptoms. In addition, a total of 24% of the sample showed clinically significant depressive symptoms (EPDS > 11) after one or both postpartum periods — 9% following the first pregnancy, 7.5% following the second pregnancy, and 3.5% following both pregnancies. Moreover, of the women who experienced clinically significant depressive symptoms following the first pregnancy, 28% had clinically significant depressive symptoms following their second pregnancy. In addition, the study showed that psychosocial stress between pregnancies can affect postpartum depression, and all four stress factors that were tested were shown to be mediators associated with postpartum depressive symptoms between pregnancies. The largest indirect effect was parenting stress and perceived stress, with life event count having the weakest indirect effect. Approximately one-third of the total effect of depression following the first pregnancy on depression following the second pregnancy was mediated by parenting stress and perceived stress that was measured one year following the first pregnancy.

Strengths: Strengths of this study include its prospective design and longitudinal analysis. The use of participant interviews with community partnership methods enhanced retention and rapport, and provided information regarding outcomes. In addition, the study had a relatively large sample size, and this sample was predominantly low income and ethnically and racially diverse, helping to apply the results to a wider range of women. Confounding factors of interpregnancy interval, relationship status, and prior birth were controlled for in the statistical analysis.
Weaknesses: Weaknesses of this study include its use of the EPDS as a screening tool as opposed to determining clinical diagnoses. Also, the study did not control for other stressors that may have occurred during the interval between the two pregnancies, for example changes in employment, or health or developmental issues in the first child. While the sample size is considerable for this type of an investigation, the total number of women who completed the study remains small. In addition, the study did not take into consideration major confounders such as traumatic deliveries, preterm delivery or neonatal complications, which could be associated with higher rates of PPD. The study also does not consider the fact that many of these pregnancies were likely to be short-interval pregnancies, and that this could influence pregnancy risk and consequently depression rates. Finally, while the study corrected for interpregnancy intervals in their covariate analysis, it is unclear how this was done and further clarification would be beneficial.

Relevance: This research is relevant to the field of women’s mental health as postpartum depression affects approximately 15% of women and risk factors for PPD are yet to be clearly elucidated. This study aids clinicians by providing prospective data from a diverse sample related to the pattern of depressive symptoms between subsequent pregnancies and related stress mediators. Clinicians may want to screen women more carefully in subsequent pregnancies, especially those with clinically significant depressive symptoms following their first pregnancy. In addition, it may help to be proactive in identifying psychosocial life stressors during the interpregnancy interval and working to provide increased support for these women. Finally, the information is beneficial to assist clinicians in educating and assisting pregnant women.

PUBLICATION #2 — Women’s Mental Health
Risk of mortality with concomitant use of tamoxifen and selective serotonin reuptake inhibitors: Multi-database cohort study
Donneyong MM, Bykov K, Bosco-Levy P, Dong YH, Levin R, Gagne JJ
BMJ 2016; 30;354:i5014
Annotation  (Shelly Kucherer)

The finding: This population-based cohort study looked at mortality rates in women who were taking tamoxifen along with an SSRI, comparing potent inducers of the 2D6 system with other SSRIs. Women were placed into one of two cohorts: women who started an SSRI after tamoxifen were placed in cohort 1 (n=6067) and women who were taking an SSRI prior to taking tamoxifen were placed in cohort 2 (n=8465). It was noted that when the databases within each cohort were pooled, the death rates were 58.6 per 1000 person years among users with SSRIs that were potent inhibitors of 2D6, and 57.9 per 1000 person years among patients using other SSRIs along with tamoxifen (hazard ratio 0.96, 95% confidence interval 0.88 to 1.06). When looking at the pooled data, there was no increase in all-cause mortality between women taking tamoxifen and an SSRI with potent inhibition at CYP2D6 compared to other SSRIs.

Strengths: This study is the first to look at the differences in mortality when comparing patients on tamoxifen who are taking SSRIs that are potent inhibitors of 2D6 compared to other SSRIs. The two-cohort design helps to minimize confounding by indication of SSRI initiation. Multiple databases were combined for this study from women covered by commercial and publically funded health insurance to help make results generalizable to the US population. Finally, statistical analyses were done looking at time of concomitant tamoxifen and SSRI exposure to look at a potential exposure duration-response effect.
Weaknesses: As this data is being pulled from collected databases and not patients directly, it is unclear how compliant patients were with their prescriptions. Cause of death was not available, limiting what conclusions may be drawn regarding the effects of the drug interaction on mortality. With a claims-based study, potential confounders from smoking, obesity, and socioeconomic status further limit the conclusions, though are offset by the matched characteristics between the cohorts. Mean follow-up was 2.4 years, so effects of long-term concomitant use of SSRIs and tamoxifen are unclear. One of the databases may have had incomplete data on deaths outside of the hospital (this is due to Social Security Administration excluding information for about 4.5% of deaths in this database); this was predicted to be similar between exposure groups. In the primary analysis, patients remained under observation even if they no longer took both medications during follow up. Ultimately, authors reported this could produce a bias towards the null hypothesis; however, this also allowed for looking at outcomes affected by exposure if these occurred much later. Finally, claims were able to identify patients with breast cancer, but it was difficult to distinguish patients with breast cancers of different stages, though they were able to conduct a sensitivity analysis excluding patients with stage 0 cancer, which did not affect the results.

Relevance: This study shows no difference in risk of death for patients taking CYP2D6-inhibiting SSRIs versus other SSRIs along with tamoxifen. This is important as there have been previous concerns that medications that inhibit the 2D6 isoenzyme could reduce the effectiveness of tamoxifen due to decreasing the bioavailability of endoxifen, an active metabolite of tamoxifen. Breast cancer remains the leading cause of cancer-related mortality among women worldwide, and almost half of breast cancer patients report depression, anxiety, or both within the first year after diagnosis. Given the effects of depression and anxiety on quality of life and morbidity in breast cancer patients, physicians can continue to consider potent 2D6-inhibiting SSRIs as a viable treatment option in women for whom these medications may provide symptom relief.