Higher breastfeeding performance index is associated with lower risk of illness in infants under six months in Ethiopia
© Haile and Biadgilign. 2015
Received: 21 March 2015
Accepted: 23 November 2015
Published: 27 November 2015
Breastfeeding performance index is an explanatory attempt to summarize key breastfeeding practices by summarizing the different dimensions of breastfeeding practices into a single summary variable. Breastfeeding performance index is used to assess optimal breastfeeding practices by constructing a single composite index that includes timely initiation of breastfeeding, prelacteal feeding, current breastfeeding status, bottle feeding, any liquid given(except medicine) in the last 24 h, formula given in the last 24 h, any solid food given in the last 24 h. This study aimed to assess optimal breastfeeding practices of 0–6 month infants using breastfeeding performance index (BPI) and its association with childhood illness in Ethiopia.
A secondary data analysis was carried out based on the Ethiopia Demographic and Health Survey (EDHS) 2011 data. The BPI was created using seven components by giving equal weight for all components during scoring. The data were described using descriptive statistics and analyzed using multivariable logistic regression.
The prevalence of low, medium, and high BPI was 18.41, 57.96 and 23.63 % respectively. The mean BPI score was 4.38 (SD 1.25; 95 % CI 4.31, 4.45). Multivariable logistic regression analysis showed that those infants who had low BPI score were 2.22 times (AOR = 2.22; 95 % CI 1.20, 4.11) and medium BPI category had 2.15 times at higher odds (AOR = 2.15; 95 % CI 1.23, 3.75) of developing diarrhea compared to infants in the highest BPI category. Being in the lower BPI category was significantly associated with higher odds of having fever (AOR = 1.73; 95 % CI 1.06, 2.80). Being in the medium index category was also associated with higher odds of having short and rapid breaths (AOR = 2.02; 95 % CI 1.01, 4.04).
More than 80 % of the infants did not receive optimal breastfeeding practices based on the Breastfeeding Performance Index. Lower BPI was statistically associated with diarrhea, fever and short and rapid breaths illness in the last 2 weeks. This study implicates the importance of optimal breastfeeding to reduce childhood illness.
Optimal breastfeeding practices rank among the most effective interventions to improve child health . Studies showed that globally 60 % of infant and young child deaths attributed to lack of optimal breastfeeding practices, of which more than 13 % of under five deaths can be prevented by promoting exclusive breastfeeding [1, 2]. Lack of optimal breastfeeding practice is responsible for 45 % of neonatal infectious deaths, 30 % of diarrheal deaths and 18 % of acute respiratory deaths in children under 5 years . In Ethiopia, about 18 % of the infant deaths have been attributed to poor feeding practices .
Annually, it is estimated that optimal breastfeeding of children under 2 years of age has the potential to prevent 1.4 million deaths in children under five of age in the developing world . Optimal breastfeeding practices especially for infants < 6 months of age is not a single practice rather it constitutes several dimensions including timely initiation, colostrum feeding, exclusiveness, and whether or not the infant is given prelacteal feeds . This multidimensional nature of optimal breastfeeding creates difficulty to assess its effect on health and nutrition of the infants. Measuring the effect of optimal breastfeeding for infants less than 6 months of age has been a great challenge because of absence of a single summary index that measures all the different breastfeeding recommendations of infants less than 6 months all at a time. For this reason in the ground, it is clearly need to develop composite indicators that can measure breastfeeding practices. Breastfeeding performance index (BPI) is an explanatory attempt to summarize key breastfeeding practices into single summary variable by summarizing the different dimensions of breastfeeding practices. According to study done by Senarath showed that creating a composite index to assess the overall breastfeeding performance among infants less than 6 months of age is feasible and applicable . BPI can be effectively used to identify target groups for breastfeeding promotion interventions. The index could also be useful in establishing the association between breastfeeding and infant morbidity and evaluating the overall effect of interventions to promote breastfeeding. However, a detailed examination of the individual infant feeding indicators in the subpopulations with low BPI would still be needed when designing a public health interventions . In addition, most previous researches on association between breastfeeding practices and health outcomes of infants have only considered the individual breastfeeding indicators like exclusive breastfeeding and bottle feeding practices.
Although there are studies about optimal breastfeeding practices in Ethiopia, there is no study which document optimal breastfeeding practices using composite summary index. This study aimed to assess optimal breastfeeding practices of infants less than 6 months of age using BPI and its association with childhood illness using 2011 Ethiopian Demographic Health Survey (EDHS) data.
Study setting and design
The sample for the 2011 EDHS was designed to provide population and health indicators at the national (urban and rural) and regional levels. The 2011 EDHS sample was selected using a stratified, two-stage cluster design and enumeration areas (EAs) were the sampling units for the first stage. A representative sample of 11,565 households was selected for the 2011 EDHS. However the analysis for this study was done for infants aged 0–6 months only. This is a quantitative cross-sectional study based on secondary data of Ethiopian DHS 2011. The detailed sampling methodology is presented in the DHS sampling of DHS country report .
The Ethiopian DHS 2011 data were downloaded from Measure DHS website in SPSS format. After understanding the detail data, further coding of the data was done by the investigators. Data on a total of 1204 infants were included in the analysis and information on a wide-range of potential variables such as socio-demographic characteristics, economic variables, breastfeeding practices and childhood illness were extracted.
Variable definition and measurement
Timely initiation of breastfeeding (≤1 h or > 1 h) was measured by asking the mother to recall when she had put her infants to her breast after delivery. Presence of prelacteal feeding (yes or no) was assessed by asking the mother to recall whether the indexed infant had taken anything other than breast milk within the third days of birth for some perceived benefit or cultural reasons. Bottle feeding practice (yes or no) was assessed by 24 h recall method whether the infant received something by bottle in the last 24 h. Current breastfeeding (yes or no) assessed by asking the mother to recall whether the infant breastfed in the last 24 h or not. Solid intake (yes or no), liquid intake (yes or no), formula intake (yes or no) was assessed by asking the mother to recall whether the infant received either of them in the last 24 h. Childhood illness i.e., diarrhea, cough, short and rapid breaths and fever (yes or no) of the infants was assessed by asking the mother to recall if the infant had any type of illness in the last 2 weeks preceding the survey.
Breastfeeding performance index (BPI)
Breastfeeding performance index was constructed using seven items including timely initiation of breastfeeding, prelacteal feeding, current breastfeeding status, bottle feeding, any liquid given(except medicine) in the last 24 h, formula given in the last 24 h, any solid food given in the last 24 h. All inappropriate breastfeeding practices were scored as zero while recommended breastfeeding practices were scored as 1. The index was constructed using equal weight method as recommended by Senarath et al. for simplicity and understanding. Senarath et al. found that the index constructed using equal weight method and principal components analysis were highly correlated . All the breastfeeding practices weigh equal value because this study considers all the feeding practices are equally important. There is currently neither standard guidance nor an empirical basis for weighting the various dimensions of infant and child feeding relative to one another. In the absence of such guidance, we gave equal weight to each component. The scores of each individual practice included in the index construction are then summed to give a possible range of 0–7 and the summed score divided in to three categories. The classification was based on a value recommendation of Arimond and Ruel on feeding index construction to minimize the differences between actual percentages and 33 %; when a choice is necessary, lump into the middle category so as not to dilute the contrast between the extremes . The bottom 33 % of infants was referred to as ‘low BPI’, the next 34 % as ‘medium BPI’, and the top 33 % ‘high BPI. The BPI was divided into 3 categories (tertiles) in the following manner: a sum score of 0–3 categorized as low BPI, sum scores of 4–5 categorized as medium BPI, and sum scores of 6–7 were classified as high BPI. Those infants who fall in high BPI score were considered as optimally breastfed while infants with low and medium BPI score were considered as sub-optimally breastfed.
Data were analyzed using STATA version 12. The “svy” command in STATA version 12 was used to weight the survey data. Descriptive statistics (mean, SD, percentages) were used to characterize the socio-demographic and childhood illness variables. Binary and multivariable logistic regression models were employed to determine the association between breastfeeding performance index and childhood illness. Variables which had statistically significant association in bivariate logistic regression at p value < 0.05 with childhood illness were considered to enter in the final multivariable model with BPI to control for their confounding effect on the association between BPI and childhood illness. Crude Odds Ratios (COR) and Adjusted Odds Ratios (AOR) were presented with 95 % confidence intervals. The association of BPI with childhood illness was declared significant at p value < 0.05 in the final multivariable logistic regression model.
The data were downloaded and used after the purpose of the analysis was communicated and permission was taken from Measure DHS Organization. The original DHS data were collected in confirmation with international and national ethical guidelines.
Characteristics of the respondents
Characteristics of mothers having infants 0–6 months and their infants in Ethiopia: EDHS 2011
Place of residence
Educational status of the mother
Sex of the infant
Place of delivery
Mode of delivery
Four times and above
Maternal age category
Size of the child at birth
Larger than average
Smaller than average
Had fever in last 2 weeks
Had cough in last 2 weeks
Problem in the chest or blocked or running nosea
Short and rapid breathsa
Diarrhea in the last 2 weeks
As shown on Table 1, about 17.62 % of infants had fever while 21.43 % the infants had cough in the last 2 weeks preceding the survey. Among those who had cough in the last 2 weeks preceding the survey, 26.34 % had chest problem, 46.66 % had nose problem and 25.56 % had both chest and nose problem. The prevalence of diarrheal diseases in the last 2 weeks preceding the survey was 10.44 %.
Breastfeeding performance index
Feeding practices among infants aged 0–6 months and breastfeeding performance index (BPI) scoring system, Ethiopia, 2011
Timely initiation of breastfeeding
Pre lacteal feeding
Formula milk/ other milka
Solid foods givena
In the binary logistic regression model, BPI was significantly associated with diarrhea, fever and short and rapid breaths illness in the last 2 weeks preceding the survey. However, cough symptom in the last 2 weeks was not significantly associated with BPI.
Association between breastfeeding performance index and childhood illness status of 0–6 months infants in Ethiopia, 2011
Short and rapid breaths
Cough in the last 2 weeks
Fever in the last weeks
COR (95 % CI)
AORa (95 % CI)
OR (95 % CI)
OR (95 % CI)
COR (95 % CI)
AORb (95 % CI)
2.25 (1.22, 4.16)*
2.22 (1.20, 4.11)*
1.03 (0.49, 2.18)
1.44 (0.95, 2.20)
1.79 (1.11, 2.89)*
2.18 (1.25, 3.79)*
2.15 (1.23, 3.75)*
2.02 (1.01, 4.04)*
1.01 (0.70, 1.47)
1.49 (0.97, 2.29)
1.44 (0.94, 2.21)
The overall breastfeeding performance of mothers having infants less than 6 months of life was assessed by using breastfeeding performance index which comprises seven components. The prevalence of low, medium, and high BPI was 18.41, 57.96 and 23.63 % respectively. A study from Egypt reported that the proportions of infants belonging to low, average and high BPI categories were 27.1, 41.7 and 31.3 %, respectively . The prevalence of low BPI category of this study is low as compared to the Egypt study. But in this study the prevalence of medium BPI category was high as compared to Egypt study. On the other hand, the prevalence of high BPI category was low as compared with the Egyptian study. This difference might be attributed to the difference in socio-cultural matter of childcare between the two countries.
According to this index the prevalence of optimal breastfeeding (high BPI) was 23.63 %. A recent study done in Ethiopia has documented that the prevalence of optimal breastfeeding as 24.6 % . The mean score of the BPI was found 4.38 (±1.25) which is similar with the study done in Timor-Leste with mean BPI score of 4.4. In this study when we compare the prevalence of individual recommended components of BPI with the prevalence of high BPI, the prevalence of recommended individual practices of the index were higher. This could be because mothers do not practice all the recommended practices at the same level. This implies that using only one breastfeeding practices aspect cannot indicate the whole practice so that optimal breastfeeding could not be assessed by one dimension of breastfeeding only in infants less than 6 months of age. This study also showed that as age approaches to 6 months the overall BPI score deceases. This is because as age increases, it is expected that a large proportion of infants were no longer exclusively breastfed and/or receiving complementary feeding with bottle. This resulted significant lowering of the overall BPI score when the age increases.
Lower breastfeeding performance index was significantly associated with higher odds of having diarrhea. Those infants who found in the medium BPI category were 2.15 times at higher odds to have diarrhea illness as compared to those infants who found in the highest category. There were studies which showed that breastfeeding is associated with lower incidence of diarrhea [12–15]. A similar study conducted in Timor-Leste showed that high 2-week period prevalence of diarrhea was found in low BPI group as compared to the average and the high BPI group . The odds across feeding groups showed a steadily increasing risk of diarrhea as the relative amount of breast milk decreased . A pooled relative risk estimate from 23 studies showed that breastfeeding reduce the risk of diarrhea among infants ≤ 6 months . This could be due to the immunological, hygienical and nutritional advantages of breastfeeding.
Breastfeeding performance index was also associated with lower fever symptom in the last 2 weeks preceding the survey. Being in the lower BPI category was 73 % at higher odds to have fever in the last 2 weeks. BPI was marginally associated with having a symptom of short and rapid breaths. Consistent finding has been also reported by other studies. Breastfeeding reduced the risk of hospitalization for respiratory infection by 57 % for infants ≤ 6 months . Studies showed that breastfeeding was associated with lower incidence of respiratory tract infection [13, 14, 18] and lower BPI was associated with 2 week prevalence of acute respiratory infection in Timor-Leste .
Public health implication
Breastfeeding has the single largest potential impact on child mortality of any preventive intervention like water and sanitation . Breastfeeding is an effective child health intervention that does not require extensive health system infrastructure. Improvements in rates of exclusive and continued breastfeeding can contribute to the reduction of child mortality inequalities in developing countries . The findings of this study support the hypothesis that good breastfeeding performance protects against infant morbidity. This implicates the importance of improving all dimension of breastfeeding in the first 6 months of life for preventing infants from illness. Along with this, it is clearly depicted that BPI may be useful to identify target groups for breastfeeding promotion interventions. The index could also be valuable in establishing the association between breastfeeding and infant morbidity and evaluating the overall effect of interventions to promote optimal breastfeeding.
The study has two basic limitations. The first limitation is that the study could not able to whether the morbidity results to score lower BPI or the morbidity comes because of the lower BPI. Recall bias and the secondary nature of data sources might impose some risk of bias for the actual estimation of the outcome of interest.
More than 80 % of the infants did not experience optimal breastfeeding practice based on breastfeeding performance index. BPI was statistically associated with diarrhea, fever and short and rapid breaths illness in the last 2 weeks. This study implicates the importance of optimal breastfeeding to reduce childhood illness.
The authors acknowledge measure DHS for granting access to Ethiopian DHS 2011 data.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Jones G, Steketee R, Black R, Bhutta ZA, Morris S, Bellagio/Child/Survival/Study/Group. How many child deaths can weprevent this year? Lancet. 2003;362(9377):65–71.View ArticlePubMedGoogle Scholar
- WHO. Global strategy for infant and young child feeding, Geneva, Switzerland: A joint WHO/UNICEF statement. Geneva: WHO; 2003. http://www.who.int/nutrition/publications/gs_infant_feeding_text_eng.pdf.Google Scholar
- WHO. Global health risks: mortality and burden of disease attributable to selected major risks. 2009. http://www.who.int/healthinfo/global_burden_disease/GlobalHealthRisks_report_Front.pdf.Google Scholar
- Federal Ministry of Health, Ethiopian PROFILES and AED(Academy for Educational Development). Why nutrition matters? PowerPoint presentation on Ethiopian PROFILES analysis, Ethiopia PROFILES, Addis Ababa. Washington DC: Academy for Educational Development and LINKAGES project; 2005.Google Scholar
- The Lancet Maternal and Child Nutrition. Executive summary of the lancet maternal and child nutrition series. 2008. http://siteresources.worldbank.org/NUTRITION/News%20and%20Events/21616235/LancetNutritionProsperityExecSumENG.pdf.Google Scholar
- Federal Ministry of Health(FMOH). National strategy for infant and child feeding, Addiss Ababa Ethiopia. 2004. http://motherchildnutrition.org/nutrition-protection-promotion/pdf/mcn-national-strategy-for-infant-and-young-child-feeding-ethiopia.pdf.Google Scholar
- Senarath U, Dibley M, Agho K. Breast-feeding performance index: a composite index to describe overall breast-feeding performance among infants under 6 months of age. Public Health Nutr. 2007;10(10):996–1004.View ArticlePubMedGoogle Scholar
- Central Statistical Agency (CSA). Ethiopia: demographic and health survey 2011. Addis Ababa, Ethiopia and Calverton. Maryland: CSA and ORC Macro; 2011.Google Scholar
- Arimond M, Ruel M. Progress in developing an infant and child feeding index: an example using the Ethiopia demographic and health survey 2000, FCND discussion paper no14. Washington, DC: International Food Policy Research Institute (IFPRI); 2002. http://ageconsearch.umn.edu/bitstream/16388/1/fc030143.pdf.Google Scholar
- El-Gilany A-H, Badawy K, Kelimeler A, Yazar S. Breastfeeding performance index at Age of 6 months in Mansoura, Egypt. TAF Prev Med Bull. 2013;12(3):225–30.Google Scholar
- Tamiru D, Belachew T, Loha E, Mohammed S. Sub-optimal breastfeeding of infants during the first 6 months and associated factors in rural communities of Jimma Arjo Woreda, Southwest Ethiopia. BMC Public Health. 2012;12:363.PubMed CentralView ArticlePubMedGoogle Scholar
- Diallo F, Bell L, Moutquin J, Garant M. The effects of exclusive versus non-exclusive breastfeeding on specific infant morbidities in Conakry. Pan Afr Med J. 2009;3(2):2.Google Scholar
- Story L, Parish T. Breastfeeding helps to prevent two major infant illnesses. Internet J Allied Health Sci Prac. 2008;6(3):2–5.Google Scholar
- Lamberti LM, Walker CLF, Noiman A, Victora C, Black RE. Breastfeeding and the risk for diarrhea morbidity and mortality. BMC Public Health. 2011;11 suppl 3:S15.PubMed CentralView ArticlePubMedGoogle Scholar
- Koyanagi A, Humphrey JH, Moulton L, Ntozini R, Mutasa K, Iliff P, et al. Effect of early exclusive breastfeeding on morbidity among infants born to HIV-negative mothers in Zimbabwe. Am J Clin Nutr. 2009;89(5):1375–82.View ArticlePubMedGoogle Scholar
- Scariati PD, Grummer-Strawn LM, Fein SB. A longitudinal analysis of infant morbidity and the extent of breastfeeding in the United States. Pediatrics. 1997;99(6), e5.View ArticlePubMedGoogle Scholar
- World Health Organization, Horta BL, Victora CG. A systematic review on the benefits of breastfeeding on diarrhoea and pneumonia mortality. 2013. http://apps.who.int/iris/bitstream/10665/95585/1/9789241506120_eng.pdf.Google Scholar
- Oddy WH, Sly P, DeKlerk N, Landau L, Kendal GE, Holt PG, et al. Breastfeeding and respiratory morbidity in infancy: a birth cohort study. Arch Dis Child. 2003;88:224–8.PubMed CentralView ArticlePubMedGoogle Scholar
- Roberts TJ, Carnahan E, Gakidou E. Can breastfeeding promote child health equity? a comprehensive analysis of breastfeeding patterns across the developing world and what we can learn from them. BMC Med. 2013;11:254.PubMed CentralView ArticlePubMedGoogle Scholar