Determinants of underweight, stunting and wasting among schoolchildren

This school based cross-sectional survey was conducted among randomly selected schoolchildren aged 7-14 years old from randomly selected three schools in Dale

Woreda. Dale woreda is found in the Sidama zone of Southern Nations and Nationalitities Peoples’ Regional State of Ethiopia. The woreda is located about 326 Km South from Addis Ababa, capital city of Ethiopia. The Woreda has a total area of 28,444 hectares; total population of 222,068 with 37,027 households. The Woreda is characterized by 1% high land and 99% mid-altitude agro-ecologies and produces a variety of crops and livestock. The area is known for its coffee production. At the time of the research, the Woreda had 36 kebeles, 7 health centers, 29 health posts and 1 hospital.

The study was conducted at three schools in October 2012 and 450 students were sampled using single population proportion formula:

Where: Z = Standard normal variable at 95% confidence level (1.96)

P = Anticipated proportion, (23.1%)

d = 0.05 (5% margin of error)

To minimize errors arising from the likelihood of non-compliance, 10% of the sample size was added. Finally, a design effect (DE) of 1.5 is used to minimize bias arising from not using simple random sampling technique.

A multi-stage sampling technique was used to select a representative sample of schoolchildren from the study area. The list of schools existing in Dale Woreda was obtained from Woreda’s education office, and three schools (Debub Kegie Millennium, Soyama and Degara) were selected using lottery method. List of all students aged 7-14 years and whose grade 1-8 was obtained from each school head. A total of 1397, 1648 and 1550 schoolchildren were found in Debub kegie millennium, Soyama and Degara schools, respectively. The number of schoolchildren aged 7-14 years from each school and grade included in the study was determined by using proportional allocation (PPS) to size. Finally, systematic sampling system was applied to select the study participants from each grade and sex using the respective class rosters for 2011/2012 academic year as the sample frame; the sampling interval for the systematic sampling system was calculated by dividing the total number of students in the roster to the specific students to be selected from each class (Figure 1).

A standardized questionnaire was developed based on known and hypothesized risk factors. Household food insecurity status was assessed using the nine Food and Nutrition Technical Assistance (FANTA) scale guideline questions [6]. The questionnaire was constructed in English and translated into Sidama (local language). The questionnaire was pretested among 45 schoolchildren (10% of the sample size). The socio-demographic and dietary interviews were done at the household level.

Parents and children whose age is 8 years and above were respondents. An oriented health extension worker recorded observations on physical situations for each child in the school setting. The data collectors all had education levels of grade 10 or above and were native to the study area.

After data collection had been completed, some of the variables were categorized for analysis purpose. Categorized variables include: age of a child which is grouped into two (7-9 years and 10 and above years), family size grouped into three: small, medium and large, maternal educational status grouped into three: no formal education, primary education completed and secondary education and above, household food insecurity status grouped into four: food secure, mild food insecure, moderate food insecure and severe food insecure households.

For the anthropometric data digital portable calibrated SECA weighing scale was used to measure the weight of studied children. The children were weighed wearing lightly clothed, without shoes and with empty pockets. The calibrated SECA balance scale has intervals/sensitivity of 0.1 kg with a capacity of 130 kg.

Height was measured to the nearest 0.1 cm precision and length up to 2 meters using the same device that has a scale and sliding headpiece. Weighing scale was calibrated to zero before taking every measurement. Clinical examination of study participants was done for variables such as the presence or absence of clean clothing, dental caries, trimmed fingernails, presence of dirt on fingernails, wearing shoes, paleness on the conjunctiva and palms, edema and for any more body abnormalities by the principal investigators and an assistant nurse. All the observations were recorded in the recording format.

Stool samples were collected and preserved using WHO standard operating procedures for the parasitological examination of feces [7]. Accordingly, study participants provided labeled plastic cups with serial numbers, soft tissue paper, and clean wooden applicator stick. And they were instructed to bring 2 g or about a thumb of fresh stool sample of their own. Students who brought inadequate stool sample were re-instructed to bring another sample. Samples collected from study participants were preserved in a tube containing 10% formalin in 0.85% saline, and were transported to Hawassa university microbiology laboratory for microscopic examination following WHO standard operating procedures [7]. Ten percent (10%) of the total fecal specimen (450) was re-checked in the laboratory by a senior laboratory technologist.

Dietary assessments were conducted at the household level and the respondents were children’s parents and children themselves. Trained data collectors filled the 24 hour recall dietary history with specific probes, to help the respondent remember all foods consumed throughout the day. To account for any day of the week effects on food and/or nutrient intake, weekends, weekdays and market days were proportionately represented in the survey.

Ethical clearance for this study was obtained from the institutional review board of Hawassa University. We obtained a written informed parental consent. The students’ privacy during the interview, stool collection, and anthropometric measurements were maintained and data obtained from them was kept confidential.

All statistical analyses were done using SPSS for windows version 16 statistical package. Data entry for anthropometric indices was made using WHO Anthro plus version 1.0.4. The nutrient and energy content of foods consumed by the index child in the preceding 24-hours was calculated using the Ethiopian food composition table [8] and micro soft excel.

All continuous data were checked for normality using the kolmogorov-smirnove test. Descriptive statistical tests were applied to indicate the prevalence of under-nutrition as frequencies and percentages. To test the presence as well as strength of association between under nutrition and factors, binary and multivariate logistic regression tests were used. Variables used for the regression model include: sex of a child, age of a child, maternal educational status, family size, family monthly income, infection with ascariasis only, hookworm only, trichuriasis only, household food security status, and dietary intake interms of energy, carbohydrate, protein, Fe, Zn and Vit A.

Factors showed significant association in the univariate analysis and factors identified in different literatures as predictors were entered into the multivariate analysis.

Under-nutrition was defined for a child, who had less than-2 z-scores (SD) from the NCHS median reference population values [9].