Determinants of Severe Acute Malnutrition among Children age 6-59 Months Old in Two Public Hospitals, North West Ethiopia: a Case Control Study

Introduction Globally sever acute malnutrition affects 16.4 million under five children and more than one quarter of those children live in Africa. In Ethiopia, about 3% of children are severely wasted and continues to be persistent over the past 15 years. To implement an effective intervention, it is essential to identify predictors predispose to it. This study therefore, aimed to identify determinants of severe acute malnutrition among under five children in selected public health facilities, Northwest Ethiopia. Methods Institution based; age matched case control study was conducted on 104 cases and 208 controls. Bivariate and multivariate analyses were done using conditional logistic regression to identify predictors. Variables having P-value ≤ 0.2 during binary analysis were entered into multivariate analysis. P value < 0.05 was considered as statistically significant. Results Children from households of large family size(AOR=2.7, 95% CI: 1.06 – 6.9), having monthly income less than 1500 birr (AOR = 5.17, 95% CI: 1.7-15.3), which are food insecure (AOR = 2.9, 95% CI:1.17-7.28)), which didn’t receive any nutrition information (AOR= 3.47,95% CI: 1.14 - 7.10), didn’t practice exclusive breastfeeding (AOR = 2.69, 95% CI: 1.18 - 6.10), and practice infrequent hand washing (AOR= 7.6 95% CI:2.44-23.6) as well as children who had history of diarrhea two weeks prior to the survey (AOR 3.2, 95%CI:1.4-7.2) were more likely to suffer from severe acute malnutrition. Conclusion Family size, monthly income, food security status, exclusive breastfeeding practice, access to information on child feeding, hand washing practice and history of diarrhea were identified to be predictors of severe acute malnutrition. Due emphasis should be given to promoting family planning, improve household livelihoods and food security, strength awareness creation on exclusive breastfeeding and frequent hand washing practices as well as prevention of diarrhea.

* Corresponding author 12 adoniadamtew@gmail.com (AD) 13 14 15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  4   90 Study conducted in deferent part of the world shows that the predicators of acute malnutrition 91 were inadequate care for children like poor infant and young child feeding, household food 92 insecurity (7)(8)(9) incomplete/lack of immunization (10,11). poor environmental health 93 condition (inadequate and unsafe water supply, and poor environmental sanitation) and 94 mothers habit of less frequent hand washing and absence of latrine (7). 95 Though acute malnutrition is one of the public health problems and persistent over time in 96 Ethiopia particularly in the study area, available studies on the determinants of SAM are 97 limited. To implement an effective and efficient intervention, which can reduce the high 98 burden of sever acute malnutrition, it is essential to understand factors predisposing to it in 99 different part of the country. Therefore, this study is intended to assess the determinant   Source and study population 115 All children 6-59 months of age with SAM or without malnutrition that have been admitted 116 and treated at therapeutic feeding units (TFU) and/or other pediatric units of the two hospitals 117 were the source population. The study population were 312 children age 6-59 months who 118 were selected for this study from both hospitals during the study period.

119
Inclusion and exclusion criteria 120 Children in the age of 6-59 months who were admitted in the two hospitals due to severe 121 acute malnutrition with their care takers/mothers were included into the study as cases.

122
Children aged 6-59 months with normal nutritional status and who had visited pediatrics 123 units for different health care services during the study period were included as controls.

124
Children in the age of 6-59 months with disability (physical deformity and handicapped) and 125 children who had sudden shock (unconscious) during the study period were excluded.

126
Children with secondary undernutrition due to other pathological disorders and with other 127 causes of edema were also excluded. Sample size and sampling procedure 131 The sample size was calculated using two population proportion formula, by taking maternal 132 autonomy in decision making as major associated factor from other study and proportion of 133 mother who had no autonomy in decision making to be 85.2% among cases and 69.4% 134 among controls and detecting OR of 2.545 (12), with 95% CI, a power of 80%, a two to one 135 ratio of control to case (2:1) and 10% non-response rate. By referring the above, the sample 136 size was 276 (92case and 184 controls). But 104 cases and 208 controls were included 137 assuming it could increase the power of the study.

138
The sample size was proportionally allocated to the two hospitals based on the average 139 monthly follow rate of under five Children with severe acute malnutrition. Out of the total 140 sample, 66 cases and 132 controls were allocated to Felege Hiwot referral hospital, while 38 141 cases and 76 controls were allocated to Debre Tabor referral hospital. Children in the age of 142 6-59 months admitted due to SAM in the therapeutic feeding units of the two hospitals during 143 the study period were included until the calculated sample size was saturated. The controls 144 were match to the case within ±2 months. Once a case child was admitted, his/her 145 mother/caretaker was interviewed immediately and then the first two controls matched with 146 one case of a similar age in months were randomly selected immediately after the admission 147 of a case.

148
Data collection and quality assurance procedures 149 The data were collected from all study participants using interviewer administered structured 150 questionnaire. The questionnaires were originally prepared in English and then translated to   Eggs; Legumes, Pulses or nuts; milk and milk products (14).  To assure the quality of the data, properly designed, pretested questionnaire was used.

169
Training was given for data collectors and supervisor for two days.   (Table 1).

Discussion
Factors contributing to SAM are multifaceted and identifying the determinants of SAM in the study area can be very important to implement effective intervention. In this study, the determinants identified for SAM were family size, monthly income, household food security status, exclusive breastfeeding practice, hand washing practice and diarrhea history two weeks prior to the survey.
Accordingly, children from households with large family size (>5) were 2.7 times more likely to be affected by SAM as compared to children from household with smaller family size. The finding was in line with a study done in Sudan, which revealed that having more than four family members in the household was associated with acute malnutrition (15). This might be because of the fact that increased number of families placed a heavy burden on scarce household resources, particularly on financial and food. That might make difficult to fulfill the dietary need of the children. Besides, the increased number of family might reduce the time and quality of care received by the children.
Children from families who had monthly income of less than 1500 birr were nearly 5 times more likely to develop SAM as compared to children from families who had monthly income of 1500 birr and above. This finding was comparable with the study conducted in Jimma Zone, Ethiopia; which points out that children whose family monthly income of less than $50(1450 Ethiopian Birr) were 6 times more likely to be affected by SAM (16). This can be explained by the fact that households with low income could not afforded to buy food for consumption which results in inadequate diet that leads to child malnutrition.
The odd of severe acute malnutrition in children increased when the household was food insecure.
Children from food insecure households were 3.7 times more likely to be affected by SAM than children whose families were food secured. This is consistent with a study done in west Gojjam, Northwest Ethiopia (17) and Haromaya District ,Eastern Ethiopia (18). In contrary, a study from Gambela town, Western Ethiopia and Sekela District , Western Ethiopia revealed that there was no significant association between household food insecurity and acute malnutrition in children aged under 5 years (19,20) . In this study, household food insecurity was found to be significant determinant of SAM and this might be due to limited availability of food or no economic access to purchase it which might lead to reduced quantity and quality of diet. Therefore, food insecure households might not satisfy the dietary need of household member, especially those under five children because they are at greater risk of malnutrition due to their higher nutrient demand for their growth. (21,22) In this study exclusive breastfeeding practice was one of the determinants of SAM among children.
Children who were not exclusively breastfed were nearly 4 times more likely to suffer from SAM.
Similar findings were also documented in other studies done in Oromia Region and East rural Ethiopia (7,23). The study done in EAG States, India also indicated that children who were exclusively breastfed were found to be 16 percent less likely to be mild wasted and 48 percent less likely to be not wasted (24). This could be explained by the fact that children who were not exclusively breastfed had lower chance of preventing infections as breast milk has many immunological properties that are likely to protect against infections in infancy. A study done in Tanzania confirmed that exclusive breastfeeding reduce the risk of diarrhea dysentery (25).
Besides, there might be contamination of bottles and foods that were early introduced to the child which contributes to higher risk of diarrheal disease in children.
Access to nutritional information on child feeding practice was also another determinant of SAM identified in this study. Children whose mothers/caretakers didn't have any nutrition information on child feeding practice were nearly 3 times more likely to develop SAM than those children whose mothers/caretakers had information on child feeding practice. This finding is also supported by other study which is done in Botswana (26). Lack of information on child feeding might lead to inappropriate child feeding practice which then could affect the nutritional status of the children.
Hand washing practice was another factor that was found to be a significant determinant of SAM.
Children whose mothers didn't wash their hands before serving food, after defecation, before feeding the child and after cleaning the child were 7 times more likely to develop SAM as compared to children whose mothers washed their hands in each activity. Similar findings were documented in studies conducted in Oromia region, West Ethiopia and Machakel district , Northwest Ethiopia (7,27). This might be due to the fact that poor hand washing practice might lead to contamination of foods and that might increase a risk of infections and diarrheal diseases that in turn might lead to poor appetite and poor absorption of nutrient and finally might expose the child to SAM.
Moreover, history of the diarrhea two weeks prior to the survey was identified to be significant predictor for occurrence of SAM. The odd of SAM increased by three fold in children who had diarrheal disease two weeks prior to the survey than those who didn't. Similar findings were reported from studies conducted in Shashogo ,southern Ethiopia and Oromia, West Ethiopia (7,12).This could be due to the fact that diarrheal disease might lead to loss of appetite, loss of nutrients from the body and poor absorption of nutrients consumed.

Limitations of the study:
Since the questions were relied on the memory of the mothers/ caretakers, this might introduce recall bias. To minimize this bias the recalling period was made shorter for some variables like history of diarrhea within two weeks prior to survey and about food consumption during last 24 hours which is a reasonable recalling period. However, for some of the variables like exclusive breastfeeding practice and vaccination received might introduce recall bias.
There might be also selection bias because controls were selected from health facilities.

Conclusion
The findings of this study confirmed that family size, monthly income, food security status , exclusive breastfeeding practice, access to information on child feeding, history of diarrhea two weeks prior to the survey and hand washing practice were significant determinants of severe acute malnutrition . due attention should be given to improve the knowledge and practice of parents towards exclusive breastfeeding and family planning by integrate breastfeeding promotion and support throughout the maternal and child health continuum, particularly in the prenatal and postpartum periods. It is also important to strengthen prevention and control of diarrheal disease through promotion of good hygiene (hand washing) and exclusive breastfeeding practices in the community. Creating income generating activities to improve the income and food security status of the poor households is also highly recommended. Since determinants of SAM were identified to be multifaceted and it cannot be addressed by a single sector only, the government should build strong multi-sectorial collaboration (Health, WASH and Agricultural sectors) to address the problem of severe acute malnutrition.