Knowledge, Attitude, and Practice of Exclusive Breastfeeding Among Mothers Attending Masaka District Hospital Kigali/Rwanda: a Cross-section Study.

Background: Exclusive breastfeeding (EBF) for 6 months is considered a major public health intervention to reduce the escalating child mortality of neonates and infants in low and middle-income countries. In most East Africa countries, exclusive breastfeeding did not meet the recommendation of WHO/UNICEF that a baby should be fed only breast milk for the rst 6 months. This study is aimed to provide knowledge, attitudes, and practice (KAP) towards EBF and identify factors associated with the practice of exclusive breastfeeding. Methods: A cross-sectional study was conducted from January to April 2020. A total of 364 mothers were interviewed using a questionnaire. Univariate statistical analysis was used to dene variables using frequencies and percentages. Multiple logistic regression was also applied to identify the factors associated with EBF practice. Results: In the current study, the majority of mothers, 84%, presented good knowledge, 87.0% also showed a good attitude toward EBF. Nearly 95.9% of the respondents understood the benets of EBF practice for at least six months, 86.0% also responded that breast milk should be started immediately after birth within an hour. Besides, 87.6% knew the importance of the rst breast milk or colostrum. 92.3% of mothers understand that breastfeeding increases mother-infant bonding, 92.3% of mothers understand that breastfed babies are healthier than formula-fed babies, and 94.8% believed that breast milk is the ideal food for babies. Furthermore, 76.4% of the surveyed mothers exclusively breastfed their infants for the rst 6 months after delivery A married women (OR=.270; 95%CI=.160-.457, p<.001), had a high school degree(OR= .150; 95CI=.073-.307; p<.001), mothers whose had babies more than three (OR=.090; 95%CI=.035-.233, p<.001), good knowledge score(OR=2.535; 95%CI=1.399-4.594, P=0.002) and positive attitude score OR=2.051; 95%CI=1.149-3.662, P=0.015)

Mothers who are mentally ill.

Sample Size and Sampling Technique
We have used a single population proportion formula by taking a proportion of prevalence of EBF of the previous study, 68.6% [13] and we used a 95% con dence interval. To adjust the non-response rate of study participants, we were added 10% of the sample size. Sample size determination using the Fischers formula: Henced the total sample size required for this study with a 10% continuity correction was 364.
where is the level of signi cance which can be obtained as 1 − con dence interval, is a proportion of prevalence EBF, which is equal to 68.6, D is the maximum acceptable difference (margin of error) which is equal to 5%, and /2 is the value under the standard normal table for the given value of con dence level which is equal to 1.96.

Data collection procedure and techniques
Data were collected during the study period via face-to-face interviews with women on a randomly selected working weekday. The researchers approached the mothers who visited MDH and those who accepted to participate in the study were interviewed until the nal sample number was reached. The interviews were performed in the waiting rooms of the hospital using a questionnaire. Also, the interviews lasted 10 minutes on average and privacy were maintained during the interviewing period. This was done to maximize the chances of the participants feeling comfortable and able to answer the questionnaire.
Before starting the interview, the researchers described the survey topic and the organization of the questionnaire to the participants and assured them that the interview was completely anonymous, and that the data collected would be saved private, and that the written informed consent was collected. Besides, the research team explained to the women that they could stop the interview at any time without penalty, the participation was voluntary, and that no payment would be given.

Instrument
The questionnaire was organized into ve sections: Section A: sociodemographic data, Section B: data on exclusive breastfeeding knowledge, Section C: Data on the attitude of respondents towards EBF, Section D: data on the practice of respondents towards EBF, and Section E: Reasons of not exclusive breastfeeding and information on breastfeeding practices.
Structured and semi-structured questionnaires were used for the data collection in the form of a closedended (multiple choice) question style. The items included in the questionnaire were chosen based on previously published investigations of a validated standardized questionnaire, the questionnaire used in the present [17,[19][20][21][22], or because they were considered interesting, or modi ed to meet our objectives of the study by the research team. The questionnaire will include a set of questions organized under ve main sections was aimed at gathering information from mothers about exclusive breastfeeding. Section A of the questionnaire included a set of variables that targeted information about mothers social and demographic characteristics namely age, nationality, religion, marital status, educational status, partner's level of education, occupation, residence, parity, gravidity, antenatal care follow-up, type of birth, gestational age, child's body weight at birth, place of birth, and rooming-in.
Besides, there were 20 items to measure the knowledge of the participants about exclusive breastfeeding. Three possible responses for each item (Yes, No, and I don't know). Correct responses were scored as one, and zero for other options. By summing up all the awarded points, the overall score was determined for each respondent and these were translated to percentages. Each respondent will be given an information grade based on his total percentage score. All mothers who score >70% in the knowledge test were considered to have good knowledge and those scoring <70% were considered as having poor knowledge according to FAO guidelines. [23].
Maternal attitudes towards EBF were determined by responses to 12 questions. Correct responses were scored as one, and zero for other options. The total score for each respondent was calculated by summing up all the awarded marks, and these were converted to percentages. The attitude score was calculated. Attitudes were measured with questions that required "Agree", "Neutral" or "Disagree" responses. The total score for each respondent was calculated by summing up all the awarded marks, and these were converted to percentages. All mothers who scored >70% in the attitude test will be considered to have a positive attitude and those scoring <70% will be considered to a negative attitude [23].
Informed consent forms were attached to all questionnaire for the participants to read and sign if they were willing to participate in the study. Questionnaires and informed consent forms were drafted in Kinyarwanda. A pilot study was conducted after ethical approval was allowed to assess the questionnaire's comprehensibility, and modi cations were accordingly made. The pilot sample met this study's inclusion, and exclusion criteria, and the pilot sample size was 36.

Data quality control
Three enumerators (data collectors) with a minimum of a Diploma in Health or Nutrition quali cation were recruited from those residing in the study area, and who speak Kinyarwanda (the local language) uently. The enumerators also had previous experience in survey data collection. The enumerators underwent 2 days of training to cover the following: the explanation of the study objectives, interview techniques, and research ethics.

Data analysis and presentation
The data was checked, cleaned, and coded, and analyzed using SPSS version 16.0. All data were analyzed using descriptive statistics of the frequencies and percentages for categorical variables. Cross tabulation and chi-square tests were used to determine univariate associations. To determine factors associated with the practice of EBF, multiple logistic regression was executed. The dependent variable of the logistic model was EBF practice. Only variables that were signi cantly associated with the practice of EBF in the cross-tabulation analysis were included in the logistic regression model. The ndings are presented as odds ratios and their respective con dence intervals at 95 %. The p-value < 0.05 was considered statistically signi cant in all analyses.

Results
Socio-demographic characteristics of the study participants Table 1.Shows the Socio-demographic characteristics of mothers. This study surveyed 364 mothers who attended Masaka district Hospital between January to April 2020, most of them were in the range of 26 to 35 years of age (51.1%), most of them were married (76.1%), had a primary level of education(47.5%), were unemployed/housewife(67.6%), was delivered at a public hospital or health centers(98.4%), were delivery in full-term 37-42week (91.5%), were room in(84.9%), and were visited at least three times in antenatal care follow up(34.3%). The results concerning the knowledge of the study participants are described in replied that breast milk should be started immediately after birth within an hour, about (87.6%) understand the importance of the rst breast milk or colostrum, (41.5%) know the right time to start complementary feeding, about (91.5%) know breast milk alone is enough for an infant during in the rst 6monthsof life, while (72.5%) of mothers responded that EBF can be used to prevent early pregnancies. However, A child who is breastfed is less likely to get sick compared with formula-fed babies, (86.8%) of mothers knew that EBF for six months protects their child from diarrhea. Nearly all (96.4%) of the participants understand that breast milk is a natural food for babies that contain all nutrients, (82.1%) know breastfeeding reduces the risk of malnutrition and obesity in children, about 78.0% admits that breast milk may protect the baby against infection and chronic diseases, (96.2%) know breast milk is safe and hygienic and always available in need. Nearly (97.0%) know exclusively breastfed infants grow healthy and strong. More than three-quarters (77.5%) of the respondents knew the effects of bottle breastfeeding for the baby, A large number (95.1%) of respondents were aware of cleaning the breasts before breastfeeding and (97.5%) of breastfeeding mothers should eat a healthy diet to produce enough breastmilk to baby, and also understand that Frequent sucking help for milk production (83.5%), More than two-thirds (72.8%) of the respondents know that to give a newborn herbal medicine is dangerous. Nonetheless, from the total study participants, based on knowledge score, (84%) of the respondents were grouped as having good knowledge, and (16%) of the study participants were categorized as having poor knowledge.
Attitudes towards exclusive breastfeeding (N=364) The EBF of the mother's table concerning the attitude is presented in Table 3. The majority of mothers 94.5% agree that introducing colostrum within an hour after delivery is important for the baby, most of the mothers 93.4% agree that breast milk for a newborn within an hour after birth is important and 93.7% of mothers understand that breastfeeding is better than arti cial feeding. However, 81.0% of participants reported that is di cult for breastfeeders to take care of their family, and 92.3% of mothers agree that breastfeeding increases mother-infant bonding. Furthermore, 92.3% of mothers agreed to prefer breastfed babies are healthier than formula-fed babies, and 89.6% agree that women should breastfeed in public places. Almost all respondents 94.8% believe that breast milk is the ideal food for babies, about half the mothers 50.5% agree that starting complementary feeding to a child before 6 months is important, while 87.9% prefer to feed your baby breast milk alone for the rst 6 months. 66.2% of mothers agree breastfeeding will make mother's breasts sag, and only 45.3% of participants agree that breastfeeding affects their beauty. However, from the total study participants, based on attitude score, 87% of the respondents were grouped as having positive and 13% of the study participants were categorized as having a negative attitude. Exclusive breastfeeding practices of study participants (N=364)  Cultural belief and the need to introduce herbal medicine for cultural 6 7.0 Cross-tabulation of characteristics of mothers who do and do not practice EBF (364) The characteristics of mothers who either do or do not practice EBF are presented in Table 5. The majority of the surveyed mothers were between the ages of 26 to 49 years. There was no association found between age and the practice of EBF χ2(2) = .760,p > .05. The characteristics of mothers who do and do not practice EBF are presented in Table 5. EBF was found to be statistically associated with marital status χ2 (1) = 43.022, p < .001. The educational level of both mothers and their partners was statistically signi cant to be associated with the practice of EBF χ2 (2) = 28.009, p < .001, χ2(2) = 10.461,p = .005 respectively. It was observed that mothers practiced EBF according to their level of education.
Unemployment for mothers stood at 246(67.6%) while 118(32.4%) were employed. The occupation of the mothers was not found to have an association with EBF χ2 (1) = .312, p > .05. The association with parity, antenatal care, and practice of EBF was statistically signi cant. However, gravidity and religion were not statistically signi cant. There was a statistically signi cant association with good knowledge score, positive attitude, and EBF practice χ2 (2) = 7.349, P = .007, χ2 (2) = 37.400, P < .001 respectively as illustrated in table 5.

Discussion
To our best knowledge, this was the rst study that to investigate knowledge, attitude, and practices towards EBF among mothers attending health facilities in Rwanda and it can help as the baseline towards policy change. The study site, as the main district hospital has a particular bene t of providing policy guidance in lower health facilities. Women with good knowledge and a positive attitude on EBF had a higher prevalence of EBF practice and there was a strong belief that breast milk alone for the rst 6 months of infant life is enough for child growth. The overall aim was to provide information about mothers' knowledge and attitudes towards practice of EBF and to identify factors predicting EBF. The authors were pleased to nd that the prevalence of EBF among women of reproductive who had at least one child aged 6 months or younger at MDH in Rwanda was low 76.4 % compared to the WHO recommended EBF coverage of 90 % [9] and the national target of EBF coverage > 90% [14]. This gure was close to that seen in another study conducted in Nigeria75.6% [2] however, it was better than the gures reported in the previous studies conducted in India 36.25% [24], in Zimbabwe 36% [25], in the United Arab Emirates only 16.9% [26], and in the United Kingdom was 26.2% [27]. This value indicates that improving EBF practice for children is more probably due to the improved awareness of lactating mothers which is resulted from the expansion of health information on the importance of EBF by the health personnel at any level of health institutions. It is of utmost public health importance that optimal breastfeeding practices, particularly EBF, are encouraged and practiced promoting the growth, survival, and health of children. Furthermore, the more frequently reported reasons for why mothers stopped breastfeeding were maternal perceptions of breast milk production was insu cient to grow children, HIV, and unintended pregnancies, baby gets hungry and being thirsty, cultural belief and the need to introduce herbal medicine for cultural, work schedule, early and single motherhood, and poverty". Finally, maternal perception of having an inadequate breast milk production which corroborates ndings in other studies [19,26].
Furthermore, Our ndings of this study show that the majority had 84.1% had good knowledge, 82.1% had a positive attitude towards EBF were high. We found that mothers had good knowledge which was similar to previous studies conducted in India 92.5% [24], in Zimbabwe 89% [25], and in Nigeria 94.0% but was higher than studies conducted in the United Arab Emirates 51.2% [26], and in Ghana 45.8 [17]. However, 82.1% of them presented a positive attitude, the nding was in line with previous studies conducted from Nigeria which is 84.7%, in the United Kingdom was 84.7% [27], Ghana was 82.6% [17], and Mizan Aman town was 89.5% in Ethiopia [21], in Debre Birhan, Ethiopia which found 97.5 % [28]. However, it was higher than previous studies conducted in Dabat Health Center, Northwest Ethiopia 76% [22] Contrary to this, the nding is higher than a study conducted in China14.50% [1]. Good knowledge and a positive attitude were more likely to practice EBF for 6 months. The higher EBF rate in Rwanda could be further explained by the fact that Rwanda has prioritized and intensi ed intervention with a focus on health promotion efforts to increase community knowledge and skills on EBF interventions and promote health-seeking behavior [14]. Mothers who had higher knowledge were also likely to have a positive attitude than their counterparts concerning the practices of EBF. Similar to our ndings, studies that report high maternal knowledge on EBF also report a high prevalence of the practice of exclusive breastfeeding [17]. This nding is also in line with the study conducted in China that reported that positive maternal attitudes toward breastfeeding are associated with continuing to breastfeed longer and having a greater chance of successful breastfeeding, Besides that, maternal education plays a role in attitudes toward breastfeeding [1]. Despite high baseline attitudes and knowledge levels in these women, their scores increased signi cantly following training. This suggests that it is likely that the training which includes changing cultural and personal breastfeeding attitudes, was accountable for their high scores [27].
According to a multivariate analysis, practicing exclusive breastfeeding demonstrated the factors to be associated with EBF included married, maternal with a college degree or higher level of education, multiparity, having good knowledge, and a positive attitude toward EBF. However, mothers with a college degree or higher level of education were more likely to report a higher practice of EBF than their counterparts. Mothers with higher levels of education may be able to comprehend that it should be practiced for up to six months and appreciate the bene ts of EBF to their infants and more motivated to practice it. This result is consistent with other previous studies conducted among women, which con rmed the positive impact of a high level of education on the subject's knowledge related to health topics [17,19,26]. This is contrary to the result obtained from a study conducted in Debre Berhan District, Central Ethiopia [28]. Mothers with a higher level of education, their chance of practicing EBF signi cantly reduced. This could be because educated women have better job opportunities than illiterate mothers, so they don't have enough time to maintain EBF practice. Besides, maternal education was associated with knowledge of the importance of health bene ts of breast milk despite the perceived inconveniences or discomfort of breastfeeding in public and may be able to comprehend and understands the bene ts of EBF. Besides, less educated mothers had more positive attitudes toward formula use and were likely to introduce formula earlier than more educated mothers [1]. Mothers who had one child were less likely to exclusively breastfeed their babies when compared to mothers with two or more children. This nding was consistent with study ndings in Gwanda District, Zimbabwe study where they concluded that multiparity was a major factor of EBF [25]. Another factor that was found associated with the practice of EBF was having good knowledge of EBF. Mothers who had good knowledge were more likely than their counterparts with poor knowledge in EBF to report practicing it. Similar to our ndings, studies that report high maternal knowledge on EBF also report a high prevalence of the practice of EBF [17,37]. In addition, positive maternal attitudes were found to be signi cantly associated with the practice of EBF in MDH.
This nding was consistent with a study in positive maternal attitudes toward breastfeeding are associated with continuing to breastfeed longer and having a greater chance of successful breastfeeding [1].
However, the major reasons for not breastfed exclusively "Breast milk was insu cient." This should be one of the contributing factors to the prevalence of stunting among children under 5 years in Rwanda (0-59 months) and still unacceptably high as 38% and 15% of children who were stunted at the age of two months was caused by poor nutrition during pregnancy [11]. It has been shown that other non-exclusive breastfed infants have signi cantly higher stunting rates compared to EBF children [31]. Stunting has been documented in several studies as a contributing factor for children's poor motor and cognitive development. Cohort studies have shown stunting before age 2-3 years predicts poorer cognitive and educational outcomes in later childhood and adolescence [31,32]. Recent research by Black et al found that in low-income countries, suboptimal breastfeeding results in approximately 800,000 infant deaths annually, or 11.6 percent of deaths among children under ve years of age [31]. Rwanda has made signi cant reductions in under-ve mortality from 196 per 1000 live births in 1993 to 26 per 1000 live births in 2020 [33]. Further, changes in attitudes can be accomplished by informing the protective effects of EBF practices ''starting within the rst hour of birth, longer-duration breastfeeding is associated with protection against childhood infections, increases in intelligence and reductions in the prevalence of overweight and diabetes [24]. Breastfeeding protects against breast cancer and improves birth spacing'' [34]. An 8% global rise in EBF to six months, however, it is estimated to have decreased child mortality by 1,000,000, decreased fertility by 600,000, and saved billions of dollars in breast milk replacement in nations [35]. Married mothers were likely to have good practices more likely than single mothers. Similar ndings have been reported previously in Ethiopia [36]. Furthermore, good practice increases with the number of ANC visits this nding was similar to the studies conducted in Tanzania and Zimbabwe respectively [25,37] Apart from mortality, a lack of EBF has been associated with considerable morbidity. Mix fed children has a higher prevalence of diarrhea and respiratory infections than EBF children [31,37]. Appropriate complementary feeding recommended by the WHO to start at 6 months of age when breast milk is not su cient to maintain a child's energy, and nutrition requirements should be encouraged [37]. Therefore, breastfeeding strategies need to be improved and scaled up, especially in the early initiation of breastfeeding and EBF. There is also a need to raise community awareness and education about the bene ts and adverse effects of mixed feeding [38]. EBF-support policies such as the Baby-Friendly Hospital Initiative (BFHI) and the Infant, and Young Child Feeding Initiative (IYCF) that have been supported by the Government of Rwanda do not seem to operate optimally and there is a need to ensure effective implementation of these initiatives. Besides, beyond the normal facility channel, creative strategies to increase women's awareness and knowledge of breastfeeding in general and in EBF are needed. One solution that may be suggested to Rwanda will be to use community EBF promotion peer counselors or women's groups that have been shown recently in community trials to increase the length of EBF [14]. It was noted that there was a high antenatal attendance during a previous pregnancy, and the majority of the women delivered at a health facility. However, there are opportunities to counsel women on EBF during antenatal and postpartum periods, as few women reported receiving advice on EBF.
Teaching knowledge without change in attitude is a failure, which agencies involved in health education should seriously consider. Thus, there is a strong need that when knowledge is conveyed it should be done in a way so that the attitude also changes for the good and help to improve the practices of EBF. We may have given the right information, but our research points out those efforts are needed to improve attitudes.

Study Recommendations And Limitations
The recommendation includes that in a future investigation should be done on both breastfeed and nonbreastfed mothers to detect lacking and area to work it will increase their knowledge, attitude, and encourage EBF practices. However, health education sessions should promote teaching mothers about the bene ts of EBF. Also, health education should be provided at the community level to obtain an optimal EBF rate. Rwanda has a higher rate of EBF but it is still below the WHO recommendation, it is time to discover cultural and traditional practices that lead to the suboptimal practice of EBF. EBF promotion should be given attention in health planning: health care providers and decision-makers should comprehensively address these issues to improve EBF practices in the community. Improving access to recommend child feeding information during routine child and maternal health services and strengthening antenatal and postnatal nutrition therapy. There is a need for greater efforts to promote and support the healthy practice of exclusive breastfeeding. The planning of public health interventions to promote longer and increase the number of mothers who adhere to achieve the better development of children, and also to implement the policymakers, an intervention that could improve knowledge, attitude, behaviors, and practices of women concerning EBF.
The limitation of a cross-sectional study design is discussed in this research. However, the limitation of the study is that it was conducted in government hospitals involving mothers that went for antenatal and postnatal care hence, the ndings of this study may not be representative, of the situation of EBF in the entire community. Another limitation is that the questionnaire was long, which led to the failure to report complete data concerning some questions. The use of convenience non-random collection sampling technique interfered with the representativeness of the collected data and self-reporting may have presented recall bias to the study Finally, this study considered a sample of mothers who attending MDH which could lead to different results than those of analogous samples in the other health facilities.

Conclusions
The ndings of this study the knowledge, attitude, and practice of mothers towards EBF was found to relatively high. The prevalence of EBF is still below WHO's recommendation. The ndings of this study highpoint to the strength that policymakers and healthcare providers direct their efforts to provide evidence-based information and recommendations to women on the bene ts of breastfeeding. These results may be supportive of policymakers and managers as they plan educational interventions on breastfeeding during both pregnancy and hospital admissions during delivery. Implementing these efforts is fundamental to growing rates of exclusive breastfeeding for the rst six months of a child's life and the subsequent improvement in both the women's and the child's health outcomes.