Study area, design and period
A community based cross-sectional study design was employed in Ambassel district from January to July 2018. It is found at 460 km away from the capital of Ethiopia, Addis Ababa. It has 1 urban and 23 rural Kebeles. The estimated population of the district is 165,000. Of these 77,480 (54%) were males and 66039 (46%) were females. Based on the reports of the district health office report in 2015/16, there are 11,669 children aged 6–59 months old. The major agricultural products cultivated in the district were grains, cereals, fruits vegetables and animal reared include ox, cow sheep, goat and hens.
Source and Study population
All mothers/care-givers having a child in the age range of 6–59 months who have been living in the district were the source population. All mothers/caregivers having a child in the age range of 6–59 months who have been living in the selected Kebeles were the study population.
Sample size calculation
The sample size was calculated using Epi Info version 126.96.36.199 by considering 95% confidence interval, 80% power, 5% margin of error and the estimated prevalence of wasting 18.6% (15) and design effect of 1.5. Thus, considering 5% non-response rate, the total number of study participants was 367.
One Kebele from better rainfall and geographic assets and one from low rain fall and geographic assets were randomly selected and included in the study. The sample size was proportionally allocated to each Kebeles. Before the actual data collection, list of the households from each Kebele was prepared with the help of health extension workers. Then, study participants were included in the study through random number tables. Apart from the rainfall distribution the two divisions were regarded as being very similar with regard to socio-demographic characteristics. Study participants were selected using systematic sampling technique.
Food security: if the household didn’t experience any of the questions from FAO/FANTA
Dietary diversity: is defined as the number of different food groups that are consumed over a given period of time (16). Dietary diversity is considered optimal, if an individual consumed diet at least the mean from food groups
Agricultural biodiversity: include the variety and variability of plants and animals that are necessary for food production and accessibility. It is the basis for food chain which contributes to food and livelihood security (8). Agricultural biodiversity is considered as good, if the household has edible plant or animal species above the mean
Data collection methods
Data were collected by open data kit (ODK) software through face to face interview with mothers/care-givers having a child aged from 6–59 months old using a structured questionnaire. The collected data were sent to the common server which was created before the start of data collection. Data collectors were Health extension workers and development agents. One MPH student who has training or experience on anthropometric measurement was recruited as a supervisor.
The age of the child was ascertained through reviewing the child health card, birth certificate, or baptism card. In situations where the mother/care-giver did not have the documents to ascertain the age of the child, they were asked to identify a child from the neighborhood who was born almost at the same time. In cases where there was more than one child in this age group, only one child was randomly selected and included in the survey.
Height was measured in a standing position, using a free standing height stadiometer. Height measurement was taken when a child was in a bare feet and heels, buttocks, shoulders and the back of the head touched the stadiometer. Two measurements were taken to the nearest 0.1 cm and the average was recorded to ensure data accuracy (17). The head was held comfortably erect with the lower border of the orbit of the eye being in the same horizontal plane as the external canal of the ear. The head piece of the measuring board was then pushed gently, crushing the hair and making contact with the top of the head.
Weight was measured using an easily portable weighing scale (SECA Germany). The scale was adjusted after weighing every five child by setting it to zero. The child dressed light cloth during weight measurement. Two measurements were taken to the nearest 0.1 kg for each child and the average was recorded.
Dietary diversity was taken by a repeated 24-hour recall. The 24-hour recall was administered to each child and repeated on a separate day. No prior notice of the repeat visit was given to care givers in case they altered their intake. The repeated 24 hour dietary recall was internationally used and validated (18). DDS was calculated by summing the number of food groups consumed by the child as reported over the 24-hour recall period. All the foods and the liquids consumed a day before the study were categorized into 7 food groups. Consuming a food item from any of the groups was assigned a score of 1 and if no food item taken a score of 0 was given. Accordingly, a DDS of 7 points was computed by adding the values of all the groups. Then it was categorized as low (≤ 3), medium (4–5) and high (6–7)(19).
The household food insecurity access scale (HFIAS) is a continuous measure of the degree of food insecurity in the household in the past four weeks (30 days). The nine item scale were constructed to capture three larger dimensions of household food insecurity: anxiety and uncertainty about household food access (item 1); insufficient quality (items 2–4) and insufficient food intake and its physical consequences or hunger (items 5–9). First, a HFIAS score variable was calculated for each household by summing the codes of each frequency of occurrence question. The maximum score for a household is 27 (the household response to all nine frequency of occurrence questions was often coded with response code of 3; the minimum score is 0 (the household responded not to all occurrence questions, frequency of occurrence questions was skipped by the interviewer and subsequently coded as 0). The higher the score, the more food insecurity the household experienced. The lower the score, the less food insecurity a household experienced.
Agricultural biodiversity was assessed by a questionnaire adapted from different literatures (12, 19). It was measured by determining the variety of food plants grown, animals reared for food and food items obtained from natural habitats. A list of all food items grown, all animals reared, hunted and other food items obtained from natural habitats through gathering or trapping was established for each household.
Data quality assurance
The questionnaire was translated to Amharic and back to English for consistency. Pre-test was conducted on 5% of the sample size out of the study area to check for clarity prior to the actual data collection. Data collectors and supervisors were selected based on their interest and experience on data collection. Three day’s training was given on the overall data collection procedure for data collectors and a supervisor to minimize systematic error. On spot checking and correction was made for incomplete questionnaire by a supervisor. Overall data collection process was controlled by principal investigator. Anthropometric measurements were conducted by two data collectors. Calibration of height and weight measurement scale was done at every five measurement.
To minimize bias of respondent memory lapses, an interview was held following a standardized schedule. First, mothers/care-givers were asked to mention all the foods and drinks that a child had consumed during the previous day, including snacks and drinks. Then they were asked to describe the foods and beverages consumed in more detail, including ingredients, cooking methods of mixed dishes and the place and time of consumption. Finally, the amounts of all foods, beverages, ingredients of mixed dishes consumed were estimated using the food photo manual which contained photos of life sized food portions.
First data were down loaded from the common server in csv form and transported to SPSS version 23 for analysis. After cleaning, descriptive statistics was computed and frequencies, proportions, percentages, mean and standard deviations were report. By checking the normality of the data, correlation between two continuous variables was analyzed with regression analysis and the strength of the correlation was measured with the Pearson correlation coefficient. To identify predictor variables for wasting, multiple linear regression models were computed by considering the Z-score values as a continuous outcome of interest. A p-value of < 0.05 and adjusted odds ratio with its 95% confidence intervals were used to represent the statistical significance. The assumptions of linearity and model fitness were checked by scatter plot of standardized residuals against standardized predicted value and the normality probability plot (P-P plot). Multi-collinearity diagnosis was checked by using variance inflation factor (VIF) and all variables have a variance of < 10. Anthropometric measurements were analyzed using WHO Anthro 2005 software.
An ethical approval letter was obtained from the Ethical Review Committee of College of Medicine and Health Sciences, Wollo University. A support letter was obtained from Wollo University, which in turn was issued a support letter to Ambassel district health office and the respective selected Kebeles in order to conduct the study. After the purpose and objectives of the study has been informed, written informed consent was obtained from each mother/care-givers. Participants were informed as participation is on voluntary basis. All the information was kept confidential, and no individual identifiers were used during data collection.