Early initiation of breastfeeding practice and associated factors among mothers of children aged less than six months of old in Mizan-Aman town, southwest Ethiopia, 2018


 BackgroundEthiopian government implemented baby-friendly hospital initiative and community integrated management of childhood illnesses program. Despite early initiation of breastfeeding taken as a key tool for tackling neonatal mortality, EIBF is still low and most of the neonatal mortalities were existed due to delayed initiation of breastfeeding in Ethiopia in general and the practice is not well documented in South West Ethiopia in particular. Therefore, this study aimed to assess early initiation of breastfeeding practice and associated factors among mothers of children aged less than six months of old in Mizan-Aman town, southwest Ethiopia.MethodsA community-based cross-sectional study was employed from April 15 to May 15, 2018. A total of 487 recently delivered mothers were included. The data was collected through face to face interview by using a pre-tested and structured questionnaire. Binary and multivariable logistic regression analysis were employed and p-value < 0.05 was identified as statistically significant factors, and the quality of the data were assured, checked, coded, cleaned and entered in Epi-Info version 3.5.3 and exported to SPSS version 25 for the analysis. ResultThe prevalence of timely initiation of breast feeding was 296(64.50%) in Mizan- Aman Town. Mothers who had < 24 months birth spacing history 1.85(AOR: 95% CI: 1.22, 2.81), mothers’ income level between 1001-1500 Ethiopian Birr 2.21 (AOR: 95% CI: 1.12, 4.37), Primipara mothers 2.00 (AOR: 95% CI: 1.24, 3.23) and home delivery 2.76(AOR: 95% CI: 1.24, 6.14) were important positive predictors for timely initiation of breast feeding. Furthermore, Government employee and merchant mother by occupation was found to be protective factors.Conclusion and recommendationThe practice of early initiation of breast feeding was suboptimal and still below the national average. Intervention at the community and facility level should paid special attention.

left untreated, more than 60 countries will miss the target for neonatal mortality (12 deaths or fewer per 1,000 live births) by 2030 [2]. In Ethiopia, 2019 EMDHS reveals that the neonatal mortality rates were 30 deaths per 1,000 live births [3].
Worldwide in 2017, it is estimated that 78 million newborns delayed more than one hour to be put to the breast. which means that 42% or only two out of ve newborns, the majority born in low-and middleincome countries, were initiated breast feeding within the rst hour of life. The practice varies worldwide from (35%) in the Middle East and North Africa to (65%) in Eastern and Southern Africa. Ethiopia found in eastern Africa region, the practice of early initiation of breast feeding in 2016 was 73.3%, slight improvement has been made from 2005 practice 66.2% [4].
Early initiation of breast feeding has bene cial effect on baby's as well as maternal health like: save more than 800,000 children each year, increase cognitive development, prevent overweight and obesity, boosts immunity system and protect against all half of diarrheal and one third of respiratory infectious diseases [5]. Indeed, mothers bene ted from breast feeding involving: a lower risk of 20,000 breast cancer death, ovarian cancer, improve birth spacing, post-partum hemorrhage, depression, type 2 diabetes and saved $300 billion to the global economy each year [5][6][7].
The longer newborns delayed for breast feeding, the greater their risk of death. New studies revealed that when compared with newborns who initiated breastfeeding within an hour of birth, the risk of dying during neonatal period is 33% higher for those who initiated 2-23 hours after birth, and was two times higher for those who initiated one day or longer after birth [8].
There are many factors which determine the practice of early initiation of breast feeding like: outdated practices in health facilities, lack of knowledge about breastfeeding after a caesarean section, cultural practices like prelacteal feeding, Massive missed opportunity, skilled birth attendants, place of deliveries, Programme and policy-related factors and Access to antenatal care [4,[9][10][11]. Despite Ethiopia have been implemented different programs on optimal breast feeding as key tool to tackle neonatal and infant mortality, yet not meet the desired outcome [12][13][14]. Although the importance of early breast-feeding practice, limited information is documented in southwest Ethiopia. Therefore, this study was attempted to ll this information gap and come up with recommendation on possible intervention for timely initiation of breast-feeding improvement and associated factors in Mizan -Aman town, southwest Ethiopia.

Study setting and study period
The study conducted in Mizan-Aman town, the capital town of Bench Maji Zone, is situated 561 Km south west of Addis Ababa capital town of Ethiopia, from April 15 to May 15/2018.The town is administratively structured by 7 kebeles and has a total population of 49,590 of which 26,392 are males and 23,197 are females. Out of all female population 11, 554 of them are women in the reproductive age group (15-49 years). About (1,582) or 3.9% of the total population is accounted by children less than one years of age. It has one public hospital, one health center, three health post, 10 private health institution and 17 drug stores.

Study design
A community based cross-sectional study was employed in Mizan-Aman town Sampling size determination The sample size was determined using single population proportion formula. The following assumption were considered: 95% level of con dence, 5% margin of error and 50% of women estimated to be initiated breast feeding within one hour. Since population size was 1582, which is less than 10,000, sample size was adjusted using correction formula, 2 design effects were considered and by adding 5 % non-response rate, the nal sample size became 487.

Sampling procedure
Since it is multistage among the two sub-cities (Mizan and Aman sub-city), Mizan sub-city was selected by lottery method. Among ve kebeles in the selected sub city, three of them were selected by lottery method. proportional allocation to size was done in each 3 selected kebeles. Then after a systematic random sampling technique was employed at the kebele level by using health extension family record book as a sample frame. Following random selection of the rst household with a woman who ful lled the selection criteria, subsequent households with women meeting the criteria were selected from every three house until the desired sample size was reached in each kebele. For households with more than one eligible woman, one of the women was chosen using a lottery method.

Data collection tools and procedures
The data collection tool was adapted from Ethiopian Demographic and Health Survey (EDHS) 2016 document [15] and translated into local Amharic languages. Training was given for data collectors and supervisors how to make face to face interview and assuring the quality of data. pre-test was done in 5% of the sample size in Tapi town.

Operational de nition
Early initiation of breastfeeding: Is putting newborns to the breast within the rst hour of life [1].
Level of information about breastfeeding: Those mothers who mentioned ≥2 components of breastfeeding information (1. Bene ts of breast feeding 2. positioning of the baby 3. exclusive breast feeding 4. managements of breast problem 5. Expression of breast milk) were considered as having good level of information and mentioned ≤1 components of information were considered as the counterpart [15].

Data quality control
Prior to data collection period, data collectors trained about the objective of the research. The PI was given the training about the objective of the study and data collection system by using semi-structured questionnaires in a one-day period. The questionnaires would prepare in English, translated into Amharic and back translated into English to check consistency. Pre-test carried out on Tapi town to familiarize the interviewer with the tools and to check the coherence. To keep the quality of data, principal investigator was checking the questionnaires for its completeness in each day.

Data processing and analysis
The data was cleaned, coded and entered in to Epi Info version 3.5.3 and exported to SPSS version 25 statistical package for analysis. Descriptive statistics was computed to determine the magnitude of early initiation of breast feeding. Furthermore, bivariate logistic regression and multivariate analysis with 95% con dent interval were done in order to determine predictors of early initiation of breast feeding. Variable with a P value of < 0.05 was taken as the criterion for statistical signi cance.

Results
A total of 487 mother-child pairs were included in the study, resulting in a response rate of 459(94.3%).

Knowledge of mother about breastfeeding
Page 10/17 The majority, 274(59.7%) of mothers have good level of information about breast feeding, while 185(40.3%) have poor level of information. The majority, 413(90%) of respondents mentioned bene cial of breast-feeding, while only 31(6.5%) of mothers mention about expressed breast milk. .135) were important positive factors for timely initiation of breast feeding. indeed, gov't employee and merchant mother by occupation and were found to be protective factors for timely initiation of breast feeding, respectively (Table 3). Discussion our study revealed that Breastfeeding practices were sub-optimal in the study setting due to the delayed initiation of breast feeding. The prevalence of timely initiation breast feeding in Mizan-Aman town is found to be 64.5%. This gure is consistent with study ndings which were conducted in Debre Berhan town, Kenya, Lesotho and Jamaica (62.6%, 62.2%, 65.3% and 64.7% respectively) [16,11,17,21].

Factors associated with timely initiation of breastfeeding
The practice of early initiation of breast feeding in our study area was higher than study from in Zimbabwe (58.3%) ,south Gondar (54.7%),Axum Town (41.6%) and Goba district (52.4%) respectively [17,18,19,20].this difference could be due to study setting, some of them were from rural side, population character, when the study subjects are those mother of children more than 6 months of age they might not recall well when to start to initiate breast feeding, information and health service utilization and socio economic difference between the referenced subject and the study place.
Mothers' income level between 1001-1500 Eth Birr were two times (AOR: 95% CI: 1.115, 4.368) more likely initiated breast feeding within one hour than mothers' income > 1500 Eth Birr. While, this study is contradicted with Debre Berhan town [16], those mothers' income level > 1969 Eth Birr were more likely initiated early breast feeding. This difference might be due to having higher income and being sophisticated city women, empowers the women to make medical decision. they can afford and prefers elective Cesarean section delivery in order to avoid labor pain. This in turns leads to delayed initiation of breast feeding. Indeed, this is supported by evidence in this study, gov't employee and merchant mother by occupation, the majority of them have gotten > 1500 ETH Birr/month, were 0.074 (AOR: 95% CI: .006, .867) and 0.079 (AOR: 95% CI: .007, .933) less likely to initiate breast feeding within one hour.
Our study revealed that the odds of timely initiation of breast feeding among primipara mothers were two times (AOR = 2.002 ,95% CI: 1.241, 3.229) higher than their counterpart. This is similar with report from Malawi [21], while contradicted with report from south Gondar [16]. Since primipara mothers are less experienced, might fear of complication. Therefore, for the sake of their infant health, they might seek frequent medical advice from the care providers. This implies that having good level of information about newborn care will result putting their newborn to breast early.
Our study showed that the odds of early initiation of breast feeding among home delivered mothers were nearly three times (AOR = 2.755, 95% CI: 1.237, 6.135) higher than as compared as had institutional delivery. This is contradicted with report from Gurage zone Gunchere woreda [25], Arsi Tiyo woreda [26], Malawi [21], Bahirdar city [22] and Motta town [24], respectively. This might be due to more births take place in health institutions with skilled providers doesn't mean necessarily will result optimal breastfeeding practice. Rather, missed opportunities, having less committed skilled attendant and having not appropriately trained staff with essential newborn care will result negative effect between institutional delivery and early initiation of breast feeding. This implies that quality of care should be improved in order to have positive relation between institutional delivery and putting newborns to breast within onehour practice.

Strength and limitation of the study
This study is a community based and tried to represent and made generalization by involving relatively adequate number of study subject. Tried to reduce recall bias by involving only those mother of infant age less than 6 months. But it will share the cross-sectional design limitations.

Conclusion And Recommendation
The practice of early initiation of breast feeding was sub-optimal and below the national average in study place. Mothers who had < 24 months birth spacing history, mothers' income between 1001-1500 Birr, Primipara mothers and home delivery. Risky groups like; multipara mothers, high income level mothers and having ≥ 24 months birth spacing practice shall be given special attention and provided intensive health education program at the community and facility level. Improving quality of care in health facilities through providing training and support for staff members on maternal and newborn care. Furthermore, improve access to skilled breastfeeding counselling for all mothers are recommended. The Future researcher shall be employed follow up and mixed study. Permission letters was obtained from Mizan-Aman town health unit and respective kebeles. Study participant mothers were asked by explaining the objective of the study and its signi cance. Since the research topics is less sensitive/ no more principal risk would be potential harm unless breach of con dentiality, and then verbal/unsigned informed consent were obtained after clearing up verbal version of a consent form (information sheet) and subjects give their verbal consent in place of written consent to participate. Mothers who are unwilling were exempted from the study. Con dentiality was maintained anonymously and not communicated for other purposes.