Determinants of early days of newborn feeding malpractice among mothers of children less than one years of age in Mizan-Aman Town, South-west Ethiopia, 2020

Back ground: Neonatal death remained a global public health issue, especially in Ethiopia due to suboptimal breast-feeding practices. While several interventions have been made to comply with the WHO guideline, sub-optimal feeding practices are commonly seen in Ethiopia. The goal of this study was therefore to identify gaps in optimal breast-feeding practice in South West Ethiopia. Objective: To assess the prevalence and determinant of early days of newborn feeding mal-practices among recently delivered women in Mizan-Aman Town, southwestern Ethiopia, 2020. Methods: A community-based cross-sectional study was undertaken on recently delivered women from March-April/2020. A total of 487 mothers-child (<12month) pair were selected by using multistage randomized sampling technique and the data was collected through face to face interview by using structured questionnaire. The result was analyzed via SPSS version 26. Multivariate logistic regression analysis was used to assess determinant of newborn feeding mal-practices and a p<0.05 was deem to be signicant. Result: Prevalence of prelacteal feeding and colostrum avoidance was approximately 21.9% and 15.5% respectively. The most common prelacteal food was Rue / Tenadam/49 (10.1 %). The key reasons were: 49(10.1 %) Cultural activity followed by 45 (9.3 %) clean-up of the intestine / throat / mouth of babies. Determinants of pre-lacteal feeding were found to be: mothers recognizing risks of prelacteal feeding, multipara mothers, having ≥ 4 number of children and infant’s birth order 4-6. Likewise, exposure to infant formula advertisement, absence of home to home health education, multipara mothers and spontaneous vaginal birth were the determinants of colostrum avoidance. Conclusion: The study found that there was a sub-optimal breast-feeding practice, one in four newborns. As a result, activities to discourage pre-lacteal feeding and to improve early initiation of colostrum feeding practices shall be carried out.


Introduction
Breast milk is a fundamental right, nutritious, has an optimum temperature, does not require any preparation and is easily accessible to infant food [1]. Colostrum is the rst milk, yellowish in color, thick and produced just 3-4 days after birth. It provides all the essential nutrients for baby growth and wellbeing, immune factors and helps the newborn to get through meconium. As such, colostrum is considered to be the rst and infant immunization [2].
Exercising optimal breast-feeding bene ts both infants and mothers. It protects the infant from acquiring infectious morbidity and mortality, overweight, diabetes and chronic disease as well as can prevent mothers breast cancer, better birth spacing, reduce the risk of diabetes and ovarian cancer. Therefore, making the breast-feeding practice to universal level, could save about 823,000 child death and 20,000 breast cancer death every year [3].
Despite the recommendation of the World Health Organization (WHO) that early initiation of colostrum feeding and exclusive breastfeeding as a golden standard for early infant feeding, about 78 million newborns worldwide waited for more than one hour to breastfeed, two out of ve newborns were given uids or solids other than breast milk and excluded from colostrum feeding in the early days of their life [4,5].
Globally, neonatal death has remained the lion share of (45%) among all under-ve deaths in 2018 [6].
Similarly, every day, about 4,000 infants and young children die due to Colostrum avoidance and introduction of prelacteal feds, worldwide [7]. As a result, many infants suffer from insu cient breast milk, lactation failure and insu cient weight gain. Likewise, 45% of neonatal infectious deaths, 30% of diarrheal deaths and 18% of acute respiratory deaths among under ve children associated with inappropriate feeding practices [8,9,10].
While Ethiopia has a strong experience in early initiation of colostrum feeding (73.3%), it does not always comply with WHO / UNICEF recommendations [4]. At the national level, one in four infants (25.29 per cent) received pre-lacteal feed during the rst three days of life [11] and colostrum avoidance practices ranged from 6.3% up to 67.2% in Tigray Aksum and Meshenti, north west Ethiopia [12,13]. As a result, 18% of infant deaths could be due to suboptimal breastfeeding practices [14].
Studies conducted in Ethiopia have shown that there are multifactorial determinants of early newborn feeding mal-practices such as: sociodemographic factors, home and caesarean section delivery, low level of breastfeeding information, lack of postnatal care and breast-feeding advice [11,12,15,16].
Since 2004, Ethiopia has adapted the Guideline for Infant and Young Child Feeding Practices (IYCF) [17] and the 2013 National Nutrition Policy to promote optimal breastfeeding practices through multi-sectoral integration [18]. While remarkable advances have been made with regard to the optimal promotion of breast-feeding at the facility and at the community extent, the level of current practice is not well recorded in the study area. The goal of this study was therefore to assess the magnitude and determinants of prelacteal feeding among recently delivered mothers in Mizan-Aman Town, southwestern Ethiopia.

Methods And Materials
Study area and period The study was conducted from 3 March to 3 April 2020 in the town of Mizan-Aman, the capital of the Bench Sheko zone. The town is situated 561 km from Addis Ababa, administratively structured by ve kebeles and has a total population of 49,590, of which 11,554 are women in the 15-49 age group and 1,582 are children under the age of one. The key economic activity in the town is cash crop farming and gold mining. It has one hospital, a health center, and three health posts.

Study Design And Population
A community-based cross-sectional study was employed. Mothers of children less than one years of age that are living in randomly selected kebeles were considered as study population.

Sampling Procedures And Size
The sample size was determined using a single population proportion formula, assuming a 50 percent, 95 percent CI, 5 percent marginal error prevalence for prelacteal feeding. Correction formula was used since the source of population was < 10,000, and the sample was multiplied by a design effect of 1.5 and nally, the nal sample size was 487 by adding 5 percent non-response rate. Multi-stage randomized sampling technique was used to capture participants in the sample. The town of Mizan-Aman has two sub-city and ve kebeles. The sub-city of Mizan has 3 kebeles and the sub-city of Aman has 2 kebeles. Of these, a total of 3 kebeles were selected using the lottery process. In order to achieve the sample size of each 3 kebeles, a proportional allocation to the size was made Following the sampling fraction / interval, the mother / child pair was chosen on a three-house basis using systematic random sampling by reference to the community health extension worker family record book as a sampling frame.

Operational De nition
Prelacteal feeding; is de ned as giving uid or semisolid before breast feeding to an infant during the rst three days after birth [19]. Colostrum avoidance; includes; pumping and discarding colostrum during the rst ve days after birth [19].

Data Collection
The tool was adapted from the 2016 Ethiopian Demographic and Health Survey (EDHS) document [19] and structured questionnaires were used to collect data via face-to -face interviews. The questionnaire was rst prepared in English, translated into Amharic. Data quality was ensured by hiring nurse practitioners as data collectors, offering preparation for data collectors and conducting pre-tests in 5% of the sample size of the mother / child pair in Debrework Area.

Data Analysis
The data was washed, encoded and entered in Epi Info version 3.5.3 and exported to SPSS version 26 of the statistical package for analysis. Descriptive statistics have been computed to determine the magnitude of the suboptimal breast-feeding practice. In addition, a bivariate logistic regression technique has been used. Subsequently, these variables with p < 0.2 were shifted to a multivariate analysis and evaluated using a stepwise multivariate logistic regression technique to monitor the effects of confounding and to classify predictors of suboptimal breast-feeding practices. A P-value of < 0.05 was found to be statistical signi cance at a 95% con dence interval.

Result
Socio-demographic characteristics of the respondents A total of 487 mother-to -child pairs were included in the study, resulting in a response rate of 485 (99.6%). The majority of women, 184 (37.9%), were 20-24 years of age and the mean (± SD) age of mothers was 24.5 (± 5.5) years of age. In addition, of the total respondents, 459 (94.6 per cent) were married, 291 (60 per cent) had a family size ≥ 4, 284 (58.6 per cent) were housewives by occupation, and 167 (34.4 per cent) attended primary school. Moreover, about 229(47.2%) of infants were rst-order / rstchild for their family (Table 1).

Colostrum Avoidance Practice
Although all mothers have breastfed ever their current infant, 163(35.5%) newborns were put to the breast more than one hour and 75(15.5%) of women were discarded colostrum for their infants within the rst ve days after birth. The main reasons for colostrum avoidance were; breast feeding problem 50(10.3%), maternal medical illness 19(3.9%) and a reason for their child health 6(1.23%) ( Table 3). Of the total respondents, within the rst three days after birth, 106(21.9%) were given prelacteal feds other than breast feed for their babies. Water with "Tenadam"/rue 49(10.1%), followed by plain water 33(6.8%), glucose with water 10(2.1%) and formula milk 9(1.9%) were the most popular types of prelacteal food.

Factor Associated With Prelacteal Feeding
According to the bivariate logistic regression analysis, the following study variables were signi cantly associated with pre-lacteal feeding (p < 0,05): mothers in age groups 20-24 years old and 35-39 years old, mothers in income > 1,500birr / month, infants in birth order 4-6, multipara mothers, mothers assisted by traditional birth attendants and mothers who understood the risk of prelacteal feeding In addition, these variables with p < 0.2 from the bivariate analysis were evaluated using a multivariate analysis. As a result, multivariate logistic regression analysis; multipara mothers, infants' birth in order 4-6, having ≥ 4 children and mothers who understood the risk of pre-lacteal feeding were determinants of pre-lacteal feeding. Multipara Mothers were 3.56(AOR: 95% CI: 1.036, 12.227) times more likely to give prelacteal feds when compared to the counterpart. The odds of prelacteal feeding among mothers who knew the risk of prelacteal feeding were almost seven times, (AOR = 6.91, 95% CI: 3.002, 15.904), higher than the counterpart. Furthermore, infant's birth order between 4-6 were 12.28(AOR: 95% CI: 1.528, 98.639) times more likely to give prelacteal feeding when compared to infant's birth order one. The odds of prelacteal feeding among mothers having ≥ 4 number of children were nearly ve times, (AOR = 5.2, 95% CI: 1.033, 26.141), higher than, those mothers having ≤ 3 number of children (Table 5). The odds of colostrum avoidance among mothers who gave birth through spontaneous vaginal delivery were three times, (AOR: 95% CI: 1.245, 7.750), higher than, those mothers who gave birth through CS (Table 6).

Discussion
This study found that breastfeeding activities were sub-optimal in the setting of the study due to the widespread implementation of prelacteal feds and colostrum prevention. The prelacteal feeding and colostrum-avoidance prevalence is found to be 21.9% and 15.5%, respectively, in Mizan-Aman-town.
Prevalence of prelacteal feeding in this study was relatively comparable with studies in Wolayta sodo (20.6%), Tigray (24.7%) and Dabat district (26.8%) [16,20,21]. However, the gure we have obtained are far higher than those found in other studies in Sidama (5.5%), in Mettu (14.2%) and in Debrebirhan (14.2%) [22,23,24]. The observed variance may be due to variations in culture, health service access and population character. The gures we have obtained are much lower than those found in other studies in South Sudan (53%), Vietnam (73.3%), Egypt (58%), Amhara (47.2%) and the Harare region (43.5%) [25,26,27,19,28]. This may be due to religious and cultural traditions that have a lion's share in the provision of pre-lacteal food in the regions mentioned above, most of which have been in rural areas. In addition, in the early days of infants, the effects of grandparents would be greater.
The magnitude of colostrum-avoidance in this analysis was consistent with studies in Raya Kobo district (13.5%), Tigray district (15%) and Mettu, southwestern Ethiopia (17.5%) [29,20,23]. While colostrum avoidance practice is higher in Mizan-Aman than in Aksum City (6.3%), Wolayta Sodo (8.5%) and Kombolicha City (11.4%) [12,16,30]. This distinction may be attributed to the use of socio-cultural, maternal health care programs and the number of institutional deliveries. The gure we have obtained is much smaller than that seen in other studies in Dabat, Northwest Ethiopia (47.1%), Debrebirhan (20.3%) and South Sudan (38.8%) [21, 24,25]. This disparity may be due to socio-cultural differences and they felt that colostrum would cause their babies to have abdominal crap and have bad sprit and should be discarded.
This study showed that those mothers who recognized the risk of prelacteal feeding were almost seven times, (AOR = 6.91, 95% CI: 3.002, 15.904), higher than the counterpart. The report is contrary with study done in Mettu, Sidama district and Raya Kobo district [23,25,29]. This may be due to the disparity in, as the mothers are conscious of the disadvantage of Prelacteal Feeding; the mothers are affected by maternal or exhaustive medical illness, CS childbirth and breast-feeding issues, in particular the lack of milk secretions within the rst three days of birth. They were concerned about the survival of their child for this cause. This, in fact, leads to the need for pre-lacteal feeding of their babies. The other possible justi cation may be that, even though mothers have obtained ample information from health facilities, community health workers and mass media about breastfeeding, the in uence of local community members, particularly grandparents, may lead them to give prelacteal feeding.
This study showed that infant's birth order between 4-6 were 12.28(AOR: 95% CI: 1.528, 98.639) times more likely to give prelacteal feeding when compared to infant's birth order one. The odds of prelacteal feeding among mothers having ≥ 4 number of children were nearly ve times, (AOR = 5.2, 95% CI: 1.033, 26.141), higher than, those mothers having ≤ 3 number of children. This is consistent with study from Harare Region, Eastern Ethiopia [28]. Although contradicting the Nepal report, which was the rst order of infant births [31]. This may be due to the gap in; mothers of infants whose births order 4-6 and mothers who have ≥ 4 number of children were relatively old and multi-para. As a result, compared to younger mothers in this study, relatively elderly and multiparous mothers were more likely to provide prelacteal feeding. Moreover, dominantly in uenced by cultural experience and unable to change their behavior as opposed to young people.
In this nding, multipara mothers were 3.56 (AOR: 95% CI: 1.036, 12.227) and 10.045 (AOR: 95% CI: 1.456, 19.289) times more likely than their counterparts to offer prelacteal feeding and colostrum avoidance. The study coincides with the three African countries in the regions of Burkina Faso, Uganda and South Africa [32] and Tigray region [20]. This may be attributed to cultural resemblance. Multipara mothers are relatively older and are unable to learn about infant feeding and may have a potential to have an effect on cultural traditions. This, in turn, contributes to the need for prelacteal feeding and colostrum avoidance of their infants.
The odds of colostrum avoidance among mothers who gave birth through spontaneous vaginal delivery were three times, (AOR: 95% CI: 1.245, 7.750), higher than, those mothers who gave birth via CS. This is contrary to a report conducted in Sudan [25]. This difference is due to the massive missed opportunity and less devotion of skilled attendants to provide essential newborn care, particularly initiation of breast feeding and counseling, since most of the mothers were delivered at the health institution via spontaneous vaginal birth. This means that institutional delivery is not a guarantee of optimum breastfeeding, but rather a high commitment of trained birth attendants delivering critical newborn care.
Among mothers who do not have home health education, the odds of colostrum avoidance were almost ve times higher than counterpart (AOR = 5,474, 95% CI: 2,876, 10,419). The study is consistent with the study carried out in the town of Aksum and the district of Raya Kobo [12,29]. Intensive health education on the possibility of sub-optimal breastfeeding will bring about behavioral improvement, thus enhancing the effective infant feeding practice.
The odds of colostrum avoidance among mothers who have exposure to infant formula advertisement were nearly seven times, (AOR = 7.476, 95% CI: 3.549, 15.752), higher than the counterpart. The study is in line with report from South Sudan [33]. Infant formula advertising has weakened breastfeeding and misled mothers to discard colostrum and to adopt prelacteal feed for their children. This suggests that the vigorous promotion of behavioral modi cation practices on colostrum feeding has gone undone.

Conclusion
Early days of newborn feeding mal practice is given one in four infants. This makes breastfeeding practices sub-optimal in the town. Mothers recognizing risks of prelacteal feeding, multipara mothers, having ≥ 4 number of children and infant's birth order 4-6 were found to be determinants of prelacteal feeding. Likewise, exposure to infant formula advertisement, absence of home to home health education, multipara mothers and spontaneous vaginal birth were the determinants of colostrum avoidance. Ethical clearance was secured from wollo university, college of medicine and health science to Mizan-Aman town health unit. Permission letters was obtained from Mizan-Aman town health unit and respective kebeles. After giving clear and deep understanding about the aim of the study, written consent was obtained from each respondent before the interview is conducted. Besides, guardian consent was obtained for those participants age below 18. Anonymous data was taken and the con dentiality of participant's information was secured.

List Of Abbreviations
Availability of data and materials