Design and Sample
This was a community based matched case-control study. The sample comprised 50 cases and 100 controls children from Satar community aged between 6 months to 59 months. To find 50 cases 664 children were screened. Children below five years from Satar (Santhal) community residing in Jhapa district of Nepal whose guardian gave consent were included in our study. The cases were the severely acute malnourished children below the age of 5 years ascertained by using the weight for height Z-score < − 3SD below the median according to WHO 2006 growth standard and the controls were sex and age-matched children aged 6–59 months from same community without malnutrition [19]. Those children whose guardian refused to provide consent or were not willing to participate in the study and children who were severely ill or were with co-morbidities were also excluded from the study.
Study Duration And Selection Of Participants
The study duration was from September 2019 to February 2020. The sample size was calculated assuming the prevalence of controls who were bottle fed at 18% with an odd ratio (OR) of 4.56 from the study in Bara, Nepal [20]. As this study considered 5% significance (α = 0.05) and 90% power to determine the sample size using formula for case control study,

The sample size calculated for each group was 43 children. Since the ratio of selecting cases and controls was 1:2 and after adding 10% non-response rate overall 50 cases and 100 controls samples were finalized. Multi-stage sampling technique was done. First, the Jhapa district was purposively selected. It consists of 8 urban and 7 rural municipalities. An indigenous group of Satar population was extracted from each municipality according to Central Bureau of Statistics, 2011 report. Among these fifteen municipalities of Jhapa District four municipalities were selected randomly, which consist of two urban and two rural areas. In these municipalities, anthropometric measurements (weight and height) of the children aged 6–59 months were taken and compared with WHO 2006 simplified growth standards to categorize the children as a case (weight for height < − 3SD) or a control (weight for height > − 2SD).
The cases and controls were matched in a 1:2 ratio with similar age intervals of (6–8, 9–12, 12–14, 15–17, 18–20, 21–23, 24–26, 27–29, 30–32, 33–35, 36–38, 39–41, 42–44, 45–47, 48–50, 51–53, 54–56, and 57–59) months and sampling frames of cases or controls were prepared for each municipality. The sample population was screened unless 50 cases of severe acute malnutrition were found and the required controls were randomly selected from the sampling frames through simple random technique. Only one younger child was selected per household during data collection.
Questionnaire
A semi-structured questionnaire was developed. The questionnaire includes age and sex of a child, religion of the family, mother's educational level, father's educational level, maternal age at the time of delivery, family size, economic status and birth interval, colostrum feeding, initiation of breast feeding, exclusive breast feeding, initiation of complementary feeding, frequency of breast feeding, bottle feeding. Food insecurity was evaluated using the validated Household Food Insecurity Access Scale (HFIAS).
HFIAS: In this study, household food insecurity was assessed using Household Food Insecurity Access Scale (HFIAS) developed by the Food and Nutrition Technical Assistance (FANTA) project. The questionnaire consists of nine occurrence questions that represent a generally increasing level of severity of food insecurity (access), and nine “frequency-of-occurrence” questions that are asked as a follow-up to each occurrence question to determine how often the condition occurred. The frequency-of-occurrence question is skipped if the respondent reports that the condition described in the corresponding occurrence question was not experienced in the previous four weeks. Some of the nine occurrence questions inquire about the respondents’ perceptions of food vulnerability or stress (e.g., did you worry that your household would not have enough food?) and others ask about the respondents’ behavioral responses to insecurity (e.g., did you or any household member have to eat fewer meals in a day because there was not enough food?). The questions address the situation of all household members [21].
Date of Birth: Majority of the mothers of under 5 children in indigenous community of Satar were unable to show child health card or birth certificate for verifying date of birth. Therefore, the date of birth of the child was asked and its validity was verified through sibling method (birth order of child), local fist and festival calendar. The reliability was checked with other supportive questions like age of mother at pregnancy, and age of mother at present.
Assessment Of Anthropometric Measurements
Weight and recumbent length/height of the children were taken during daytime and were conducted for the 3 times to finalize the final measurement to record.
Height measurement: If a child was less than 2 years old, recumbent (lying down) length was measured. If the child was aged 2 years or older and was able to stand, standing height was measured. The height measurement was taken on flooring that was not carpeted and against a flat surface such as a wall. The child was measured without shoes, bulky clothing, hair ornaments, and unbraid hair that interfered with the measurement.
Weight measurement: A digital scale was used for the accurate measurement. The scale was placed on firm flooring (such as tile or wood). If the child was less than 2 years old or was unable to stand, the mother of that child was asked to stand on the scale along with her child in her arms. Then, the mother was asked to step on the scale to be weighed alone. She needed to adjust any long garments that covered the display. If the child was 2 or more than 2 years or was able to stand, he/she was asked to remove shoes and heavy clothing, such as sweaters. The children were asked to stand with both feet in the center of the scale and stand still, pointing their head firmly.
Validity And Reliability
Validity and reliability of the study was ensured by using Nepali translated semi-structured questionnaires and HFIAS used in NDHS 2016 [22]. Guide, co-guides and subject experts made necessary changes as per the suggestion from the departments. Furthermore, the questionnaire was pretested on 15 children (5 children with SAM and 10 children without SAM) which was 10% of the sample size. Pretesting was done in similar setting in Biratnagar, Nepal, and necessary correction was made in the questionnaire. During data collection, respondents were assured for confidentiality of information, and that will not be used other than the study purpose. The data was collected by investigator herself.
Statistical analysis
All interviewed questionnaire was indexed and kept in file. Master chart and coding list was prepared before entering the data. Collected data were entered into Microsoft Excel, 2007 and were converted into Statistical Package for Social Sciences (SPSS 11.5 version) for statistical analysis. Data were entered after every day’s work, and were checked for validity.
All the numerical data variables were checked for normality using the Kolmogorov-Smirnov test, to assume parametric nature of the data. The independent t-test was used for parametric data. Descriptive statistics were conducted on demographic data collected. Bivariate and multivariate conditional logistics regression analysis was conducted to compare different characteristics between cases and controls taking the age group and sex matching into consideration. Statistical significance was tested with 95% confidence interval and p < 0.05 was considered significant. Further to identify the strong predictors for determining SAM, confounders were adjusted by applying multi-collinearity diagnostic test between the independent variables before logistic regression was applied. Decisive criteria were set out to be a variance inflation factor (VIF) value of > 3. The variables within the criteria were only used for logistic regression.
Ethical Clearance:
It was obtained from the Institutional Review Committee (IRC) of BPKIHS Nepal. The purpose of the study and the procedure was well explained and written informed consent was gained before starting the data collection. The participants were also informed that their participation was voluntary, and can withdraw at any moment. Moreover, they were assured regarding their anonymity, and the confidential treatment of their responses. All information were kept confidential, and the consent forms were number coded for identification.