Article Text

Download PDFPDF

Review: depression and anxiety during pregnancy are strong indicators of postpartum depression
  1. Catherine ECS Williams, BDS, PhD, FDS RCPS(Glasg), MPaed Dent
  1. Specialist Registrar (Paediatric Dentistry), Bart’s and The London NHS
    Trust, London, UK

Statistics from

OpenUrlCrossRefPubMedWeb of Science

Q What are the risk factors for postpartum depression?


Embedded ImageDesign:

Systematic review with meta-analyses.

Embedded ImageData sources:

Nineteen databases of medical, psychological, and social studies, searched to 2002.

Embedded ImageStudy selection and analysis:

Prospective studies (with diagnoses made 2 weeks postpartum or later, to avoid including women with postpartum blues) on non-psychotic depression starting within a year of giving birth were eligible. For inclusion, the following factors had to be clearly defined: diagnostic criteria, assessment methods, and risk factors. Also, the statistical relation between postpartum depression and any variables had to be clearly stated. Recent studies were analysed with respect to two previous meta-analyses to ensure agreement with findings. Effect sizes were calculated and reported as defined by Cohen, where an effect size of 0.2 indicates a small relation, 0.4 indicates a moderate relation, and 0.8 a strong relation.

Embedded ImageOutcomes:

Postpartum depression (according to DSM-IV).


The search identified two previous meta-analyses (n>14 000) plus additional studies including a further 10 000 women. Major predictors of postpartum depression were depression and anxiety during pregnancy, stressful life events, low social support, and a previous history of depression (effect sizes: depression: 0.75; anxiety: 0.68; life events: 0.61; social support: −0.64; previous history: 0.58). Neuroticism and the state of the marital relationship were moderate indicators (effect size of both: 0.39) and socioeconomic status or pregnancy related obstetric complications were low risk factors (effects sizes: −0.14 and 0.26). Maternal age (in women over 18 years), educational attainment, parity, and relationship length had no significant association with postpartum depression.


Depression and anxiety during pregnancy are the strongest indicators of postpartum depression.


Postpartum depression is the most common complication of childbearing occurring in 10–15% of recently delivered mothers, with well documented negative health implications for the mother, child, and family.1 The literature investigating possible risks for postpartum depression is vast and of variable quality.

This article is a synthesis of the recent literature from 1990–2002. It includes the results from two meta-analyses by O’Hara and Swain2 and Beck3 involving 12 000 subjects. It adds results from more recent studies involving a further 10 000 subjects.

The findings of this review have important implications for all health professionals involved in the care of recently delivered mothers.

Based on the research included in the synthesis, risk factors which are the strongest predictors of postpartum depression have been shown to be (in decreasing order of effect reported in terms of Cohen’s d) depression during pregnancy, anxiety during pregnancy, experience of stressful life events during pregnancy or early puerperium, low levels of social support, and previous history of depression.

The findings of this review reinforce the importance of exploring postpartum depression risk factor status when taking the mother’s history. Consideration could be given to the use of the Edinburgh Postnatal Depression scale (EPDS) for screening all new mothers in the immediate postpartum period. The efficacy of the EPDS is documented, such as the recent study by Dennis.4 Successful identification of risk factors allows timely implementation of appropriate preventive and treatment interventions.

The paper highlights the possible increased risk for postpartum depression in a specific group of mothers. Recent immigrants, who may be physically and culturally separated from their support systems, may be at risk due to the documented lack of social support.


Supplementary materials

  • Web-only Commentary

    The Commentary is available as a downloadable PDF (printer friendly file).

    If you do not have Adobe Reader installed on your computer,
    you can download this free-of-charge, please Click here


    Files in this Data Supplement:


  • For correspondence: Dr E R Robertson, University Health Network/Toronto General Hospital, 657 University Avenue, Toronto, Ontario M5G 2N2, Canada;

  • Sources of funding: not stated.


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.