In This article

• Suicidal ideation can follow different trajectories: In this study of 38 women, researchers identified three patterns during pregnancy and during the postpartum period, including stable low endorsement, stable more frequent endorsement, and changing endorsement over time.
• Patterns may persist or change after delivery: Individuals with consistently low or more frequent endorsement generally remained in the same trajectory group from pregnancy into the postpartum period, although trajectories varied among participants.
• The EPDS and PHQ-9 classified participants differently: The proportion assigned to each trajectory depended on the screening instrument used, suggesting that the choice of measure can influence how suicidal or self-harm ideation is captured.
• The screening items assess somewhat different experiences: The EPDS asks about thoughts of self-harm over the past week, whereas the PHQ-9 assesses thoughts of being better off dead or self-harm over the past two weeks. These differences are important when interpreting results in clinical practice and research.
• More research is needed: Larger, more diverse studies and approaches such as ecological momentary assessment could help clarify how suicidal ideation changes across pregnancy and postpartum and identify when additional assessment may be needed

Suicidal ideation (SI) is a public health concern, yet its course during pregnancy and the postpartum remains poorly characterized. Estimates suggest that approximately 5-7% of individuals experience SI during the perinatal period; however, prevalence estimates vary across studies and may differ based on psychiatric history and other sample characteristics.

In clinical settings, two of the most common depression screening instruments are the Edinburgh Postnatal Depression Scale (EPDS) and the Patient Health Questionaire-9 (PHQ-9). Within each measure, the last item is used to assess self-harm ideation or SI.

Prior research has identified factors that may contribute to SI risk during the perinatal period, including negative life events and psychological factors such as guilt, shame, or worthlessness. However, less is known about the course of SI over time, including how it emerges, persists, or changes across the perinatal period. Previous research has identified that, unlike depression, periods of SI during the perinatal period may be less likely to be episodic or transient. However, few studies looking at perinatal SI have included follow up through the postpartum period, limiting ability to effectively characterize trajectories of SI across the entire perinatal period.

Data From the UPWARD-S Study

To address this knowledge gap, the current study examined SI endorsement on item 9 of the PHQ-9 and item 10 of the EPDS in 38 women across pregnancy and up to six months postpartum. The goal of this analysis was to understand the course of SI over the perinatal period and to compare whether the trajectories were similar across the two measures.

Participants with histories of major depressive disorder were recruited for a larger study called Understanding and Preventing Depressive Relapse During Pregnancy (UPWARD). Participants were not eligible to participate in UPWARD if they endorsed SI during screening. To be inclusive of individuals who experience SI during pregnancy, UPWARD-S intentionally recruited individuals that endorsed SI during pregnancy, were at least 16 weeks pregnant and had a history of major depressive disorder (MDD), current dysthymia, or current subsyndromal depression. Participants were excluded if they had bipolar disorder or a psychotic disorder, active mania, or an active substance use disorder within the last twelve months.

After enrolling in the study, participants underwent a clinical interview to confirm a history of MDD and were eligible if they completed a baseline interview within 1 week of enrollment. The baseline interview included the completion of the Columbia Suicide Severity Rating Scale (C-SSRS) to assess levels of current SI and suicidal behaviors. Following the successful completion of a baseline interview, participants completed self-report surveys every two weeks including the EPDS and PHQ-9. The final sample included 38 individuals with twenty-six timepoints across pregnancy and postpartum.

Key Findings

1. In both pregnancy and postpartum periods, this study identified three distinct trajectories of SI.
2. During pregnancy, the three trajectories of SI included stable low endorsement (consistently low SI endorsement across pregnancy), declining endorsement (SI endorsement decreased across pregnancy), and stable more frequent endorsement (SI remained consistently high across pregnancy).
3. During postpartum, the three trajectories of SI were identified, including stable low endorsement (consistently low SI following delivery), increasing frequency of endorsement (SI increased following delivery), and stable more frequent endorsement (SI endorsement remained consistently high following delivery).
4. The two measures captured a different percentage of participants in any of these 3 trajectories, depending on the measure used. For example, 31.6% of individuals were classified into the stable low SI group during pregnancy with the EPDS. On the other hand, 60.5% of individuals were classified in the stable low SI group with the PHQ-9.

5.     Individuals in the stable low and stable more frequent groups generally remained in the same trajectory group from pregnancy into the postpartum period for each measure.

What Do the Findings Tell Us?

While suicidal ideation is reported in a subset of women during pregnancy, previous studies have not addressed the course of SI in individual patients.  The UPWARD-S study provides more nuanced information regarding SI across the perinatal period.

1. The endorsement of SI is dynamic over the course of the perinatal period, and the trajectories of SI can be empirically grouped into different groups that are defined by the nature of change over time. For some individuals, SI trajectories may remain consistent across pregnancy and postpartum, but for others SI trajectories are more variable.
2. The difference in subgroups identified in trajectories of SI across pregnancy and the postpartum highlight that the perinatal period cannot be categorized as one continuous risk window.
3. The difference in proportions of those identified using the EPDS and PHQ-9 suggests that the instruments may influence how patterns of SI are captured. Attention to the differences in what item 9 of the PHQ-9 and item 10 of the EPDS assess (i.e., suicidal ideation vs self-harm ideation, respectively) is important when using these measures for clinical or research purposes.

A limitation of the study is its sole reliance on the EPDS and PHQ-9 as measures of SI. While both assessments include items that measure suicidality, the EPDS item 10 is not exclusive to suicidality, but also assesses for thoughts of self-harm. Additionally, the two measures assess different time periods with the EPDS measuring thoughts within the past week and the PHQ-9 assessing thoughts within the past two weeks. 

While the current investigation represents one of very few intensive longitudinal studies on SI across pregnancy and the postpartum period, future research in larger more diverse cohorts is ultimately needed to determine trajectories of SI during the perinatal period. Incorporating approaches such as ecological momentary assessment (EMA) may also help capture greater nuance in how changes in SI in real-world settings. Together, these approaches may help us better understand when SI changes, for whom, and how measurement tools influence observed trajectories.

—Hannah Dineen, BA