Timing of Postpartum Depression Onset May Predict Symptom Pattern

By | March 9, 2019

Postpartum depression (PPD) is a common and complicated illness, affecting about 15% of women in the perinatal period. Recent research from the PACT consortium (Postpartum Depression: Action Toward Causes and Treatment) has sought to classify the many potential subtypes of this heterogenous illness and has found emerging differences in time of onset, severity, comorbid anxiety, and suicidal ideation.

This previous research utilizing PACT data suggested that three distinct subtypes of PPD could be distinguished based on symptom type and severity and that time of onset plays a major role in symptom development (see previous blog post on this article: Heterogeneity of PPD). The divergence of symptoms among women who begin to experience PPD symptoms before delivery compared to those who only experience symptoms after delivery suggests that time of onset may play an important role in treatment considerations.

To expand upon subtype symptomatology, a second study was conducted using PACT data from 663 women between the ages of 19 and 40 in seven of the 19 PACT international sites. All data sets included prospective information about perinatal depression onset and complete Edinburgh Postnatal Depression Scale (EPDS) scores. To examine co-occurring anxiety symptoms and account for fluctuations in biological factors like hormones across the perinatal period, researchers used principal component and common factor analyses to break the EPDS into three symptom dimensions using the research domain criteria (RDoC) established by the National Institutes of Mental Health. The scores for each symptom dimension – depressed mood, anxiety, and anhedonia – were examined cross-sectionally during each trimester of pregnancy and during 3 postpartum periods (0-4 weeks, 4-8 weeks, and 8+ weeks).

Five subtypes of perinatal depression were identified by isolating symptom patterns in women that differed in severity, time of onset, and type of symptoms:

Subtype 1: Severe anxious depression most typically emerged in the first trimester of pregnancy or more than 8 weeks after birth. Ninety-eight percent (98%) of women with this subtype were categorized in the moderate to severe or very severe EPDS categories (mean EPDS score 20.2) and suffered severe depression and anxiety symptoms but did not have comparably high rates of anhedonia (inability to feel pleasure). Nearly all women in this category (99%) endorsed thoughts of self-harm. Many women reported a history of depression before pregnancy, although data regarding pre-pregnancy depression status were not collected.

Subtype 2: Moderate anxious depression shared a similar onset pattern as severe anxious depression. Women experienced moderate anxiety and depression symptoms (mean EPDS score 16.0), and most also endorsed thoughts of self-harm (76%).  Many of these women also reported depression symptoms prior to pregnancy.

Subtype 3: Anxious anhedonia, like subtype 1, resulted in high rates of moderate to severe and very severe EPDS scores (mean EPDS score 19.2). Symptoms emerged most frequently within 8 weeks postpartum, with rare onset during pregnancy, and carried high anxiety and high anhedonia, but few thoughts of self-harm.

Subtype 4: Pure anhedonia showed lower EPDS scores (mean 14.9) but about half of the women endorsed thoughts of self-harm (47%). Few women had onset during the immediate postpartum period (0-4 weeks), but otherwise time of onset was spread evenly across all pregnancy and postpartum periods of study.

Subtype 5: Resolved depression showed an onset of symptoms during the perinatal period which had resolved at time of EPDS assessment.  Symptoms typically emerged during the third trimester of pregnancy, and this subtype was characterized much lower EPDS scores (mean 4.1) than other subtypes.

The prevalence of these subtypes varied according to time of assessment during pregnancy or the postpartum period (as shown in Figure 3), with anxious subtypes (1, 2 and 3) predominating during the postpartum period.  Looking at this sample across all time points, half of the women had either the severe (32%) or moderate (19%) anxious depression subtypes. 26% had the subtype of resolved depression,12% had anxious anhedonia, and 11% had pure anhedonia.

Several clinically-important findings were also noted. During pregnancy, onset of depression symptoms in the second or third trimesters was associated with better outcomes in postpartum EPDS assessments. Conversely, onset of symptoms during the first 8 weeks of the postpartum period was associated with higher rates of severe depression (almost 4 times higher than women who reported onset during pregnancy). Considering the enormous hormonal fluctuations taking place during the transition from pregnancy to the postpartum period, it is possible that these hormonal shifts could have exacerbated severity in the early postpartum period; thus, this finding may indicate important biological differences in subtypes. Comorbid anxiety and anhedonia were additionally associated with both pregnancy and obstetric complications.

Limitations noted by the researchers include the cross-sectional secondary analysis of existing data, which provides less reliable information than prospective longitudinal data collection and may introduce ascertainment bias. The seven sites selected may have had interstudy differences in selection criteria, recruitment settings, and variables collected, and the population studied may not be generalizable due to a high proportion of white, married, and highly-educated participants in the current study. The PACT dataset utilized also included little information regarding mental health diagnoses prior to pregnancy and other factors that may have influenced mood and anxiety.

Despite these limitations, the findings make an important contribution toward improving the care of pregnant and postpartum women. This study expands our understanding of the multiple forms of postpartum depression seen in the clinical setting by using a novel approach to examine onset and quality of depressive symptoms during several perinatal periods. It generates hypotheses for future work and highlights the critical importance of tailoring therapies based on the subtype and symptom severity of women who present with clinically distinct phenotypes of perinatal depression. With further examination of these phenotypes, physicians will be able to provide faster, more efficient identification and treatment of postpartum depression.

Taylor Church, BS

Postpartum Depression: Action Towards Causes and Treatment Consortium. Heterogeneity of postpartum depression: a latent class analysis. (2015). Lancet Psychiatry, 2(1), 59-67.

Putnam, K. T., Wilcox, M., Robertson-Blackmore, E., Sharkey, K., Bergink, V., Munk-Olsen, T., et al. Meltzer-Brody, S. (2017). Clinical phenotypes of perinatal depression and time of symptom onset: analysis of data from an international consortium. Lancet Psychiatry, 4(6), 477-485.

MGH Center for Women's Mental Health