Short communication
Validation of the Edinburgh Depression Scale during pregnancy

https://doi.org/10.1016/j.jpsychores.2010.07.008Get rights and content

Abstract

Background

Untreated depression during pregnancy may have adverse outcomes for the mother and her child. Screening for depression in the general pregnant population is thus recommended. The Edinburgh Depression Scale (EDS) is widely used for postpartum depression screening. There is no consensus on which EDS cutoff values to use during pregnancy. The aim of the current study was to examine the predictive validity and concurrent validity of the EDS for all three trimesters of pregnancy.

Methods

In a large unselected sample of 845 pregnant women, the sensitivity, specificity, and validity of the EDS were evaluated. The Composite International Diagnostic Interview (depression module) was used to examine the predictive validity of the EDS. The anxiety and somatization subscales of the Symptom Checklist 90 (SCL-90) were used to examine its concurrent validity. Only women with a major depressive episode were considered as cases.

Results

The prevalence of depression decreased toward end term: 5.6%, 5.4%, and 3.4%. The EDS scores also decreased toward end term, while the SCL-90 subscale anxiety scores increased. The EDS showed high test–retest reliability and high concurrent validity with the SCL-90 anxiety and somatization subscales. The area under the receiver operating characteristic curve was high and varied between 0.93 and 0.97. A cutoff value of 11 in the first trimester and that of 10 in the second and third trimesters gave the most adequate combination of sensitivity, specificity, and positive predictive value.

Conclusions

The EDS is a reliable instrument for screening depression during pregnancy. A lower cutoff than commonly applied in the postpartum period is recommended.

Introduction

Being young and female are two important risk factors for depression [1]. This has repeatedly been shown in various epidemiological studies indicating that up to 10%–15% of young women (18–40 years old) suffer from depressive symptoms [2]. Pregnant women in Western countries consult health professionals (e.g., midwife, family doctor, and obstetrician) up to 10 times more than non-pregnant women of comparable age. These consultations are mainly used to evaluate somatic/obstetric parameters, with often inadequate attention being paid to factors of mental health and well-being [3]. This is remarkable in view of the fact that symptoms of depression during gestation are considered to be as common and severe as those with a postpartum onset or as those observed in non-childbearing women of similar age. When left untreated, antenatal depressive symptoms are likely to continue into the postpartum period, often in a more aggravated form [4]. While there is a plethora of studies on the detection and treatment of postpartum mental health problems, only recently has more attention been paid to mood and anxiety symptoms during pregnancy [5], [6]. These recent studies have repeatedly shown that mental health problems during pregnancy (e.g., depression, anxiety) are associated with impaired obstetric and infant outcomes. Antenatal stress and depression, for example, have been linked to preterm birth and impaired physical health [7], [8], [9], as well as to future emotional, behavioral, and cognitive problems in the offspring [10], [11], [12]. Taken together, these data not only illustrate the pathophysiological link between pregnancy, stress, and maternal/infant health outcomes but also stress the importance of frequent routine maternal mental health screening during pregnancy.

Early screening for antenatal depression is also recommended by national bodies, such as the National Institute for Health and Clinical Excellence (“Antenatal and postnatal mental health: clinical management and service guidance”). Therefore, there is a well-defined need for an easy-to-administer self-report scale that can be used by such healthcare professionals as midwives, obstetricians, health visitors, and family doctors.

The Edinburgh Postnatal Depression Scale (EPDS) is an example of such a brief self-rating scale, although it was originally designed for screening for postpartum depression. The EPDS is among the most widely used screening tools for depression in women and has been translated in over 50 languages. It has been validated in childbearing, non-childbearing, and menopausal women, in men, and in the elderly. This has resulted in a new name: the Edinburgh Depression Scale (EDS) [13], [14]. Recently, an internet version of the EDS has also been validated [15].

There have been relatively few studies on the psychometric qualities of the EDS as a screening tool for depression in pregnant women [16], [17], [18], [19], [20]. Sample size and sample selection issues, however, have made the results of these studies rather inconclusive. As a result, studies on depression during pregnancy often apply different cutoff points. This has, in turn, resulted in different reported depression prevalence rates (e.g., from 6.8% to 18.4%) [21]. In the absence of a consensus on appropriate trimester-specific EDS cutoff values during pregnancy, the current study was designed to evaluate both the predictive validity and the concurrent validity of the EDS in a large unselected sample of pregnant women at the three trimesters of pregnancy.

Section snippets

Participants and procedure

Between 2002 and 2004, at their first (12 weeks' gestation) obstetric control visit, 1507 pregnant women from five community midwifery practices in and around the city of Eindhoven were invited to participate in a large antenatal thyroid screening study. Seventy-nine percent (n=1191) consented to participate. Only Caucasian women with appropriate knowledge of the Dutch language (n=1113) were eligible. Of these women, 1085 (97%) completed the questionnaires. In total, 113 women were lost to

Results

At 12, 24, and 36 weeks' gestation, the numbers of women suffering from a major depressive episode according to the CIDI were 47 (5.6%), 46 (5.4%), and 29 (3.4%), respectively. The reliability values of the EDS indicated by Cronbach's α coefficient per trimester were 0.82, 0.83, and 0.84, respectively. With regard to test–retest reliability, the correlations between the EDS scores were as follows: between 12 and 24 weeks' gestation, r=0.61 (P<.01); between 12 and 36 weeks, r=0.55 (P<.01); and

Discussion

This is the first study in which the EDS was validated (trimester-specific) in a large unselected sample of pregnant women against a CIDI-based syndromal diagnosis of depression. Our findings confirm the use of the EDS as a valid measure for detecting depression in a pregnant population. The fact that the EDS also showed high test–retest reliability and relatively high concurrent validity confirmed good psychometric properties overall.

Previous EDS validation studies in pregnancy recommend a

References (36)

  • PhillipsJ et al.

    Validation of the subscales of the Edinburgh Postnatal Depression Scale in a sample of women with unsettled infants

    J Affect Disord

    (2009)
  • LewinsohnPM et al.

    Age-cohort changes in the lifetime occurrence of depression and other mental disorders

    J Abnorm Psychol

    (1993)
  • VolleberghWA et al.

    The structure and stability of common mental disorders: the NEMESIS study

    Arch Gen Psychiatry

    (2001)
  • EvansJ et al.

    Cohort study of depressed mood during pregnancy and after childbirth

    Bmj

    (2001)
  • GavinNI et al.

    Perinatal depression: a systematic review of prevalence and incidence

    Obstet Gynecol

    (2005)
  • DayanJ et al.

    Prenatal depression, prenatal anxiety, and spontaneous preterm birth: a prospective cohort study among women with early and regular care

    Psychosom Med

    (2006)
  • SuriR et al.

    Effects of antenatal depression and antidepressant treatment on gestational age at birth and risk of preterm birth

    Am J Psychiatry

    (2007)
  • RahmanA et al.

    Impact of maternal depression on infant nutritional status and illness: a cohort study

    Arch Gen Psychiatry

    (2004)
  • Cited by (0)

    View full text