Exercise for postnatal psychological outcomes: a systematic review and meta-analysis

Abstract

BackgroundPostnatal psychological morbidity is an important and prevalent issue. Postnatal anxiety affects 3–43% of mothers and postnatal depression (PND) 10–15% of women in the year after giving birth. Both can affect the social and cognitive development of the child. In 2007, the UK National Institute for Health and Clinical Excellence recommended exercise as part of the treatment for women with a new mild or moderate episode of PND. These guidelines were based on two very small methodologically flawed randomised controlled trials (RCTs). A systematic review by Daley and colleagues (2009) assessed the effectiveness of exercise as an adjunctive treatment for PND. Their findings gave some support for the use of exercise. However, only 238 patients were included from five trials, most of which contained important methodological flaws. MethodsWe undertook a systematic review and meta-analysis of RCTs and quasirandomised controlled trials with interventions designed to increase aerobic physical activity. Studies with populations of mothers up to 1 year postpartum measuring any of the following outcomes were included: depression, anxiety, self-efficacy for exercise, quality of life, mother and infant bonding, and child social development. Trials of co-interventions such as physical activity and social support were included; however, trials comparing two types of physical activity were excluded. Medline, Embase, Cochrane Library, Psych Lit, and SportDiscus were searched from inception to March 13, 2012, with the search terms “exercise”, “physical activity”, “postnatal”, “postpartum”, “mother”, “birth”, “perinatal”, “depression”, “anxiety”, “self-efficacy”, “quality of life”, “mother and infant bonding”, and “child development”. No language restrictions were used. ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform were searched for trials in progress. OpenGrey and OAIster were searched for grey literature. Study selection, data abstraction, and quality assessment were done independently and in duplicate. The Cochrane Collaboration's method for assessing risk of bias was used to establish study quality. Weighted mean differences (WMD), standardised mean difference (SMD), and 95% CIs were calculated for individual studies and pooled. Findings5771 articles were identified from initial searches, 61 full text articles were obtained, and eight trials were eligible, with a total of 1354 participants. Four studies involved exercise classes or a set exercise routine; four consisted of tailored exercise advice sessions. Two studies had an unclear risk of bias and six a high risk. A previous review by Daley (2009) assessed quality using a quantitative measure; included trials had a mean risk of bias of 4·6 of a maximum score of 7, suggesting a moderate overall risk. Both reviews showed moderate to high risk of bias, mainly due to incomplete reporting of bias criteria. Women receiving exercise interventions in the postnatal period had a significantly reduced Edinburgh postnatal depression score (six trials included): WMD 2·22 (95% CI 0·48–3·96; χ2 =20·27, p=0·001, I2=75%). SMD for depression score for all trials was 0·39 (0·14–0·64; χ2 =17·93, p=0·01, I2=61%). SMD for trials with set exercise classes or routines was 0·73 (95% CI 0·35–1·11) and for trials with tailored exercise advice was 0·18 (95% CI 0·01–0·35). There were insufficient data for other outcomes to allow meta-analyses. InterpretationThe results suggest a moderate effect of exercise in reducing the symptoms of PND. However, the mean difference of 2·22 was less than the clinically significant difference of 4 units. Substantial heterogeneity was present, suggesting that caution should be used in interpretation of this result. We used a random effects model to address the heterogeneity. Subgroup analysis suggested that structured exercise classes or routines were more effective than tailored advice. These results will inform the development of more effective exercise interventions for postnatal depression and inform policy. FundingNational School of Primary Care Research.