Support for breastfeeding mothers
There was evidence of effectiveness of both professional and lay support in promoting continued breastfeeding, especially where baseline breastfeeding initiation rates were intermediate, between 60% and 80%. But methodological problems in the included studies make it difficult to draw sound practical recommendations on this topic, or to extrapolate the review findings to communities in under-resourced settings.
RHL Commentary by Pellegrini L, Sguassero Y
1. EVIDENCE SUMMARY
The current WHO recommendation for mothers is to practice exclusive breastfeeding from birth of the baby to six months of age with no supplemental liquid or solid foods other than medications or vitamins. After that safe, appropriate and adequate complementary foods may be introduced but on-demand breastfeeding should continue until the baby is two years of age or beyond (1).
The present systematic review (2) is an updated version of a previous review (3). The primary objective of the review was to assess the effectiveness of different supportive interventions (i.e., by lay persons or professionals, or both) provided during antenatal and/or postnatal periods to:
- pregnant women intending to breastfeed;
- women who have delivered and who express intention to breastfeed; and
- women who breastfeed their babies for different durations of time (4–6 weeks, four months, etc.) up to 12 months.
The secondary objectives were:
- to explore the association between baseline breastfeeding prevalence and effectiveness of supportive interventions; and
- to compare effectiveness of interventions carried out in the antenatal and postnatal periods versus those taking place in the postnatal period alone.
Only randomized or quasi-randomized controlled studies were eligible. Loss to follow-up greater than 15% was considered as an exclusion criteria. Studies on interventions performed exclusively during the antenatal period were also excluded.
The outcomes sought were: effect of interventions on the duration of any type of breastfeeding (partial or exclusive breastfeeding) at different points in time; neonatal or infant morbidity; and maternal satisfaction with feeding or care recommendations.
The search methods used to identify relevant studies were comprehensive, but it should be highlighted that databases containing literature from developing countries such as the Latin American and Caribbean Literature on the Health Sciences (a BIREME System's cooperative database), Index Medicus for the WHO Eastern Mediterranean Region and the African Index Medicus were not explored. The searches were based on the strategy used by the Cochrane Pregnancy and Childbirth Group. No language restrictions were applied.
Quality assessment of included studies was done according to the following criteria: concealment of allocation, method randomization, rate of withdrawals and drop-outs, blinding, and intention-to-treat analysis.
The size of treatment effect was calculated as relative risk (RR) with 95% confidence intervals (CI). The heterogeneity of results was assessed using the I2 statistic (4, 5). In case of heterogeneity, the sources were explored according to pre-specified subgroup and sensitivity analyses.
A total of 34 studies were included. Twenty-five studies were conducted in developed countries (such as Canada, United Kingdom, and USA) and nine had been carried out in Bangladesh, Brazil, India, Mexico and Nigeria. The majority of the supportive interventions were offered to women intending to breastfeed. Baseline features of the target populations and details of training and qualifications of supporters varied across the included studies. Breastfeeding support was provided by professionals in 18 studies, by lay persons in nine studies, and by both professionals and lay persons in six studies.
There is no reference in the review to the overall quality of the included studies, although details about the studies are provided in the "Table of Included Studies". In general, the method of randomization was not reported in the studies. Only 14 out of 34 included studies were adjudged to present adequate allocation concealment (implementation of random allocation sequence without foreknowledge of intervention assignments). Blinding was also a problematic area and the outcome assessors could have been aware of the groups mothers were assigned to. The pooled estimates should be interpreted with caution due to the high degree of heterogeneity among the included studies.
When comparing all forms of extra support versus usual care, a beneficial effect on the duration of any type of breastfeeding before the last outcome assessment up to six months was found by pooling the data of 28 studies (RR 0.91; 95% CI 0.86–0.96; I2=53.6%). Sensitivity analysis based on 14 of these studies with adequate concealment allocation did not show any difference. However, when studies were grouped according to baseline rates of breastfeeding initiation [i.e., high (>80%), intermediate (60%–80%), and low (<40%)], the studies with high (RR 0.91; 95% CI 0.81–1.01; I2=64,7%) and low (RR 0.88; 95% CI 0.69–1.02; I2=76.2%) rates showed no significant effect compared with studies with intermediate initiation rates (RR 0.92; 95% CI 0.85–0.98; I2=43,4%). Results were inconsistent when measuring the impact at different points in time – e.g., RR 0.88 (95% CI 0.78–1.00) and I2=53,7% before 4–6 weeks, versus RR 0.83 (95% CI 0.69–0.99) and I2=67,3% before two months. Nevertheless, when measuring the impact on stopping exclusive breastfeeding before the last outcome assessment, pooled data from 20 studies suggested some benefit (RR 0.81; 95% CI 0.74–0.89; I2=92.2%). It appeared that this positive effect was greater during the first four months.
There was no significant overall effect on stopping any breastfeeding before the last outcome assessment up to six months when comparing extra professional support versus usual care (RR 0.94; 95% CI 0.87–1.01; I2=49.8%). Conversely, a beneficial effect on interrupting exclusive breastfeeding was found (RR 0.91; 95% CI 0.84–0.98; I2=81.5%).
The overall effect on stopping any breastfeeding before the last outcome assessment up to six months was significant when lay support was compared with usual care (RR 0.86; 95% CI 0.76–0.98; I2=75.6%). Similarly, a beneficial effect on stopping exclusive breastfeeding was found (RR 0.72; 95% CI 0.57–0.90) even if the level of heterogeneity was higher, i.e., I2=96.3%. Face-to-face strategies tended to be more effective than support provided over the telephone (RR 0.91; 95% CI 0.86–0.96; I2= 53.6%).
A beneficial effect was found in prolonging exclusive breastfeeding when combining the data of six trials using the WHO/UNICEF training course (RR 0.69; 95% CI 0.52 –0.91; I2=97.9%). Only one small trial assessed the impact of La Leche League training (RR 0.52; 95% CI 0.39–0.69).
Data on child health outcomes and mothers satisfaction were scarce.
Based on the available evidence, the following factors should be taken into account in arriving at any conclusions:
- The intervention groups were heterogeneous (different approaches, length, focus and quality of training, etc).
- The “usual care” offered to control groups varied widely and was not always well-described.
- The reporting of outcome was either lacking altogether or not consistently provided
- The timing of end-point assessments varied broadly among the included studies.
All the above described factors may have seriously limited the ability of the review authors to conduct a meta-analysis appropriately. Moreover, regarding the underlying rationale for doing a subgroup analysis (i.e. to identify if the intervention behaves differently, or behaves the same, in different types of populations), great caution is recommended because of the danger of selective reporting of subgroups or selective availability of data from subgroups. It is possible that such selective availability may be biased and, therefore, any subgroup analyses based on less than a complete dataset might be especially likely to lead to false conclusions.
2. RELEVANCE TO UNDER-RESOURCED SETTINGS
2.1. Magnitude of the problem
Each year, new scientific and epidemiological evidence contributes to our knowledge of breastfeeding's role in the survival, growth, and development of children as well as the health and well-being of mothers (6). On the other hand, current international trends on exclusive breastfeeding (7, 8), suggest that further research is still needed to know why breastfeeding rates are not optimal.
Data from Maternidad Martin (12), a Baby Friendly Hospital in Rosario, Argentina, since 1996, showed that 512 breastfed babies out of 3575 live born in 2006 returned to the emergency room within the first week of life. 113 babies were hospitalized. Jaundice was diagnosed in most of the cases (86%) and a weight loss greater than 10% was also associated. These data may suggest that early breastfeeding support could prevent inadequate breast milk intake during the first days of life and, consequently, could diminish neonatal morbidity and hospitalization.
Poor breastfeeding practices, including low initiation rates, short duration of breastfeeding and early introduction of complementary feeding are major public health problems, particularly in under-resourced settings (9). However, it is important to point out that inadequate initial breastfeeding support may increase neonatal morbidity rates during the first week of life in both low- and high-income populations. The available evidence suggests that poor breastfeeding techniques (e.g. improper positioning of the baby and poor latch on) cause insufficient intake of breast milk within the first days of life. This difficulty, in turns, leads to health problems in the newborn such as dehydration, jaundice, and infection, requiring hospitalization in neonatal care unit (10, 11). A recent study conducted in Ghana,(13) Africa, demonstrated that 16% of neonatal deaths could be saved if all infants were breastfed from day one and 22% if breastfeeding starts within the first hour after birth. In view of these data, the slogan of the World Breastfeeding Week in 2007 focused on the importance of appropriate professional supportive interventions to promote early initiation of breastfeeding, i.e., within the first hour of life.
2.2. Applicability of the results
Nine out of 34 studies included in the review were conducted in developing countries, and in communities with intermediate level of breastfeeding, provision of support was helpful in promoting continued breastfeeding. However, there was variability between studies, primarily in terms of how the interventions were implemented and the time points at which data were collected. Hence, it is not easy to draw sound practical recommendations on this topic, or to extrapolate the review findings to communities in under-resourced settings, where access to health-care services is not equitable and health-care resources are often lacking.
2.3. Implementation of the intervention
Local assessment of exclusive breastfeeding trends can serve as a basis for planned actions aimed at improving breastfeeding practice. In the province of Santa Fe, Argentina, surveys are annually conducted to determine the breastfeeding rate at four, six, 12, 24 months of age and the rate of weaning. During the last two surveys, exclusive breastfeeding at six month has consistently been around 23% (14).
In Rosario, Argentina, the initial rate of breastfeeding (i.e. after discharge at 48 hours of life) at Martin Maternity has been in the order of 99% during the last ten years. Furthermore, follow-up of 144 babies born at this maternity hospital during 2004, whose mothers received a face-to-face professional support to continue breastfeeding based on the WHO/UNICEF training course in lactation counseling (15), showed a rate of exclusive breastfeeding of 74% at six months (16). One explanation of these high success rates may be the fact that the intervention was delivered immediately after birth and regularly on mothers’ demand through out the first year of life. This example illustrates the potential beneficial impact of early professional supportive intervention, particularly within the first month.
Developing and implementing long-term breastfeeding early support systems in under-resourced settings is a major challenge. The following, among others, could be the possible reasons:
- insufficient commitment on the part of governments and public health authorities to develop lactation counseling training at all public health-care levels as a central strategy for improving child health and nutrition,
- not allocating “protected time” to put into practice supportive breastfeeding interventions and to develop counseling skills (a time consuming activity),
- scarcity of child and maternal health-care resources, and
- misuse of the available health-care resources.
Further research is needed to assess the effectiveness of interventions to support breastfeeding in developing countries. Future trials should explore appropriate and feasible strategies to tackle the local problems encountered by women who intend to breastfeed for up to six months in under-resourced communities. Women's perspectives and satisfaction should also be considered as outcomes of interest.
Another key element that needs additional consideration is the cost–effectiveness of different training approaches as a platform for success of lactation counselling programmes at the country level.
Ancillary assessment of the impact of professional and lay breastfeeding support may help to set up effective supportive networks for breastfeeding mothers at both health care and community levels.
Sources of support: Centro Rosarino de Estudios Perinatales, Rosario, Argentina.
- Guiding principles for complementary feeding of the breastfed child. Washington, DC: Pan American Health Organization; 2001.
- Britton C, McCormick FM, Renfrew MJ, Wade A, King SE. Support for breastfeeding mothers. Cochrane Database of Systematic Reviews 2007; Issue 1. Art. No.: CD001141; DOI: 10.1002/14651858.CD001141.pub3.
- . Sikorski J, Renfrew MJ, Pindoria P, Wade A. Support for breastfeeding mothers. Cochrane Database of Systematic Reviews 2002; Issue 1. Art. No.:CD001141; DOI:10.1002/14651858.CD001141.pub3.
- Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539-58.
- Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60.
- Quantifying the benefits of breastfeeding: a summary of the evidence. Washington, DC: Pan American Health Organization; 2002.
- The WHO global data bank on breastfeeding and complementary feeding. Web site: http://www.who.int/research/iycf/bfcf/bfcf.asp?menu=00 (accessed 14 February 2008).
- UNICEF Global database: breastfeeding indicators. Web site: http://www.childinfo.org/eddb/brfeed/test/database.htm (accessed 14 February 2008).
- Collaborative WHO study team on the Role of Breastfeeding on the Prevention of Infant Mortality. Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries: a pooled analysis. Lancet 2000;355:451-5.
- Newman J, Pitman T. The ultimate breastfeeding book of answers. Roseville, CA: Prima Publishing; 2000.
- Martinez JM. El primer mes del bebé. Programa de Actualización Pediátrica Continua, Módulo 3, 2005. Sociedad Argentina de Pediatría.
- Pellegrini L. Prevención y manejo de la patología del primer mes de vida. Maternidad Martin, Rosario, Argentina. 34 Congreso Nacional de Pediatría, Cordoba, Argentina, 2006.
- Edmond KM, Zandoh C, Quigley MA, Amenga-Etego S, Owusu-Agyei S, Kirkwood BR. Delayed breastfeeding initiation increases risk of neonatal mortality. Pediatrics, 2006;117(3):e380-6.
- The importance of exclusively breastfeeding for the first six months. Washington, DC: Pan American Health Organization; 3 August 2004. Press release, available at: http://www.paho.org/English/DD/PIN/pr040803.htm (accessed 14 February 2008).
- Módulo de capacitación en lactancia materna para el equipo de salud del Ministerio de Salud de la Provincia de Santa Fe. Santa Fe, Argentina: Subsecretaría de Salud y Dirección Provincial de Promoción y Protección de la Salud.
- Consejería en lactancia maternal: curso de capacitación. Geneva: World Health Organization, 1993 (WHO/CDR/93.6. UNICEF/NUT/93.4).
This document should be cited as: Pellegrini L, Sguassero Y. Support for breastfeeding mothers: RHL commentary (last revised:24 September 2007) The WHO Reproductive Health Library; Geneva: World Health Organization.