Introduction
Mother-to-child HIV transmission is the primary mode of infection for infants during pregnancy, birth, or breastfeeding.(Reference Dagnew and Teferi1) Mothers living with human immunodeficiency virus can strive always to breastfeed with negative outcomes for their health and the health of their kids.(Reference Daniel Baza and Markos2) However, the mixed feeding practices for HIV-infected mothers increase the risk of HIV transmission by 3-4-fold.(Reference Daniel Baza and Markos2) Reducing this transmission is a critical global public health challenge faced by researchers, healthcare professionals, policymakers, and HIV-infected women worldwide.(Reference Ejara, Mulualem and Gebremedhin3,Reference Genetu, Yenit and Tariku4)
Globally, around 36.7 million people, primarily in sub-Saharan Africa (71%), are living with HIV and each year, 600,000 infants globally are infected with HIV, averaging 1,600 infections per day.(Reference Mutawulira, Nakachwa, Muharabu, Wilson Walekhwa and Kayina5,Reference Zewdu, Bekele, Bantigen and Wake6) Nearly half (42.5%) were infected during pregnancy, labour, and breastfeeding, especially where mixed feeding is predominant in sub-Saharan Africa, strain of economic burden causes postnatal transmission.(Reference Daniel Baza and Markos2,Reference Ejara, Mulualem and Gebremedhin3,Reference Ekubagewargies, Mekonnen and Siyoum7)
In low- and middle-income countries (LMICs), WHO advises HIV-infected mothers on combined antiretroviral therapy (cART) to breastfeed infants for 12–24 months, supported by the heightened risks of morbidity and mortality in formula-fed babies due to infections and malnutrition.(Reference Abuogi, Noble, Smith, Committee On, Adolescent and Section On8) Limited access to clean water and the high cost of formula milk in impoverished populations underscore the importance of this recommendation.(Reference Fassinou, Songwa Nkeunang, Delvaux, Nagot and Kirakoya-Samadoulougou9,Reference Bansaccal, Van der Linden, Marot and Belkhir10) Maternal knowledge of proper newborn safe feeding procedures, including when and how to start, significantly affected the transmission rate of HIV.(Reference Zewdu, Bekele, Bantigen and Wake6,Reference Temesgen, Negesse and Getaneh11) The 2016 Ethiopian Demographic and Health Survey disclosed significant HIV-related insights, with 1.5 million new cases and 680,000 reported deaths, and notably 74% of Ethiopian women are aware of HIV transmission through breast milk.(Reference Daniel Baza and Markos2,Reference Zewdu, Bekele, Bantigen and Wake6,Reference Temesgen, Negesse and Getaneh11)
Previous systematic reviews(Reference Dagnew and Teferi1,Reference Endalamaw, Demsie, Eshetie and Habtewold12,Reference Kassa13) and primary studies(Reference Ejara, Mulualem and Gebremedhin3,Reference Genetu, Yenit and Tariku4,Reference Belay and Wubneh14–Reference Yapa, Drayne and Klein19) have highlighted key factors, including CD4 count, viral load, and ART adherence influencing the prevention of HIV transmission. The Ethiopian government promotes infant health and HIV-free survival through safe infant feeding in Option B+ care for all pregnant women.(Reference Yapa, Drayne and Klein19) However, in 2016 an Ethiopia Demographic and Health Survey (EDHS) reported, that children had low rates of dietary diversity (4.3%) and meal frequency (47.7%), with 17% practicing safe infant feeding for all HIV-exposed infants.(Reference Genetu, Yenit and Tariku4,Reference Temesgen, Negesse and Getaneh11,Reference Belay and Wubneh14,Reference Muluye, Woldeyohannes, Gizachew and Tiruneh17,Reference Astewaya, Tirhas and Tessema20,Reference Girma, Wendaferash, Shibru, Berhane, Hoelscher and Kroidl21) In Ethiopia, as of the updated guidelines in 2018, the Prevention of Mother-To-Child Transmission (PMTCT) guidelines recommend breastfeeding as the safest option for HIV-positive mothers, particularly those who have achieved high viral load suppression. However, it is important to note that the gradual introduction of mixed feeding is highly advised for lactating women in such cases.(Reference Dagnew and Teferi1,Reference Kebede and Kebede22) However, several economic and peer support challenges caused a significant risk of HIV transmission with a 3.54% to 4.23% rate practiced before 6 months of mixed feeding.(Reference Ejara, Mulualem and Gebremedhin3,Reference Zewdu, Bekele, Bantigen and Wake6)
Previous research suggests that various maternal factors impact on PMTCT to infants including maternal education (12 instances),(Reference Girma, Wendaferash, Shibru, Berhane, Hoelscher and Kroidl21) employment status (7 instances),(Reference Ndubuka, Ndubuka, Li, Marshall and Ehiri23) maternal knowledge, and attitude (5.9 to 15.2 instances),(Reference Ndubuka, Ndubuka, Li, Marshall and Ehiri23) HIV disclosure status (6.2 times) were identified as significant hindrances to achieving zero transmission.(Reference Misganaw, Naghavi and Walker25) Therefore, this study aimed to estimate levels of HIV-exposed infants’ safe feeding practices and maternal enriching factors in Northeast Ethiopia.
Methods
Study area and period
The study was conducted between April 1 and June 20, 2023, in the North Wollo zone, Amhara region, Northeast Ethiopia. The zone is centred on Woldia and is located 521 km from Addis Ababa and 360 km from the regional capital, Bahir Dar. It shares borders with the South Gondar zone in the West, the South Wollo zone in the South, the Afar region in the East, the Tigray region in the Northeast, and the Waghimra zone in the North West sides. The projected population for 2023 was estimated at 1,763,246, with 50.2% females and 13% residing in urban areas. The zone consists of 14 districts, including three town administrative areas. Healthcare facilities in the zone include 6 public hospitals, 69 health centres, 309 health posts, 10 private medium clinics, 42 primary clinics, and 33 pharmacies. Among these, five hospitals and 22 health centres provide ART services.(Reference Lang’at, Ogada and Steenbeek31)
Study design
A multi-centre, institution-based, descriptive cross-sectional study was conducted among 314 HIV-infected mothers with their dyads less than 24 months.
Source population
All mothers attending antiretroviral therapy (ART) services with infants ≤24 months under Option B+ care in selected health institutions form the sampled population.
Inclusion criteria
From April to June 2023, all HIV-infected women who had infants less than 24 months of age and were receiving treatment at the ART unit were eligible for this study.
Exclusion criteria
Mothers who were severely ill and unable to communicate their HIV status, as well as their children’s HIV status, were excluded from the study
Sample size determinations
The sample size was determined using the single population proportion formula using a 95% confidence level and a 5% margin of error, as well as the prevalence of infant feeding practice, which was found to be 75.2%. The formula used to calculate the sample size is as follows: n = (Zα/2) 2 [p (1−p)]/d². Where: n = required sample size Zα/2 = critical value for the normal distribution at a 95% confidence interval, which is equal to 1.96 p = prevalence (75.2%), d = margin of error (5%). Using the given values, the calculation for the sample size is as follows: n = (1.96)² [0.752(1-0.752)]/(0.05)², which results in n = 286. However, after accounting for a non-response rate of 10%, the sample size is adjusted by non-response rate by adding10% non-response rate as 286 + (0.10 × 286) = 314. Hence, the final sample size was found to be 314 HIV-infected mothers with their dyads were interviewed for final analysis.
Sampling procedure
In the North Wollo zone, there were 27 health facilities providing Option B+ services (PMTCT), comprising 5 public hospitals and 22 health centres. To select the sampled participants, 30% of the facilities were first randomly selected from the total 27 Option B+ services giving centres by using a lottery method. This resulted in a sample of eight health facilities for final sample size selection based on their serving-giving population over the past 3 months. In each health facility, a 3-month file of service was given population divided by our sample size, and we determined K =, then we selected a total of 314, using systematic sampling technique within each facility using the formula (K = N/n, 837/314 = 3) where N represents the total client population (837) and n denotes the required sample size (314). The final sample of 314 participants was selected using a k = 3 interval from each health facility based on their population level.
Dependent variable
This study evaluated WHO-recommended infant feeding practices (Yes/No), emphasising safety for up to 2 years. Unsafe practices include early breastfeeding substitution, introducing complementary foods at 6 months or earlier, or mixed feeding before 6 months. In LMICs, WHO recommends 12–24 months of breastfeeding for HIV-infected mothers cART, and gradual introduction of complementary feeding with stressing strict adherence to ART prophylaxis given for infants.(Reference Dagnew and Teferi1,Reference Ejara, Mulualem and Gebremedhin3,Reference Sendo, Mequanint and Sebsibie16)
Independent variable
Maternal Age, Marital status, Occupation, Educational status, Monthly income, Knowledge of vertical transmission, Disclosure of HIV status, Place of delivery, Antenatal follow-up, Stage of HIV, CD4 Count, Breast problem, Counseling practice, Sex of the child, Age of child
Operational definition
Minimum dietary diversity
The Children’s Complementary Food Dietary Diversity Score (DDS) was based on seven food groups: grains/roots/tubers, legumes/nuts, dairy products, flesh foods (meats/fish/poultry), eggs, vitamin A-rich fruits and vegetables (VAFV), and other fruits and vegetables (OFV). The DDS, ranging from 0 to 7, measured dietary diversity by assigning one point to each food group. We considered minimum dietary diversity as consuming food from at least four different groups (DDS ≥ 4).(Reference Esubalew, Atenafu and Abebe24) Exclusive breastfeeding; Exclusive breastfeeding involves giving only maternal breast milk to the infant for the first 6 months, while replacement feeding entails providing a diet with all necessary nutrients for infants not receiving any breast milk. Mixed feeding occurs when infants under 6 months are given liquids or foods alongside breast milk.(Reference Dagnew and Teferi1) Complementary feeding; HIV-exposed infants require careful feeding, following guidelines for exclusive breastfeeding for the first 6 months, followed by the introduction of complementary foods while breastfeeding up to 12 months. ART is crucial in reducing HIV transmission risk, and decisions on infant feeding should involve healthcare providers, considering specific circumstances and ongoing monitoring for the well-being of these infants during the transition to complementary feeding.(Reference Misganaw, Naghavi and Walker25,26) Mean Good knowledge; Good knowledge is defined as respondents who scored equal to or greater than the mean score of maternal safe infant feeding related knowledge question list in WHO guidelines.(26,27)
Data collection procedure and quality control
The questionnaire, initially in English, was translated into Amharic. Six data collectors and three supervisors underwent a 1-day training on study objectives, privacy, and confidentiality. Data collection used a pre-tested semi-structured questionnaire, with a pilot study conducted on a 5% sample size for adjustments. Daily supervision by supervisors and the principal investigator ensured consistency, completeness, clarity, and accuracy in the data collection process.
Data process and analysis
Collected data underwent editing, entry, and coding with EPI info v7.2.5.0 software, then analysed using SPSS v25. Results were presented through frequency tables, figures, and percentages. Maternal knowledge of infant feeding, assessed with nine structured questions, produced a mean score. Bivariate and multivariable logistic regressions were performed with independent variables having P-value<0.25 in bivariate analysis. Collinearity effects and data normality were checked, by applying a stepwise backward elimination procedure. Categorical variables with adjusted odds ratios and 95% confidence intervals determined safe infant feeding at P<0.05. Model fitness was assessed using the Hosmer–Lemeshow goodness-of-fit test. Maternal knowledge was categorised as poor or good based on the mean score, and the study questions were adapted from a previously published article.(Reference Genetu, Yenit and Tariku4,Reference Zewdu, Bekele, Bantigen and Wake6,Reference Temesgen, Negesse and Getaneh11,Reference Belay and Wubneh14,Reference Muluye, Woldeyohannes, Gizachew and Tiruneh17,Reference Kebede and Kebede22,Reference Esubalew, Atenafu and Abebe24,Reference Bekere, Garoma and Beyene28–Reference Umeobieri, Mbachu and Uzochukwu30) Cronbach’s alpha yielded reliability coefficients of 0.76 for maternal knowledge and 0.82 for infant feeding practices, signifying good internal consistency. Tables containing all maternal knowledge and safe infant feeding-related questions were incorporated into the study.(Reference Ejara, Mulualem and Gebremedhin3,Reference Mutawulira, Nakachwa, Muharabu, Wilson Walekhwa and Kayina5,Reference Belay and Wubneh14,Reference Andare, Ochola and Chege15,Reference Muluye, Woldeyohannes, Gizachew and Tiruneh17,Reference Ndubuka, Ndubuka, Li, Marshall and Ehiri23,Reference Lang’at, Ogada and Steenbeek31,Reference Naturinda, Akello and Muwonge32)
Result
Socio-demographic characteristics
The study included 314 HIV-infected women with infants from eight public health institutions, achieving a 100% response rate. The mean age for mothers and children was 32.4 years (SD±4.4) and 7.8 months (SD±2.9), respectively. Among live-birth infants, 63.7% were female, and 36.6% were male. The majority of participants (72.3%) lived with their spouses, and 38.2% had no formal education. Urban residents accounted for 60.2%, with 73.6% being housewives. The majority (90.3%) completed their fourth antenatal care (ANC) visit, while 4.6% had a history of home delivery. Additionally, 270 pregnant women had a CD4 count of ≤50 cells/mm3.
Maternal and obstetrics characteristics
Nearly half of the 165 respondents (52.5%) had fewer than two children, and the majority of the 304 participant women (96.8%) received counselling on infant feeding options. Among the mothers who gave birth, 283 (90.1%) delivered at health institutions, with 277 (88.2%) having a spontaneous vaginal delivery. During ANC care, 142 women (45.2%) became aware of their HIV status, while 139 (44.3%) knew about their pregnancy before initiating ANC. Additionally, 210 respondent mothers (66.9%) disclosed their HIV status (Table 1).
Table 1. Socio-demographic characteristics of HIV-positive mothers and their infants attending ART service
Maternal knowledge for safe infant feeding practice
Over half of 165 respondents (52.5%) had <2 children, and 96.8% of 304 women received feeding counselling. For those giving birth, 90.1% delivered at health institutions, with 88.2% having spontaneous vaginal delivery. During interviews, mothers mentioned HIV transmission: 63.37% (200) during delivery, 7.9% (24) during breastfeeding, and 32.4% (97) during breast pain, oral ulcers of infants, and mother problems (Table 2).
Table 2. Maternal knowledge of infant feeding practice among HIV-positive women attending ART service
Maternal practice for safe infant feeding
Almost all (99%) of mothers received safe infant feeding demonstrations and counselling during ANC from healthcare providers. During the interview, 203 (64.49%) of them also practiced demonstrated how breastfeeding after the discussion (Table 3).
Table 3. Maternal practicing related questions for HIV exposed infant feeding
Mothers’ status during an interview
During the study, 67.2% of mothers practiced safe infant feeding, with a mean knowledge score of 70.3%. The majority of mothers (83.3%) had a CD4 count greater than 500 cells/mm3, and 146 (46.5%) were classified as WHO clinical stage II. Among the infants, 81.2% had not experienced any oral ulcers.
Level of safe infant feeding practice
The overall prevalence of safe infant feeding practices was 67.2% (95% CI: 61.7, 72.9), whereas the remaining 32.8% of participants used mixed or unsafe infant feeding options. Among women who used unsafe infant feeding options mainly reported having breast problems with (n = 101), the most common breast problem reported was engorgement (58.4%), followed by sore nipples 19(18.8%), cracked nipples 18(17.8%), and burning or tingling 5(4.9%).
Factors affecting infant feeding practice
During the final multivariable logistic regression of this report, variables with a P-value<0.25 on bivariate analysis were considered candidates for multivariable regression. These included mother and infant age, residence, marital status, education, number of children, ANC visits, place of delivery, timing of breastfeeding, maternal knowledge score of infant feeding practices, HIV disease progression, presence of long-term illness, and infant mouth ulcers.
After controlling certain confounding factors, five variables were significantly associated with safe infant feeding during the PMTCT. These include being maternal age with 25–35 years (adjusted odd ratio (aOR) = 2.9, 95% CI: 1.2, 7.6), completing high school education (aOR = 9.2, 95% CI: 1.3, 6.8), having a good knowledge score for infant feeding (aOR = 8.2, 95% CI: 2.1, 32.7), and urban residency (aOR = 2.2, 95% CI: 1.1, 4.5) are maternal enriching factors as compared with their respective counter groups. On the other hand, infant mothers living with their spouses had an 83% reduced likelihood of safe infant feeding practices compared to those who were never in a union (aOR = 0.17, 95% CI: 0.36, 0.80) but having baby mothers (Table 4).
Table 4. Factors affecting infant feeding practices among HIV-positive mothers attending ART service
* Indicated statistical significant variables after association.
Discussion
At the end of the study periods, the overall safe infant feeding practice among mothers for their dyads was found to be 67.2%). This finding is consistent with previously reported 63.43% in Gondor Hospital,(Reference Belay and Wubneh14) 63.8% in Samra Hospital,(Reference Zewdu, Bekele, Bantigen and Wake6) 63.99% in Debre Markos,(Reference Temesgen, Negesse and Getaneh11) and 63.8% in Bahir Dar Hospital.(Reference Naturinda, Akello and Muwonge32) This might be related to healthcare providers using similar guidelines for therapeutics, and counselling principles across different study settings similar contextual factors, such as cultural norms, available resources, and healthcare policies, may have influenced safe infant feeding practices across the included healthcare facilities. Conversely, the final report of safe infant feeding practices is higher than previously found at 25.5% in Gondar Hospital,(Reference Esubalew, Atenafu and Abebe24) 49.3% in Wolaita Soda Hospital,(Reference Daniel Baza and Markos2) and 18.2% in Kenya Hospital,(Reference Andare, Ochola and Chege15) but lower than previously reported 86.4% in Gondar hospital.(Reference Genetu, Yenit and Tariku4) These differences may result from variations in maternal healthcare utilisation across different Ethiopian facilities and the discount could stem from differences in study settings, access to information, and technology, influencing community awareness. Maternal workload and limited time for childcare may contribute to these disparities.
The final report of this study indicated that the mean maternal knowledge score on infant feeding was found to be 70.3%, which is lower compared to previous reports of 86.4% and 65.3% in,(Reference Bekere, Garoma and Beyene28) 50.1% found in Botswana,(Reference Ndubuka, Ndubuka, Li, Marshall and Ehiri23) 83.6% in Nigeria.(Reference Umeobieri, Mbachu and Uzochukwu30) This indicates the existence of gaps in mother-to-child transmission and the actual practice in Ethiopia and the healthcare providers sometimes struggle to influence maternal behavioural changes in implementing recommended infant feeding choices which is an urgent need for better counselling for pregnant women on infant feeding options to eliminate transmission.
Regarding maternal enriching factors for safe infant feeding factors identified, accordingly, mothers of HIV-exposed infants within 30–35 years of age were two-fold times more likely to adopt safe infant feeding practices compared with counter groups. The findings of this study are consistent with previous findings in Gondar town,(Reference Sendo, Mequanint and Sebsibie16) SNNPR regions,(Reference Bekere, Garoma and Beyene28) Amhara region,(33) Northern Kenya,(Reference Yapa, Drayne and Klein19) and southern Nigeria.(34) The possible reasons for the similarity might be due to middle-aged mothers who had exposed infants are more likely to adhere to any recommended medical practice
Consistent with previous study findings in Gondar referral hospital,(Reference Sendo, Mequanint and Sebsibie16) North America, and Nigeria,(34) married women with HIV-exposed infants were 83% less likely to adopt the safe way of feeding practices compared to divorced women. The possible explanation may be that married women may face unsupportive partners upon disclosing their HIV status, affecting the adoption of safe feeding practices. HIV disclosing can lead to social challenges, including from husbands, friends, and community members when transitioning from exclusive breastfeeding.
The finding of the current study also indicated that getting information or counselling about safe infant feeding from the correct health professionals was significantly associated with adopting the recommended infant feeding practices that counter group. This is consistent with the study done in Addis Ababa,(35) Woldia town,(Reference Girma, Wendaferash, Shibru, Berhane, Hoelscher and Kroidl21) and Oromia regions.(Reference Umeobieri, Mbachu and Uzochukwu30) This could be because many women find that receiving skilled information on infant feeding options may not be enough for informed decision-making, and it helps them choose appropriate feeding methods, improve adherence, and opt for safer options like exclusive breastfeeding or complete avoidance of breastfeeding.(Reference Dagnew and Teferi1)
Consistent with study previous studies done in the Amhara region(33) and Addis Abeba City,(36) mothers with good knowledge about safe infant feeding are more likely to practice it. This might be related to preventing mother-to-child transmission and promoting safe infant feeding, early adoption is encouraged. HIV-positive women need customised counselling to make informed choices about feeding options based on local circumstances, ensuring optimal growth for their babies. Moreover, the findings of this study reveal that individual women with permanent urban residency were 2.2 times more likely to adopt safe infant feeding practices compared to rural dwellers. This finding aligns with previous studies conducted in Gondar town(Reference Sendo, Mequanint and Sebsibie16) and Nigeria.(Reference Fassinou, Songwa Nkeunang, Delvaux, Nagot and Kirakoya-Samadoulougou9) The possible reason is that the majority of urban residents had previous exposure, and those in urban areas tend to be more receptive to the training and guidance provided by medical professionals. They exhibit an eagerness to acquire and apply information, which may contribute to their higher adoption of safe infant feeding practices.
Limitations of the study
This cross-sectional study design limited the ability to establish a cause-and-effect relationship, and there was a possibility of recall bias as mothers were expected to remember the feeding patterns of their children since birth.
Conclusion
This study’s findings have important implications for public health interventions targeting infant feeding practices among HIV-positive mothers. The majority (67.2%) followed safe guidelines, but mixed feeding (32.8%) increased HIV transmission risk. Predictors included age, marital status, residency, access to information, and knowledge of recommended options and targeted interventions through healthcare providers give accurate information and counselling services, for mothers to make informed decisions about infant feeding.
Acknowledgements
We would like to express our deepest gratitude to all the midwives, data collectors, and hospital administrators in the North Wollo zone health institutions for their invaluable and unwavering collaboration during the data collection process. Their dedicated efforts and cooperation were instrumental in the successful collection of data for this study.
Authorship
BD contributed to conceptualisation, writing the original draft, writing review and editing, investigation, and conceptualisation. ML contributed to writing review and editing, formal analysis, and methodology. FK contributed to formal analysis, methodology, writing review and editing, data curation, software editing, and investigation. All authors have read and agreed to the published version of the manuscript.
Financial support
The author(s) declare there is no financial support from any organisation for this research
Competing interests
No any conflict of interest for this manuscript
Ethical consideration
The study was conducted in compliance with the Declaration of Helsinki, adhering to relevant guidelines. The Institutional Review Board of Woldia University College of Health Science approved the study after a thorough review of the procedures, objectives, and public health considerations (Ethical Approval Assigned No. 095, dated 17/5/2023). Official letters were obtained from the Zone Health Department, and the significance and objectives of the study were communicated to the selected health facilities. Informed consent was obtained from all participants, ensuring their voluntary participation. Confidentiality was strictly maintained during data collection and throughout all stages of the study.
Open access
