About 90% of pediatrics HIV infection is Mother-To-Child Transmission (MTCT), which may occur during pregnancy, delivery, and breastfeeding [
Breastfeeding especially in the first 12 months of life can significantly prevent malnutrition, infectious diseases, and mortality compared to nonbreastfed infants [
According to the World Health Organization (WHO) guideline, breastfeeding, especially early initiation, and exclusive breastfeeding were the most critical factors in improving child survival. Moreover, the WHO’s global public health direction for all infants is to breastfeed exclusively (EBF) for the first six months and then introduce nutritionally adequate and safe complementary foods while breastfeeding continues for up to 2 years of age or beyond [
Therefore, the decision on infant feeding practice in the era of HIV is a big challenge for caretakers and health care providers [
In Ethiopia, studies on infant feeding practice and its predictors among HIV positive mothers have been conducted in different parts of the country with a time variation. But, the results of these studies were inconsistent. Therefore, the aim of this systemic review and meta-analysis is to estimate the national pooled prevalence of exclusive breastfeeding and mixed feeding practices among HIV positive mothers and of its association with counseling on feeding option and HIV status disclosure to the spouse in Ethiopia.
The protocol of this systematic review and meta-analysis has been registered in the International prospective register of systematic review and meta-analysis (Prospero) with a registration number of
For reporting of the findings, the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guideline was employed (Additional file
All available articles were searched in PubMed, Google Scholar, EMBASE, and the Web of Science. Moreover, studies were searched from the reference lists of included studies and the Ethiopian institutional research repositories. Searching was employed using the following searching terms including “infant”, “neonate”, “child”, “infant feedings”, “Infant feeding practices”, “child feeding practices”, “exclusive breastfeeding practices”, “mixed feeding practices”, “HIV”, “Human Immune Deficiency Virus”, “Mother”, “HIV positive mothers”, “Human Immune Deficiency Virus Positive mothers”, “HIV infected mothers”, “HIV exposed infants”, “HIV infected infants”, “predictors”, “factors”, “ barriers”, “risk factors”, “prevalence”, and “Ethiopia”. The search string was developed using “AND” and “OR” Boolean operators. The searching was done from September 10 to October 28/2018. As for the search of articles in PubMed, we used this searching string (Additional file
Studies which met the criteria, (1) observational studies including cross-sectional, case-control, and cohort studies, (2) studies that report the prevalence and/or predictors of infant EBF and mixed feeding practices among HIV positive mothers, (3) studies done in Ethiopia, (4) published and unpublished studies at any time, (5) studies that have been written in English language, and 6 studies that reported extractable data to compute the odds ratio of counseling and HIV disclosure status to the spouse, were included in this study. Abstracts, studies without full texts, conference papers, editorials, letters, protocols, program evaluation reports, systematic reviews, trials, and qualitative studies were excluded. Additionally, studies with a high risk of bias/scored less 50% of critical appraisal checklist were excluded.
The infant receives only breast milk without any other liquids or solids, not even water, except for oral rehydration solution or drops or syrups of vitamins, minerals, or medicines in the first 6 months [
It is providing other liquids and/or foods together with breast milk for the infant within six months of age. This could be water, other types of milk, or any type of solid food [
In the beginning, all studies retrieved from the databases were imported to EndNote version 7 citation manager. Next, duplicates were checked and removed. Then, two independent authors (GM and CA) screened the title and the abstracts of retrieved articles. The discrepancies between the authors were solved by discussion and consensus. Consequently, two authors (CA and GM) reviewed the full texts and extracted the data. The first author of the article, year of publication, study area, design, population, region, sample size, extractable data that helps to compute odds ratio of counseling and HIV disclosure status to the spouse, prevalence of EBF, and mixed feeding practices were extracted. Any disagreement between investigators was solved by discussion and repeating of the procedures. Consequently, the data were exported to excel spreadsheets for further analysis.
Two independent authors (GM and CA) assessed the quality of the studies using the Joanna Briggs Institute (JBI) quality appraisal checklist [
To estimate the national pooled prevalence of EBF and mixed feeding practices among HIV positive mothers, a weighted inverse variance random effects model [
First, the national pooled prevalence of EBF and mixed feeding among HIV positive mothers were performed separately. Second, the regional pooled prevalence of EBF and mixed feeding practices were done. Third, the pooled OR of counseling and HIV disclosure status to the spouse was performed separately. To compute OR, primarily, number of EBF and mixed feeding practices among counseled and noncounseled mothers were extracted from the cross tabulation of the included studies. In addition, the number of EBF and mixed feeding practices were also extracted from mothers who disclose/not disclose their status to the spouse. Then, the extracted data was exported from excel spreadsheet to Stata version 11 to analyze the pooled OR of counseling and HIV disclosure status to the spouse. Finally, the Preferred Reporting Items of Systematic Review and Meta-Analysis (PRISMA guideline) was used to report the findings of this systematic review and meta-analysis [
A total of 3264 articles were retrieved from different databases of which 3177 were from PubMed, 60 from Google Scholar, 12 from the Web of Science, 4 from EMBASES, 6 from the Ethiopian university research repositories, and 5 from reference lists of the included studies. However, 79 articles were removed due to duplicates, 3129 due to the irrelevant titles and abstracts, 10 due to the study designs, and 20 due to study areas. Finally, 18 articles were included in this systematic review and meta-analysis (Figure
A PRISMA flow diagram of articles screening and process of selection.
In this study, a total of 18 studies [
General characteristics of included studies that reported the prevalence and/predictors of exclusive breast feeding among HIV positive mothers (n=18).
Author/year of publication | Region | Study design | Study area | Populations | Sample size | No. of outcome | Prevalence (%) | Quality |
---|---|---|---|---|---|---|---|---|
Maru y et al./2009 [ | Addis Ababa | Cross-sectional | Addis Ababa | Young infant | 327 | 100 | 30.6 | Low risk |
Esubalew F et al./2018 [ | Amhara | Cross-sectional | Gondar Hospital | 6-18 months | 420 | 107 | 25.5 | Low risk |
Wakwoya EB et al./2016 [ | Amhara | cross-sectional | Debre Markos Hospital | < 2 yrs. | 260 | 201 | 77.3 | Low risk |
Girma Y et al./2014 [ | Tigray | Cross-sectional | Mekelle Town | < 2 yrs. | 207 | 187 | 90 | Low risk |
Ejara D et al./2018 [ | Oromia | Cross-sectional | Bishoftu towns | <18 months | 283 | 242 | 85.5 | Low risk |
Demssie DB et al./2016 [ | Oromia | Cross-sectional | Shashemene Referral Hospital | <2 yrs. | 146 | 141 | 96.6 | Low risk |
Modjo KE et al./2015 [ | SNNPRS | Cross-sectional | SNNPRS Hospital | <2 yrs. | 436 | 210 | 48.2 | Low risk |
Wondie T et al./2012 [ | Addis Ababa | Cross-sectional | Addis Ababa | <12 months | 116 | 65 | 56 | Low risk |
Demlew MZ et al./2014 [ | Addis Ababa | Cross-sectional | Addis Ababa | <2 yrs. | 356 | 240 | 68 | Low risk |
Muluy D et al./2012 [ | Amhara | Cross-sectional | Gondar health center | <2 yrs. | 209 | 175 | 83.8 | Low risk |
Bekere A et al./2014 [ | Oromia | Cross-sectional | West Oromia | 0-6 months | 118 | 85 | 72 | Low risk |
G/Hiwot A et al./2014 [ | Tigray | Cross-sectional | central zone | <2 yrs. | 219 | 198 | 90.4 | Low risk |
Ali Y/2015 [ | Amhara | Cross-sectional | Wollo Zone | 2-11 months | 373 | 279 | 74.8 | Low risk |
Hailu C/2005 [ | Oromia | Cross-sectional | Jima town | 12 months | 643 | 86 | 13.4 | Low risk |
Tadese F/2017 [ | Amhara | Cross-sectional | Bahir Dar town | 12 months | 230 | 173 | 75.2 | Low risk |
Mengstie A et al./2015 [ | SNNPRS | Cross-sectional | SNNPRS hospital | <2 yrs. | 87 | 49 | 56.3 | Low risk |
Mebratu T/2014 [ | Addis Ababa | Cross-sectional | Addis Ababa city | 12 months | 221 | 30 | 13.4 | Low risk |
Ketema Z/2016 [ | Oromia | Cross-sectional | Adama health facility | 12 months | 193 | 164 | 84.7 | Low risk |
All studies [
Publication bias was not observed as the funnel plot was symmetrical with visual inspection, and Egger’s test value was 0.32 (Figure
Funnel plot for publication bias, logp, or LNP(log of proportion) represented in the x-axis and standard error of log proportion in the y-axis.
On the whole, 18 studies [
Forest plot of the prevalence EBF with 95% CI. The midpoint and the length of each segment showed the prevalence and 95% CI, respectively. The diamond shape showed the combined prevalence of each region.
In the subgroup analysis by the region, the highest prevalence of EBF practices was observed in Tigray 90.12% (95% CI: 87.39, 93.03), and the lowest in Addis Ababa 41.92% (95% CI: 15.50, 68.34). Moreover, the prevalence of EBF practices in other regions includes Oromia (70.43%), Amhara (67.29%), and SNNPRS (50.88%). The detailed descriptions were illustrated in Figure
Out of 18 studies, fifteen [
General characteristics of included studies that reported the prevalence and/predictors of mixed feeding practices among HIV positive mothers (n=15).
Author/year of publication | Study area | Region | Study design | Population | Sample size | Number of outcome | Prevalence (%) | Quality |
---|---|---|---|---|---|---|---|---|
Maru y et al./2009 [ | Addis Ababa | Addis Ababa | Cross-sectional | Young infant | 327 | 50 | 15.30 | Low risk |
Wakwoya EB et al./2016 [ | Debre Markos Hospital | Amhara | Cross-sectional | < 2 yrs. | 260 | 37 | 14.20 | Low risk |
Girma Y et al./2014 [ | Mekelle Town | Tigray | Cross-sectional | < 2 yrs. | 207 | 13 | 6.30 | Low risk |
Ejara D et al./2018 [ | Bishoftu town | Oromia | Cross-sectional | <18 months | 283 | 23 | 8.30 | Low risk |
Demssie DB et al./2016 [ | Shashemene Referral Hospital | Oromia | Cross-sectional | <2 yrs. | 146 | 1 | 0.70 | Low risk |
Modjo KE et al./2015 [ | SNNPRS Hospital | SNNPRS | Cross-sectional | <2 yrs. | 436 | 151 | 34.60 | Low risk |
Muluy D et al./2012 [ | Gondar H/c | Amhara | Cross-sectional | <2 yrs. | 209 | 22 | 10.50 | Low risk |
Bekere A et al./2014 [ | West Oromia | Oromia | Cross-sectional | 0-6 months | 118 | 29 | 24.60 | Low risk |
G/Hiwot A et al./2014 [ | Central zone | Tigray | Cross-sectional | <2 yrs. | 219 | 13 | 5.90 | Low risk |
Ali Y/2015 [ | Wollo Zone | Amhara | Cross-sectional | 2-11 months | 373 | 44 | 11.80 | Low risk |
Hailu C/2005 [ | Jima town | Oromia | Cross-sectional | <12 months | 643 | 521 | 81.00 | Low risk |
Tadese F/2017 [ | Bahir Dar town | Amhara | Cross-sectional | <12 months | 230 | 25 | 10.90 | Low risk |
Mengstie A et al./2015 [ | SNNPRS hospital | SNNPRS | Cross-sectional | <2 yrs. | 87 | 31 | 35.60 | Low risk |
Mebratu T/2014 [ | Addis Ababa city | Addis Ababa | Cross-sectional | 12 months | 221 | 180 | 81.50 | Low risk |
Ketema Z/2016 [ | Adama health facility | Oromia | Cross-sectional | 12 months | 193 | 12 | 6.30 | Low risk |
In this study, the overall pooled prevalence of mixed feeding practices among HIV positive mothers was 23.11% (95% CI: 10.10, 36.13) (Figure
Forest plot of the prevalence of mixed feeding by region with 95% CI. The midpoint and the length of each segment revealed the prevalence and 95% CI of each study whereas the diamond shape showed the combined prevalence.
In the subgroup analysis by the region, the highest pooled prevalence of mixed feeding was observed in Addis Ababa (48.38%), followed by SNNPRS (34.76%), Oromia (24.17%), Amhara (11.81%), and Tigray (6.09%) (Figure
The studies reported different predictors which significantly associated with infant feeding practices of HIV positive mothers of which, counseling on feeding options is the most frequently reported predictors.
In this systematic review and meta-analysis, six studies [
Forest plot of odds ratio of counseling with 95%CI. The midpoint and the length of each segment revealed OR and 95%CI. The diamond shape showed pooled OR.
Five studies [
Forest plot of odds ratio of HIV disclosure status to the spouse with 95% CI. The midpoint and the length of each segment showed OR and 95% CI, respectively. The diamond shape revealed pooled OR.
This systematic review and meta-analysis aimed to estimate the national pooled prevalence of exclusive breastfeeding and mixed feeding practices and of its association with counseling and HIV disclosure status to the spouse. The overall pooled prevalence of EBF and mixed feeding practices among HIV positive mothers were 63.43% and 23.11%, respectively. Counseling on feeding options and HIV disclosure status to the spouse were significantly associated with exclusive breast feedings practices.
Regarding the prevalence of EBF practices, the result is in line with the study conducted in Kenya (57.7%) [
On the other hand, parents from other countries perceived that breast milk alone is not an enough source of nutrition. For instance, in South Africa, there was a belief that breast milk is not enough for their infant [
Concerning mixed feeding practices, the result was in line with the study done in Nigeria (13%) [
Subgroup analysis revealed that there was a significant variation among regions. In subgroup analysis, the lowest prevalence of EBF practices was observed in Addis Ababa. This variation could be (1) low fear of discrimination in Addis Ababa compared to other regions. According to the 2016 EDHS report, in Addis Ababa there was low fear of discrimination (18%) [
Concerning predictors, in this systematic review and meta-analysis, HIV positive mothers who have been counseled on feeding options were exclusively breastfeeding their infants nearly four times more likely as compared to none counseled mothers. This finding was in line with the study done in Nigeria [
This was due to the fact that counseled mothers on feeding options would have higher knowledge and awareness on feeding options and PMTCT compared to noncounseled mother [
Another factor that was significantly associated with EBF practice was HIV disclosure to the spouse. Even though there are a number of predictors of EBF practices, HIV status disclosure to the spouse is prominent. In this study, we found that the prevalence of exclusive breastfeeding practices among HIV positive mothers who disclose their status to the spouse was nearly six times more likely than mothers who did not disclose their status. This finding was in agreement with the study conducted in Nigeria [
Most mothers report exclusive breastfeeding, but there are still almost a quarter of mothers who mix feed. Counseling on feeding options and HIV disclosure to the spouse were significantly associated with exclusive breastfeeding practices. Therefore, counseling on feeding options and HIV disclosure status to the spouse should be strengthened.
Even though the this systematic review and meta-analysis aimed to estimate the national prevalence of EBF and mixed feeding practices and its association with counseling and HIV disclosure status, it is not representative for all regions as data was not found in Afar, Somali, Benshangul-Gumuz, Gambella, and Harari. This study also did not determine other factors.
Acquired Immunodeficiency Syndrome
Antenatal Care
Adjusted Odds Ratio
Antiretroviral Therapy
Exclusive Breastfeeding
Human Immune Deficiency Virus
Mixed Feeding
Prevention of Mother-to-Child Transmission
Southern Nations, Nationalities and Peoples Region State.
All data generated or analyzed during study are included in this systematic review and meta-analysis.
The authors declare that they have no conflicts of interest.
Getaneh Mulualem Belay conceived and designed the study. Getaneh Mulualem Belay and Chalachew Adugna Wubneh established the search strategy, extracted the data, assessed the quality of included studies, did the analysis, and finally wrote the review. All authors had prepared the manuscript. Finally, the authors read, modified, and agreed on the final prepared manuscript.
Our special gratitude goes to the authors of included studies who helped us to do this systematic review and meta-analysis.