Statistics from Altmetric.com
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.
Q Do preventive psychosocial and psychological interventions reduce the risk of postpartum depression (PPD)?
Cochrane Pregnancy and Childbirth Group trials register (January 2004), Cochrane Depression, Anxiety, and Neurosis trials register (October 2003), Cochrane Central Register of Controlled trials (October 2003), Medline (1966–2004), CINAHL (1982–2004), EMBASE/Excerpta Medica (1980–2004), secondary references, and experts.
Study selection and assessment:
randomised controlled trials (RCTs) of preventive psychosocial or psychological interventions with a primary or secondary aim of reducing the risk of PPD in pregnant women and new mothers (<6 wks postpartum). Methodological quality of individual studies was assessed.
main outcome was PPD.
15 trials (n = 7697) met the selection criteria. Most trials (n = 10) defined PPD as scores >12 on the Edinburgh Postnatal Depression Scale (EPDS). Meta-analysis showed that, in general, psychosocial and psychological interventions did not prevent PPD (table⇓). 6 pre-specified subgroup analyses were reported. (i) Type of psychosocial intervention: antenatal and postnatal classes (2 trials), lay home visits (1 trial), early postpartum follow up (1 trial), and continuity of care (1 trial) did not reduce PPD; home visits by a health professional did reduce PPD (table⇓). (ii) Type of psychological intervention: psychological debriefing (5 trials) and interpersonal psychotherapy (2 trials) did not reduce PPD. (iii) Mode of delivery: individually based interventions were marginally beneficial (table⇓), whereas group based interventions had no effect (4 trials). (iv) Intervention onset: postnatal only interventions had a beneficial effect (table⇓), whereas interventions initiated antenatally and continued postnatally did not (4 trials). (v) Duration of intervention: neither single contact (4 trials) nor multiple contact (11 trials) interventions reduced PPD. (vi) Study sample: interventions directed at women at high risk had a beneficial effect (table⇓ ), whereas those directed at the general population did not (8 trials).
Generally, preventive psychosocial and psychological interventions do not reduce risk of postpartum depression. However, home visits by professionals, interventions initiated postnatally, and those directed at high risk women may reduce risk of postpartum depression.
The systematic review by Dennis and Creedy provides a unique contribution to the literature on the prevention of PPD. The finding that psychosocial and psychological interventions, in general, do not prevent PPD might be a result of study heterogeneity and the quality of the 15 RCTs. Nevertheless, 3 types of intervention appear to reduce PPD: home visits by health professionals, interventions initiated postnatally, and those directed at high risk women. Home visits by health professionals often include multiple interventions that are not described in detail in the trials reviewed. Nurses therefore must rely on theories and other research to guide home visiting practices. Interventions to change negative thinking are indicated as a strategy for women at risk of depression1 and have been used successfully in other populations2 and in the management of PPD.3 Interventions initiated prenatally and continued after the birth were not effective, whereas those that began after the birth were effective. This might be explained by the fact that interventions started prenatally are anticipatory in nature, whereas those that begin after birth can focus on actual problems that contribute to depression. The finding that interventions for high risk women were more successful than those for the general population is relevant for decision making about programming and resource allocation. Targeted interventions are likely to have lower costs than those offered universally. Future research should address the methodological limitations of previous studies (eg, inadequate sample sizes and high attrition), clearly describe details of the intervention and its intensity, and establish which specific interventions work for women with which characteristics and why.
For correspondence: Professor C-L Dennis, Faculty of Nursing, University of Toronto, Toronto, Ontario, Canada.
Source of funding: no external funding.