Minimum Acceptable Diet Practice and its Associated Factors among Children’s Aged 6–23 Months in Rural Communities of Goncha District, North West Ethiopia, 2020

Introduction: After the rst 6 months breast milk is no longer sucient to meet the nutritional needs of the infant. Therefore, complementary foods should be added to the child’s diet. Feeding children’s with diversied diet is practiced improperly in developing countries including Ethiopia particularly in the rural community. Interventions are intended to undertake the nutrition problems in children in the country. However, the progress was not satisfactory, particularly; minimum acceptable diet has increased from 3% to 7% in a decade (2005-2016). Objective: To determine the proportion of minimum acceptable diet practice and its associated factors among children’s aged 6-23 months in rural communities of Goncha district, 2020 Methodology: Community based cross-sectional study was employed at rural communities of Goncha district from June 15 to July 15 2020. Multi stage sampling technique was used to select study subjects, and interview administered structured questionnaire was used to collect the data. Data was entered by Epi Data version 4.0.2 and exported to SPSS 20 for analysis. Bivariate and multivariable logistic regression analysis was used to see the association between minimum acceptable diet and independent variable. Then, P-value <0.05 with 95% CI on multivariable logistic regression analysis were used to identify the independent predictor of outcome variable Result: A total of 430 mothers who have children aged 6-23 months were included in the analysis with 98% of response rate. About 12.6% of children’s aged 6-23 months received the recommended minimum acceptable diet. Children whose mothers who had formal education [AOR= 2.7, 95%CI (1.133, 6.231)], institutional delivery [AOR= 4.5, 95%CI (1.986, 10.362)], media exposure [AOR=2.6, 95%CI (1.303, 5.291)] and higher household wealth index [AOR= 2.5, 95%CI (1.139, 5.90)] were signicantly associated with minimum acceptable diet. Conclusion: The practice of minimum acceptable diet in the study area was inadequate and very low according to notional and world health organization’s recommendation. So, strengthening institutional delivery, improving the wealth of the community and exposure to media, and nally

Minimum acceptable diet (MAD) is an indicator for evaluating child feeding practices presented via World Health Organization. It is a combination of the minimum dietary diversity and minimum meal frequency (3). Infant and young children should have a minimum acceptable diet (MAD) to ensure appropriate growth and development, otherwise they are vulnerable to under nutrition especially stunting and micronutrient de ciencies, and increased morbidity and mortality (1,4). In addition to lack of adequate and balanced diet, there are a number of people who are attacked by diseases that occur due to proteinenergy malnutrition and due to lack of disease protecting foods. Children, pregnant women and lactating mothers are most vulnerable by the problem (5).
Although Ethiopia is a manufacturer of a diversity of agricultural products, still the countries is known in the world with the highest number of malnourished population. Malnutrition is primarily seen among the rural residents and the prevalence among children was 48.5% (6). The Ethiopian government has been try to improve child feeding practices by implementing the national nutrition program of child feeding practices and a multi-sectorial plan of nutrition intervention to end child under nutrition in Ethiopia by 2030 (7). Despite this the progress was not satisfactory, particularly national prevalence of MAD practice was 7%. The problems are higher in rural areas and also signi cant variation exists between regions.
Children in urban areas (19%) are more likely to feed than those in rural areas (6%) and the proportion of children who receive the minimum acceptable diet is highest (27%) in Addis Ababa and lowest (2%) in Amara region (2,8).
Even though studies were conducted about the determinants of the optimal complementary feeding practices in Ethiopia, however, insu cient information was documented about minimum acceptable diet practice and its associated factors independently, and most of the data were not representative especially for rural communities. Also, as far as researcher's knowledge concerned, no documented data were accessible speci cally in the study area. Therefore, this study was planned to assess minimum acceptable diet practice and its associated factors among children's aged between 6-23 months in rural communities of Goncha district, North West Ethiopia, 2020 Methodology Study area and period The study was conducted from June 15 to July 15 2020 in rural communities of Goncha district, which is located in East Gojjam Zone, Amhara region. The full name of this woreda is called as Goncha Siso Enesie, is one of the 21 woredas of East Gojjam zone and located 154 km East of Bihar Dar, the capital city of Amhara region, and 335 km far from Northwest of Addis Ababa. Goncha Siso Enesie is boarded on the South by Enarge ena Enawuga, on the West by Hulet Eju Enesie, on the North by Abay River which separates it from North Gonder zone, and on the East by Enebisie sar midir. This woreda is administratively structured by 43 Keble's (41 rural and 2 urban Keble's). Of these rural Keble's 12 are lowlands and 29 are highlands, and major town in the woreda is Gind Weyin. Almost all of the district's population consists of subsistence farmers heavily depending on rains for their agriculture. The district is one of the areas in the region, known to experience chronic food insecurity due to variable rainfall pattern, and the population especially who are resided in lowland Keble's are users of government safety net program. In the woreda there are 8 governmental health centers with ratio of one health center to ve Keble's and 43 health posts. Due to lack of accessibility to transportation, most of rural community is forced to go longer time with foot to get health care service, and they are exposed to deaths related to treatable diseases. Goncha woreda is inhabited by total populations of 210,423, of these 108,909 are females and 42,569 are in the reproductive age group

Sample size determination
The number of sample required for the rst objective of this study was calculated using single population proportion formula; by considering the following assumptions: By adding 2 design effect and 5% non-response rate, N = 439 Where ni = initial sample size required for the study Z = standard normal distribution (Z = 1.96) with con dence interval of 95% p = 8.6% taken from proportion of MAD in Dembecha district (9)  To ensure the adequacy of sample size, Epi info was used to calculate sample size for factors associated with minimum acceptable diet Finally the maximum sample size was selected from sample size calculated from rst objective (proportion of minimum acceptable diet). So, the total sample size required for this study was = 439 Sampling procedure Multi stage sampling technique was used to select the study subjects. First total rural Keble's in Goncha district were strati ed in to highland (dega) and lowlands (kola) based on their predominating agroecological characteristics (classi cation obtained from Goncha woreda health o ce). Second, from the two strata, 8 from 29 highland Keble's and 4 from 12 lowland Keble's were selected by using lottery method. Then, total sample size was allocated proportionally and sampling interval-"K" was determined. K = N/n = 1840/439 = 4.19 ~ 4 Where, K = sampling interval N = total number of 6-23 month age children's in the selected Keble's n = total sample size required in this study Finally, 439 children's aged 6-23 months were selected by using a systematic sampling method based on sampling frame created by Health Extension Worker's (HEW's) record. After randomly identi ed the rst child, we preceded to the second child every 4 interval. proportion of children's aged 6-23 month who had at least minimum meal frequency and minimum diversi ed diet during the previous day (10) Minimum meal frequency proportion of breast feeding and non-breast feeding children's aged 6-23 months who receive soft, solid and semi-solid foods (but also including milk feeds for non-breast feed child) in the last 24 hours. Breast feed Infants aged 6-8 months 2 times in the last 24 hours; breast feed infants and young children aged 9-23 months 3 times in the last 24 hours. For non-breast feeding infants and young children aged 6-23 months at least 4 times in the last 24 hours (3)

Dependent variable
Minimum dietary diversity: proportion of children's aged 6-23 month who receive ve or more food groups out of the eight food groups in the last 24 hours. These foods groups used for this indicator are: breast milk, grains, roots and tubers; legumes and nuts; dairy products (milk, yogurt); Flesh foods (meat, sh, poultry and liver/organ meats); eggs; vitamin A-rich fruits and vegetables; and other fruits and vegetables. Quality and quantity of any amount from those groups can be considered as su cient to count (10) Maternal knowledge on IYCF practice knowledge of mothers on infant and child feeding practice was measured based on ten knowledge questions. Each correct answer (yes) earned one point, and any wrong answer (no) got zero. The calculated knowledge score ranged from 0 to 10 points. Those who score above the mean (5.7 ± 2.6 standard deviation) was categorized as knowledgeable and those who score below the mean was categorized as not knowledgeable (11) Exposure to media media exposure was categorized as satisfactory or unsatisfactory. Mothers who listen to radio, or watched television at least once a week was satisfactory media exposure, and otherwise unsatisfactory (12) Data collection tool and procedure Primary data was collected using interviewer-administered and structured questionnaire from selected mothers who have had child aged 6-23 months through face to face interview. The data collection tool has six parts; Socio-demographic characteristics, maternal and child health characteristics, maternal knowledge on IYCF practice, twenty four hour recall child feeding practice, household level characteristics and wealth index related characteristics. Twenty four hour recall method for dietary diversity and meal frequency questionnaire was adopted from WHO IYCF indicator assessment tool with some modi cation to t with the context (4). the other ve parts of the questionnaire were adopted from EDHS (2016) and other related literatures (11, 13) 8 diploma midwife and nurse professionals for data collection and 2 1st degree public health o cers as supervisor were participated in the survey.

Data quality control
Questionnaires were prepared in English and were translated in to local language (Amharic), and again it was translated back in to English to keep its consistency. Data collectors were trained on the aim of the research, content of the questionnaire and how to conduct interview to increase their performance in eld activities for one day prior to data collection period. The collected data was checked regularly by supervisors, and principal investigators were followed through phone contact either with supervisor or data collectors for its completeness and consistency. Finally the collected data was cleaned, coded and entered through epidata version 4.0.2.

Data processing and analysis
Data was entered by using Epi Data entry client version 4.0.2 and exported to SPSS 20 statistical package for analysis. Data cleaning was performed to check for consistencies and values. Dietary diversity score was computed out of eight from eight food groups, and household economic status was measured by constructing wealth index through principal component analysis (PCA) and ranked as low, middle and high wealth index. Descriptive statistics (frequency, mean, standard deviation and Percentage) were used to describe socio-demographic characteristics of the study population. Tables and graphs were used for data presentation. Bivariate logistic regression analysis was done to see the association of independent variables on dichotomous outcome variable (MAD, 1 = met minimum acceptable diet, 0 = not met minimum acceptable diet). To control the effect of potential confounders and to identify the independent effect of the explanatory variable on minimum acceptable diet, variables with p value less than 0.25 in bivariate logistic regression were considered as candidate for multivariable logistic regression analysis. Before going on multivariable logistic regression analysis multi colliniarity was checked by co linearity diagnostic tests and variables with variance in ation factor (VIF} less than ten, and variables that ful ll sample size assumptions were entered in to the model. At this level, model tness was checked with hosmer and lemeshow goodness of t at p value ≥ 0.05. Finally, variables with p value less than 0.05 and 95% CI were considered as the predictive for outcome variable. The strength of associations and statistical signi cances between independent variables and outcome variables were expressed using OR and 95% of con dence interval respectively.

Result
Socio-demographic characteristics of study participant Among a total of 439 sampled subjects, 430 children's aged between 6 and 23 months with mother were enrolled in the study that makes a response rate of 98%. Among children's aged 6-23 months, 217 (50.5%) were females and 167(38.8%) were categorized in the age group between 18 and 23 months. The mean age of children was 15.3 ± 5.4 (SD) months, and half of children's aged 6-23 months were in the birth order of second to fourth 218(50.7%). Out of mothers of children aged 6-23 months, majority of 365 (84.9%) mothers were married. With regard to educational status of children's parents, About 283 (65.8%) and 262(60.9%) of mothers and fathers had no formal education respectively. Finally, from total children's parents, about three fourth 320 (74.4%) of fathers were farmers followed by merchants 56(13.0%), and less than half 182 (42.3%) of mothers were found to be house wife's and about 173(40.2%) of mothers were farmers (see Table 1). Greater than three fourth, 341(79.3%) of children did not received monthly growth monitoring and promotion service (see Table 2)  Table 3).

Factors in uencing minimum acceptable diet practice
To identify factors associated with minimum acceptable diet practice, bivariate and multivariable logistic regression analysis was done. On binary logistic regression analysis mother education, father education, agro-ecology, family size, place of delivery, postnatal care visit, pre lacteal fed for a child, decision making in household, exposure to media, knowledge of mothers on IYCF practice, wealth index, antenatal care visit and time of initiation of complementary food for a child with p value less than 0.25 were considered as candidate for multivariable logistic regression analysis. Before going on multivariable logistic regression analysis, antenatal care visits and time of initiation of complementary food for a child were not ful lled the sample size assumptions of logistic regression and they were removed, and eleven variables were selected for nal analysis. Finally, model tness was checked by hosmer-lemeshow goodness of t (which was p-value = 0.37) and after that variables with p value less than 0.05 and 95% CI i.e. mother education, place of delivery, exposure to media and wealth index were found to be the potential predictors of meeting minimum acceptable diet.

Discussion
The nding of this study revealed that 12.6% with 95% CI (9.5, 15.7) of children's aged 6-23 months were received the recommended minimum acceptable diet. The result of this study was higher compared to study conducted in Northwest Ethiopia, Dembecha (8.6%) (9). The variation might be due to difference study period. The above study was conducted in populations were only orthodox religion followers during the fasting season in which feeding habits might be reduced either in food diversity, especially animal source foods or meal frequency which under estimate the nding when compared to other periods. Also the above study was conducted in dry season what we call it "winter" in which nutritional availability of most fruits and vegetables might be low compared to seasons especially "summer" a period in which this study was conducted.
The nding was also higher than EDHS report of 2016, only 7% of children's aged 6-23 months received minimum acceptable diet (2). The difference might be due to EDHS was done nationwide on a larger sample size. In addition, the EDHS were conducted on culturally different population, which may underrate child feeding practices while this study was conducted on an almost culturally homogenous population with similar feeding practices. The results of this study was higher than studies conducted in Ethiopia multilevel analysis report of EDHS 2016 (6.1%), Malawi (8.36%), Nigeria (7.3%) and Philippines (6.7%) of children aged 6-23 months received the recommended minimum acceptable diet (8,9,(14)(15)(16).
The reason for high percentage of feeding practice in this study area might be due to variation in study design, data collection period, and nutrition education with media and health extension workers might play a major role for increasing community awareness towards appropriate child feeding practice (17).
The result of this study was inconsistent with study conducted in North Shoa (12.5%) and Northern Ethiopia (11.9%) of children's meet MAD (18,19). The nding of this study was lower than study done in wolayita Sodo town (Southern Ethiopia) 21.1% of children's consumed minimum acceptable diet (20). The variation might be because of difference study settings and study period; this study was conducted in the rural communities where as the above study was conducted in urban communities, as communities from rural area are less likely to feed a minimum acceptable diet than people reside in urban area (2). Also the difference might be due to higher non educated mothers were participated in this study, on the contrary higher numbers of educated participants were included in the above study. Also the study was lower than study done in Bensa District, Sidama region, 16% of children's consumed minimum acceptable diet (21). The variation might be due to different in sample size and socio demographic characteristics.
This nding was very low compared to national and world health organization recommendations of minimum acceptable diet. This implies that majority of children's resided in the study area might be more vulnerable to under nutrition, especially stunting and micronutrient de ciencies, and that also increased morbidity and mortality (1). The result was also lower than study conducted in different countries; Ghana, Uganda and Kenya in which 29.9%, 23.9% and 48.5% of children's received recommended minimum acceptable diet, respectively (22)(23)(24). Lower nding in this study area might be due to differences in study design, sample size, study period and difference in socio demographic characteristics. Hence, most of the above studies were systemic reviews from demographic health surveys. Also the nding was low compared to 2020 global nutrition report (18.9%) (25). The variation might be due to difference in sample size and socio demographic characteristics Mother educational status was signi cantly associated with minimum acceptable diet. Based on this study, mothers who had formal education were 2.7 times more likely to provide minimum acceptable diets for their children compared to mothers who had no formal education. This nding was supported by a study done in Dembecha (9). This might show that education enables mothers to know the bene ts of the practice of child feeding and plays an important role in meeting minimum acceptable diet standards.
However, this nding was not supported by the study done in North Shoa, Oromia region and multilevel analysis report of EDHS 2016 (8,18) This study found that children's born in health facility were 4.5 times more likely to receive minimum acceptable diet than those who born at home. This result was similar with study done in Northwest Ethiopia (9). This might be due to health professional counseling on appropriate child feeding after delivery on health facility increases mothers awareness on practice of minimum acceptable diet; Hence mother's awareness on appropriate child feeding practice who got from health professionals have had a better child feeding practices than their counterparts (17) This study revealed that children's whose mothers who were exposed to media, i.e. watched television, listen to radio every day or once a week, has 2.6 times more likely to meet minimum acceptable diet than those children's of mothers who watched television and listen to radio less than once a week or not at all.
This nding was similar to other ndings in North Shoa, Oromia region and multi-level analysis report of EDHS 2016 (8,18). This might be due to the fact that the currently Ethiopian ministry of health and its partners promote child feeding practices through radio, television and family health cards. This might enhance the mother's awareness on feeding a minimum acceptable diet to their children. Also, this might be because of mothers who have been exposed to the media having had better opportunity to access information on appropriate child feeding practices. This could improve the mother's capacity to challenge unfavorable information towards child feeding practices in the community and increase appropriate child feeding habits. On the other hand, this nding was not supported by the study conducted in Dembecha (9) The present study also revealed the association between household economy and minimum acceptable diet. Children's born from mothers with high wealth index were 2.5 times more likely to receive the recommended minimum acceptable diet than children's born from mothers with low wealth index. This result was nearly similar with study done in the Philippines (26), in which children's born from mothers in the middle wealth index were more likely to meet minimum acceptable diet compared to those children's born from mothers in the poorest wealth index,. The possible explanation of this signi cance association might be due to the limited food purchasing power to provide diversi ed diet to their children in peoples with lower wealth index, and also mothers in high wealth index were more likely provide nutritious food to their children compared to mothers from low wealth index households who were more focus on the quantity of food (10). This nding was not supported by other studies conducted in Dembecha, North Shoa and multilevel analysis report of EDHS 2016 (8,9,18) Conclusion Minimum acceptable diet practice among children's aged 6-23 months in the study area was low, almost one from eight children meet the recommended minimum criteria. So, child feeding practices in the study area were not achieved the national and world health organization's infant and young child feeding recommendation. Determinant Factors which signi cantly affect meeting of minimum acceptable diet were educational status of mother, place of delivery, exposure to media and household wealth index. This implies that the problems are range from individual to household level, and even may go through at large in the community level. First of all, we would like to thank Hawassa University College of medicine and health science department of Midwifery for giving the chance to conduct this thesis. Also our acknowledgment goes to Goncha woreda health o ce and health extension workers who are working in selected Keble's, for giving valuable information regarding study area. Our deepest gratitude also goes to data collectors and supervisors for their hard work and commitment during the process of data collection. Last but not least, we would like to thank study participants for their participation and giving their valuable time.