1. Objectives

To evaluate the effect of antenatal breastfeeding education on breastfeeding initiation and duration

2. How studies were identified

The following databases were searched in March 2016:

• Cochrane Pregnancy and Childbirth’s Trials Register
• CENTRAL (The Cochrane Library 2016, Issue 3)
• MEDLINE
• Embase
• Scopus
• CINAHL

Reference lists were also searched and the authors directly contacted investigators

3. Criteria for including studies in the review

3.1 Study type

Randomized controlled trials, including cluster randomized trials

3.2 Study participants

Pregnant women, their partners, or both

3.3 Interventions

Antenatal breastfeeding education was compared to routine care, a different type of breastfeeding education, or to different methods of providing antenatal breastfeeding education

(Antenatal breastfeeding education was defined as breastfeeding education delivered during pregnancy, in any form, to women or their partners. Interventions examining intrapartum or postpartum breastfeeding education in addition to antenatal education were excluded)

3.4 Primary outcomes

• Duration of any breastfeeding
• Duration of exclusive breastfeeding
• Proportion of women achieving any breastfeeding at three and six months
• Proportion of women exclusively breastfeeding at three and six months
• Initiation of breastfeeding

Secondary outcomes included maternal satisfaction, breastfeeding complications (mastitis, breast abscess), infant growth (head circumference, weight), neonatal sepsis, and doctor’s visits and hospital admissions for the child

4. Main results

4.1 Included studies

Twenty-four randomized controlled trials, enrolling 10,056 women, met the inclusion criteria for this review

• Twenty studies provided relevant data for analysis, involving a total of 9789 women
• Fourteen randomized controlled trials and five cluster-randomized trials contributed data for analyses, with four trials comparing multiple treatment arms
• Sixteen studies meeting the inclusion criteria compared a single method of breastfeeding education to routine care, and in 13 of these studies routine care involved some form of breastfeeding education or support
• Two studies meeting the inclusion criteria compared one type of breastfeeding education to another, three studies compared multiple methods of providing education to a single method, and two studies compared different combinations of multiple interventions
• Interventions employed breastfeeding education sessions, printed material, video, peer counselling, lactation consultation, text and voice phone messages, Internet-based education, and breastfeeding support
• Women were recruited at a range of gestational time points, from ≥12 weeks’ to >36 weeks’ gestation

4.2 Study settings

• Australia (3 trials), Canada (2 trials), China, Denmark, Iran, the Netherlands, Nigeria, Singapore, the United Kingdom of Great Britain and Northern Ireland (2), and the United States of America (11 trials)
• Trials were predominantly conducted in high-income settings, and most studies recruited women who were accessing usual antenatal care

4.3 Study settings

4.3.1 How the data were analysed
Five comparisons were made: i) one type of breastfeeding education versus standard/routine care; ii) one type of breastfeeding education versus a different type of breastfeeding education; iii) multiple methods of breastfeeding education versus a single method; iv) varying combinations of multiple methods of providing breastfeeding education; and v) multiple methods of breastfeeding education versus routine care. For dichotomous data, relative risks (RR) and 95% confidence intervals (CI) were calculated, while for continuous data, mean differences (MD) and 95% CI were calculated. Data from cluster-randomized trials were reported in the included studies as odds ratios (OR) with adjustment for clustering, and were not included in pooled analyses with randomized controlled trial data. Where clinical or statistical heterogeneity occurred, random effects meta-analysis was used. To investigate potential sources of heterogeneity, the following subgroup analyses were planned, but due to insufficient data, not performed:

• Type of intervention
• Study setting
• Maternal education
• Maternal occupation

Sensitivity analyses excluding studies rated as poor-quality for allocation concealment or high attrition were also planned

4.3.2 Summary of effects
One type of breastfeeding education versus standard/routine care
Duration of any breastfeeding
Two studies reported on this outcome, both employing breastfeeding education session interventions. One trial involved 165 women and found no difference in breastfeeding duration between those who received education and those who received standard care (MD 0.0 weeks, 95% CI [-2.78 to 2.78]), while the other involved 16 women and found a non-statistically significant increase in breastfeeding duration of 6.2 months (95% CI [-10.84 to 23.24 months]).

Duration of exclusive breastfeeding
No identified trials reported on this outcome.

Proportion of women breastfeeding at three and six months
Pooled analysis demonstrated no evidence of a difference between those who received breastfeeding education and those who received standard care in the proportion of women achieving any breastfeeding at three months (RR 0.98, 95% CI [0.82 to 1.18], 2 studies/654 women) or at six months (RR 1.05, 95% CI [0.90 to 1.23], 4 studies/1636 women). No difference between treatment groups was found in the proportion of women exclusively breastfeeding at three months (RR 1.06, 95% CI [0.90 to 1.25], 3 studies/822 women) or at six months (RR 1.07, 95% CI [0.87 to 1.30], 3 studies/2161 women).

Initiation of breastfeeding
In pooled analysis, initiation of breastfeeding was not affected by the provision of breastfeeding education in comparison to standard care (RR 1.01, 95% CI [0.94 to 1.09], 8 trials/3505 women).

Additional outcomes
Nipple pain and nipple trauma were reduced with lactation consultation in comparison to standard care in one trial of 70 women (nipple pain: MD -19.80 visual analogue scale points [lower points being less pain], 95% CI [-23.23 to -16.37], p<0.00001; nipple trauma: MD 38.65 nipple trauma index points [higher points being less trauma], 95% CI [32.95 to 44.35], p<0.00001). Breastfeeding problems were not different between treatment groups in one trial of 1168 women, and other pre-specified secondary outcomes were not reported in any trials.

One type of breastfeeding education versus a different type of breastfeeding education
Duration of breastfeeding
No identified trials reported on the duration of any or exclusive breastfeeding.

Proportion of women breastfeeding at three and six months
No significant difference in the proportion of any breastfeeding at three months was found between those receiving group or individual discussion (RR 2.84, 95% CI [0.61 to 13.18], 1 trial/74 women). At six months, no significant difference between those receiving practical skills education or attitude education was found in the proportion of any breastfeeding (RR 1.09, 95% CI [0.94 to 1.28], 1 trial/590 women) or exclusive breastfeeding (RR 1.03, 95% CI [0.61 to 1.73], 1 trial/590 women).

Initiation of breastfeeding
In one trial of 614 women, no difference in the rate of initiation of breastfeeding was found between practical skills education and attitude education groups (RR 1.02, 95% CI [0.99 to 1.06]).

No additional outcomes were reported on in the identified trials.

Multiple methods of breastfeeding education versus a single method of breastfeeding education
Duration of any breastfeeding
Breastfeeding education plus commitment versus breastfeeding education alone had no significant effect on the duration of any breastfeeding in one trial of 18 women (MD 8.00 days, 95% CI [-6.84 to 22.84]).

Proportion of women achieving any breastfeeding at six months
Video plus breastfeeding education session in comparison to the provision of pamphlets had no effect on the proportion of women achieving any breastfeeding at six months (RR 1.59, 95% CI [0.86 to 2.94], 1 trial/175 women).

No additional outcomes were reported on in the identified trials.

Different combinations of multiple methods of providing BF education
Proportion of women breastfeeding at three and six months
Any breastfeeding (OR 0.82, 95% CI [0.58 to 1.16], 1 trial/698 women) and exclusive breastfeeding (OR 0.79, 95% CI [0.57 to 1.09], 1 trial/698 women) at three months did not differ between those receiving booklet plus 24-hour free lactation consultation versus booklet plus phone number for breastfeeding questions. Exclusive breastfeeding at three months did not differ between those given a booklet plus video plus lactation consultant versus a booklet plus video (OR 1.40, 95% CI [0.70 to 2.80], 1 trial/150 women). However, at six months the difference favouring booklet plus video plus lactation consultant reached borderline statistical significance (OR 2.50, 95% CI [1.00 to 6.25], p=0.05; 1 trial/169 women).

No additional outcomes were reported on in the identified trials.

Multiple methods of BF education versus standard/routine care
Duration of any breastfeeding
Breastfeeding education plus commitment versus standard care had no statistically significant effect on the duration of any breastfeeding in one trial of 16 women (MD 14.20 days, 95% CI [-2.97 to 31.37]). Lactation consultation plus incentive did not significantly increase breastfeeding duration in comparison to standard care in one trial of 48 women (median 12 versus six weeks, data not shown).

Proportion of women exclusively breastfeeding at three and six months
In one trial involving 159 women, provision of a booklet plus video plus lactation consultant versus routine care significantly improved the odds of breastfeeding at three months (OR 2.60, 95% CI [1.25 to 5.40], p=0.01), the booklet and video alone had no statistically significant effect (OR 1.80, 95% CI [0.80 to 4.05]). At six months, booklet plus video plus lactation consultant resulted in a borderline statistically significant improvement in exclusive breastfeeding (OR 2.40, 95% CI [1.00 to 5.76], p=0.05; 1 trial/175 women), while booklet and video alone had no effect (OR 0.90, 95% CI [0.30 to 2.70]). In a further trial involving 390 women, monthly education sessions plus a weekly cell phone message statistically significantly increased the likelihood of exclusive breastfeeding at three months (OR 1.80, 95% CI [1.10 to 2.95], p=0.019), and at six months (OR 2.40, 95% CI [1.40 to 4.11], p=0.0015).

Initiation of breastfeeding
Monthly education sessions plus a weekly cell phone message statistically significantly increased rates of breastfeeding initiation in comparison to standard care (OR 2.61, 95% CI [1.61 to 4.24], p=0.00011; 1 trial/380 women).

No additional outcomes were reported on in the identified trials.

5. Additional author observations*

Twenty-two of the 24 included studies were conducted in high-income settings limiting the generalizability of the findings to low- and middle-income regions. No studies were blinded and two-thirds of studies (n=16) had inadequately reported allocation concealment methods. For the pooled comparisons between antenatal breastfeeding education and standard care, the GRADE approach was used to evaluate the quality of the evidence, although no downgrading was applied for lack of blinding. Initiation of breastfeeding and the proportion of women achieving any breastfeeding at six months was rated as high quality, the proportion of women exclusively breastfeeding at three and six months was graded as moderate quality, and the proportion of women achieving any breastfeeding at three months was rated as low quality.

Overall, the evidence summarised in this review indicates that in comparison to standard care, additional antenatal breastfeeding education has no effect on the initiation of breastfeeding, the proportion of women breastfeeding three or six months, or the duration of breastfeeding. However, in most trials standard care included some form of breastfeeding education, and in trials involving multiple types of breastfeeding education, some improvements were observed.

Further research on the effect of antenatal breastfeeding education on breastfeeding outcomes should be conducted in low- and middle-income settings, and should ideally include comparator groups with little to no antenatal breastfeeding education.