Brief report
Are we overpathologising motherhood?

https://doi.org/10.1016/j.jad.2009.05.004Get rights and content

Abstract

Background

The rate of psychiatric disorders in motherhood is often estimated on the percentage of women scoring high on validated self-report mood measures such as the Edinburgh Depression Scale. Screening for possible current or likely future distress also uses self-report mood measures, as well as additional psychosocial questions.

Methods

This paper critically questions whether such prevalence rates, and percentage of women with high risk status following screening, are being overestimated. The properties of the Edinburgh Scale are examined, along with the validity of diagnostic criteria. In addition a consideration as to the percentage of women classified as ‘at-risk’ is considered.

Results

The properties of the Edinburgh Scale show that around 50% of women scoring high are not in fact depressed. Revised estimates of prevalence rates are therefore given that take the properties of the scale into account which are more conservative than current estimates. Repeat testing of the scale after just two weeks to help differentiate transient from enduring distress will also lower the possibility of overpathologising motherhood, as will the use of correct cut-off scores. The DSM IV diagnostic criteria for depression are also questioned in relation to perinatal women and men. Finally, classifying women to be ‘at-risk’ based upon the presence of a single risk factor is questionable given that the majority of women with risks do not become depressed, and also the rate of women reported to have at least one risk (up to 88%) is so high as to negate the usefulness of this concept.

Conclusions

Current estimates of the prevalence of perinatal distress, and of women with risks, are an overestimation of the true rates. The clinical practice of using the presence of a single risk factor, or a single high score on a self-report mood scale, to form part of the assessment to determine whether or not to actively intervene may also overpathologise the situation. A more thorough understanding of these issues will improve our assessment procedures so that resources can be appropriately targeted to those women, and their families, who really need specialist mental health intervention.

Introduction

Perinatal distress, a term that encompasses disorders of depression and anxiety, either during pregnancy or the first 12 months postpartum, is serious and reportedly prevalent. Rates of depression are typically reported to be around 10–15% (e.g., Buist et al., 2008, Baker et al., 2005), and these rates will be higher when anxiety disorders are also included (Wenzel et al., 2003). The impact of depression on the foetus, infant, toddler and partner has been shown to be significant (e.g., Deave et al., 2008, McGrath et al., 2008, Matthey et al., 2000, Trapolini et al., 2008). Anxiety has also been shown to have detrimental effects (e.g. on the foetus: Glover and O'Connor, 2002).

That this condition has been recognised and identified is clearly beneficial to the well-being of families. Initiatives to implement routine screening for these conditions are also beneficial (e.g. Matthey et al., 2004), if this can help ameliorate, or prevent, such mood disorders in women.

The purpose of this paper however is to question whether we are now overpathologising ‘motherhood’. This is not to question the seriousness of perinatal mental health difficulties, nor the need to accurately identify women or men with significant difficulties or risks early in the process. Instead this paper puts forward an argument that many women (and possibly men) are being incorrectly identified as having clinically significant distress, or have risks that predict they may develop this level of distress. If this is in fact the case, it is likely that a more accurate assessment in the perinatal period will allow clinical services to better target those women or men who are experiencing significant distress, or are likely to become significantly distressed, and thus use the limited health resources more efficiently.

Section snippets

Do ‘high scores’ on self-report mood scales represent likely depression?

Self-report mood measures, such as the Edinburgh Depression Scale (Cox et al., 1987), the Beck Depression Inventory (Beck et al., 1961), and the CES-D (Radloff, 1977) are often used to report probable rates of depression in perinatal women. Clinically other information is also used to determine each individual's emotional experience.

For the EDS, which is the most commonly used measure, the validated cut-off score is 13 or more for probable major depression in English-speaking postpartum women.

The use of a single high self-report score to determine probable distress

Cox and Holden (2003) stated that:

“One high score (on the EDS) may indicate only that the woman is feeling temporarily over-whelmed by her circumstances or that she is tired and miserable on a particular day. Two high scores separated by 2 weeks, plus an interview, will usually confirm depression” (p. 61).

Yet only two studies, to my knowledge, have followed this advice when reporting the rate of probable depression based upon high EDS scores. Ballestrem et al. (2005) stated that “(doing the

Incorrect cut-off scores on scales used to determine ‘probable distress’

Many studies use incorrect cut-off scores on the EDS (e.g., 12 or more instead of the validated 13 or more; or the antenatal use of the lower postnatal cut-off score; Matthey et al., 2006). This can overestimate the rate of high scorers by 30%. Unfortunately there continue to be publications using, or recommending, incorrect cut-off scores on the EDS (e.g., Campbell et al., 2008, Rychnovsky and Brady, 2008, Sword et al., 2008), thus inflating the rate of probable depression. Using such

Diagnostic criteria may overpathologise normal experiences of parenthood

The ‘gold standard’ for diagnosing depression in the perinatal period is whether or not the individual meets the criteria for major depression as set out in DSM IV (or the similar ICD).

DSM IV diagnoses have the following nine symptoms:

  • 1.

    Depressed mood

  • 2.

    Anhedonia

    • 3.

      Weight or appetite change

    • 4.

      Sleep disturbance

    • 5.

      Observable psychomotor activity

    • 6.

      Fatigue

    • 7.

      Lack of self-worth

    • 8.

      Concentration difficulties

    • 9.

      Thoughts of death.

For major depression the client needs to have at least 5 symptoms, with either #1 or #2 present.

At-risk criteria

The arguments so far have focused on the interpretation of scale scores or symptoms. Also relevant to considering whether we are overpathologising motherhood is the classification of women as being ‘at-risk’ of developing postnatal distress or depression.

While there are different ways of determining such a classification (e.g., Austin et al., 2005, Carroll et al., 2005, Matthey et al., 2004), all of which are based on well-documented risk factors for pnd, in many instances the percentage of

Conclusion

This paper has put forward the argument that rates of probable postnatal distress, or risks for becoming distressed, are usually substantially overinflated due to weaknesses in our measures, procedures, or interpretations. Five pieces of evidence are described supporting this argument:

  • 1)

    Approximately half the women who score high on the most commonly used self-report scale, the EDS, actually meet criteria for depression in validation studies.

  • 2)

    No differentiation is usually made between women with

Role of funding source

There was no external funding involved with this manuscript.

Conflict of interest

There is no known conflict of interest.

References (35)

  • MattheyS. et al.

    Paternal and maternal depressed mood during the transition to parenthood

    J. Affect. Disord.

    (2000)
  • McGrathJ.M. et al.

    Maternal depression and infant temperament characteristics

    Inf. Behav. Dev.

    (2008)
  • RychnovskyJ.D. et al.

    Choosing a postpartum depression screening instrument for your pediatric practice

    J. Pediatr. Health Care

    (2008)
  • AustinM.-P. et al.

    Antenatal screening for the prediction of postnatal depression: validation of a psychosocial Pregnancy Risk Questionnaire

    Acta Psychiatr. Scand.

    (2005)
  • BakerL. et al.

    Prevalence of postpartum depression in a Native American population

    Matern. Child Health J.

    (2005)
  • BallestremC.-L.V. et al.

    Contribution to the epidemiology of postnatal depression in Germany — implications for the utilization of treatment

    Arch. Women Mental Health

    (2005)
  • BeckA.T. et al.

    An inventory for measuring depression

    Arch. Gen. Psychiatry

    (1961)
  • BoyceP. et al.

    Edinburgh Postnatal Depression Scale: validation for an Australian sample

    Aust. N. Z. J. Psychiatry

    (1993)
  • BuistA.E. et al.

    Postnatal mental health of women giving birth in Australia 2002–2004: findings from the beyondblue National Postnatal Depression Program

    Aust. N. Z. J. Psychiatry

    (2008)
  • CampbellA. et al.

    Aboriginal and Torres Strait Islander women's experience when interacting with the Edinburgh Postnatal Depression Scale: a brief note

    Aust. J. Rural Health

    (2008)
  • CarrollJ.C. et al.

    Effectiveness of the Antenatal Psychosocial Health Assessment (ALPHA) form in detecting psychosocial concerns: a randomised controlled trial

    Can. Med. Assoc. J.

    (2005)
  • CooperP.J. et al.

    Prediction, detection, and treatment of postnatal depression

    Arch. Dis. Child.

    (1997)
  • CoxJ. et al.

    Perinatal Mental Health: a Guide to the Edinburgh Postnatal Depression Scale (EPDS)

    (2003)
  • CoxJ. et al.

    Detection of postnatal depression: development of the 10 item Edinburgh Postnatal Depression Scale

    Br. J. Psychiatry

    (1987)
  • DeaveT. et al.

    The impact of maternal depression in pregnancy on early child development

    BJOG

    (2008)
  • DennisC.-L. et al.

    Depressive symptomatology in the immediate postnatal period: identifying maternal characteristics related to true – and false – positive screening scores

    Can. J. Psychiatry

    (2006)
  • EdwardsB. et al.

    Antenatal psychosocial risk factors and depression among women living in socioeconomically disadvantaged suburbs in Adelaide, South Australia

    A.N.Z.J.P.

    (2008)
  • Cited by (58)

    • Partner relationship, social support and perinatal distress among pregnant Icelandic women

      2017, Women and Birth
      Citation Excerpt :

      Perinatal distress can be expressed as stress,5 anxiety and/or depressive symptoms2,5–7 and can be detected as a screen positive level of self-reporting scales or by assessment of experienced clinicians. Perinatal distress is common among pregnant women in relation to other pregnancy problems and has been found to be 12–15% on self-report scales.6 In an Icelandic study 9.7% of women were reported to experience distress at the 16th week of pregnancy.7

    • Factors influencing maternal distress among Dutch women with a healthy pregnancy

      2015, Women and Birth
      Citation Excerpt :

      We have used self-reported screening instruments to examine the levels of maternal distress. We did not use diagnostic instruments for maternal distress, which implies that we identified only those women who are more likely to develop maternal distress but not necessarily suffer from maternal distress.75,76 The number of women in our study at risk for maternal distress, do not necessarily represent women diagnosed with maternal distress.73,74

    View all citing articles on Scopus
    View full text