The effectiveness of exercise in the management of post-natal depression: systematic review and meta-analysis

DALEY Amanda, JOLLY Kate, MACARTHUR Christine
Journal article citation:
Family Practice, 26(2), April 2009, pp.154-162.
Oxford University Press

The Chief Medical Officer and NICE have both advised health professionals to consider exercise as an intervention to treat post-natal depression, but no systematic review of its effectiveness has been conducted. This review covered five randomised and quasi-randomised trials, and found a statistically significant large effect size in favour of exercise over no exercise. However, all the trials were small and there was considerable heterogeneity so the results should be treated with caution. When the results of a trial that included exercise as a co-intervention with social support was removed, heterogeneity was no longer present and the overall effect size fell substantially (standardised mean difference -0.81 to -0.42). Further research is required.

Extended abstract:


DALEY Amanda; JOLLY Kate; MACARTHUR Christine;


The effectiveness of exercise in the management of post-natal depression: systematic review and meta-analysis.

Journal citation/publication details

Family Practice, 26(2), April 2009, pp.154-162.


The five trials in this review suggest that physical exercise can have a beneficial effect on women with post-natal depression, but the findings need to be interpreted with caution because of the limitations of the evidence base. More research is needed.


Some 10-15% of women experience PND in the year after delivery, with significant associated difficulties for the child and family as a whole. Some women are reluctant to take antidepressants after giving birth, and there is limited availability of psychological therapies. However, based on the increasing amount of evidence about its beneficial effects on depression in general, the National Institute for Health and Clinical Excellence (NICE) advises that exercise should be considered as a treatment for women with mild to moderate PND. No systematic review has yet been published on this topic.


What sources were used?

The following databases were searched: CINAHL (Cumulative Index to Nursing and Allied Health Literature); the Cochrane Library; EMBASE (Excerpta Medica); Medline; PsycINFO; Science Citation Index and Social Science Citation Index [presumably searched simultaneously through Web of Science]; and SPORTDiscus. The National Research Register, Current Controlled Trials and were searched for ongoing and completed but unpublished trials. The reference lists of relevant studies were checked, together with those of previous reviews, and authors in the field were contacted. 

What search terms/strategies were used?

Both free text and MeSH (Medical Subject Heading) terms were used, namely: PND; post-partum depression; exercise; physical activity; physical training; pram pushing; walking. No language or date restrictions were applied, and searches were carried out in March 2008. 

What criteria were used to decide on which studies to include?

Eligible studies were randomised or quasi-randomised controlled trials, published or unpublished, that compared any type of exercise treatment (or combination of exercise and other treatment) for PND with other treatments or no treatment. Participants could be diagnosed using a range of methods from the ‘clinical judgement of a health professional’ to clinical interview. Exercise was defined as ‘any planned, structured and repetitive bodily movement’, interventions had to last at least six weeks, and depression outcomes had to be measured using a validated instrument. 

Who decided on their relevance and quality?

The searches, conducted by the lead author, produced a total of 362 studies which she screened by title and abstract for relevance. Twenty-seven studies were acquired in full text and assessed independently by the first two authors. This filtering process is summarised in Figure 1. Quality assessment of the studies selected for review was conducted independently by the first two authors using a modified version of the Delphi list criteria that is set out in Table 2. 

How many studies were included and where were they from?

Five trials, published between 2003 and 2008, were reviewed and are summarised in Table 1. They were conducted in Australia (2), Canada (1, reported in 2 papers), England (1, conducted by two of the review authors); and Taiwan (1). 

How were the study findings combined?

Data were extracted independently by the first two authors using a standardised form on: publication; participants; study design; number of cases; recruitment procedures; nature and length of intervention; comparison group; outcome assessment; intervention adherence; potential sources of bias; and relevant confounders. A meta-analysis for PND was carried out using a random effects model and employing STATA software.

Findings of the review

The Australian, Canadian and English studies were RCTs, and the Taiwanese study a quasi-randomised trial. All were small in size, covering a total of 221 participants (114 in intervention groups, 104 in comparison groups), and they offered various types of treatment: group pram walking (Australia); whole stretching classes at a hospital plus phone calls to encourage compliance (Taiwan); individual exercise consultations plus phone calls (England); and individual home-based exercise administered by an exercise physiologist (Canada). 

In four trials participants were eligible if they were also taking antidepressants, but the Canadian trial specifically excluded women who were taking antidepressants or had received psychotherapy in the previous year. Exercise was generally compared with usual care. However, one trial compared exercise plus social support with usual care, and one compared exercise with social support. All the interventions lasted twelve weeks. 

The meta-analysis shows that, compared with no exercise, exercise reduces PND symptoms. The effect size was large (standardised mean difference (SMD) -0.81) but the 95% confidence intervals were wide (-1.53 to -0.10) and there was significant variability in results between studies. When the trial comparing exercise plus social support with exercise was removed, the effect size fell to SMD -0.42 and there was no significant variability between studies.

Authors' conclusions

‘This systematic review and meta-analysis provides some support for the role of exercise in the management of PND.’ The trials were generally of a good methodological standard, and reported high levels of attendance and adherence to treatment (although this may be attributable to the fact that participants were all volunteers). However, their limited number and small size mean that ‘the estimates may be imprecise’ and should be treated with caution. Moreover, the largest of the trials did not show a significant difference in depressive symptoms between the study groups at follow-up. The authors suggest that NICE guidance issued in 2006 on the basis of only two very small trials ‘may have been somewhat premature’ although their findings do add ‘a little more support’ to its recommendation. 

Further and larger scale research is needed to explore what types of exercise (e.g. aerobic, non-aerobic, combination of the two) may be particularly effective, and future trials should include longer follow-up periods.

Implications for policy or practice

The authors note that in three of the five trials participants were receiving other treatments, and conclude that ‘if exercise is to be recommended to women with PND, it should be as an adjunctive treatment. We found no evidence to support the effectiveness of exercise as a replacement for standard treatment.’

Subject terms:
intervention, outcomes, physical exercise, postnatal depression;
Content type:
research review
Journal home page
ISSN print:

Key to icons

  • Free resource Free resource
  • Journal article Journal article
  • Book Book
  • Digital media Digital media
  • Journal Journal

Give us your feedback

Social Care Online continues to be developed in response to user feedback.

Contact us with your comments and for any problems using the website.

Sign up/login for more

Register/login to access resource links, advanced search and email alerts