Nutritional status of primary school students and its association on academic performance in north-central Ethiopia: Cross sectional study

Background: Under-nutrition is a fundamental factor which resulted in delayed cognitive development and under achievement of academic performance later in children’s life. Despite the great struggle to prevent malnutrition it is still a major global burden which mainly affects children. The extent of under-nutrition in Ethiopia has been established well but there are insufficient evidences about its relationship with educational achievement of school age children. Objective : To assess the correlation of nutritional status with academic performance among public primary school children Method: - An institution based cross-sectional study was conducted over 399 primary school children in Debre-Tabor Town, Amhara Regional State North-Central Ethiopia, 2018 after selecting them with simple random sampling. The nutritional status was assessed using anthropometric measurement of WHO 2007 reference as Height for Age Z-score (HAZ), Weight for Age Z-score (WAZ) and Body Mass Index for Age Z-score (BAZ). Academic performance was assessed by calculating two semesters overall subjects average score. Descriptive statistics, Pearson’s correlation test and linear regression were used for statistical analysis. Data were analyzed using SPSS version 24 and nutritional status was analyzed using WHO AnthroPlus version 1.0.4 (5-19 years). Result: -The prevalence of stunting, wasting, underweight and overweight was 22%, 32.4%, 24.3% and 1.3% respectively. The mean academic performance for study participants was 71.65(±12.63 SD). There was statistically significant positive relationship between academic achievement and HAZ. Age of child and absenteeism were associated negatively while head of house, attending preschool and breakfast habit associated positively with educational achievement of the study participants. Conclusion:- The prevalence of under-nutrition among study participants was high. Considering correlation of absenteeism, skipping breakfast, having preschool education and HAZ, nutritional interventions should be considered at prenatal,

infancy and preschool period to sustain positive impact and reducing negative factors. Key Terms: Nutritional status, Academic performance, primary school, child1.

Background
Primary education is an instrument in order to meet United Nation Education, scientific and Cultural Organization (UNESCO) goal "quality education and lifelong learning for all by 2030" and "end poverty by 2030". In addition it is one of the most potential approach to realize Ethiopia's current Gross Transformation Program (GTP) development vision to become a middle income economy by 2025 [1]. Progress had achieved over Millennium Development Goals (MDG) and Education for All (EFA) goals. But, still 58 million primary school age children are out of school worldwide due to various reason including nutritional challenges [2].
Stunting and under-nutrition have immediate and long term effects that ranges from immediate school dropout to the worse deprivation of growth and development, and subsequent cognition which negatively affects educational achievements [3]; [4].More specifically, malnutrition remains as a key crisis in adolescents' school performance due to inappropriate eating practices [5]. Moreover, primary school children who are suffering from malnutrition has under achievement of academic performance which potentially affect their excellence of life in the future [6].
Ethiopian elementary or primary education is free with first(one to fourth grade) and second(five to eight grade) cycles lasting for 8 years [7]. And in line with the targets set in Education Sector Development Program IV (ESDP IV) for dropout and repetition rates of 1% in all primary grades, the factors of poor academic performance in primary school children need to be investigated with more powerful design [8].
Various researches have been consistently describing the impact of well nourishment for its positive effect over intelligence, academic achievements learning and cognitive performance, maintaining education there by preserving the nation for grave effect of malnutrition [9] ; [10]; [11]; [12]. Furthermore, globally the death of 45% under five children and 38% of early school leaving are attributed to malnutrition. Moreover, there is estimation of over 450million children will be affected by stunting which is impact of chronic under nutrition by 2025 [13]; [14,15]. So nutritional intervention is important for children attending primary school since it enhance efforts to reduce levels of malnutrition and supplement better academic achievement in the preschool years.
Consistently, the USA school breakfast program was proven the increment of academic achievement over program utilizers [16]. Another study also shown Children having frequent breakfast had perform well in IQ test scores while those who occasionally have breakfast perform low [17]. In Zimbabwe also having frequent meals and receiving a regular breakfast, lunch and dinner have shown positive effects on children's learning achievement [4].
Only Africa shares 38% stunted, 27% wasted and 24% overweight children in under five children [18]. More specifically from 59 million continental prevalence East African region constitute 24% and aggressively suffering from stunting compared to west Africa region which was 19.2% (19.2%) [18]. Despite greater demand to improve students' academic performance in developing countries various hampering factors were identified like under nutrition and children health status, which adversely affects ability to learn and poor performance [5,19]; [20].
Various factors of malnutrition like food insecurity, skipping breakfast, overweight, and underweight were shown strong association with decreased school performance in UK, Brazil and USA [12]; [21]; [16]. More specifically, Educational underachievement was seen in 31.5% of obese, 30.0% of underweight and 22.9% of overweight children respectively [22].
In addition to underlying nutritional problems, a various study found that boys suffer more from poor educational achievement (23%) than girls (18.6%) [22]; [6]. Another study in Morocco also showed more profound effect of malnutrition over boy's school grade average than girls. This study reveals 30.1% of girls and 38.7% of boys rated as poor for mathematics and 17.1% of girls and 37.3% of boys rated as poor to average [5]. Another studies also reviel the impact of socio demographic factors including sex, absenteeism, and distance from school, grade attended, parental socio economic status and parental involvement in their children's schooling were found to be significantly associated with primary school children's academic performance [23][24][25].
African malnourished school age children are at risk of repeating grades and dropping out of school with an achievement of 0.2 to 1.2 years less in school education [26]. Moreover, more than a quarter of children in sub-Saharan Africa are too thin which is resulted in impaired mental development and low educational achievement [24]. By 2025 an additional 11.7 million children will be stunted in sub-Saharan Africa when compared to the 2010 data [13].
Ethiopian Demographic and Health Survey (2016 EDHS) report shows that the prevalence of under-five severe malnutrition were 18% for stunting, 3% for wasting and 7% for underweight respectively [27]. The Cost of Hunger in Africa (COHA) summary report revealed that more than 2 out of every 5 children in Ethiopia are stunted with 16% repetitions in primary school children and an achievement of 1.1 years less in schooling [28].
Even though researches done in South Gondar and Goba Town conclude that nutritional status did not show an association with academic performance [20,25]. Other pocket studies done in various areas of Ethiopia consistently reporting association and effect of under nutrition over mathematics score [20];and overall academic performance [24]; [23].
Like as it was investigated in various developed and developing countries, the effect of nutritional status to the school performance need to be consistently identified in our context for convincing planner and diverting emphasis since factors differ across various socio-demographic factors. With a diverse role of nurses even as school nurses, the output of this study probably shifts attention of school nurses for specific nutritional intervention at school level parallel to other routine service and nurses working in policy design for resource and effort mobilization. By far this study benefits the regional education and health office policy makers and zonal planners to intend appropriate nutritional interventions through SFP and NEP at school in order to deal with the impact of malnutrition on academic performance among primary school children.
The aim of the study is to assess the association of nutritional status with academic performance. specifically estimate the level of various type of malnutrition in school children, estimate school performance and their association.

Design and setting
An institution based cross-sectional study was conducted from March 30 to April 30, 2018 in Debre Tabor Town, North-Central Ethiopia which is located 665kms North of Addis Ababa which is the capital city of Ethiopia and 74kms of Bahir Dar Town which is the capital city of Amhara Regional State. The 2017 population projection the total population of the town showed 96,973 [29]. As information obtained from the town education sector, the town has two preparatory schools, fourteen elementary schools of which seven governmental, and nine Kindergarten schools. The total population of public primary school children was 6,935 of which 49.14% were males.
Sampling procedure and sample size Single population proportion formula was used to estimate sample size by considering at 95% confidence level, 0.05 margin of error and average academic performance of 62.25% in the study area [25]. The required sample of 399 was obtained after adding 10% of nonresponse rate. First simple random sampling was used to select three schools from seven schools, and then proportional allocation of sample to population size of selected school and then final systematic random sampling of every nine student were taken as techniques to obtain final study participants.

Study instrument
A structured questionnaire was administered to assess the children's parental sociodemographic and economic status which is adapted from previous studies [20,25]. The nutritional status was assessed using anthropometric measurement indices of the WHO standard reference 2007 as HAZ score < -2SD (stunted), WAZ score < -2SD (wasted) and BAZ score < -2SD (underweight) or thin), BAZ > +1SD (overweight) and BAZ > +2SD (obese). The child's height and weight was measured with meter and digital weight scale while the academic performance was assessed by calculating the two semesters overall subjects average score from the individual child's grade report of school roster. An English version of the questionnaire was translated into Amharic to collect data and back to English by another person to check its consistency for the analysis

Data collection
Six trained Health extension workers as data collector and two trained BSc nurses as supervisor were recruited for the process. Consent was obtained from their parents and informed assent was obtained from children for their authorization to get their anthropometric measurements. The data collection was done at each section of grade by arranging convenient time and mobilization of measuring equipment's. The previous grade also obtained from schools record officers for having statistical tests with anthropometric measurements.

Data quality control
Random selection of student who participated in the study was used as quality assurance.
Moreover, training of data collectors, conducting pretest over 5% of population and day to day discussion about the progress with supervisors and data collector were activities to maintain quality of the study. Using of standard tool to measure and interpret the measurement were also measures to keep quality.

Variables and Measurements
The study outcome variables are nutritional status indicator: Stunting, Underweight and Wasting. The children's body weight was measured to the nearest 0.1kg with light closing and bare foot using a digital weighing scale by trained data collectors. The child stood with both feet in the center of the scale after checking the weight measurement scale indicator against zero reading after weighing every child. This means that the scale reads zero when no weight is placed on it. Their height was measured using meter to the nearest 0.1cm without shoes in a standing position. Child stood on flat surface with feet flat, together and against the wall. Legs, arms and shoulders were straight, at sides and level respectively. The completeness and consistency of the data collection was checked before the study participants leave. The overall activities were monitored by the principal investigator.
Academic performance was the dependent variable for the study and operationally measured by taking two semester's average score result of the total subjects obtained from the individual child's semester grade report and considered as good if it is more than the mean academic performance and poor if less than the mean academic performance.
Students grade report cards taken as measure of academic performances(Zaini MZ et al.,

2005).
Nutritional status: is the child's state of the body which is determined by anthropometric measurement indices as stunted, wasted, underweight or overweight based on the WHO standard reference 2007 [30].
Anthropometric measurements: is the nutritional status measurement method of primary school children to identify whether they are stunted, wasted underweight or overweight.
Malnutrition: is a poor nutritional status of primary school children which is expressed in anthropometric indices when the Z-scores for the WHO standards of 2007 are less than or equal to minus two (Z-scores < -2SD) or more than plus one for BAZ (z-score > +1 SD).
Stunting: indicates the nutritional status of the study participants when their height for age z-score is less than or equal to minus two (HAZ < -2SD) for the WHO standard reference 2007.
Well nourished: indicate children's nutritional status which is measured as HAZ greater than minus two standard deviation (HAZ> -2 SD) to the WHO reference 2007. Underweight: indicates the child's nutritional status among the study participants which is measured as body mass index for age z-score less than or equal to minus two (BAZ<-2SD) of the WHO standard reference 2007.
Wasting: is the nutritional status of primary school children aged 7-10 years and expressed in weight for age z-score less than or equal to minus two (WAZ < -2SD) of the WHO standard reference 2007.
Overweight: is the school age children nutritional status which is measured in terms of body mass index for age z-score greater than or equal to plus one (BAZ ≥ +1 SD) of the WHO standard reference 2007.
Primary school children: -Children who attend primary school(one to eight grade) and are in the age group of 7-14 years as set by ESDP IV [8].

Data analysis
Data was checked for completeness, coded, entered, cleaned and checked using Epi-Data 3.1 and analyzed using SPSS version 24. The nutritional indicators HAZ, WAZ and BAZ were calculated from measurements using WHO AnthroPlus version 1.0.4 software and compared with the WHO references 2007 data. The anthropometric indices measurement below-2 standard deviation (-2SD) of the WHO median for WAZ, HAZ and BAZ were considered as wasted, stunted and underweight respectively. Child was considered as overweight for BAZ above +1 SD from the WHO reference median.
A descriptive analysis was conducted to get summary data on frequencies, means and standard deviations of the child and their parental socio-demographic and economic characteristic variables. Pearson's correlation test was performed to assess the relationship between nutritional status and academic performance of the study participants. Linear regression was performed to determine the socio-demographic and economic predictors of academic performance. The statistical significance was declared at P value ≤ 0.05.

Results
Socio-demographic and economic characteristics of the study participants A total of 399 public primary school children were enrolled with 100% response rate. Out of these participants, (52%) were males and (48%) were females with mean age of 12 (±1.9 SD) and 11.8 (±1.8 SD) respectively. Nearly three fourth of the study participants (73%) were in the age group 11-14. The large proportion of the participants (62%) was came from second cycle (5-8 grades). Almost all of the children were Orthodox Christian (96%) in religion, from urban (99%) and from married families (73%). Majority of the participants, (42%) were headed by both father and mother. More than half of the participants, 56% were from a family size of less than five.
Majority of the participants were born from no formal education mothers (45%), government employee fathers (37%), monthly income of >2000 ETB (62%) and house wife mothers (55.4%). Only half of the study participants, 55% have attended preschool.
Regarding breakfast, almost all of the participants (97%) ate breakfast and more than half, 55% of them ate breakfast always. Most of the participants, 97% travels for less than 30 minutes to go to school with an average time of 14.4minute (± 8.4 SD) for males and 15.3minute (±8.8 SD) for females respectively. Most of the participants, (71%) were absent from school and 86% of them absent from the school for less than 5 days. ( Table 1) Nutritional status of primary school children The anthropometric assessment of the study participants revealed that the overall prevalence of any form of malnutrition was 56.14%. Out of this, 88 (22%) were stunted, 34

Discussion
This study attempted to assess the correlation of nutritional status with academic performance and to determine the socio-demographic and economic predictors of academic performance among public primary school children in Debre-Tabor Town, Amhara regional state North-Central Ethiopia.
The overall prevalence of any form of malnutrition for the current study was 56% which is higher than the finding of Hawa Gelan in 2017 in Oromia, Ethiopia (34.6% ) [24]. However this finding is less than the finding in Sri Lanka in 2015 (60.2%) [6]. This variation could be due to difference in the two countries socio-economic differences.
The prevalence of wasting in this study was 32.4% which was almost similar with a finding in India (33.3%) [37]. However this finding was higher than a finding in Fogera, Ethiopia (30.7%), Hawa Gelan in Oromia, Ethiopia (12.7%), Sidama, Ethiopia (28.2%) and that of Kenya (9.7%) [24,34,35,38]. The finding of wasting in this study was lesser when compared with a finding in Sri Lanka (50.4%) [6]. The discrepancy in acute malnutrition might be difference in children's health status, nutrient composition (balanced diet), diet size and frequencies of meal. Acute conditions like diarrhea and upper respiratory problems are also key factors for children's acute malnutrition since it reduces their appetite and impede their immunity. Children from low socio-economic status face problems on having regular breakfast, lunch and dinner which resulted in wasting.
Again in this study the prevalence of underweight was 24.3% which was almost similar with a finding in Bahir Dar, Ethiopia (24.8%) and the regional state of the study area (24%) [27,39]. This finding was higher than the findings in Adama, Ethiopia (21.3%), Oromia, Ethiopia (14.2%), Kenya (14.9%) and that of Lambani, India (14.66 %) [11,24,32,34]. In contrary this finding was lesser than findings in Fogera, Ethiopia (59.7%) and Sri Lanka (33.7%) [6,35]. The inconsistency might be due to differences in socio-economic status and cultural beliefs about foods. Children from rural areas are more at risk for malnutrition than urban children. Additionally meal frequencies affect children's nutritional status in which children having regular meals at morning, lunch and night will have better status than those having any of the three meals.
The prevalence of overweight in this study was 1.3%. This result is less than a finding in Adama in 2016, Ethiopia (3.3%), Nigeria in 2013 (37%), Malaysia in 2014 (18.1%) and that of Brazil in 2014 (22%) [21,32,40,41]. This disparity might be due to variation in individual nutrient intake and food preferences. Genetic makeup of the children also plays a role in their nutritional status. In addition differences in socio-economic status could be the reason as Ethiopia is a lower income country. Lower income countries are at risk of food insecurity which affects children's diet composition, diet size and frequency of meal.
In this study the mean academic performance for study participants were (71.65±12.63 SD) which is a higher achievement when compared with a study done in Hawa Gelan in Oromia, Ethiopia (67.2% ±15.4% SD), Tach-Gaynt, South Gondar, Ethiopia (62.25%), Hawassa, Ethiopia (35.1 ± 4.1SD), Sri Lanka in 2015 (<40%) and Nigeria (66.19 ± 17.00 SD) [6,24,25,42,43]. The variation could be difference in assessment of academic performance, learning methods, learning resources accessibility, tutors given and grading or scoring systems. Additionally differences in children's psychology like stress, emotions and their physiology like health status contributes for this discrepancy.
In this study males perform better (72.02±11.57 SD) academically when compared with female students (71.24±13.73 SD). This finding is consistent with a research finding in North-Central and South Central Ethiopian that males have performed 2.39 times better than females in academic performances [23,25]. In contrary findings in Nigeria, Morocco, India and Asia revealed that girls have better educational performance than boys [5,6,22,42]. It may be related with social status of female in the community and recognition of their learning. Moreover, lack of role model and high percent of students were not passing through preschool training may contribute to the disparities.
In this study primary school children from high monthly income parents were perform better (72.91±1234 SD) than those from low (69.53±14.46 SD) income parents.
Educational achievement was poor for those who absent for more than five days The result of this finding shows that there was a statically significant positive relationship between HAZ and academic achievement (p-value=0.009) which is in line with the finding done in Goba town and Hawa Gelan in Oromia, Ethiopia and Sri Lanka [6,20,24]. It is also consistent with a finding in Morocco in 2016 [5]. This indicates that as children become well-nourished their height for age z-score increases to the appropriate growth and development stage which in turn resulted with positive educational achievement.
In this finding WAZ and BAZ did not show a statistical significant relationship with academic performance of the study participants which is consistent with a finding in various areas of Ethiopian: Tach-Gynt, South Gondar; Gurage Zone; and Goba town [20,23,25]. However this is not consistent with the finding in Oromia, Ethiopia and Sri Lanka in which both WAZ and BAZ had significant positive associations with learning achievement [6,24]. Even though, it is beyond the scope this study the discrepancy could be variation in nutrients taken by children which may have either synergistic or antagonistic effects with each other. At this point, this finding sought more powerful study design that can indicate causal relationship among factors.
In this study age of child and absenteeism have a statistically significant negative association with educational achievement. Predictors including head of house, attending preschool and breakfast habit associated positively with educational achievement of the school age children. In this study monthly income did not show significant association with academic performance which is opposing to Hawa Gelan finding that study participants from high monthly income households perform 2.85 times better than those from low monthly income households [24]. In this study breakfast habit and attending preschool were significantly positively associated with academic performance. This finding is conflicting to the finding in Goba town, Oromia that breakfast habit and attending preschool did not show significant association [20]. Variables including sex of the child, residence, maternal education, paternal education, parental occupation and family size were not significantly associated with academic performance which is parallel to the finding of Goba town in Oromia, Ethiopia [20]. Absenteeism was significantly associated with school age children's academic performance which is parallel to the finding of Tach-Gynt in South Gonar, Ethiopia [25]. This study revealed that as age of child increases by a unit of year in children aged [11][12][13][14] years their educational achievement decreases by 5.07 when compared to those aged 7-10 years (B: -5.07, 95% CI: -7.86, -2.28). These variations could be due to more attention for pubertal period changes rather than learning and attending classes as age increases

Strengths And Limitations
Using wider range of student class grade from two to eight and large sample size can be taken as strength of this study. But, this study was confined to public schools and fails to provide evidence about causal relationship due to limitation of chosen design. Moreover, it would have been better and more powerful if the study was incorporated measurements of biochemical parameters that may help to describe and rationalize some of discrepancies with other studies. Limited number of variables under study may not show the whole factor of school performance and the absence of follow-up and control group also decrease the power.

Conclusions
The present study revealed that the prevalence of under-nutrition among school age children is high in the study area with an overall prevalence of any form of malnutrition 56.14%. The prevalence for each nutritional indicator was 22% for stunting, 32.4% for wasting, 24.3% for underweight and 1.3% for overweight. The mean academic performance of the study participants were 71.65±12.63 SD. Study participants who have breakfast, from monthly income of >2000 ETB, well-nourished and overweight perform better academically when compared with those not have breakfast, from <1000 ETB, stunted and underweight respectively. Males and those in the age group of 7-10 years perform well academically than females and 11-14 years children (72). A significant positive relationship was seen between HAZ and academic achievement of the study participants. In this study age of child, head of house, attending preschool, breakfast habit and absenteeism were variables that significantly associated with educational achievement of the study participants. Accordingly the following recommendation made to the relevant bodies.
The zonal health and education bureau should work in collaboration with other organizations (NGOs) to combat malnutrition and to strengthen the positive effect of nutrition on educational achievement of primary school children. And also, they should implement nutritional interventions through school feeding programs (SFP) by mobilizing funds and work intensively over preschool programs.
The school teachers should encourage students to have breakfast, attend class attentively, motivate female students and create opportunity to discuss with family about student status.
Since the majority of studies done in primary school children nutritional status and educational achievement relationship were cross-sectional which does not show the causal effect of one on the other. We recommend researchers to conduct further study with strong study design to investigate the true relationship between nutritional status and academic achievement of school age children. Ethical clearance was obtained from Addis Ababa University school of Nursing and midwifery Ethical committee. A formal written letter was given to Debre-Tabor Town Education Sector, Health bureau and to the respective primary schools and health centers.
Then a written consent was obtained from the parents of the participant children and informed assent was obtained from children for their authorization to get their anthropometric measurements. Codes were given to the study participants in order to keep their confidentiality.

Competing interest
The authors declare that they have no competing interest TE: contribute to conception, design write the proposal, participate in data collection supervision, and analyze the data. FA: had revised the proposal, participate in data collection and research report. DG: had revised the proposal, participate in data collection and research report. AA: had revised design, the proposal, and research report and prepare the manuscript. DT: had revised the proposal and research report. All authors read and approved the final manuscript.   Table 3 :-Descriptive result for academic achievement of study participants in Debre-Tabor Town, Amhara Regional State, North-Central Ethiopia, 2018, (N= 399).