A systematic review of systematic reviews of interventions to improve maternal mental health and well-being

Midwifery 29 (2013) 389–399

Contents lists available at SciVerse ScienceDirect

Midwifery journal homepage: www.elsevier.com/midw

A systematic review of systematic reviews of interventions to improve maternal mental health and well-being Fiona Alderdice, BSSc, PhD (Chair in Perinatal Health and Well-being)n, Jenny McNeill, BSc, MSc, PhD (Lecturer in Midwifery Research), Fiona Lynn, BA, MSc, PhD (Improving Children’s Lives Research Fellow) School of Nursing and Midwifery, Queens University Belfast, Medical Biology Centre, Lisburn Road, UK

a r t i c l e i n f o

abstract

Article history: Received 5 August 2011 Received in revised form 24 May 2012 Accepted 27 May 2012

Objective: to identify non-invasive interventions in the perinatal period that could enable midwives to offer effective support to women within the area of maternal mental health and well-being. Methods: a total of 9 databases were searched: MEDLINE, PubMed, EBSCO (CINAHL/British Nursing Index), MIDIRS Online Database, Web of Science, The Cochrane library, CRD (NHS EED/DARE/HTA), Joanne Briggs Institute and EconLit. A systematic search strategy was formulated using key MeSH terms and related text words for midwifery, study aim, study design and mental health. Inclusion criteria were articles published from 1999 onwards, English language publications and articles originating from economically developed countries, indicated by membership of the Organisation for Economic Co-operation and Development (OECD). Data were independently extracted using a data collection form, which recorded data on the number of papers reviewed, time frame of the review, objectives, key findings and recommendations. Summary data tables were set up outlining key data for each study and findings were organised into related groups. The methodological quality of the reviews was assessed based on predefined quality assessment criteria for reviews. Findings: 32 reviews were identified as examining interventions that could be used or co-ordinated by midwives in relation to some aspect of maternal mental health and well-being from the antenatal to the postnatal period and met the inclusion criteria. The review highlighted that based on current systematic review evidence it would be premature to consider introducing any of the identified interventions into midwifery training or practice. However there were a number of examples of possible interventions worthy of further research including midwifery led models of care in the prevention of postpartum depression, psychological and psychosocial interventions for treating postpartum depression and facilitation/co-ordination of parent-training programmes. No reviews were identified that supported a specific midwifery role in maternal mental health and well-being in pregnancy, and yet, this is the point of most intensive contact. Key conclusions and implications for practice: This systematic review of systematic reviews provides a valuable overview of the current strengths and gaps in relation to maternal mental health interventions in the perinatal period. While there was little evidence identified to inform the current role of midwives in maternal mental health, the review provides the opportunity to reflect on what is achievable by midwives now and in the future and the need for high quality randomised controlled trials to inform a strategic approach to promoting maternal mental health in midwifery. & 2012 Elsevier Ltd. All rights reserved.

Keywords: Maternal mental health Well-being Midwifery Review

Introduction Pregnancy is characterised by physiological, social and emotional changes and demands which can impact on maternal wellbeing. Empirical studies suggest that 15–25% of women experience high anxiety or depression during pregnancy (Ross and

n

Corresponding author. E-mail addresses: [email protected] (F. Alderdice), [email protected] (J. McNeill), [email protected] (F. Lynn). 0266-6138/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.midw.2012.05.010

McLean, 2006; Lee et al., 2007; Figueiredo and Conde, 2011) and that anxiety and depression levels are higher during pregnancy compared with the postpartum period. Both antenatal anxiety and depression have been found to be associated with poor obstetric and neonatal outcomes (Mancuso et al., 2004; Marcus and Heringhausen, 2009; Dunkel-Schetter and Glynn, 2010; Dunkel-Schetter, 2011). In addition to short term morbidity, maternal mental illness can have an adverse impact on family functioning, and the cognitive, emotional, social and behavioural development of infants (Dennis and Hodnett, 2007; Talge et al., 2007).

390

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Women are in regular contact with health-care services during the perinatal period and midwives are in a key position to educate and support women about mental health and well-being and identify women at risk: those experiencing increased stress, women at increased risk of developing mental health problems or women with existing mental illness. The final report from Midwifery 2020, identifies the midwife as the key health professional to promote well-being for women with uncomplicated pregnancies and it outlines a pivotal role for midwives in co-ordinating the journey through pregnancy for all women. The co-ordinating role of the midwife ensures women are referred to other services when appropriate and that holistic care is provided to optimise each woman’s birth experience regardless of risk factor (Midwifery 2020 Final Report, 2010). This is further highlighted by the CMACE report (Lewis, 2011), which recommends that midwives should be able to refer directly to psychiatry services to avoid women being lost in the system. However clarity is required on what effective interventions exist that would enable midwives to offer appropriate support and coordination of care within the area of maternal mental health and well-being. Gaps in knowledge also need to be identified, so that we provide an evidence based approach to the ongoing research and development of the role of midwifery in supporting maternal mental health and well-being. This systematic review of systematic reviews was conducted as part of a larger review study on the public health role of the midwife for Midwifery 2020 (McNeill et al., 2010). The review explored the education, support and screening roles of midwives through the reproductive pathway starting before conception through pregnancy, childbirth and the postnatal period. Looking towards 2020, the review explored interventions based on the everyday role of the midwife, which could be built on over the next decade to further develop the public health role of the midwife. The Royal College of Midwives (RCM) states that it is appropriate for midwives to gain competence in new skills, in accordance with NMC requirements, so that they can offer women a wider range of choices during maternity care including non-invasive therapies (RCM, 2007). Therefore this review also included interventions that could be conducted by midwives with specialist training. In the presence of time and financial constraints, a systematic review of systematic reviews provides a coherent appraisal and summary of reviews, allowing the findings of individual reviews to be compared and contrasted, facilitating a broad scope of mental health interventions. The specific research aim of this review was to identify which non-invasive interventions in the perinatal period would enable midwives to offer effective care to women within the area of maternal mental health and well-being.

Methods Search strategy A total of 9 databases were searched: MEDLINE, PubMed, EBSCO (CINAHL/British Nursing Index), MIDIRS Online Database, Web of Science, The Cochrane library, CRD (NHS EED/DARE/HTA), Joanna Briggs Institute and EconLit. A systematic search strategy was formulated including key MeSH terms and related text words under the headings of Midwifery, Study Aim, Study Design and Mental Health (see Fig. 1). Inclusion criteria were articles published from 1999 onwards, English language publications and articles originating from economically developed countries, indicated by membership of the Organisation for Economic Co-operation and Development (OECD). If the review did not clearly state the search strategy or include

MeSH terms Midwifery Midwifery / Obstetric nursing Community Health Nursing Pregnancy / Pregnant women Prenatal care Care, postnatal Study aim Evidence-based practice Health promotion Costs and cost analysis Cost-Benefit Analysis Study design Review Meta-Analysis Mental health Depression Depression, postpartum Mental health Mental disorders Stress, psychological / Anxiety Stress disorders, post-traumatic Mood disorders

Additional, associated text words Midwife/ves / certified nurse midwife Community midwife/ves Antenatal care / services (Supportive) Intervention(s) Prevention(s) Cost evaluation Cost effectiveness / economic evaluation Literature review Systematic review Pre-existing mental illness Well-being / wellbeing / wellness Postpartum psychosis Fig. 1. Search terms.

search terms and databases accessed they were not deemed eligible for inclusion. The key terms for inclusion in the search strategy were discussed and agreed with the study Advisory group and the Midwifery 2020 Public Health Stream working group. Interventions that could be used by or co-ordinated by a midwife were broadly agreed to include education, screening and support. Ambiguity as to the suitability of inclusion in regard to the role of the midwife was discussed within the project team (which included a midwife and 2 others with significant experience of maternity care research). If consensus could not be reached within the study team inclusion was agreed with the Advisory group. The following exclusion criteria were applied: reviews related specifically to obstetric interventions in pregnancy, rather than midwifery, and those which were not directly related to mental health and well-being of women from antenatal to the postnatal period. Reviews of interventions requiring specialist long term professional training such as psychotherapeutic interventions were also excluded unless part of a broader review. The initial search was conducted in November 2009, with a final update of the search carried out in October 2010. The initial search was conducted as part of a larger review of reviews, the full methods and results of which can be found in McNeill et al. (2010). A broader literature review on maternal mental health in the antenatal and postnatal period, beyond the remit of the current review of reviews on the role of the midwife in maternal mental health and well-being, can be found in the NICE guidelines (National Collaborating Centre for Mental Health, 2007).

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Data extraction Data were independently extracted using a data collection form, which recorded data on the number of papers reviewed, time frame of the review, objectives, key findings and recommendations. One review author independently assessed all the potential studies meeting the inclusion criteria and this was verified by a second author. Data were then extracted by the primary author using the data collection form. Any ambiguity or disagreement was resolved through discussion. Summary data tables (Table 1) were set up outlining key data for each study and findings were organised into related groups. Data quality assessment The methodological quality of the reviews was assessed based on assessment criteria in Smith et al. (2011). For a review to be rated as ‘high quality’ there would be evidence of search strategy, selection and inclusion criteria, assessment of publication bias and heterogeneity noted in methods or results. A review was rated as ‘medium quality’ if there was evidence of search strategy, selection and inclusion criteria but no assessment of bias or heterogeneity; while a rating of ‘low quality’ was given if there was evidence of a search strategy but no evidence of the other quality assessment criteria. One reviewer independently assigned a quality rating to each review and this was then verified in discussion with the other two reviewers.

391

Lee, 2008; Dennis and Allen, 2008). Dennis et al.’s (2007) Cochrane review of psychological and psychosocial interventions for the treatment of antenatal depression found only one trial which examined the efficacy of interpersonal psychotherapy for the treatment of depressed pregnant women. While the psychotherapy group had a reduction in EPDS scores immediately post-treatment in comparison to a parenting education group, the trial was too small to draw any conclusions. It was also conducted by a trained therapist and, therefore, beyond the everyday role of the midwife. Dennis and Allen (2008) conducted a high quality review looking at interventions other than non-pharmacological/psychosocial/psychological interventions delivered by a professional or lay person. One small three arm trial (two others did not have the methodological quality required for inclusion) looking at massage and depression-specific acupuncture sessions was identified and neither treatment was found to be effective in reducing depressive symptoms. A review by Beddoe and Lee (2008) suggests that there is some evidence from a number of poorly designed studies that pregnant women have health benefits from mind–body therapies, for example, psycho-education, relaxation, yoga and meditation in conjunction with conventional prenatal care. Existing studies suggest that these interventions could reduce perceived stress and anxiety, increase birth weight and shorten labour (Beddoe and Lee, 2008). However, the evidence is limited by small sample size and lack of a control group and should be interpreted with caution.

Postnatal Results Postnatal screening for mental health problems The search identified 2,497 abstracts which were screened for relevance to the area of maternal mental health and well-being and use of a review methodology. Full text was obtained for 194 reviews and 32 were identified as examining interventions that could be used or co-ordinated by midwives in relation to some aspect of maternal mental health and well-being from the antenatal to the postnatal period and had a documented search strategy. Full details of the search stages can be found in Fig. 2. The 32 reviews were organised under antenatal and postnatal headings which are identified in Fig. 2. Six more generic reviews covered more than one mental health issue and they have been introduced into one of seven sections below where they provide the most data. Eight of the reviews also included a meta-analysis. The key findings of these reviews can be found in Table 1.

Antenatal Antenatal screening for mental health problems A Cochrane review by Austin et al. (2008) on antenatal psychosocial assessment for reducing perinatal mental illness did not support the use of screening tools to identify mental illness in pregnancy. The Edinburgh Postnatal Depression Scale (EPDS) is widely used to screen for postnatal depression symptoms and a cut off score of 9–10 is often used to indicate women who are at risk of postnatal depression. Two small studies were found, both of which had significant methodological limitations, there was no evidence that using these screening instruments led to improved maternal mental health as measured by the EPDS. Interventions to treat antenatal depression and anxiety Three reviews were identified that explored treatments for antenatal depression and anxiety (Dennis et al. 2007; Beddoe and

One review of postnatal screening for mental health problems was identified (Hewitt and Gilbody, 2009; also reported in Hewitt et al. 2009) and a further general review provided data in this area (Matthey, 2004). Hewitt and Gilbody (2009) identified four screening studies in their review, all using the EPDS. Two of the studies looked at EPDS threshold scores at 6 weeks and two looked at EPDS scores at 16 weeks. The meta-analysis showed a significant reduction in the EPDS which suggests a reduction in risk of postnatal depression. However, they noted that the use of a formal screening instrument to identify postnatal depression was confounded in a number of studies, as the screening intervention was included with enhancement of care, such as counselling or interviewing training. Hewitt and Gilbody (2009) conclude that there is insufficient evidence that the EPDS is effective in the postnatal detection of postnatal depression and the subsequent improvement in maternal and infant health and well-being. Matthey (2004) provides evidence on detection and treatment of postnatal depression (including anxiety) and suggests both anxiety and depression should be assessed in new mothers and fathers. However, the review was rated as low quality as it lacked reporting of selection and inclusion criteria, assessment of publication bias and heterogenity. Interventions for the prevention of postpartum depression Of the eight reviews in this area (Ciliska et al., 2001; Austin, 2003; Bick, 2003; Ogrodniczuk and Piper, 2003; Dennis and Creedy, 2004; Lumley et al., 2004; Dennis, 2004a; Boath et al., 2005), the Cochrane review by Dennis and Creedy (2004) provided the highest quality data. Within this review, study methodological quality was good and all but one involved an intervention from a health professional. The review summarised the results of 15 trials involving 7,697 women that were conducted in four countries in a wide variety of circumstances.

392

Table 1 Included reviews of mental health and well-being (n¼ 32). Author (s) (year)

Antenatal Screening for mental health problems Austin et al. (2008)

Intervention

Mental health outcomes

Main results/findings

Metaanalysis (yes/no)

Quality level

2 (–2008)

Antenatal psychosocial assessment

Psychiatric symptomology

No significant effects on depression, anxiety Trend towards raised level of clinical awareness of risk factors.

No

High

Studies included psycho-education (3) relaxation (7) and yoga and medication (2) High heterogeneity in design, small sample size, poor quality—results inconclusive One small trial including massage therapy and acupuncture. Results inconclusive. Interpersonal psychotherapy compared to a parenting education programme was associated with a reduction in the risk of depressive symptoms immediately posttreatment

No

Low

No

High

No

High

Anxiety and depression scores Interventions for treating antenatal-depression and anxiety Beddoe and Lee (2008) 12 (1980–2007)

Dennis and Allen (2008)

1 (1966–2007)

Dennis et al. (2007)

1 (1966–2006)

Postnatal Postnatal screening for mental health problems Hewitt and Gilbody (2009) 4 (–2007)

Interventions for the prevention of postpartum depression Austin (2003) 5 (1995–2001)

Mind–body interventions during pregnancy

Perceived stress, mood and perinatal outcomes

Non-pharmacological/psychosocial/ psychological interventions for the treatment of antenatal depression Psychosocial and psychological interventions for treating antenatal depression

Antepartum/postpartum depression Maternal mortality

Antenatal and postnatal identification of depressive symptoms (all used EPDS)

Reduction in EPDS score postpartum and cost effectiveness

Meta-analysis showed beneficial effect of using the EPDS in reducing EPDS scores (OR 0.61 CI 0.48–0.76). Difficult to interpret as 2 out of 4 studies also included enhancement of care.

Yes

High

Targeted antenatal group interventions

Depression scores, DSM-IV depression classification

No

Medium

Care provided by midwives (and other care workers) to prevent or reduce postnatal psychological health problems A broad range of interventions for preventing postnatal depression, including psychological and social support interventions

Depression scores, questions on psychological health

One study showed significant effects on depression scores for primaparous women 5/12 studies showed positive effects on preventing postnatal depression

No

Medium

No

High

No

High

No

High

Antepartum/postpartum depression Maternal mortality

Bick (2003)

12 (1990–2002)

Boath et al. (2005)

21 (1966–2003)

Ciliska et al. (2001)

20 (1980–1998)

Home visiting in the prenatal and postnatal period

Not specified in methods range of maternal and infant outcomes reported

Dennis (2004a)

29 (1966–2003)

Non-biological interventions for the prevention of postpartum depression

Postpartum outcome assessment

Postnatal depression

9/21 studies showed positive short term effect of intervention on preventing postnatal depression. Evidence of effect in one trial for extending specialist midwifery care. Varying study quality. No negative effect of home visiting reported. Positive effects reported include physical and mental health, development of healthy habits and knowledge and service utilisation of mother and baby. Results inconclusive. Varying study quality. Interventions include interpersonal psychotherapy, cognitive-behavioural

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Number of papers included (date range)

Dennis and Creedy (2004)

15 (1966–2004)

Psychosocial and psychological interventions for preventing postpartum depression

Postpartum depression/psychosis

Maternal mortality

42 (1980–2003)

Non-pharmaceutical and nonhormonal interventions to reduce postnatal depression

Depression characterised by caseness or probably caseness by diagnostic interview or standard measure

Ogrodniczuk and Piper (2003)

19 (1990–2003)

Interventions for preventing postpartum depression

Postpartum depression

Postnatal debriefing Gamble et al. (2002)

3 (1982–2000)

Postnatal debriefing and non-directive counselling

Postpartum psychological morbidity

Gamble and Creedy (2004)

19 (1966–2003)

Content and process of postpartum counselling

Description of interventions

Maternal outcomes Lapp et al. (2010)

9 (–Feb 2010)

Postnatal interventions for PTSD

PTSD symptoms after childbirth

High variability in studies in terms of design, definition of debriefing and measurement of outcomes. Overall no evidence of effect. Most studies recommended an interpersonal approach to counselling but findings descriptive rather than evaluative Descriptions of intervention general and non-specific Six RCTs and one pilot RCT study of debriefing, one case report on CBT and

Yes

High

Yes

High

No

Medium

No

Medium

No

Low

No

High

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Lumley et al. (2004)

therapy, psychological debriefing, antenatal classes, intrapartum support, supportive interactions, continuity of care, antenatal identifications and notification, early postpartum followup, flexible postpartum care, educational strategies, relaxation with guided imagery. There was insufficient evidence to recommend any particular intervention. Women receiving psychosocial interventions were equally likely to develop depression as those receiving standard care (RR 0.81, CI 0.65–1.02). Identifying mothers at risk assisted prevention of postpartum depression (RR 0.67, CI 0.51–0.89) Interventions with only postnatal component more beneficial than those also incorporating antenatal component (RR 0.76, CI 0.58–0.98) Individually based interventions more effective than group based (RR 0.76, CI 0.59–1.00) No preventive effect of psychological debriefing (RR 0.57 CI, 0.31–1.04) Postnatal counselling intervention provided to women with depression or probable depression will reduce depressive symptoms and depression substantially with an NNT of 2 or 3. Community postnatal midwifery care had a NNT of 14. Continuity of care, doula support in labour, nurse support in labour, postnatal debriefing by a midwife, interventions to enhance mother–infant intervention could not be recommended as strategies for reducing postnatal depression. Results were mixed 13/19 studies were midwifery based and several provided support for the role of midwife based interventions for women both in the antenatal and postnatal period

393

394

Table 1 (continued ) Author (s) (year)

Number of papers included (date range)

Intervention

Mental health outcomes

Main results/findings

one pilot study of eye movement desensitisation and reprocessing. Overall inconclusive evidence of effective interventions. No overall evidence of effect. 2/8 studies showed evidence of effect (one midwife-led counselling, the other midwife-led debriefing). Methodological quality variable.

8 (1990–2006)

Postnatal debriefing interventions

Maternal mental health problems

Treatment of postpartum depression Bledsoe and Grote (2006)

11 (1997–2004)

Treatment for women with depression during pregnancy and postpartum period

Depressive symptomatology

Boath and Henshaw (2001)

30 (1964–2000)

Dennis (2004b)

21(1966–2003)

Treatment of women with postnatal depression Non-biological interventions to treat postpartum depression

Self-reported depression and depressive symptomatology Postpartum depression up to one year postpartum

Dennis and Hodnett (2007)

10 (1966–2006)

Postnatal psychosocial and psychological interventions

Postpartum depression Maternal mortality

Daley et al. (2007)

5 (–2006)

Exercise

Postpartum depression

Daley et al. (2009)

5 (–2008)

Exercise

Postnatal depression

Freeman (2006)

4 (1965–05)

Omega-3 fatty acids

Major depressive disorder

Stevenson et al. (2010)

6 (1950–2008)

Group cognitive behavioural therapy (CBT)

Postnatal depression

Parental training programmes to improve maternal mental health Barlow et al. (2003) 26 (1970–2002) Group-based parenting programmes

Psychosocial Health (anxiety, depression, social support, self-esteem, relationship with partner)

Meta-analysis of treatment v control effect size for all interventions 0.673 (p o 0.001). However the effect size was greatest for medication with psychological therapies. None of the 3 counselling and educational interventions had a significant effect Result inconclusive. Studies of varying quality and design. A range of interventions were identified IPT, CBT, peer and partner support, non-directive counselling, relaxation massage therapy, infant sleep interactions, infant mother relationships therapy, maternal exercise. The results were inconclusive and the studies of varying quality. Psychological and psychosocial interventions were effective in decreasing depressive symptomatology, however measured, within the first year postpartum (RR 0.70, CI 0.6–0.81) Results inconclusive. Studies of varying quality. Exercise reduced postnatal depression (RR  0.8, CI  0.64 to  0.35). Significant heterogeneity reduces the integrity of the findings Two studies showed no evidence of effect Two pilot studies showed potential efficacy 3/6 studies showed that group CBT was effective in reducing depression when compared to routine primary care, usual care or waiting list initiatives Meta-analysis of five outcomes Parenting programs effective in improving depression (Mean difference  0.26, CI  0.40 to  0.11)

Quality level

No

Medium

Yes

Medium

No

Medium

No

High

Yes

High

No

Low

Yes

High

No

Low

No

High

Yes

High

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Rowan et al. (2007)

Metaanalysis (yes/no)

Effective in improving anxiety/stress (mean difference  0.4, CI  0.6 to  0.2) No evidence of effect in improving social/support Effective in improving self-esteem (MD  0.3, CI  0.5 to 0.1) Effective in improving relationship with partner (MD  0.4 CI  0.7 to  0.2) Generic Austin and Priest (2005)

Detection and treatment of perinatal mood and anxiety disorders

Perinatal outcomes impact on offspring

Borja-Hart and Marino (2010)

6 (2003–2008)

Omega-3 fatty acids

Reduction in antenatal, perinatal and/ or postpartum depression, as measured by depression scales (EPDS, HAM-D, BDI, MADRS)

D’Souza and Garcia (2004)

9/48 (1990–2003)

Disadvantaged childbearing women

Perinatal outcomes

Subgroup: mental health problems

Leis et al. (2009)

6 (1989–2005)

Home based interventions to prevent and treat postnatal depression

Maternal reports of health status

Matthey (2004)

Number unclear (2002–2003)

Detection and treatment of postnatal depression

Morrell (2006)

37 (1966–2005)

Instruments for detecting postnatal depression Summary of treatment strategies Prevention and treatment of postnatal depression

Maternal reports of health status

Studies reporting treatment effects and impact of depression and anxiety on offspring were outlined. Overall review findings inconclusive due to limited quality assessment All studies showed a positive effect on depression scores, with 3/6 reporting a statistically significant positive effect on depression scores compared to placebo. However, study limitations included small sample size, variable doses of omega-3 fatty acids, short study durations and lack of a control group. Findings inconclusive. Promising interventions include: Professional or lay social support to help in treatment of postpartum depression Parenting programmes Community based postnatal are delivery specially trained community midwives 4/6 studies showed evidence of treatment effect. The four studies looked at psychological treatments Inconclusive in relation to screening and treatment of postnatal depression. Inconclusive due to methodological and reporting limitations of studies

No

Low

No

Low

No

Low

No

High

No

Low

No

Medium

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Number unclear (1995–2005)

395

396

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Number of abstracts screened n=2,497

Number of abstracts filtered out n=2,303 Duplicates (n=451) Non-English (n=22) Non-OECD (n=59) Not a review (n=194) Not a midwifery-led intervention (n=1,577)

Number of full text obtained n=194 Number of ineligible studies n=162 Not a systematic review (n=57) Not a midwifery-led intervention (n=105)

Total number of eligible studies n=32 Antenatal screening for mental health problems (n=1) Interventions to treat antenatal anxiety and depression (n=3) Postnatal screening for mental health problems (n=1) Interventions for the prevention of postpartum depression (n=8) Postnatal debriefing (n=4) Treatment of postpartum depression (n=8) Parent training programmes (n=1) Generic (n=6)

Fig. 2. Search results and filtering process.

Antenatal classes focusing on postpartum depression were shown to have no preventive effect. The evidence for interpersonal psychotherapy and lay support was uncertain. Preliminary evidence of the effect of early postpartum follow up suggested there is no preventive effect on postpartum depression. There was preliminary evidence to suggest that intensive postnatal nursing home visits with at risk mothers was protective during the first 6 weeks postpartum (Armstrong et al., 1999). The review by Ciliska et al. (2001), although less rigorously conducted, also supported this finding. Five of the reviews in this area explored a broader evidence base including, psychological and social support interventions, interpersonal therapy, postnatal stress debriefing, information and discussion/education, reconfiguring midwifery and other services, individual home based care, hormonal prevention, antidepressant prevention and dietary interventions (Bick, 2003; Ogrodniczuk and Piper, 2003; Lumley et al., 2004; Dennis, 2004a; Boath et al., 2005). Dennis (2004a) provided a broad review of 29 non-biological interventions using quasi-experimental and RCT designs and the conclusions were similar to that of the Cochrane review (Dennis and Creedy, 2004). Lumley et al. (2004) conducted a

broad systematic review and meta-analysis of 42 trials excluding pharmaceutical or hormonal intervention studies. The review concluded that only postnatal interventions for women at risk for depression have a substantial impact on postnatal depression. Therefore postnatal counselling interventions by a variety of professionals for women at risk of depression were effective but universal interventions were considered to be ineffective based on review evidence. Community postnatal midwifery care also significantly reduced depressive symptoms. However continuity of care, doula support in labour, nurse support in labour, postnatal debriefing by a midwife, interventions to enhance mother–infant intervention could not be recommended as strategies for reducing postnatal depression. The review by Boath et al. (2005) included 21 RCTs, too diverse to be meta-analysed and there was no analysis of bias or heterogeneity. Fifteen of the trials looked at psychological and social support and included similar trials to that reported in Dennis and Creedy (2004) although the two reviews had different exclusion criteria. One small trial found under the heading of reconfiguring midwifery and other services looked at an extended period of specialist midwifery care which was associated with a significantly lowered EPDS score at 4 months postpartum in comparison to controls. The reviews by Ogrodniczuk and Piper (2003) and Bick (2003) identify a number of midwifery interventions. The review by Bick (2003) included 12 studies and focused specifically on midwifery care for the prevention or reduction of postnatal psychological health problems. Bick concluded that significant benefits to postnatal psychological well-being have been found following the implementation of new models of midwifery-led care although further evidence is needed to substantiate these findings. Ogrodniczuk and Piper (2003) identified 19 studies looking at interventions for pregnant women in general and interventions that targeted high risk women. Of the 19 studies in the review 13 involved some form of midwifery care either provided directly by a midwife or someone in a midwifetype role. While overall the results were inconclusive, there were a number of promising midwifery based interventions based on individually tailored case management by midwives and continuity of care identified. This review did not provide sufficient information on individual studies to assess overall quality. The final review in this area was Austin (2003) which specifically looked at targeted group antenatal prevention of postnatal depression and identified five studies. One of the studies used a midwife educator to facilitate the group while the four others used clinical psychologists (n ¼3) and another used a psychiatric nurse. Two out of the five interventions reported significant effects; one small study using interpersonal therapy and the other involved six weekly meetings with a psychiatric nurse which was found to be effective in primparous women. Overall the studies were small and methodological problems were identified in defining women at risk of postnatal depression. Postnatal debriefing Three reviews of postnatal debriefing were identified (Gamble et al., 2002; Gamble and Creedy, 2004; Rowan, Bick and Bastos, 2007). A further review by Lapp et al. (2010) which looked at the management of post traumatic stress disorder after birth, also focused predominantly on debriefing and counselling interventions. Gamble et al. (2002) were specific in their definition of debriefing as a single debriefing session or non-directive counselling session to reduce depression and trauma symptoms in women in the first few days following birth. While the implication of included studies was that counselling should be offered a few days/weeks after birth, this was not always clear. Gamble and Creedy (2004), Rowan et al. (2007) and Lapp et al. (2010) conducted broader reviews in terms of

F. Alderdice et al. / Midwifery 29 (2013) 389–399

definition, reflecting the more generalised definitions of such interventions in current research literature. Also, it was noted that many existing randomised controlled trials (RCTs) and cohort studies set up to examine debriefing have used depression as an outcome rather than PTSD (Gamble et al., 2002). Rowan et al. (2007) identified eight RCTs of debriefing or counselling intervention in childbirth settings, seven within the UK. Results of one randomised controlled trial reported possible short term harm, two trials indicated a positive association related to a psychological intervention, while the others showed that there were no differences in outcomes. Lapp et al. (2010) identified seven debriefing or counselling interventions after birth; overall, the reviews concluded that there was insufficient evidence to support the use of formal postnatal debriefing. Timing of the intervention was not always clear and some studies describe the provision of a counselling service any time after birth and even years later (Gamble and Creedy, 2004). Treatment for postpartum depression There were eight reviews of treatment for postpartum depression of varying quality (Boath and Henshaw, 2001; Dennis 2004b; Bledsoe and Grote, 2006; Freeman, 2006; Dennis and Hodnett, 2007; Daley et al., 2007, 2009; Stevenson et al., 2010). One review by Borja-Hart and Marino (2010) provided evidence on the broader area of perinatal depression and three further reviews were identified that provided information on prevention and treatment of postpartum depression (Austin and Priest, 2005; Morrell, 2006; Leis et al., 2009). A range of treatment approaches including pharmacological, psychological, psychosocial and hormonal were identified in the review by Boath and Henshaw (2001). However there were limited data presented on the individual studies in regard to methods and recruitment and review findings was inconclusive. Dennis (2004b) also conducted a broad review of interventions for interpersonal therapy, support interventions, mother–infant interventions and maternal exercise. The review identified 21 papers. Definitive conclusions could not be drawn due to methodological limitations; lack of RCTs, small sample size or lack of true control group. However Dennis and Hodnett’s (2007) Cochrane review of 9 trials (reporting outcomes on 956 women) found any psychosocial or psychological intervention compared to usual postpartum care was associated with a reduction in the likelihood of continued depression. However, definite conclusions could not be reached about the relative effectiveness of the different treatment approaches due to poor design, diversity of settings and small sample sizes. The findings of this review are also supported by Bledsoe and Grote (2006); although their review included less rigorous studies. Trials selecting participants based on a clinical diagnosis of depression were just as effective in decreasing depressive symptoms as those that enroled women who met inclusion criteria based on selfreported depressive symptoms. All interventions were face to face and provided by a health professional except for one trial that provided telephone-based peer support. Psychosocial interventions included peer support and non-directive counselling. A range of psychological interventions were included, for example, cognitive behaviour therapy (CBT), that generally require specialist training beyond the everyday role of the midwife. A Health Technology Assessment review by Stevenson et al. (2010) reported on six studies of group CBT for postnatal depression. Three studies showed the treatment to be effective in reducing depression when compared to routine primary care but there was no adequate evidence on which to assess group CBT and how it compared with other treatments for postnatal depression. Morrell (2006), in her review of interventions to prevent or treat postnatal depression, found that the 37 studies identified were

397

inconclusive. Studies reviewed by Morrell (2006) had variable recruitment times ranging between a few days after birth through to 18 months postpartum. Leis et al. (2009) review looked at RCT’s of home based psychological interventions to prevent or treat postnatal depression and the studies overlapped with those reported in Morrell’s review (2006). The review by Austin and Priest (2005) provided more generic reporting of treatment effects but the review findings were inconclusive due to the lack of focus and quality of the review. Other interventions that have been used to treat mental illness include exercise, diet and complementary therapies. Daley et al. (2007) found two trials exploring the role of exercise in treating postpartum depression. This was later updated with Daley et al. (2009) reviewing five studies looking at exercise interventions with other treatments/no treatment in women with postnatal depression. Due to heterogeneity of included studies, it was unclear whether exercise reduces symptoms of postnatal depression. Freeman (2006) conducted a review of Omega-3 fatty acids and perinatal depression and found four studies; two evaluating efficacy in postpartum depression prophylaxis and two evaluating its use as acute treatment. The included studies were small and the results were inconclusive. A 2010 review by Borja-Hart and Marino identified seven studies of the use of Omega-3 Fatty Acids for the prevention or treatment of perinatal depression. Only six were reviewed as one trial was discontinued early due to the relapse of depression symptoms. As with the review by Freeman (2006) the results were inclusive and limited by sample sizes, variable dosing and study duration. Parent training programmes to improve maternal mental health One high quality Cochrane review was identified by Barlow et al. (2003), which aimed to establish whether parent-training programmes can improve maternal psychosocial health. Twenty-six studies were included in the review, with a total of 64 assessments of a range of psychosocial outcomes: depression, anxiety, stress, selfesteem, psychiatric morbidity. Data sufficient to combine in metaanalysis existed for only five outcomes (depression, anxiety/stress, self-esteem, social support, and relationship with spouse/marital adjustment). Meta-analyses showed statistically significant results favouring the intervention group for depression, anxiety/stress, selfesteem, and relationship with spouse/marital adjustment. All of the programmes reviewed were successful in producing positive change in maternal psychosocial health in the short term. However, long term consequences need to be explored and the authors highlight that there is insufficient evidence to reach any firm conclusions regarding the role of parenting programmes in the primary prevention of mental health problems. D’Souza and Garcia (2004) highlighted parenting programmes as a promising intervention for women with mental health problems in their review of women with mental health problems, as a subgroup of disadvantaged women (9 out 48 studies in the overall review). The review included systematic reviews and original studies and included no quality assessment.

Discussion The 32 reviews identified in this review were diverse in quality and in content. Many provided some evidence of effect but findings have been confounded by poor design and quality of included studies. Overall no review identified an intervention in the perinatal period that could be definitively recommended in clinical practice. While some reviews identified promising interventions, the samples of included studies were small and the range of interventions was too heterogeneous to establish which intervention would be most effective. A number of interventions were highlighted that warrant further consideration, for example, midwifery led models of care for

398

F. Alderdice et al. / Midwifery 29 (2013) 389–399

preventing depression, psychological and psychosocial interventions for treating postnatal depression, mind–body interventions in pregnancy and use of exercise. Facilitation/co-ordination of parent-training programmes was also considered to be promising in preventing maternal mental health problems. However, more research is required on these interventions to ensure a good evidence base to inform training, policy and practice about the relative effectiveness of these different treatment approaches. It is clear from the reviews identified that most of the research in this field has focused on intervention for depression rather than stress and anxiety. Also many of the interventions identified would require additional midwifery training, for example, those involving psychological and psychosocial interventions. Those interventions related specifically to midwifery did not appear to be highly intensive or to differ much from usual care but the reviews highlighted the importance of tailoring the intervention to the individual needs of the woman. More measures of maternal well-being should be included in future randomised controlled trials of midwifery models of care to further elucidate these findings. In the absence of clear evidence, the priority is to raise awareness of the importance of maternal mental health and the potential role of the midwife in promoting maternal well-being. A recommendation of this review is for midwives to avail of the opportunity to inform and influence a strategic research approach to perinatal mental health interventions and to drive this strategy by building on their core role. Future research needs to address the major gap in the literature on effective interventions in maternal mental health and well-being in pregnancy as this is the point of most intensive contact and provides considerable opportunity to improve maternal well-being. As the evidence continues to mount in relation to the impact of maternal stress and anxiety on the fetus and woman’s health and well-being in the short and long term (Dunkel-Schetter, 2011), there is an urgency to consider how we can best support women in pregnancy to maximise well-being. This is not just in relation to the screening and diagnosis of mental ill health antenatally, but more generally into the stress and strains in pregnancy and how women cope with them. Consideration needs to be given to what makes women anxious in pregnancy and what can be done to support her in routine care (Alderdice and Lynn, 2010). Midwives are ideally placed to introduce simple interventions to enhance self care or to educate women about the symptoms and changes to expect in pregnancy to help relieve stress. However, research is urgently needed in the area of stress reduction interventions in pregnancy before they can be introduced in practice. While this reviews of reviews provides an important overview and coherent appraisal of current knowledge, the reviews included varied in quality and scope, which imposes limitations on interpreting the findings. Consideration needs to be given to the differences in populations in each review and how relevant the population studied is to the population where the intervention will be implemented. The consequence of this diversity is a narrative review with limited synthesis, which is a common problem within systematic reviews. This is reflected in this study, with less than a quarter of included systematic reviews reporting a meta-analysis. In addition, reviews may provide information about the effectiveness of interventions but it is also pertinent to consider how realistic or practical the interventions are and the impact on service users. Within the context of this review, the intervention would need to align with both the expectations of women and the scope of midwifery practice. Despite these limitations, a systematic review of reviews can provide reassurances on the conclusions of individual reviews and highlight the best quality reviews in a single document with definitive summaries to inform clinical practice (Smith et al., 2011) and help direct future research strategies. This review highlighted the lack of effective interventions currently available for use by midwives to support women’s maternal mental health and

well-being. The review also identified the need to focus more on well-being rather than solely mental illness. It should also be noted that evidence was identified only for short term outcomes related to immediate psychosocial well-being and further research is needed to evaluate longer term health outcomes and outcomes related to resources and organisation of care.

Conclusions This systematic review of systematic reviews provides a valuable summary of the current evidence and gaps in evidence related to interventions to improve maternal mental health in the perinatal period. It also provides the opportunity to reflect on what needs to be considered by midwives on their potential role in the field going forward. The review highlighted that based on current systematic review evidence it would be premature to consider introducing any of the identified interventions into midwifery training or practice. However there were a number of examples of possible interventions in the included reviews worthy of further research. A major gap in the review literature was the lack of reviews exploring a specific midwifery role in maternal mental health and well-being in pregnancy, and yet, this is the point of most intensive contact and needs a strategic research approach. Midwives play a key role in refocusing a perinatal mental health research strategy to ensure it includes high quality randomised controlled trials of interventions that can promote well-being and reduce stress for all women, in addition to supporting women with mental illness.

Acknowledgements We would like to acknowledge the NHS Education for Scotland and the Midwifery 2020 Public Health Group for funding the project and for providing strategic direction. We would also like to acknowledge the support and commitment of the Review of Reviews steering group (Mrs Elizabeth Bannon, Prof Debra Bick, Dr Helen Cheyne, Prof Mike Clarke, Mrs Joanne Gluck, Prof Billie Hunter, Dr Dermot O’Reilly). References Alderdice, F.A., Lynn, F.A., 2010. The factor structure of the Prenatal Distress Questionnaire. Midwifery, 2010, http://dx.doi.org/10.1016/j.midw.2010.05.003. Armstrong, K.L., Fraser, J.A., Dadds, M.R., Morris, J., 1999. A randomised controlled trial of nurse home visiting to vulnerable families with newborns. Journal of Paediatrics and Child Health 35, 237–244. Austin, M.-P., 2003. Targeted group antenatal prevention of postnatal depression: a review. Acta Psychiatrica Scandanavica 107, 244–250. Austin, M.-P., Priest, S.R., 2005. Clinical issues in perinatal mental health: new developments in the detection and treatment of perinatal mood and anxiety disorder. Acta Psychiatrica Scandinavica 112, 97–104. Austin, M.-P., Priest, S.R., Sullivan, E.A., 2008. Antenatal psychosocial assessment for reducing perinatal mental health morbidity. Cochrane Database of Systematic Reviews. John Wiley & Sons, Chichester, UK. Barlow, J., Coren, E., Stewart-Brown, S., 2003. Parent-training programmes for improving maternal psychosocial health. Cochrane Database of Systematic Reviews. John Wiley & Sons, Chichester, UK. Beddoe, A.E., Lee, K.A., 2008. Mind–body interventions during pregnancy. Journal of Obstetric, Gynecologic, & Neonatal Nursing 37, 165–175. Bick, D., 2003. Strategies to reduce postnatal psychological morbidity: the role of Midwifery Services. Disease Management and Health Outcomes 11, 11–20. Bledsoe, S.E., Grote, N.K., 2006. Treating depression during pregnancy and the postpartum: a preliminary meta-analysis. Research on Social Work Practice 16, 109–120. Boath, E., Bradley, E., Henshaw, C., 2005. The prevention of postnatal depression: a narrative systematic review. Journal of Psychosomatic Obstetrics and Gynecology 26, 185–192. Boath, E., Henshaw, C., 2001. The treatment of postnatal depression: a comprehensive literature review. Journal of Reproductive and Infant Psychology 19, 215–248. Borja-Hart, N.L., Marino, J., 2010. Role of omega-3 fatty acids for prevention or treatment of perinatal depression. Pharmacotherapy 30, 210–216.

F. Alderdice et al. / Midwifery 29 (2013) 389–399

Ciliska, D., Mastrilli, P., Ploeg, J., Hayward, S., Brunton, G., Underwood, J., 2001. The effectiveness of home visiting as a delivery strategy for public health nursing interventions to clients in the prenatal and postnatal period: a systematic review. Primary Health Care Research and Development 2, 41–54. Daley, MacArthur, A., Winter, 2007. The role of exercise in treating postpartum depression: a review of the literature. Journal of Midwifery and Women’s Health 52, 56–62. Daley, A., Jolly, K., MacArthur, C., 2009. The effectiveness of exercise in the management of post-natal depression: systematic review and meta-analysis. Family Practice 26, 154–162. Dennis, C.-L., 2004a. Preventing postpartum depression, part II: a critical review of nonbiological interventions. Canadian Journal of Psychiatry 49, 526–538. Dennis, C.-L., 2004b. Treatment of postpartum depression, part 2: a critical review of non-biological interventions. Journal of Clinical Psychiatry 65, 1252–1265. Dennis, C.-L., Creedy, D.K., 2004. Psychosocial and psychological interventions for preventing postpartum depression. Cochrane Database of Systematic Reviews. John Wiley & Sons, Chichester, UK. Dennis, C.-L., Allen, K., 2008. Interventions (other than pharmacological, psychosocial or psychological) for treating antenatal depression. Cochrane Database of Systematic Reviews. John Wiley & Sons, Chichester, UK. Dennis, C.-L., Hodnett, E.D., 2007. Psychosocial and psychological interventions for treating postpartum depression. Cochrane Database of Systematic Reviews. John Wiley & Sons, Chichester, UK. Dennis, C.-L., Ross, L.E., Grigoriadis, S., 2007. Psychosocial and psychological interventions for treating antenatal depression. Cochrane Database of Systematic Reviews. John Wiley & Sons, Chichester, UK. D’Souza, L., Garcia, J., 2004. Improving services for disadvantaged childbearing women. Child Care Health and Development 30, 599–611. Dunkel-Schetter, C., 2011. Psychological science on pregnancy: stress processes, biopsychosocial models, and emerging research issues. Annual Review of Psychology 62, 531–558. Dunkel-Schetter, C., Glynn, L., 2010. Stress in pregnancy: empirical evidence and theoretical issue to guide interdisciplinary researchers. In: Contrada, R., Baum, A. (Eds.), Handbook of Stress, 2nd edn. Springer publishing Company, New York. Figueiredo, B., Conde, A., 2011. Anxiety and depression in women and men from early in pregnancy to 3-months postpartum. Archives of Women’s Mental Health 14, 247–256. Freeman, M.P., 2006. Omega-3 fatty acids and perinatal depression: a review of the literature and recommendations for future research. Prostaglandins Leukotrienes and Essential Fatty Acids 75, 291–297. Gamble, J., Creedy, D., 2004. Content and processes of postpartum counselling after a distressing birth experience: a review. Birth 31, 213–218. Gamble, J., Creedy, D., Webster, J., Moyle, W., 2002. A review of the literature on debriefing or non-directive counselling to prevent postpartum emotional distress. Midwifery 18, 72–79. Hewitt, C.E., Gilbody, S.M., 2009. Is it clinically and cost effective to screen for postnatal depression: a systematic review of controlled clinical trials and economic evidence. BJOG: An International Journal of Obstetrics and Gynaecology 116, 1019–1027. Hewitt, C.E., Gilbody, S.M., Brealey, S., et al., 2009. Methods to identify postnatal depression in primary care: an integrated evidence synthesis and value of information analysis. Health Technology Assessment 13 (36), 1–230. Lapp, L.K., Agbokou, C., Peretti, C.-S., Ferreri, F., 2010. Management of posttraumatic stress disorder after childbirth: a review. Journal of Psychosomatic Obstetrics and Gynecology 31, 113–122.

399

Leis, J.A., Mendelson, T., Tandon, S.D., 2009. A systematic review of home-based interventions to prevent and treat postpartum depression. Archives of Women’s Mental Health 12, 3–13. Lee, A.M., Lam, S.K., Sze Mun Lau, S.M., Chong, C.S., Chui, H.W., Fong, D.Y., 2007. Prevalence, course and risk factors for antenatal anxiety and depression. Obstetrics and Gynecology 110, 1102–1112. Lewis, G., 2011. Centre for Maternal and Child Enquiries CMACE Saving mothers’ lives: reviewing maternal deaths to make motherhood safer: 2006-08. The eighth report on confidential enquiries into maternal deaths in the United Kingdom. British Journal of Obstetrics and Gynecology 118 (1), 1–203. Lumley, J., Austin, M-P, Mitchell, C., 2004. Intervening to reduce depression after birth: A systematic review of randomized trials. International Journal of Technology Assessment in Health Care 20 (2), 128–144. Mancuso, R.A., Schetter, C.D., Rini, C.M., 2004. Maternal prenatal anxiety and corticotropin-releasing hormone associated with timing of delivery. Psychosomatic Medicine 66, 762–769. Marcus, S.M., Heringhausen, J.E., 2009. Depression in childbearing women: when depression complicates pregnancy. Primary Care 36, 151–165, ix. Matthey, S., 2004. Detection and treatment of postnatal depression (perinatal depression and anxiety). Current Opinions in Psychiatry 17, 21–29. McNeill, J., Lynn, F., Alderdice, F., 2010. Systematic Review of Reviews: The PublicHealth Role of the Midwife. School of Nursing & Midwifery, Queen’s University /http://midwifery2020.org/documents/2020/Public_Health_Lit_Re view.pdfS. Midwifery 2020 Final Report, 2010. Delivering Expectations /http://midwif ery2020.org.uk/documents/M2020Deliveringexpectations-FullReport2.pdfS. Morrell, C., 2006. Review of interventions to prevent or treat postnatal depression. Clinical Effectiveness in Nursing 9, s135–161. National Collaborating Centre for Mental Health, 2007. Antenatal and Postnatal Mental Health: The NICE Guideline on Clinical Management and Service Guidance. NICE Guidelines CG45. The British Psychological Society & The Royal College of Psychiatrists, London. Ogrodniczuk, J.S., Piper, W.E., 2003. Preventing postnatal depression: a review of research findings. Harvard Review of Psychiatry 11, 291–307. Royal College of Midwives, 2007. Position Statement Complementary and Alternative Therapies. RCM London. Ross, L.E., McLean, L.M., 2006. Anxiety disorders during pregnancy and the postpartum period: a systematic review. Journal of Clinical Psychiatry 67, 1285–1298. Rowan, C., Bick, D., Bastos, M.H., 2007. Postnatal debriefing interventions to prevent maternal mental health problems after birth: exploring the gap between the evidence and UK policy and practice. Worldviews on EvidenceBased Nursing 4, 97–105. Smith, V., Devane, D., Begley, C.M., Clarke, M., 2011. Methodology in conducting a systematic review of systematic reviews of healthcare interventions. BMC Medical Research Methodology 11, 15. Stevenson, M.D., Scope, A., Sutcliff, A., Slade, P., Parry, G., Saxon, D., et al., 2010. Group cognitive behaviorual therapy for postnatal depression: a systematic review of clinical effectiveness, cost-effectivenss and value of information analysis. Health Technology Assessment 14, 44, http://dx.doi.org/10.3310/htal4440. Talge, N.M., Neal, C., Glover, V., 2007. Antenatal maternal stress and long-term effects on child neurodevelopment: how and why? Journal of Child Psychology and Psychiatry 48, 245–261.