Psychometric assessment of the Edinburgh Postnatal Depression Scale in an obstetric population

Highlights

• The current study provides understanding of the factor structure of the EPDS, association of subscales of the EPDS, and demographic correlates of EPDS subscales.

• Results suggest alternative use of the EPDS to track several mental health symptom categories for expectant and new mothers.

• A 3-factor model (i.e. anxiety, depression, anhedonia) of the EPDS displayed the best fit to the current data.

• Demographic correlates of EPDS subscales included history of depression, history of anxiety, race, and pregnancy status (i.e. first child or not).

Abstract

The prevalence and negative effects of perinatal depression are well known. The Edinburgh Postnatal Depression Scale (EPDS) is a common screening tool for perinatal depression and it is recommended for use by several professional organizations. The current study tested competing EPDS factor structures and assessed EPDS change from intake to 6-week follow-up, and identified demographic correlates in an outpatient obstetric sample. Using a retrospective observational study design, medical records were coded for demographic, mental health, and EPDS patient data (n = 524). Confirmatory factor analysis, t-tests, and ANOVA were utilized. Findings included: (1) a 3-factor model (i.e. anxiety, depression, anhedonia) of the EPDS displayed the best fit to the current data; (2) small declines in all 3 subscales of the EPDS from intake to 6-week follow-up appointments and; (3) demographic correlates of EPDS subscales included history of depression, history of anxiety, race, and pregnancy status (i.e. first child or not). The 3-factor structure can be used in clinical practice to assess perinatal depression in a nuanced fashion. Given that history of depression and anxiety are risk factors for perinatal depression, a thorough assessment of these items in clinical practice is needed.

Introduction

Perinatal depression (PD) is a common disorder affecting between 8.4% and 12.7% of women while pregnant (Gavin et al. 2005; Vesga-Lopez 2008) and approximately 11% of women after delivery (Ko et al. 2017). Approximately 11.9% of women experience PD at some point during the perinatal period (Woody et al. 2017). Mothers with PD are at an increased risk for several negative outcomes, including preterm birth, low birth weight (Grote et al., 2010), decreased child verbal IQ (Barker et al. 2011), and child observed and parental reported aggression (Hay et al. 2011). However, less than 50% of cases of PD are identified by health care providers in routine practice (Hewitt et al. 2009). Screening for PD is recognized as a way to identify PD and improve PD outcomes (Georgiopoulos et al. 2001). One common screening tool is the Edinburgh Postnatal Depression Scale (EPDS; Cox et al. 1987). The EPDS is a 10-item self-report scale that has been shown to have adequate reliability and validity (Cox et al. 1987).

Routine screening for PD is recommended by leading professional organizations. Postpartum Support International (PSI) recommends universal PD screening using an evidence-based tool such as the EPDS (Postpartum Support – PSI n.d). The American Academy of Pediatrics recommends incorporating the EPDS into the 1, 2, 4, and 6 month visits (Earls and Committee on Psychosocial Aspects of Child and Family Health 2010). The noted recommendations by professional organizations indicate the importance of screening for PD using a validated tool.

Obstetric healthcare providers in the current study utilized the EPDS as a means of assessing perinatal depression in patients. The utility of the EPDS can be assessed by examining the factor structure. Current understanding of the factor structure of the EPDS varies from a one (Berle et al. 2003), two (Hartley et al. 2014), or three (Coates, Ayers, de Visser 2017) factor structure. Little evidence exists for support of a one-factor total score structure. Two factor structures typically consist of depression and anxiety (Phillips et al. 2009; Vivilaki et al. 2009). Three factors structures typically consist of depression, anxiety, and anhedonia (Coates, Ayers, de Visser 2017; Kubota et al. 2014).

Clinically, there exists inconsistent diagnostic labeling of PD. For instance PSI recognizes the distinctions between Perinatal Mood and Anxiety Disorders (PMAD), PD, and Perinatal Anxiety (PA; Pregnancy & Postpartum n.d.). PSI includes diagnoses such as depression, anxiety, obsessive-compulsive disorder, posttraumatic stress disorder, and bipolar disorder in the category of PMADs (Pregnancy & Postpartum n.d.). The American College of Obstetricians and Gynecologists (ACOG) utilizes the term PD (ACOG 2018). ACOG defines PD as major or minor depressive episodes that occur during pregnancy or in the first 12 months after delivery (ACOG 2018). Further complicating inconsistent diagnostic definition, the DSM-5 classifies PD as major depressive disorder with peripartum onset (American Psychiatric Association 2013). Given measurement should follow from consistent diagnostic framing and vice versa, EPDS factor structure support may offer empirical grounding for diagnostic classification moving forward.

In light of the current screening recommendations by professional organizations and variability of the EPDS factor structure, the present study seeks to address the following aims:

• Assess the factor structure of the EPDS at intake and 6-week follow-up time points.

• Evaluate the association of subscales of the EPDS at intake and 6-week follow-up time points using the model that best fits the data.

• Assess demographic correlates of EPDS subscales at intake and 6-week follow-up time points using the model that best fits the data.