breastfeeding for 6 months versus 3-4 months withmixed breastfeeding thereafter, resulted in the recommendationto promote exclusive breastfeeding for thefirst 6 months of life [9]. More recently, the authors ofthe Lancet nutrition series published a random effectsmeta-analysis estimating the increased risk of diarrheaspecificmorbidity and mortality among childrenyounger than 2 years in relation to suboptimal breastfeedingpractices [7]. While these estimates provide confirmationof the protective effect of breastfeeding, theywere based on a limited data set, rather than a completesystematic review, and thus a more thorough andupdated revision is warranted.Building upon previous reviews, this systematic reviewand meta-analyses use carefully developed and standardizedmethods to focus on the effects of breastfeedingpractices as they relate to diarrhea incidence, prevalence,mortality and hospitalization among children 0-23months of age. Here we present a comprehensive systematicreview and meta-analysis as evidence to be utilizedby the Lives Saved Tool (LiST) to model the effectof breastfeeding practices on diarrhea-specific morbidityand mortality [10,11]. The results of our analysis willserve as the basis for generating projections of childlives that could be saved by increasing exclusive breastfeedinguntil 6 months of age and continued breastfeedinguntil 23 months of age.MethodsWe systematically reviewed all literature published from1980 to 2009 to identify studies with data assessinglevels of suboptimal breastfeeding as a risk factor fordiarrhea morbidity and mortality outcomes. We conductedour initial search on July 28, 2009 and twoupdated searches on April 8 and May 5, 2010. Allsearches were completed in Pubmed, EMBASE, the GlobalHealth Library Global Index and Regional Index,and the Cochrane central register for controlled trialsusing combinations of key search terms: breastfeeding,breast milk, human milk, diarrhea, gastroenteritis, morbidity,mortality, infant and child. To ensure the identificationof all relevant literature, we also reviewed thereferences of included papers.After initially screening for eligibility based on title andabstract, we thoroughly reviewed full publications forinclusion and exclusion criteria outlined a priori. Weincluded randomized controlled trials (RCT), cohort andobservational studies that assessed suboptimal breastfeedingas a risk factor for at least one of the followingoutcomes: diarrhea incidence, diarrhea prevalence, diarrheamortality, all-cause mortality, and diarrhea hospitalizations.Included studies were published in any languagefrom 1980 2009 and were conducted in developingcountries with a target population of children 0-23months of age. We excluded studies reporting diarrheaas a result of only one microbial cause, and those withunclear methodology or data in a form that could not beextracted for meta-analysis. We also excluded studiesreporting exclusive breastfeeding for children beyond 6months of age and those failing to restrict the allocationof diarrhea outcomes to concurrent breastfeeding status.Additionally, we excluded morbidity studies with diarrhearecall beyond two weeks and mortality studieswhere the removal of deaths occurring within the firstthree to seven days of life was not possible. For studiesreporting outcomes stratified by HIV status, we onlyabstracted data on HIV-negative infants and children.We abstracted data for each diarrhea outcome bybreastfeeding exposure levels, which were classifiedaccording to current WHO definitions (Table 1) [12,13].To allow for the comparability of breastfeeding labelsand definitions derived from studies published over multipledecades, during which time breastfeeding definitionsand terms evolved, we assigned the exposurecategories described by each study to a WHO categoryon the basis of the studys definition of that exposurecategory, not the authors category label. The majorityof discrepancies between breastfeeding label and definitionarose over the term exclusive breastfeeding. Bycurrent standards, exclusive breastfeeding does notinclude the ingestion of anything other than breastmilkand prescribed vitamins and medications, and infantsreceiving non-nutritive liquids, such as waters and teas,are classified as predominantly breastfed [12]. This distinctionwas not formally recommended until 1988when a meeting of the Interagency Group for Action onBreastfeeding first proposed the development of a set ofstandardized breastfeeding definitions [14]. WHO officiallyintegrated indicators differentiating between exclusiveand predominant breastfeeding in 1991 [12]. Assuch, for this review we assumed the exclusive breastfeedingcategory was more appropriately labelled predominantbreastfeeding for studies published prior to1991, unless the study specifically defined exclusivebreastfeeding according to the current definition.For studies that grouped exclusively and predominantlybreastfed infants into a fully breastfeeding category,we employed a conservative approach in whichfully breastfeeding exposure was treated as predominant.We excluded studies that combined exposures otherthan exclusive and predominant breastfeeding into onebreastfeeding category.In this review we did not seek to address the issue ofearly initiation of breastfeeding and prelacteal feeds.Thus, in assigning breastfeeding exposure, we did notdifferentiate between exclusive and predominant breastfeedingon the basis of receipt of prelacteal feeds duringthe first 3 days of life.Lamberti et al. BMC Public Health 2011, 11(Suppl 3):S15http://www.biomedcentral.com/1471-2458/11/S3/S15Page 2 of 12
Use the order calculator below and get started! Contact our live support team for any assistance or inquiry.
[order_calculator]