Interventions (other than pharmacological, psychosocial or psychological) for treating antenatal depression

  • Review
  • Intervention

Authors


Abstract

Background

A meta-analysis of 21 studies suggests the mean prevalence rate for depression across the antenatal period is 10.7%, ranging from 7.4% in the first trimester to a high of 12.8% in the second trimester. Due to maternal treatment preferences and potential concerns about fetal and infant health outcomes, diverse non-pharmacological treatment options are needed.

Objectives

To assess the effect of interventions other than pharmacological, psychosocial, or psychological interventions compared with usual antepartum care in the treatment of antenatal depression.

Search methods

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 January 2013), scanned secondary references and contacted experts in the field to identify other published or unpublished trials.

Selection criteria

All published and unpublished randomised controlled trials of acceptable quality evaluating non-pharmacological/psychosocial/psychological interventions to treat antenatal depression.

Data collection and analysis

Both review authors participated in the evaluation of methodological quality and data extraction. Results are presented using risk ratio (RR) for categorical data and mean difference (MD) for continuous data.

Main results

Six trials were included involving 402 women from the United States, Switzerland, and Taiwan. For most comparisons a single trial contributed data and there were few statistically significant differences between control and intervention groups.

In a trial with 38 women maternal massage compared with non-specific acupuncture (control group) did not significantly decrease the number of women with clinical depression or depressive symptomatology immediately post-treatment (risk ratio (RR) 0.80, 95% confidence interval (CI) 0.25 to 2.53; mean difference (MD) -2.30, 95% CI -6.51 to 1.91 respectively). In another trial with 88 women there was no difference in treatment response or depression remission rates in women receiving maternal massage compared with those receiving non-specific acupuncture (RR 1.33, 95% CI 0.82 to 2.18; RR 1.14, 95% CI 0.59 to 2.19 respectively).

In a trial with 35 women acupuncture specifically treating symptoms of depression, compared with non-specific acupuncture, did not significantly decrease the number of women with clinical depression or depressive symptomatology immediately post-treatment (RR 0.47, 95% CI 0.11 to 2.13; MD -3.00, 95% CI -8.10 to 2.10). However, women who received depression-specific acupuncture were more likely to respond to treatment compared with those receiving non-specific acupuncture (RR 1.68, 95% CI 1.06 to 2.66).

In a trial with 149 women, maternal massage by a woman's significant other, compared with standard care, significantly decreased the number of women with depressive symptomatology immediately post-treatment (MD -6.70, 95% CI -9.77 to -3.63).

Further, women receiving bright light therapy had a significantly greater change in their mean depression scores over the five weeks of treatment than those receiving a dim light placebo (one trial, n = 27; MD -4.80, 95% CI -8.39 to -1.21). However, they were not more likely to have a treatment response or experience a higher remission rate (RR 1.79, 95% CI 0.90 to 3.56; RR 1.89, 95% CI 0.81 to 4.42).

Lastly, two trials examined the treatment effect of omega-3 oils. Women receiving omega-3 had a significantly lower mean depression score following eight weeks of treatment than those receiving a placebo (one trial, n = 33; MD -4.70, 95% CI -7.82 to -1.58). Conversely, in a smaller trial (21 women) there was no significant difference in the change in mean depression scores for women receiving omega-3 and those receiving a placebo (MD 0.36, 95% CI -0.17 to 0.89), and women who received omega-3 were no more likely to respond to treatment (RR 2.26, 95% CI 0.78 to 6.49) or have higher remission rates (RR 2.12, 95% CI 0.51 to 8.84). Women in the placebo group were just as likely to report a side effect as those in the omega-3 group (RR 1.12, 95% CI 0.56 to 2.27).

Authors' conclusions

The evidence is inconclusive to allow us to make any recommendations for depression-specific acupuncture, maternal massage, bright light therapy, and omega-3 fatty acids for the treatment of antenatal depression. The included trials were too small with non-generalisable samples, to make any recommendations.

アブストラクト

産前(妊娠)うつ病の治療介入(薬理学的、心理社会的、心理学的以外の介入)

背景

21件の研究のメタアナリシスから、妊娠期間中を通じたうつ病の平均有病率は10.7%であり、妊娠前期の7.4%から、妊娠中期の12.8%までに及ぶことが示された。母親の治療選好と、胎児および乳児の健康上のアウトカムへの潜在的な懸念から、多種多様な非薬理学的治療の選択肢が必要である。

目的

産前(妊娠)うつ病の治療において、通常の分娩前ケアと比較した薬理学的、心理社会的、心理学的以外の介入の効果を評価すること。

検索戦略

Cochrane Pregnancy and Childbirth Group’s Trials Registerを検索した(検索日:2013年1月31日)。二次文献を調べ、他の発表済みまたは未発表の試験を同定するためにその分野の専門家に問い合わせた。

選択基準

産前(妊娠)うつ病を治療するための非薬理学的/心理社会的/心理学的介入を評価するために、許容できる質のすべての発表済みおよび未発表のランダム化比較試験。

データ収集と分析

2人のレビュー著者が、方法論的質の評価およびデータ抽出の評価に参加した。カテゴリデータはリスク比(RR)で、連続データは平均差(MD)で結果を示した。

主な結果

米国、スイス、台湾で実施された6件の試験(女性402例を対象)が選択された。ほとんどの比較で、単一試験からのデータを用いており、対照群と介入群の間に統計学的な有意差はほとんどなかった。

妊産婦向けのマッサージと一般的な鍼療法(対照群)を比較したある試験(女性38例を対象)では、治療直後に臨床的うつ病または抑うつ症状のある女性の数に有意な減少は認められなかった(それぞれリスク比[RR]0.80、95%信頼区間[CI]0.25~2.53;平均差[MD]-2.30、95% CI -6.51~1.91)。別の試験(女性88例を対象)では、妊産婦向けのマッサージを受けた女性は一般的な鍼療法を受けた女性と比較して、治療の反応率またはうつ病の寛解率に差がなかった(それぞれRR 1.33、95% CI 0.82~2.18;RR 1.14、95% CI 0.59~2.19)。

ある試験(女性35例を対象)で、うつ病の症状に特化した鍼療法では、一般的な鍼療法と比較して、治療直後に臨床的うつ病または抑うつ症状のある女性の数は有意に減少しなかった(RR 0.47、95% CI 0.11~2.13;MD -3.00、95% CI -8.10~2.10)。ただし、うつ病に特化した鍼療法を受けた女性のほうが、一般的な鍼療法を受けた女性と比較して、治療への反応率は高かった(RR 1.68、 95% CI 1.06~2.66)。

ある試験(女性149例を対象)では、女性にとって大切な人からの妊産婦向けのマッサージでは、標準的なケアと比較して、治療直後に抑うつ症状のある女性の数が有意に減少した(MD -6.70、95% CI -9.77~-3.63)。 加えて、高照度光療法を受けた女性は、低照度のプラセボ群と比較して、5週間の治療期間にわたりうつ病尺度の平均スコアが有意に大きく変化した(1試験、n = 27、MD -4.80、95% CI -8.39~-1.21)。ただし、治療への反応や高い寛解率を示す可能性が高いわけではなかった(RR 1.79、95% CI 0.90~3.56;RR 1.89、95% CI 0.81~4.42)。

最後に、2件の試験でω3オイルの治療効果を調べていた。ω3を摂取した女性は、プラセボ群と比較して、8週間の治療期間後にうつ病尺度の平均スコアが有意に低下した(1試験、n = 33、MD -4.70、 95% CI -7.82~-1.58)。反対に、ある小規模の試験(女性21例を対象)では、ω3を摂取した女性とプラセボ群で、うつ病尺度の平均スコアの変化に有意な差が認められなかった(MD 0.36、95% CI -0.17~0.89)。ω3を摂取した女性は、治療への反応率が高いわけではなく(RR 2.26、95% CI 0.78~6.49)、高い寛解率を示すわけでもなかった(RR 2.12、95% CI 0.51~8.84)。プラセボ群における女性の副作用の報告は、ω3群と同等であった(RR 1.12、95% CI 0.56~2.27)。

著者の結論

このエビデンスは、産前(妊娠)うつ病の治療として、うつ病に特化した鍼療法、妊産婦向けのマッサージ、高照度光療法、ω3脂肪酸摂取に関する推奨方針を提示するだけの決定的なものではない。選択された試験は、サンプルを一般化できず、推奨を行うには規模が小さすぎた。

訳注

《実施組織》厚生労働省「「統合医療」に係る情報発信等推進事業」(eJIM:http://www.ejim.ncgg.go.jp/)[2016.1.1]
《注意》この日本語訳は、臨床医、疫学研究者などによる翻訳のチェックを受けて公開していますが、訳語の間違いなどお気づきの点がございましたら、eJIM事務局までご連絡ください。なお、2013年6月からコクラン・ライブラリーのNew review, Updated reviewとも日単位で更新されています。eJIMでは最新版の日本語訳を掲載するよう努めておりますが、タイム・ラグが生じている場合もあります。ご利用に際しては、最新版(英語版)の内容をご確認ください。

Plain language summary

Interventions (other than pharmacological, psychosocial or psychological) for treating antenatal depression

There is not enough evidence available to determine if acupuncture, maternal massage, bright light therapy, or omega-3 fatty acids are effective interventions in treating antenatal depression.

Approximately 12% of women will suffer from depression during their pregnancy. Research suggests that women who experience significant stress, have a history of depression, lack social support, have a history of domestic violence, are not married and living alone, and have an unintended pregnancy or poor relationships may be at a higher risk than other women of developing antenatal depression. Symptoms can include overwhelming feelings of sadness and grief, loss of interest or pleasure in activities that are usually enjoyed, feelings of worthlessness or guilt, poor sleep, a change in appetite, severe fatigue and difficulty concentrating. Unfortunately, depression during pregnancy is related to poor maternal self-care behaviours, which may influence the baby's health; it also places a woman at significant risk of developing postpartum depression. Many women prefer not to take medication during their pregnancy and they are often interested in other complementary forms of treatment. The review found only six randomised controlled trials involving 402 women evaluating depression-specific acupuncture (the insertion of needles into the superficial body tissues for remedial purposes), maternal massage, bright light therapy, and omega-3 fatty acids for the treatment of antenatal depression. The included trials were too small to reach any conclusions; they also used a variety of interventions, outcome measures and comparisons. The trials provided insufficient evidence to determine if these therapies are effective treatments for antenatal depression. Further research is needed.

平易な要約

産前(妊娠)うつ病の治療介入(薬理学的、心理社会的、心理学的以外の介入)

産前(妊娠)うつ病の治療に、鍼療法、妊産婦向けのマッサージ、高照度光療法、またはω3脂肪酸摂取が有効な介入であるかどうかを判断できる十分なエビデンスはない。

約12%の女性が妊娠中にうつ病を患う。研究によると、相対的に産前(妊娠)うつ病を発症するリスクが高い女性の傾向は、著しいストレスを感じる、うつ病の既往歴がある、社会的支援が受けられない、家庭内暴力を経験している、結婚していないかひとり暮らしである、意図しない妊娠であるか人間関係が良好でないことである。症状として挙げられるのは、圧倒的な悲しみや悲嘆の感情、通常なら楽しい活動への興味や喜びの喪失、自己否定気分や罪悪感、睡眠障害、味覚の変化、重度の倦怠感、集中力の低下などである。残念ながら、妊娠中のうつ病は母親のセルフケア行動の悪化につながり、胎児の健康に影響を及ぼす可能性がある。また、産後うつ病発症の重大なリスクとなる。多くの女性は妊娠中に薬物を摂取することを好まず、しばしば他の補完療法に関心を示す。本レビューでは、産前(妊娠)うつ病の治療として、うつ病に特化した鍼療法(治療目的で表在の体組織に針を刺入すること)、妊産婦向けのマッサージ、高照度光療法、ω3脂肪酸摂取を評価するランダム化比較試験わずか6件のみ(女性402例を対象)を評価対象とした。選択された試験は規模が小さすぎるため結論に至らなかった。また、さまざまな介入、アウトカム指標、比較を扱っていた。これらの治療が産前(妊娠)うつ病の治療に効果があるかどうかを判断するには、試験のエビデンスが不十分であった。さらなる研究が必要である。

訳注

《実施組織》厚生労働省「「統合医療」に係る情報発信等推進事業」(eJIM:http://www.ejim.ncgg.go.jp/)[2016.1.1]
《注意》この日本語訳は、臨床医、疫学研究者などによる翻訳のチェックを受けて公開していますが、訳語の間違いなどお気づきの点がございましたら、eJIM事務局までご連絡ください。なお、2013年6月からコクラン・ライブラリーのNew review, Updated reviewとも日単位で更新されています。eJIMでは最新版の日本語訳を掲載するよう努めておりますが、タイム・ラグが生じている場合もあります。ご利用に際しては、最新版(英語版)の内容をご確認ください。

Background

Description of the condition

Depression is a common complication in pregnancy, exceeding rates in the general female population. A meta-analysis of 21 studies suggests that the mean prevalence rate of depression across the antenatal period is 10.7%, ranging from 7.4% in the first trimester to a high of 12.8% in the second trimester (Bennett 2004a). An earlier meta-analysis including 59 studies also found that depression was common during pregnancy with a period prevalence of 18.4% across the nine months of pregnancy, with 12.7% having an episode of major depression during this time. Not surprisingly, antenatal depression is a strong risk factor of postpartum depression, and this review identified an overall prevalence of 13% within the first 12 weeks following childbirth (n = 12,810; 95% confidence interval 12.3% to 13.4%) (O'Hara 1996). Symptoms may include overwhelming feelings of sadness and grief, loss of interest or pleasure in activities usually enjoyed, feelings of worthlessness or guilt, poor sleep, a change in appetite, severe fatigue and difficulty concentrating. Numerous studies of variable methodological quality have examined antenatal depression risk factors. A systematic review that included 57 studies found the following variables were associated with a greater likelihood of antepartum depressive symptoms: anxiety in pregnancy, increased life stress, history of depression, lack of social support including intimate partner support, history of domestic violence, unintended pregnancy, single/non-cohabitation status, and poor intimate relationship quality (Lancaster 2010). However, inconsistent findings were found related to smoking, alcohol use, illicit drug use, parity, maternal race/ethnicity, and maternal age. Five studies assessed measures of socioeconomic status and no significant associations were found. Inconsistent results were found for three sub-components of socioeconomic status: income, employment, and education. Unfortunately, there is a general lack of awareness about antenatal depression in the community (Highet 2011), which may partially explain why it often remains under-diagnosed (Flynn 2006). As it is an important public health issue, several professional organisations now recommend routine screening for antepartum depression to improve detection and treatment rates (ACOG 2006; NCCMH 2007).

Description of the intervention

Decisions about how to treat depression in pregnancy are complex. While untreated depression poses risk to mothers and fetuses, no treatment option comes without disadvantages. Pharmacotherapy such as anti-depressant medication is a common treatment for moderate to severe depression that has well-demonstrated efficacy (Lam 2009). However, information on the safety of anti-depressant medication in pregnancy comes from observational studies which cannot rule out long-term risks of fetal exposure. As a potential adjunct or alternative to anti-depressant medication, psychotherapy such as cognitive-behavioural therapy and interpersonal psychotherapy have been validated as effective treatment for general depression (Parikh 2009). A Cochrane systematic review has been completed evaluating these psychotherapies and psychosocial interventions (e.g., peer support, non-directive counselling, nurse/midwifery home visits) for the treatment of postpartum depression (Dennis 2007a). A Cochrane systematic review of psychosocial and psychological interventions for the treatment of antenatal depression has also been completed (Dennis 2007b) where only one trial was included in the review (Spinelli 2003). However, psychotherapy may take 12 weeks or more to take effect and it may not be effective on its own in women with more severe illness (Yonkers 2009). Further, access to psychotherapy may be limited especially for those in rural and remote areas. Interventions other than pharmacological, psychosocial or psychological are also available for the treatment of depression. These "alternative" treatment options, such as light therapy and herbal supplements, are commonly used in part because of a prevalent belief that “natural is better” (Ravindran 2009). There is some reason to support the use of these therapies as a survey of primary care patients found about 11% of people with depression and anxiety reported using complementary or alternative medicine therapy (Roy-Byrne 2005), a rate similar to the proportion of people with depression who use anti-depressant medication (Mojtabai 2008).

There are several alternative treatment options available for depression including physical therapies (for example, bright light therapy, physical exercise, yoga, acupuncture, sleep deprivation), nutraceuticals (for example, omega-3 fatty acids), and herbal remedies (for example, St. John's Wort) (Ravindran 2009). Light therapy is exposure to bright light where the standard “dose” is 10,000 lux (intensity) with a fluorescent light box for 30 minutes per day given in the early morning usually at home. Response typically occurs within three weeks. While the mechanism of action is still unclear, correction of disturbed circadian rhythms and modulation of serotonin and catecholamine systems are two of the theories proposed (Crowley 2012; Sohn 2005). Another treatment option is some form of physical exercise over eight to 20 weeks (average of 12 weeks), usually three times a week for 30 to 60 minutes per session. Along a similar line is yoga, which integrates physical postures, breath control, and meditation. It is practiced typically between four to eight weeks, with an average frequency of four times a week for 45 to 60 minutes per session (Ravindran 2009). Acupuncture uses special needles to pierce the skin surface at specific body points to produce particular therapeutic effects. Treatment is often four to eight weeks and the number of needles (two to 16) and sessions varies widely.

Nutraceuticals are nonprescription natural health products that are usually in concentrated forms of naturally occurring substances such as vitamins and minerals (e.g., omega-3 fatty acids, tryptophan, S-adenosyl-L-methionine (SAM-e), folic acid, inositol, amino acids, alpha-lactabumin, dehydroepiandrosterone (DHEA), and acetyl-L-carnitine) (Ravindran 2009). They are used alone or in combination to promote general well-being. Common nutraceuticals are omega-3 fatty acids, polyunsaturated fatty acids found in multiple biological systems. While the mechanisms of action are not fully understood, omega-3 fatty acids have been studied in different doses and formulations, which include estyl esters of eicosapentanoic acid (EPA) or docosahexaenoic acid (DHA) or a combination of both (Freeman 2011). Herbal remedies are another sub-category of natural health products that are derived from plants and plant extracts, such as leaves, flowers, roots, bark and berries, and sold as a non-prescription product (Ravindran 2009). This category includes St. John's wort, Crocus sativus, Lavandula angustifolia, Rhodiola rosea, and Gingko biloba to name a few.

Why it is important to do this review

Untreated (or incompletely managed) antenatal depression has been associated with a range of adverse pregnancy complications (Bonari 2004), such as pre-eclampsia (Kurki 2000), preterm labour (Dayan 2002), and prolonged labour (Hanlon 2009). Previous research has also demonstrated an association between mental illness and maternal smoking (Zhu 2002), abnormal endocrine and immune regulation (Clark 1996) which may impair fetal development and lead to low birthweight (Rondo 2003). There is also rapidly growing evidence that antenatal depression may impact child development (for example, Evans 2012). In addition, depression during pregnancy is one of the strongest risk factors for postpartum depression (Beck 2001), another well documented condition with significant research demonstrating a negative effect on infant and child developmental trajectories. Due to the limited amount of research evaluating the effectiveness of psychological and psychosocial interventions for the treatment of antenatal depression (Dennis 2007b) and the unknown risks of pharmacotherapy, alternative forms of treatment that have been evaluated for general depression should be examined. This review examined the effectiveness of these non-pharmacological/psychosocial/psychological interventions in the treatment of antenatal depression, where treatment is defined as any intervention initiated among pregnant women who have been identified with depressive symptomatology (using a self-report measure) or diagnosed with depression (using a clinical interview).

Objectives

The primary objective of this review was to assess the effects, on mothers, infants and their families, of interventions other than pharmacological, psychosocial, or psychological interventions compared with usual antepartum care in the treatment of antenatal depression.

Methods

Criteria for considering studies for this review

Types of studies

All published and unpublished randomised controlled trials examining interventions other than pharmacological, psychosocial, or psychological interventions in which the primary or secondary aim was to treat antenatal depression. Quasi-randomised trials (e.g., those randomised by delivery date, or odd versus even medical record numbers) were excluded from the analysis.

Types of participants

Pregnant women identified with antenatal depression (variously defined).

Types of interventions

Any form of standard or usual care compared to a variety of interventions (including bright light therapy, physical exercise, yoga, acupuncture, sleep deprivation), nutraceuticals (for example, omega-3 fatty acids), and herbal remedies (for example, St. John's Wort) delivered via telephone, home or clinic visits, or individual or group sessions antenatally by a professional (e.g., nurse, midwife, childbirth educator, physician, psychiatrist, psychologist) or lay person (e.g., a specially trained woman from the community, student, research assistant).

Types of outcome measures

Primary outcomes
Maternal

1. Antenatal depression (as variously defined and measured by trialists).

Secondary outcomes
Maternal

2. Postpartum depression (as variously defined and measured by trialists).
3. Maternal mortality and serious morbidity including self-harm and suicide attempts.
4. Maternal-infant attachment.
5. Anxiety.
6. Maternal stress.
7. Maternal perceived social support.
8. Maternal dissatisfaction with care provided.

Infant

9. Breastfeeding duration (variously defined).
10. Breastfeeding level (exclusive, almost exclusive, high, partial, token, bottle-feeding).
11. Infant health parameters including immunisation, accidental injury, non-accidental injury.
12. Infant developmental assessments (variously defined).
13. Child abuse and/or neglect.

Family outcomes

12. Marital/partner discord.

Search methods for identification of studies

Electronic searches

We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register by contacting the Trials Search Co-ordinator (31 January 2013). 

The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from: 

  1. monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);

  2. weekly searches of MEDLINE;

  3. weekly searches of Embase;

  4. handsearches of 30 journals and the proceedings of major conferences;

  5. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.

Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group

Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords. 

Searching other resources

We examined secondary references and contacted experts in the field. We did not apply any language restrictions.

Data collection and analysis

For the methods used when assessing the trials identified in the previous version of this review, see Appendix 1. For this update we used the following methods when assessing the reports identified by the updated search.

Selection of studies

Both review authors independently assessed for inclusion all the potential studies identified as a result of the search strategy. We resolved any uncertainties regarding the appropriateness for inclusion through discussion or consultation with a third person.

Data extraction and management

We designed a form to extract data. For eligible studies, both review authors independently extracted data using the agreed form. We resolved discrepancies through discussion or, if required, we consulted a third person. We entered data into Review Manager software (RevMan 2012) and checked for accuracy. When information regarding any of the above was unclear, we attempted to contact the authors of the original reports to provide further details.

Assessment of risk of bias in included studies

The review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third assessor.

(1) Sequence generation (checking for possible selection bias)

We have described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the method as:

  • low risk of bias (any truly random process, e.g. random number table; computer random number generator);

  • high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number);

  • unclear.

(2) Allocation concealment (checking for possible selection bias)

We have described for each included study the method used to conceal the allocation sequence and have determined whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.

We assessed the methods as:

  • low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);

  • high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);

  • unclear.   

(3) Blinding (checking for possible performance bias)

We have described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding could not have affected the results. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed the methods as:

  • low, high or unclear risk of bias for participants;

  • low, high or unclear risk of bias for personnel;

  • low, high or unclear risk of bias for outcome assessors.

(4) Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, protocol deviations)

We have described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We have stated whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes.  Where sufficient information was reported, or supplied by the trial authors, we have re-included missing data in the analyses which we undertake. We assessed methods as:

  • low risk of bias;

  • high risk of bias;

  • unclear.

(5) Selective reporting bias

We have described for each included study how we investigated the possibility of selective outcome reporting bias and what we found.

We assessed the methods as:

  • low risk of bias (where it is clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review have been reported);

  • high risk of bias (where not all the study’s pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);

  • unclear.

(6) Other sources of bias

We have described for each included study any important concerns we have about other possible sources of bias such as baseline imbalance between groups.

We assessed whether each study was free of other problems that could put it at risk of bias:

  • low risk of other bias;

  • high risk of other bias;

  • unclear risk of other bias.

(7) Overall risk of bias

We made explicit judgements about whether studies are at high risk of bias, according to the criteria given in the Cochrane Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we consider it is likely to impact on the findings. We planned to explore the impact of the level of bias through undertaking sensitivity analyses - see Sensitivity analysis.

Measures of treatment effect

Dichotomous data

For dichotomous data, we have presented results as summary risk ratio with 95% confidence intervals. 

Continuous data

For continuous data, we used the mean difference if outcomes were measured in the same way between trials. We planned to use the standardised mean difference to combine trials that measured the same outcome, but used different methods.  

Unit of analysis issues

Cluster-randomised trials

We planned to include cluster-randomised trials in the analyses along with individually-randomised trials. In this version of the review we did not identify any cluster-randomised trials, however, if such trials are included in future updates, we will adjust their sample sizes using the methods described in the Cochrane Handbook (Higgins 2011) using an estimate of the intracluster correlation co-efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster-randomised trials and individually-randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely.

We will also acknowledge heterogeneity in the randomisation unit and perform a sensitivity analysis to investigate the effects of the randomisation unit.

Cross-over trials 

We did not include cross-over trials.

Other unit of analysis issues

For trials with more than two treatment groups, where possible we combined arms and carried out pair-wise comparisons.

Dealing with missing data

For included studies, we noted levels of attrition. We explored the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis.

For all outcomes, we carried out analyses, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing.

Assessment of heterogeneity

We assessed statistical heterogeneity in each meta-analysis using the T2, I² and Chi² statistics. We have regarded heterogeneity as substantial if the T2 was greater than zero and either the I2 was greater than 30% or there was a low P value (less than 0.10) in the Chi² test for heterogeneity.

Assessment of reporting biases

Had there been 10 or more studies in the meta-analysis, we would have investigated reporting biases (such as publication bias) using funnel plots. We would have assessed funnel plot asymmetry visually. If asymmetry was suggested by a visual assessment, we would have performed exploratory analyses to investigate it.

Data synthesis

We carried out statistical analysis using the Review Manager software (RevMan 2012). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials examined the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects would differ between trials, or if substantial statistical heterogeneity was detected, we planned to use random-effects meta-analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. The random-effects summary would have been treated as the average range of possible treatment effects and we would have discussed the clinical implications of treatment effects differing between trials. If the average treatment effect was not considered to be clinically meaningful, we would not combine trials.

If we had used random-effects analyses, the results would have been presented as the average treatment effect with its 95% confidence interval, and the estimates of  T2 and I2.

Subgroup analysis and investigation of heterogeneity

Due to the limited number of included studies, we were unable to perform any subgroup analyses.

Sensitivity analysis

We planned to perform sensitivity analyses for the primary outcome in instances in which any of the following occurred:

  1. a high risk of bias associated with the methodological quality of included trials;

  2. incomplete outcome data (more that 20% missing data) for any of the included trials.

Results

Description of studies

See Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification.

Results of the search

Overall, there were 18 studies of non-pharmacological/psychosocial/psychological interventions identified from the literature search. Of the these studies, only six were suitable for inclusion in this review (Field 2009a; Freeman 2008; Manber 2004; Manber 2010; Su 2008; Wirz-Justice 2011).

Included studies

Six randomised controlled trials, reported between 2004 and 2011, were identified and met the inclusion criteria (Field 2009a; Freeman 2008; Manber 2004; Manber 2010; Su 2008; Wirz-Justice 2011). In total, 402 women were included in the meta-analysis. Of the six trials, four were conducted in the United States (Field 2009a; Freeman 2008; Manber 2004; Manber 2010) while the other two trials were conducted in Switerland (Wirz-Justice 2011) and Taiwan (Su 2008). All trials included the outcome of depression in pregnancy; one trial included an anxiety measure (Field 2009a). Due to the limited number of included studies we were unable to perform any subgroup or sensitive analyses.

Definition of depression

A diagnosis of depression is made by a clinician following examination and clinical assessment; most of the studies included in the review did not report the number of women diagnosed with depression, but instead reported scores on various scales measuring symptoms of depression such as low mood or suicidal ideation (depressive symptomatology). In all trials, depressive symptomatology was assessed using a self-report measure; only one trial included a diagnosis of depression following clinical interview as an outcome (Manber 2004). The most common self-report measure used was the Hamilton Rating Scale for Depression (HRSD) (Freeman 2008; Manber 2004; Manber 2010; Su 2008; Wirz-Justice 2011). Two trials used the Beck Depression Inventory (BDI) (Manber 2004; Su 2008) and two trials used the Edinburgh Postnatal Depression Scale (EPDS) (Freeman 2008; Su 2008). One trial incorporated the Center for Epidemiologic Studies Depression Scale (CES-D) (Field 2009a) and one trial used the Structural Interview Guide for the Hamilton Depression and Anxiety Scales (SIGH-AD) (Wirz-Justice 2011). Several trials examined a difference in mean scores (Field 2009a; Manber 2004; Su 2008), while others examined a change in mean scores (Freeman 2008; Wirz-Justice 2011). Three trials examined both treatment response (defined in trials as a 50% reduction in baseline depressive symptomatology scores) and depression remission (defined in trials as a depressive symptomatology score below a specific cut-off indicating no symptoms of depression) (Manber 2010; Su 2008; Wirz-Justice 2011). Due to the significant differences in the timing of outcome data, we included the outcome assessment point "immediately post-treatment antenatally". While two trials examined depressive symptomatology scores both antenatally and postnatally (Field 2009a; Manber 2004), one trial did not provide usable postnatal data that could be included in the review (Field 2009a).

Types of interventions

Interventions evaluated in this review included the physical therapies of acupuncture (Manber 2004; Manber 2010), maternal massage (Field 2009a), bright light therapy (Wirz-Justice 2011), and the nutraceutical therapy of omega-3 fatty acids (Freeman 2008; Su 2008). No studies were found that examined other physical therapies such as exercise, yoga, or sleep deprivation and other nutraceutical therapies such as DHEA. No herbal remedies such as St John's Wort were included in this review.

Excluded studies

The nine studies that were excluded examined the following interventions: acupuncture (Bosco 2007), omega-3 fatty acids (Doornbos 2009; Mozurkewich 2011), bright light therapy (Epperson 2004), maternal massage (Field 2004; Field 2009b), relaxation and guided imagery (Gedde-Dahl 2012), education (Hayes 2001), and early/late sleep therapy (Parry 2010). Three trials (Bosco 2007; Field 2004; Field 2009b) lacked the methodologic quality required for inclusion while three other trials (Doornbos 2009; Hayes 2001; Mozurkewich 2011) specifically focused on the prevention of postpartum depression rather than the treatment of antenatal depression. The trial by Mozurkewich 2011 is currently ongoing and the decision to exclude was based on a published trial protocol.

There are three studies meeting inclusion criteria that are awaiting classification (Field 2008; Field 2012; Rees 2008); authors have been contacted to obtain the data needed for inclusion.

Risk of bias in included studies

The methodological quality of the included trials was not strong. The sample sizes were well below 100 except for two trials (Field 2009a; Manber 2010).

Summaries of 'Risk of bias' assessments are presented in Figure 1 and Figure 2.

Figure 1.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 2.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Although randomisation was conducted, there were no details provided regarding allocation generation or concealment in four trials (Field 2009a; Freeman 2008; Manber 2004; Su 2008). The other two trials used block randomisation generated by computer (Manber 2010; Wirz-Justice 2011).

Blinding

In only two trials (Manber 2010; Wirz-Justice 2011) was it clear that the outcome assessors were blinded to group allocation.

Incomplete outcome data

Loss of participants to follow-up was very high (at least 30%) in three trials (Field 2009a; Su 2008; Wirz-Justice 2011).

Selective reporting

It was unclear from assessment of the published study reports whether or not selective reporting had occurred. Trial protocols were not assessed. Additional information from one trial author has been requested (Field 2009a).

Other potential sources of bias

All trials but two (Field 2009a; Manber 2004) systematically examined side effects.

Effects of interventions

We have set out results for each type of intervention separately. While all trials reported depression or depressive symptomatology following treatment no trials reported on infant or family outcomes. While postpartum depression was one of our secondary outcomes, to avoid repetition in both the analyses and text, we have reported results for postpartum depression together with results for the antenatal period. For our primary outcomes, outcome measures for each individual treatment approach varied resulting in no data being combined in meta-analysis.

Comparison one: maternal massage versus non-specific acupuncture (control group)

A. Maternal primary outcomes
Outcome: Clinical depression

Maternal massage compared to non-specific acupuncture (control group) did not significantly decrease the number of women diagnosed with clinical depression (DSM-IV criteria) immediately post-treatment (one trial, n = 38; risk ratio (RR) 0.80, 95% confidence interval (CI) 0.25 to 2.53) or at final assessment at 10 weeks postpartum (one trial, n = 32; RR 1.93, 95% CI 0.37 to 10.01) Analysis 1.1.

Outcome: Depressive symptomatology

Maternal massage, compared to non-specific acupuncture did not significantly decrease the number of women with depressive symptomatology as measured by the Hamilton Rating Scale for Depression (HRSD) immediately post-treatment (one trial, n = 38; mean difference (MD) -2.30, 95% CI -6.51 to 1.91) or at final assessment at 10 weeks postpartum (one trial, n = 33; MD -0.20, 95% CI -4.91 to 4.51) Analysis 1.2. Similarly, there was no beneficial effect found when depressive symptomatology was measured using the Beck Depression Inventory (BDI_ immediately post-treatment (one trial, n = 38; MD -2.20, 95% CI -5.22 to 0.82) or at final assessment at 10 weeks postpartum (one trial, n = 33; MD -0.60, 95% CI -6.23 to 5.03) Analysis 1.3.

Outcome: Treatment response

Treatment response rates (a 50% reduction in HRSD scores from baseline) did not differ between women receiving maternal massage and those receiving non-specific acupuncture (one trial, n = 88; RR 1.33, 95% CI 0.82 to 2.18) Analysis 1.4.

Outcome: Depression remission

Remission rates (HRSD score less than eight) were not significantly different between the women receiving prenatal massage and those receiving non-specific acupuncture (one trial, n = 88; RR 1.14, 95% CI 0.59 to 2.19) Analysis 1.5.

Comparison two: depression-specific acupuncture versus non-specific acupuncture (control group)

A. Maternal primary outcomes
Outcome: Clinical depression

Acupuncture specifically treating symptoms of depression, compared to non-specific acupuncture did not significantly decrease the number of women diagnosed with clinical depression (DSM-IV criteria) immediately post-treatment (one trial, n = 35; RR 0.47, 95% CI 0.11 to 2.13) or at final assessment at 10 weeks postpartum (one trial, n = 32; RR 0.64, 95% CI 0.06 to 6.39) Analysis 2.1.

Outcome: Depressive symptomatology

Acupuncture specifically treating symptoms of depression, compared to non-specific acupuncture did not significantly decrease the number of women with depressive symptomatology as measured by the HRSD immediately post-treatment (one trial, n = 35; MD -3.00, 95% CI -8.10 to 2.10) or at final assessment at 10 weeks postpartum (one trial, n = 32; MD -0.90, 95% CI -5.72 to 3.92) Analysis 2.2. Similarly, there was no beneficial effect found when depressive symptomatology was measured using the BDI immediately post-treatment (one trial, n = 35; MD -3.00, 95% CI -6.85 to 0.85) or at final assessment at 10 weeks postpartum (one trial, n = 32; MD -3.90, 95% CI -9.96 to 2.16) Analysis 2.3.

Outcome: Treatment response

Women who received depression-specific acupuncture were more likely to have a treatment response rates (a 50% reduction in HRSD scores from baseline) than those receiving non-specific acupuncture (one trial, n = 86; RR 1.68, 95% CI 1.06 to 2.66) Analysis 2.4.

Outcome: Depression remission

Remission rates (HRSD score less than eight) were not significantly different between the women receiving depression-specific acupuncture and those receiving non-specific acupuncture (one trial, n = 86; RR 1.26, 95% CI 0.67 to 2.40) Analysis 2.5.

Comparison three: maternal massage versus standard care

A. Maternal primary outcomes
Outcome: Depressive symptomatology

Massage by a significant other, compared to standard care significantly decreased the number of women with depressive symptomatology as measured by the Center for Epidemiological Studies-Depression Scale (CES-D) immediately post-treatment (one trial, n = 149; MD -6.70, 95% CI -9.77 to -3.63) Analysis 3.1.

B. Maternal secondary outcomes
Outcome: Anxiety

Massage by a significant other, compared to standard care did not significantly decreased the number of women with anxiety as measured by the State Anxiety Inventory (STAI) immediately post-treatment (one trial, n = 149; MD -1.80, 95% CI -5.04 to 1.44) Analysis 3.2.

Comparison four: bright light therapy versus placebo

A. Maternal primary outcomes
Outcome: Depressive symptomatology

Women receiving bright light therapy had a significantly greater change in their mean HRSD scores over the five weeks of treatment than those receiving a dim light placebo (one trial, n = 27; MD -4.80, 95% CI -8.39 to -1.21 ) Analysis 4.1. A similar significant change favouring the bright light therapy group was found using the Structural Interview Guide for the Hamilton Depression and Anxiety Scale (SIGH-AD) (one trial, n = 27; MD -5.00, 95% CI -10.00 to 0.00) Analysis 4.2.

Outcome: Treatment response

Women who received bright light therapy were no more likely to have a treatment response (a 50% reduction in HRSD scores from baseline) than those receiving a dim light placebo (one trial, n = 27; RR 1.79, 95% CI 0.90 to 3.56) Analysis 4.3. A similar non-significant treatment response was found with the SIGH-AD (one trial, n = 27; RR 2.06, 95% CI 0.90 to 4.74) Analysis 4.4.

Outcome: Depression remission

Remission rates (HRSD score less than eight) were not significantly different between the women receiving bright light therapy and those receiving a dim light placebo (one trial, n = 27; RR 1.89, 95% CI 0.81 to 4.42) Analysis 4.5. A similar non-significant remission rate was found with the SIGH-AD (score less than eight)(one trial, n = 27; RR 1.72, 95% CI 0.40 to 7.32) Analysis 4.6.

Comparison five: omega-3 fatty acids versus placebo

A. Maternal primary outcomes
Outcome: Depressive symptomatology

Two trials examined the treatment effect of omega-3 fatty acids. Women receiving omega-3 fatty acids had a significantly lower mean HRSD score following eight weeks of treatment than those receiving a placebo (one trial, n = 33; MD -4.70, 95% CI -7.82 to -1.58) Analysis 5.1. A similar significant difference in mean scores favouring the omega-3 group at eight weeks following treatment was found using the Edinburgh Postnatal Depression Scale (EPDS) (one trial, n = 33; MD -5.80, 95% CI -9.85 to -1.75) Analysis 5.2 and BDI (one trial, n = 33; MD -9.00, 95% CI -15.76 to -2.24) Analysis 5.3. Conversely, in a smaller trial there was no significant difference in the change in mean HRSD scores from baseline to eight weeks of treatment for women receiving omega-3 fatty acids and those receiving a placebo (one trial, n = 21; MD 0.36, 95% CI -0.17 to 0.89) Analysis 5.4. Similary, there was no significant difference in the change in mean EPDS scores from baseline to eight weeks of treatment between the two group (one trial, n = 21; MD -0.03, 95% CI -0.75 to 0.69) Analysis 5.5.

Outcome: Treatment response

Women who received omega-3 fatty acids were no more likely to have a treatment response (a 50% reduction in HRSD scores from baseline) than those receiving a placebo (one trial, n = 24; RR 2.26, 95% CI 0.78 to 6.49) Analysis 5.6.

Outcome: Depression remission

Remission rates (HRSD score less than eight) were not significantly different between the women receiving omega-3 fatty acids and those receiving a placebo (one trial, n = 24; RR 2.12, 95% CI 0.51 to 8.84) Analysis 5.7.

B. Non-prespecified outcomes
Outcome: Side effects

Women in the placebo group were just as likely to report a side effect as those in the omega-3 fatty acids group (two trials, n = 57; RR 1.12, 95% CI 0.56 to 2.27) Analysis 5.8.

Discussion

Summary of main results

Six trials were included in this review incorporating 402 women from the United States, Switzerland, and Taiwan. The body of evidence provided in this review does not allow for a strong conclusion regarding non-pharmacological/psychosocial/psychological interventions for the treatment of antenatal depression. Only six trials were included and all treatment approaches had two or less trials providing evidence for effectiveness (acupuncture = two trials; massage = one trial; bright light therapy = one trial; omega-3 fatty acids = two trials). Outcome measures for each individual treatment approach varied resulting in no data being combined for the primary outcome. The methodological quality for most of the included trials was not strong with many common limitations such as unclear randomisation procedures, small and homogeneous samples, unclear blinding of outcome assessors, and high attrition rates. These limitations render any results questionable.

One three-armed trial tested the efficacy of maternal massage and depression-specific acupuncture for the treatment of depressed pregnant women (Manber 2004). This pilot study of 61 US women found that neither the 12 20-minute massage sessions nor the 12 25- to 30-minute depression specific-acupuncture sessions were more effective in reducing depressive symptoms than non-specific acupuncture. In a larger trial by the same group (Manber 2010), maternal massage sessions again were not more effective in achieving a significant treatment response or obtaining depression remission than non-specific acupuncture. However, this trial did find that depression-specific acupuncture was more likely to produce a treatment response than non-specific acupuncture; no effect was found related to depression remission. One trial evaluated the effect of maternal massage by a women's significant other versus standard care (Field 2009a). Women who received the massage where significantly more likely to experience a decrease in depressive symptomatology than those who received standard care; there was no impact on maternal anxiety. In a trial by Wirz-Justice 2011, bright light therapy was not more effective than inactive bright light therapy in providing a treatment response or promoting depression remission. However, there was a significantly greater change in mean depression scores in those who received the active bright light therapy than those who received inactive therapy. While two trials evaluated omega-3 fatty acids for the treatment of antenatal depression, neither trial evaluated a similar outcome so their results could not be combined. In the trial by Su 2008, women who received omega-3 fatty acids were not more likely to achieve depression remission or have a treatment response. However, in this trial women who received omega-3 fatty acids had significantly lower mean depressive symptoms scores following treatment than those receiving a placebo. In the trial by Freeman 2008, there was no significant difference in the change in mean depression scores for women receiving omega-3 fatty acids and those receiving a placebo. In both trials, very few women reported a side effect and there was no differences in reporting an event between those who received omega-3 fatty acids and those who received a placebo.

Overall completeness and applicability of evidence

The treatment of antenatal depression using non-psychosocial/psychological interventions is an area that has been neglected, despite antenatal depression being a significant public health concern. This updated review added five new trials to the original review which only included one study (Manber 2004). The original trial in this review tested the efficacy of massage and depression-specific acupuncture for the treatment of depressed pregnant women. Another larger, acupuncture trial (n = 149) has been added by the same research group (Manber 2010). This trial found acupuncture specific for depression was associated with a significantly higher rate of response compared to control acupuncture (acupuncture not specific for depression). Several sample characteristics limit the generalisability of the results including the high education and socioeconomic status, predominance of Caucasians, and exclusion of comorbid mental and medical disorders. As such, the results are not generalisable to specific minority groups or pregnant women with other mental health concerns. While this trial alone does not provide strong evidence to indicate acupuncture should be used for the treatment of antenatal depression, the results are consistent with two other recent reviews that found acupuncture, including manual-, electrical-, and laser-based, is a generally beneficial, well-tolerated, and safe treatment for depression (Wang 2008; Wu 2012). The studies included in these reviews had many methodological limitations such as small sample size and unclear description of enrolment criteria, randomisation or blinding, and forms of acupuncture used. Recently, a randomised, subject- and assessor-blind, parallel-group, sham-controlled trial was conducted to evaluate the effect of acupuncture on postpartum depression among 20 Hong Kong mothers (Chung 2012). Although this trial was generally well designed, no significant differences were found between the two groups. This negative finding may possibly be the result of a type II error as the study may have had insufficient power to detect differences between the study groups due to the small sample size. Despite consistent methodological limitations, there is growing evidence to suggest that acupuncture may be beneficial for the treatment of depression and large, well-designed randomised controlled trials are warranted to evaluate the effectiveness of acupuncture with depressed pregnant women.

Very few studies have been conducted to evaluate the effect of massage on the treatment of antenatal depression. The one trial included in this review that evaluated maternal massage compared with standard treatment care (Field 2009a) found women who received the massage intervention were significantly more like to experience a decreased depressive symptomatology. However, in this trial the massage intervention was provided by the woman's significant other. Given that a lack of support and a poor intimate partner relationship are significant risk factors for antenatal depression (Lancaster 2010), the beneficial effect found in the trial may have been due to the additional attention provided by the women's significant other rather than the intervention itself. The other trial in this review that examined massage by a therapist versus non-specific acupuncture (control group) found no beneficial effect of massage (Manber 2004). In a recent systematic review of 17 randomised controlled trials of moderate quality, massage therapy was found to be effective in alleviating depressive symptoms (Hou 2010). Additional research in this area is warranted; however, standardisation of massage therapy protocols are required to provide comparable results.

In this review, only one trial was included that evaluated the effect of bright light therapy (Wirz-Justice 2011). While no beneficial effect on treatment response or depression remission was found, there was a significantly greater change in mean depression scores from baseline in those who received the active bright light therapy than those who received inactive therapy. This result is consistent with a systematic review of light therapy for non-seasonal depression that reported the efficacy of bright light against placebo conditions (Even 2008). This review noted that most of the studies included had small numbers, inadequate control conditions, and were short-term (two to five weeks). The side effects of bright light therapy include headache, eye strain, nausea, and agitation, but these are generally mild and rarely lead to treatment discontinuation (Ravindran 2009). Bright light therapy may be an attractive treatment for antenatal depression because it is low cost, home-based, and has a much lower side effect profile than pharmacotherapy. In a interesting review by Crowley 2012, there are several factors related to the pathophysiology of depression and response to light which might make bright light especially suitable for antenatal depression. For example, bright light treatment could potentially offset insufficient low levels of light exposure; pathological hormonal profiles; co-morbidities, including disturbed sleep and fatigue; and serotonergic dysregulation which has been linked to inadequate maternal behaviour. Only three studies (Epperson 2004; Oren 2002; Wirz-Justice 2011) have examined bright light therapy in depressed pregnant women. Larger randomised controlled trials are needed since bright light may be preferable to other types of treatment.These trials should have adequate control conditions, provide treatment for five weeks or longer, provide comparative effectiveness, and address co-morbidities that bright light could alleviate such as anxiety, disturbed sleep, and fatigue.

While omega-3 fatty acids are one of the most widely used non-vitamin supplements, there are insufficient data from adequately powered controlled trials to say whether omega-3 fatty acids are efficacious in the treatment of depression (Freeman 2011). Most randomised controlled trials have been small and of short duration and have yielded inconsistent findings (Freeman 2011). Further, the majority of trials have evaluated omega-3 fatty acids as an adjunctive treatment and the doses and composition of the omega-3 supplements have been extremely variable. In this review, two trials (Freeman 2008; Su 2008) evaluated the effect of omega-3 fatty acids as a treatment for antenatal depression. One trial (Su 2008) found a positive effect of omega-3 fatty acids on depressive symptomatology immediately post-treatment while the trial (Freeman 2008) did not. However, there was significant improvement from baseline in both groups in this trial, which could be attributed to the supportive therapy that all women in both groups received. It is important to note that this trial was very small (n = 21) and insufficiently powered to detect differences between study groups. Given the growing evidence that supports the role that omega-3 fatty acids plays in mental illness and health (Freeman 2011), research examining the effect of omega-3 supplementation in the treatment of antenatal depression is warranted. Future trials should determine if: (1) omega-3 fatty acids are efficacious as a monotherapy; (2) there is an optimal dose, ratio of EPA/DHA, and treatment course; and (3) the dose needs to be tailored to individuals based on dietary habits (Freeman 2011).

There were no trials evaluating other non-psychosocial/psychological interventions in the treatment of antenatal depression such as physical exercise. Exercise is a relatively low-cost intervention, with minimal side effects, and there is now evidence to support the antidepressant effects of exercise in general and clinical populations (Lawlor 2001). Limited evidence also supports a relationship between participation in exercise and a reduction in postpartum depression. For example, a review article explored the potential role of exercise, particularly pram walking, as an adjunctive treatment for postpartum depression (Daley 2007). Two small randomised controlled trials conducted in Australia were included in this review, which supported exercise as a useful treatment for women with postpartum depression (Armstrong 2003; Armstrong 2004). Given the reluctance by some women to use anti-depressant medication antenatally and the limited availability of psychosocial and psychological therapies (Dennis 2007b), exercise as a therapeutic possibility for depressed pregnant women deserves further exploration.

Authors' conclusions

Implications for practice

The evidence is inconclusive to allow us to make any recommendations for the treatment of antenatal depression. The trials included evaluated various interventions and were too small with non-generalisable samples to make any recommendations.

Implications for research

Antenatal depression is common, and perhaps more prevalent than postpartum depression (Bennett 2004a). Further, research consistently indicates that for many women, antenatal depression continues into the postpartum period leaving the woman and her child at risk for the negative outcomes associated with both antenatal (Bennett 2004b) and postpartum depression (Grace 2003). In the light of potential or perceived concerns associated with the use of anti-depressant medication during pregnancy and the potential lack of resources to access or obtain psychosocial or psychological interventions, other alternative forms of treatment are needed for antenatal depression. There are an increasing number of randomised controlled trials for alternative treatments. The quality of many trials remains an issue, with variability in diagnostic criteria, small sample sizes, limitations of blinding and placebo controls, and few systematic evaluations of side effects. Even for those treatments with reasonable evidence of efficacy, there are variations and lack of standardisation in dosage, potency, and concentration, all of which make it difficult for health professionals and women to be confident they are using the same doses as described in clinical studies. Future research needs to address these issues. There is, on balance, greater evidence and clinical experience with traditional treatments (pharmacotherapy and psychotherapy).

Acknowledgements

We would like to thank Dr Kim Allen for her contributions on the first version of this review.

As part of the pre-publication editorial process, this review has been commented on by three peers (an editor and two referees who are external to the editorial team), a member of the Pregnancy and Childbirth Group's international panel of consumers and the Group's Statistical Adviser.

The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group.  The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.

Data and analyses

Download statistical data

Comparison 1. Maternal massage versus non-specific acupuncture (control group)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Diagnosis of depression1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Immediately post-treatment antenatally138Risk Ratio (M-H, Fixed, 95% CI)0.8 [0.25, 2.53]
1.2 Final assessment postnatally132Risk Ratio (M-H, Fixed, 95% CI)1.93 [0.37, 10.01]
2 Depressive symptomatology - HRSD1 Mean Difference (IV, Fixed, 95% CI)Subtotals only
2.1 Immediately post-treatment antenatally138Mean Difference (IV, Fixed, 95% CI)-2.30 [-6.51, 1.91]
2.2 Final assessment postnatally133Mean Difference (IV, Fixed, 95% CI)-0.20 [-4.91, 4.51]
3 Depressive symptomatology - BDI1 Mean Difference (IV, Fixed, 95% CI)Subtotals only
3.1 Immediately post-treatment antenatally138Mean Difference (IV, Fixed, 95% CI)-2.20 [-5.22, 0.82]
3.2 Final assessment postnatally133Mean Difference (IV, Fixed, 95% CI)-0.60 [-6.23, 5.03]
4 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline)188Risk Ratio (M-H, Fixed, 95% CI)1.33 [0.82, 2.18]
5 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8188Risk Ratio (M-H, Fixed, 95% CI)1.14 [0.59, 2.19]
Analysis 1.1.

Comparison 1 Maternal massage versus non-specific acupuncture (control group), Outcome 1 Diagnosis of depression.

Analysis 1.2.

Comparison 1 Maternal massage versus non-specific acupuncture (control group), Outcome 2 Depressive symptomatology - HRSD.

Analysis 1.3.

Comparison 1 Maternal massage versus non-specific acupuncture (control group), Outcome 3 Depressive symptomatology - BDI.

Analysis 1.4.

Comparison 1 Maternal massage versus non-specific acupuncture (control group), Outcome 4 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline).

Analysis 1.5.

Comparison 1 Maternal massage versus non-specific acupuncture (control group), Outcome 5 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8.

Comparison 2. Depression-specific acupuncture versus non-specific acupuncture (control group)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Diagnosis of depression1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
1.1 Immediately post-treatment antenatally135Risk Ratio (M-H, Fixed, 95% CI)0.48 [0.11, 2.13]
1.2 Final assessment postnatally132Risk Ratio (M-H, Fixed, 95% CI)0.64 [0.06, 6.39]
2 Depressive symptomatology - HRSD1 Mean Difference (IV, Fixed, 95% CI)Subtotals only
2.1 Immediately post-treatment antenatally135Mean Difference (IV, Fixed, 95% CI)-3.0 [-8.10, 2.10]
2.2 Final assessment postnatally132Mean Difference (IV, Fixed, 95% CI)-0.90 [-5.72, 3.92]
3 Depressive symptomatology - BDI1 Mean Difference (IV, Fixed, 95% CI)Subtotals only
3.1 Immediately post-treatment antenatally135Mean Difference (IV, Fixed, 95% CI)-3.0 [-6.85, 0.85]
3.2 Final assessment postnatally132Mean Difference (IV, Fixed, 95% CI)-3.90 [-9.96, 2.16]
4 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline)186Risk Ratio (M-H, Fixed, 95% CI)1.68 [1.06, 2.66]
5 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8186Risk Ratio (M-H, Fixed, 95% CI)1.26 [0.67, 2.40]
Analysis 2.1.

Comparison 2 Depression-specific acupuncture versus non-specific acupuncture (control group), Outcome 1 Diagnosis of depression.

Analysis 2.2.

Comparison 2 Depression-specific acupuncture versus non-specific acupuncture (control group), Outcome 2 Depressive symptomatology - HRSD.

Analysis 2.3.

Comparison 2 Depression-specific acupuncture versus non-specific acupuncture (control group), Outcome 3 Depressive symptomatology - BDI.

Analysis 2.4.

Comparison 2 Depression-specific acupuncture versus non-specific acupuncture (control group), Outcome 4 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline).

Analysis 2.5.

Comparison 2 Depression-specific acupuncture versus non-specific acupuncture (control group), Outcome 5 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8.

Comparison 3. Maternal massage versus standard care (control group)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Depressive symptomatology - CES-D1149Mean Difference (IV, Fixed, 95% CI)-6.70 [-9.77, -3.63]
1.1 Immediately post-treatment antenatally1149Mean Difference (IV, Fixed, 95% CI)-6.70 [-9.77, -3.63]
2 Anxiety - STAI1149Mean Difference (IV, Fixed, 95% CI)-1.80 [-5.04, 1.44]
2.1 Immediately post-treatment antenatally1149Mean Difference (IV, Fixed, 95% CI)-1.80 [-5.04, 1.44]
Analysis 3.1.

Comparison 3 Maternal massage versus standard care (control group), Outcome 1 Depressive symptomatology - CES-D.

Analysis 3.2.

Comparison 3 Maternal massage versus standard care (control group), Outcome 2 Anxiety - STAI.

Comparison 4. Bright light therapy versus placebo
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Depressive symptomatology - HRSD mean change score post-treatment127Mean Difference (IV, Fixed, 95% CI)-4.8 [-8.39, -1.21]
2 Depressive symptomatology - SIGH-AD mean change score post-treatment127Mean Difference (IV, Fixed, 95% CI)-5.00 [-10.00, 0.00]
3 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline)127Risk Ratio (M-H, Fixed, 95% CI)1.79 [0.90, 3.56]
4 Treatment response post-treatment antenatally (depressive symptomatology - SIGH-AD reduction of 50% from baseline)127Risk Ratio (M-H, Fixed, 95% CI)2.06 [0.90, 4.74]
5 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8127Risk Ratio (M-H, Fixed, 95% CI)1.89 [0.81, 4.42]
6 Depression remission post-treatment antenatally (depressive symptomatology - SIGH-AD score less than 8 following treatment127Risk Ratio (M-H, Fixed, 95% CI)1.72 [0.40, 7.32]
Analysis 4.1.

Comparison 4 Bright light therapy versus placebo, Outcome 1 Depressive symptomatology - HRSD mean change score post-treatment.

Analysis 4.2.

Comparison 4 Bright light therapy versus placebo, Outcome 2 Depressive symptomatology - SIGH-AD mean change score post-treatment.

Analysis 4.3.

Comparison 4 Bright light therapy versus placebo, Outcome 3 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline).

Analysis 4.4.

Comparison 4 Bright light therapy versus placebo, Outcome 4 Treatment response post-treatment antenatally (depressive symptomatology - SIGH-AD reduction of 50% from baseline).

Analysis 4.5.

Comparison 4 Bright light therapy versus placebo, Outcome 5 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8.

Analysis 4.6.

Comparison 4 Bright light therapy versus placebo, Outcome 6 Depression remission post-treatment antenatally (depressive symptomatology - SIGH-AD score less than 8 following treatment.

Comparison 5. Omega-3 fatty acids versus placebo
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Depressive symptomatology - HRSD133Mean Difference (IV, Fixed, 95% CI)-4.70 [-7.82, -1.58]
1.1 Immediately post-treatment antenatally133Mean Difference (IV, Fixed, 95% CI)-4.70 [-7.82, -1.58]
2 Depressive symptomatology - EPDS133Mean Difference (IV, Fixed, 95% CI)-5.80 [-9.85, -1.75]
2.1 Immediately post-treatment antenatally133Mean Difference (IV, Fixed, 95% CI)-5.80 [-9.85, -1.75]
3 Depressive symptomatology - BDI133Mean Difference (IV, Fixed, 95% CI)-9.0 [-15.76, -2.24]
3.1 Immediately post-treatment antenatally133Mean Difference (IV, Fixed, 95% CI)-9.0 [-15.76, -2.24]
4 Depressive symptomatology - HRSD mean change score post-treatment121Mean Difference (IV, Fixed, 95% CI)0.36 [-0.17, 0.89]
5 Depressive symptomatology - EPDS mean change score post-treatment121Mean Difference (IV, Fixed, 95% CI)-0.03 [-0.75, 0.69]
6 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline)124Risk Ratio (M-H, Fixed, 95% CI)2.26 [0.78, 6.49]
7 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8124Risk Ratio (M-H, Fixed, 95% CI)2.12 [0.51, 8.84]
8 Side effects reported (including nausea and dizziness)257Risk Ratio (M-H, Fixed, 95% CI)1.12 [0.56, 2.27]
Analysis 5.1.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 1 Depressive symptomatology - HRSD.

Analysis 5.2.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 2 Depressive symptomatology - EPDS.

Analysis 5.3.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 3 Depressive symptomatology - BDI.

Analysis 5.4.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 4 Depressive symptomatology - HRSD mean change score post-treatment.

Analysis 5.5.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 5 Depressive symptomatology - EPDS mean change score post-treatment.

Analysis 5.6.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 6 Treatment response post-treatment antenatally (depressive symptomatology - HRSD reduction of 50% from baseline).

Analysis 5.7.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 7 Depression remission post-treatment antenatally (depressive symptomatology - HRSD score less than 8.

Analysis 5.8.

Comparison 5 Omega-3 fatty acids versus placebo, Outcome 8 Side effects reported (including nausea and dizziness).

Appendices

Appendix 1. Methods used to assess trials included in previous versions of this review

The following methods were used to assess Manber 2004.

Selection of studies

We assessed for inclusion all potential studies we identified as a result of the search strategy. We resolved any disagreement through discussion.

Data extraction and management

Each author independently extracted data from trial reports using a pilot-tested data extraction form developed by the primary author. We resolved discrepancies through discussion. We used the Review Manager software (RevMan 2003) to double enter all the data or a sub-sample. There was no need to contact authors of the original reports to provide further details.

Assessment of methodological quality of included studies

We assessed the validity of each study independently using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2005). Methods used for generation of the randomisation sequence were described for the included trial.

(1) Selection bias (randomisation and allocation concealment)

We assigned a quality score for each trial, using the following criteria:
(A) adequate concealment of allocation: such as telephone randomisation, consecutively numbered sealed opaque envelopes;
(B) unclear whether adequate concealment of allocation: such as list or table used, sealed envelopes, or study does not report any concealment approach;
(C) inadequate concealment of allocation: such as open list of random number tables, use of case record numbers, dates of birth or days of the week.

(2) Attrition bias (loss of participants, for example, withdrawals, dropouts, protocol deviations)

We assessed completeness to follow up using the following criteria:
(A) less than 5% loss of participants;
(B) 5% to 9.9% loss of participants;
(C) 10% to 19.9% loss of participants;
(D) more than 20% loss of participants.

(3) Performance bias (blinding of participants, researchers and outcome assessment)

We assessed blinding using the following criteria:
(A) blinding of participants (yes/no/unclear);
(B) blinding of caregiver (yes/no/unclear);
(C) blinding of outcome assessment (yes/no/unclear).

We assigned a rating to each trial, compared results and discussed differences until we reached agreement. We described reasons for exclusion of any apparently eligible trial in the 'Characteristics of excluded studies' table.

Measures of treatment effect

We carried out statistical analysis using the Review Manager software (RevMan 2003). We planned to use fixed-effect meta-analysis for combining data in the absence of significant heterogeneity if trials were sufficiently similar. If we had found heterogeneity, we planned to explore this by sensitivity analysis followed by random-effects if required. For dichotomous data, we presented results as summary risk ratio with 95% confidence intervals. For continuous data, we used the mean difference if outcomes were measured in the same way between trials. We would have used the standardised mean difference to combine trials that measured the same outcome, but used different methods. Had there been evidence of skewness, we would have reported this. For trials using different treatment strategies, we planned to analyse them separately and combine the results only if there was no reason to think that they differed in relevant ways. While the primary meta-analysis was based on the occurrence of depression or not (however measured by trialists), we incorporated several depression rating scales or cut-off points.

Intent-to-treat analysis

We analysed data on all participants with available data in the group to which they are allocated, regardless of whether or not they received the allocated intervention. If in the original reports participants were not analysed in the group to which they were randomised, and there is sufficient information in the trial report, we attempted to restore them to the correct group.

Assessment of heterogeneity

We planned to apply tests of heterogeneity between trials, if appropriate, using the I2 statistic. Had we identified high levels of heterogeneity among the trials (exceeding 50%), we would have explored it by prespecified subgroup analysis and performed sensitivity analysis. We would have used a random-effects meta-analysis as an overall summary if this was considered appropriate.

Subgroup analyses

Although not completed due to insufficient data, we planned to conduct subgroup analyses classifying whole trials by interaction tests as described by Deeks 2001. Two a priori subgroup analyses were planned:
(1) the effects of intervention mode (for example, individual versus group-based interventions); and
(2) the effects of intervention duration (for example, single-contact interventions versus multiple-contact interventions).

If within-trial subgroup analysis was required, we conducted a separate meta-analysis for participants in each subgroup.

Sensitivity analyses

Although not completed due to insufficient data, we planned to conduct sensitivity analysis to explore the effect of trial quality. This would have involved analysis based on an A, B, C or D rating of selection bias and attrition bias. Studies of poor quality would have been excluded in the analysis (those rating B, C or D) in order to assess for any substantive difference to the overall result.

What's new

DateEventDescription
12 April 2013New search has been performedSearch updated. Methods updated.
12 April 2013New citation required but conclusions have not changedFive new trials included (Field 2009a; Freeman 2008; Manber 2010; Su 2008; Wirz-Justice 2011). The evidence for acupuncture, maternal massage, bright light therapy, or omega-3 in treating antenatal depression is inconclusive.

History

Protocol first published: Issue 4, 2007
Review first published: Issue 4, 2008

DateEventDescription
31 March 2010AmendedSearch updated. Eleven reports added to Studies awaiting classification.
16 January 2008AmendedConverted to new review format.

Contributions of authors

C-L Dennis developed the protocol and completed the review. T Dowswell assessed risk of bias, completed data extraction, entered data and was involved in setting up the analysis for the 2013 update.

Declarations of interest

None known.

Sources of support

Internal sources

  • University of Toronto, Canada.

External sources

  • The National Institute for Health Research (NIHR), UK.

    TD is supported by the NIHR NHS Cochrane Collaboration Programme grant scheme award for NHS-prioritised centrally-managed, pregnancy and childbirth systematic reviews: CPGS 10/4001/02

Differences between protocol and review

Methods updated.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Field 2009a

Methods2-armed RCT.
Participants

200 women diagnosed as depressed based on a SCID were recruited from 2 hospital ultrasound clinics in the USA (date and site of recruitment not stated) with 149 providing follow-up outcome assessment data. Women were primarily low socioeconomic status and were approximately 57% Hispanic, 38% Black, and 5% non-Hispanic.

Inclusion criteria: (1) age greater than 18 years, (2) gestational age 16-20 weeks, (3) singleton pregnancy, (4) no pregnancy complications. Exclusion criteria: (1) medical illness, (2) metabolic or eating disorder, (3) other psychiatric condition, (4) self reported drug use, using other medications or smoked during pregnancy.

Interventions

Intervention Group (massage; n = 88 in analyses): 2 20-minute moderate pressure massages per week for a period of 12 weeks by their significant other who was taught the massage by a massage therapist and were given DVDs for at-home coaching in the massage procedure. Massage began at 20 weeks' gestation until 32 weeks' gestation.

Control Group (standard care; n = 61 in analyses): standard treatment - not described.

Outcomes

Follow-up assessments were completed at 32 weeks' gestation and 2 days postpartum.

Outcomes: Depression - CES-D; Anxiety - STAI.

NotesWe contacted the author on 11 December 2012 for more information on study methods.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe method of randomisation was not described.
Allocation concealment (selection bias)Unclear riskNot described.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskIt was unclear whether outcome assessors were blinded to group allocation.
Incomplete outcome data (attrition bias)
All outcomes
High riskThe authors report that 200 women recruited and attrition rates were 30%. It is unclear how many women were randomised to each group and whether the attrition was equally divided between the 2 groups. Published report indicates that 88 women were analysed in the intervention group and 61 women were analysed in the control group suggesting either uneven randomisation or attrition. Group denominators were not provided in results tables.
Selective reporting (reporting bias)Unclear riskAssessment from published study report; author contacted December 2012 for more information.
Other biasUnclear riskUnclear.

Freeman 2008

Methods2-armed RCT, placebo-controlled.
Participants

59 pregnant and postpartum women attending for care in Texas USA were recruited with 21 pregnant women providing follow-up outcome data.

Inclusion criteria: (1) after 12 weeks’ gestation (12-32 weeks), (2) aged 18-45 years, (3) met criteria for major depressive verified by a SCID, (4) scored > 8 on EPDS.

Exclusion criteria: (1) previous intolerance of omega-3 fatty acids, (2) current use of antidepressants or anticoagulants, (3) psychosis, (4) bipolar disorder, (5) active substance abuse, (6) active suicidal ideation.

(Note: women were also eligible for this trial if they had postnatal depression, onset within 4 weeks of the birth, up to 6 months postpartum. Data for these women have not been included in the analyses in this review).

Interventions

Intervention Group (results for 12 women recruited in pregnancy): 4 capsules per day for a total daily dose of 1.9 g of omega-3 that included 1.1 g of eicosapentaenoic acid (EPA) and 0.8 g of docosahexaenoic acid (DHA) plus 6 individual 30-minute manualised supportive psychotherapy sessions.

Control Group (results for 9 women recruited in pregnancy): 4 placebo capsules per day containing corn oil with a small amount (1%) of fish oil plus 6 individual 30-minute manualised supportive psychotherapy sessions.

OutcomesOutcomes: depression (change scores from initial assessment to final assessment) - EPDS, HRSD; maternal self-reported side effects.
NotesSmall, homogenous sample.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskPlacebo-controlled trial with identical capsules - described as double-blind but no details were provided.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskWomen and caregivers were stated to be blind to group allocation. It was unclear whether outcome assessors were blinded to group allocation.
Incomplete outcome data (attrition bias)
All outcomes
Unclear risk59 women were recruited and outcome data were available for 51 (86%). It was not clear how many pregnant women dropped out as the sample included both pregnant and postpartum women. Outcome data were available for 21 pregnant women (9 in the placebo group and 12 in the omega-3 group). Data were presented if women attended at least 1 follow-up appointment. (It is not clear in the results whether women attended 1 or 6 follow-ups therefore data are difficult to interpret).
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear riskThere was some baseline imbalance in relation to weeks gestational at recruitment: mean week gestation was 27.3 in the control group while mean week gestation in omega-3 group was 22.8.

Manber 2004

Methods3-armed RCT. Modified intention-to-treat analysis - included only participants with 1 post-randomisation evaluation; data were not collected from treatment dropouts.
Participants61 depressed pregnant US women; 54 women provided data for analyses.
Identified using DSM-IV criteria.
Interventions Intervention Group 1 (depression-specific acupuncture; n = 20): 12 25-30 minute active acupuncture sessions by an acupuncturist following the principles of traditional Chinese medicine over an 8-week period. Treatment specifically addressed depressive symptoms.
Intervention Group 2(maternal massage; n = 20): 12 25-30 minute massage sessions over an 8-week period to provide control for attention, physical contact, relaxation, and respite from daily stress.
Control group (non-specific acupuncture; n= 21): 12 25-30 minute acupuncture sessions by an acupuncturist delivered over an 8-week period. Acupuncture did not specifically address depressive symptoms.
OutcomesOutcomes: depression (immediately post-treatment and postnatally): MDD, HRSD, BDI.
NotesSmall, homogenous sample.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe method of randomisation was not described.
Allocation concealment (selection bias)Unclear riskNot described.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThis was a 3-arm trial and it was stated that there was partial blinding of women and acupuncturists for the 2 acupuncture arms. Acupuncturists delivering the interventions did not carry out the initial assessment and were advised to follow the prescribed treatment. It was unclear whether outcome assessors were blinded to group allocation.
Incomplete outcome data (attrition bias)
All outcomes
Unclear risk61 women were randomised and 54 (88.5%) were followed up; it was stated that baseline assessments were similar for those women who were not followed up. It is unclear how many women complied with intended treatment in each of the arms.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear riskWomen attended for different numbers of sessions depending on gestational age at recruitment. Baseline characteristics for each group were not described although it was stated that there were no statistically significant differences for gestational age and depression scores.

Manber 2010

Methods3-armed RCT.
Participants

150 depressed pregnant US women recruited through advertising in publications and local clinics

Inclusion criteria: (1) viable pregnancy, (2) between 12 to 30 weeks’ gestation, (3) at least 18 years old, (4) had a major depression determined by clinical interview (DSM IV), (5) scored at least 14 on the HRSD.

Exclusion criteria: (1) other serious psychiatric disorders (other than social phobia), SAD or psychosis or personality disorders, (2) suicidal, (3) abnormal thyroid or drug screen, (4) receiving other treatments for depression (pharmacotherapy, psychotherapy, herbal therapy, ECT), (5) not receiving prenatal care, (6) on bed rest or with other serious medical conditions.

Interventions

In all groups treatment was over 8 weeks, twice-weekly sessions in the first 4 weeks then weekly for a further 4 weeks. In all groups interventions were delivered by trained staff and each treatment session lasted approximately 25 minutes. In all groups staff were advised to minimise verbal interaction.

Intervention Group: (depression-specific acupuncture; n = 52). Women received acupuncture that was individualized and specific for treating depression following principles of traditional Chinese medicine. 7-12 points were needled at each session (avoiding any points not recommended in pregnancy).

Control Group 1: (non-specific acupuncture; n = 49). Women received acupuncture not specific to depression following the same regimen as the intervention group.

Control Group 2: (Swedish massage not known to relieve depression; n = 49). The sessions lasted a similar time to the acupuncture interventions.

In the data and analyses we have combined results for the 2 control conditions to form a single control group.

OutcomesOutcomes: depression (reduction in depression scores at 4 and 8 weeks after commencing treatment): HRSD; treatment response (50% reduction in depression scores from baseline); depression remission; participant and provider views; side effects.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskBlocked randomisation by computer.
Allocation concealment (selection bias)Low riskThe allocation was carried out by a study coordinator and “randomisation sequence was concealed until all the interventions were assigned”.
Blinding (performance bias and detection bias)
All outcomes
Low risk

Participants receiving acupuncture were unaware as to whether the acupuncture was depression-specific (intervention group) or non-specific (control group).

The outcome assessor was blinded to group allocation.

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskAlthough there was some loss to follow-up, intention-to-treat analysis was conducted (imputing values for women who were not assessed). A quarter of the women in each group did not receive the planned intervention or were lost to follow-up.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear riskRecruitment appeared to take 5 years. Participants recruited later may have received different care compared with those recruited earlier and there may have been staff changes that had differential effects in the study groups.

Su 2008

Methods2-armed RCT.
Participants

Women were recruited in a hospital in Taiwan.

Inclusion criteria: 36 women aged 18 to 40 years with major depression onset between 16 and 32 weeks' gestation. Screened using EPDS (Taiwanese version) by psychiatric research nurse and then interviewed by psychiatrists using the Mini-International Neuropsychiatric interview (MINI), with score of at least 18 on the 21 item HAM-D scale. Women were in good physical health.

Exclusion criteria: (1) had bipolar disorder, (2) were psychotic, (3) substance abuse or dependence, (4) diagnosis of borderline or antisocial personality disorder, (5) had taken psychotropic agents in the previous month.

Interventions

Intervention Group: (omega-3; n = 18): 5 capsules per day for a total daily dose of 2.2 g of EPA and 1.2 of DHA produced from fish oil concentrate.

Control Group: (placebo; n = 18): 5 placebo capsules per day containing olive oil.

Both active and placebo capsules were deodorized and contained orange flavour to disguise taste.

Intervention over 8 weeks.

OutcomesOutcomes: depression: HRSD, EPDS, BDI at 8 week follow-up.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described. “random assignment” after 1 week placebo lead in trial. Women who had decreased HRSD scores of 20% or more were excluded after initial phase.
Allocation concealment (selection bias)Low riskPlacebo-controlled trial with identical capsules - described as double-blind.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskIt was unclear whether outcome assessors were blinded to group allocation.
Incomplete outcome data (attrition bias)
All outcomes
Unclear risk

36 women were randomised and 12 dropped out (33%).

It was stated that an intention-to-treat was carried out for 33 women for some outcomes.

Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskNo other obvious bias identified.

Wirz-Justice 2011

  1. a

    BDI: Beck Depression Inventory
    CES-D: Center for Epidemiologic Studies Depression Scale
    DDHA: docosahexaenoic acid
    ECT: electroconvulsive therapy
    EPA: eicosapentaenoic acid
    EPDS: Edinburgh Postnatal Depression Scale
    DSM: Diagnostic and Statistical Manual for Mental Disorders
    HAM-D: Hamilton depression scale
    HRSD: Hamilton Rating Scale for Depression
    MDD: Major depressive disorder
    RCT: randomised controlled trial
    SAD: seasonal affective disorder
    SCID: Structure Clinical Interview for Depression
    SIGH-AD: Structural Interview Guide for the Hamilton Depression and Anxiety Scales
    STAI: State Anxiety Inventory

Methods2-armed RCT.
Participants

Women recruited via outpatient clinics, GP referrals and via media in Basel, Switzerland. Women were recruited between 2004-2008.

46 women randomised.

Inclusion criteria: 100 women were screened and 70 met criteria (telephone interview score of 10 or more on the EPDS). Women were then assessed by clinical interview and those diagnosed with MDD (DSM-IV) and with HDRS and SIGH-AD score of 20 or more. Women were German speaking and 18-45 years of age, medically healthy, 4-32 weeks’ gestation.

Exclusion criteria: women were excluded if they had bipolar disorder, SAD, any psychotic episode, substance abuse within previous 6 months, recent suicide attempt, primary anxiety disorder, sleep disorder or usual sleep onset after 1am or waking later than 9 am, obstetric complications. Women who started antidepressant therapy after randomisation were excluded from the analyses.

Interventions

Intervention Group: (Bright light therapy - 7000 lux white light; n = 24 randomised with 16 (66.6%) included in the analyses). Women used light box shortly after waking for 1 hour daily for 5 weeks.

Control Group (Inactive bright light - 70 lux red light; n = 22 randomised with 11 included in the analyses). Same regimen above with inactive light.

Outcomes

Outcomes: depression (change scores from initial assessment to final assessment at 5 weeks): HDRS, SIGH-ADS; treatment response (50% reduction in scores); depression remission (50% reduction in scores and score < 8).

Note: BDI, Mongomery Asberg Depression rating scale and SIGH-SAD-SR mean scores were not reported in our analyses as there appeared to be some differences between groups at baseline.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskBlock randomisation (block size 6) computer-generated.
Allocation concealment (selection bias)Unclear riskThe person carrying out randomisation was described as a non-blind research nurse, but it was also reported that light boxes were in identical, coded cartons to preserve blinding.
Blinding (performance bias and detection bias)
All outcomes
Low risk

Light boxes were provided in identical coded cartons and women in the intervention and control groups were asked about expectations of the therapy. Scores were very similar which lead authors to believe that the placebo was convincing.

It was not clear whether the researcher was blind. Women were asked not to disclose details about the light boxes to staff collecting outcome data and it was reported that all other staff and outcome assessors were blind.

Incomplete outcome data (attrition bias)
All outcomes
High riskOf 46 randomised only 27 were followed up and the reasons for loss to follow-up were imbalanced in the 2 arms of the trial. 5/6 women starting adjuvant therapy (and excluded from the analyses) were in the control group.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear riskThere was some imbalance at baseline. 4 women in the intervention group were taking antidepressants at baseline (reported to be ineffective). Recruitment was over 4 years; it was not clear whether there were changes (e.g., in staff) that may have affected outcomes.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    RCT: randomised controlled trial

Bosco 2007This study was not an RCT but rather a quasi-experimental study (randomisation was not performed) and primary outcome was a collection of "emotional complaints" that included anxiety, depression, and irritability. Study examined acupuncture.
Doornbos 2009This study was not an RCT. Further, the purpose of the study was not the treatment of antenatal depression but rather to examine the effect of omega-3 for the prevention of postpartum depression.
Epperson 2004This study was not an RCT but rather a pilot study evaluating bright light therapy.
Field 2004Uncertain randomisation with allocation and concealment strategy not described; small sample size; uncertain if outcome assessor was blinded to group allocation; standard deviations not reported. Study examined effect of massage.
Field 2009bUncertain randomisation with allocation and concealment strategy not described; high attrition (64% in control group and 51% in intervention group); uncertain if outcome assessor was blinded to group allocation; denominators were not provided in results tables. Study examined effect of combining massage therapy with group interpersonal psychotherapy.
Gedde-Dahl 2012The purpose of the study was not the treatment of antenatal depression; all study participants were healthy pregnant women (not depressed). Study examine the impact of self-administered relaxation and guided imagery techniques on the birth experience.
Hayes 2001The purpose of the study was not the treatment of antenatal depression but rather the prevention of postpartum depression. Intervention was educational.
Mozurkewich 2011The purpose of this ongoing trial is not the treatment of antenatal depression but rather the prevention of antenatal and postpartum depression. Intervention is evaluating omega-3.
Parry 2010Both pregnant and postpartum women with depression were included in this study of which only 7 pregnant women – cross-over design so no real control group as standard care. Study examine early versus late wake therapy.

Characteristics of studies awaiting assessment [ordered by study ID]

Field 2008

MethodsRCT. 2 arms with individual randomisation.
Participants57 women with prenatal depression and their partners recruited at an urban hospital in the USA.
Interventions

The intervention group were allocated to receive twice weekly massage from their partners over 16 weeks from 20 weeks' gestation.

The control group received standard care.

OutcomesDepression, anxiety, leg and back pain at the end of the intervention.
NotesDenominators for the intervention and control group were not provided. We have contacted the author (11th December 2012) for further information. tfield@med.miami.edu

Field 2012

MethodsRCT, 2 arms with individual randomisation.
Participants84 pregnant women 18-22 weeks' gestation recruited at 2 hospital ultrasound clinics in the USA.
InterventionsYoga and massage sessions over 12 weeks by trained therapists. Women in the control group received standard care.
OutcomesDepression, neonatal outcomes including gestation at delivery and birthweight.
NotesDenominators for the intervention and control group were not provided. We have contacted the author (11th December 2012) for further information. tfield@med.miami.edu

Rees 2008

  1. a

    DHA: docosahexaenoic acid
    DSM: Diagnostic and Statistical Manual for Mental Disorders
    EPA: eicosapentaenoic acid
    EPDS: Edinburgh Postnatal Depression Scale
    DSM: Diagnostic and Statistical Manual for Mental Disorders

MethodsRCT, 2-arm trial with individual randomisation.
Participants

26 women attending a perinatal depression clinic at a hospital in Sydney, Australia.

Inclusion criteria: women with a current episode of major depression or dysthymia, according to DSM-IV criteria confirmed by clinical interview, more than 21 years of age, during third trimester of pregnancy or up to 6 months postpartum. Women with bipolar disorder, psychosis, drug and or alcohol abuse, obsessive compulsive disorder, eating or personality disorders were excluded, along with women with unstable medical conditions, diabetes, receiving anticoagulants or with fish allergy. Women receiving antidepressants, psychological therapy, taking fish oil supplements or with diet high in fish oil (3 oily fish portions per week) were also excluded.

26 women randomised. This study included both pregnant and postpartum women; separate outcome data for pregnant woman were not reported.

Interventions

Intervention group: (includes both pregnant and postpartum women).

6 g day of fish oil (containing DHA and EPA omega-3 fatty acids).

Control group: placebo (Sunola oil).

Outcomes

EPDS scores weekly.

HDRS scores weekly.

MDRS scores weekly up to 6 weeks after treatment.

Side effects.

This study was assessed as eligible for inclusion, however, we have not included data from this study in the review as separate data were not available for pregnant women.

NotesWe have contacted the author (17 December 2012) to try to obtain data for pregnant women.

Ancillary