Nutritional Status of Dalit Children: A Comparative Study with Non-Dalit Children in Eastern Terai of Nepal

In bi-variate analysis the significant difference were observed among children's nutri onal status between castes as well as a number of socio-demographic variables however in mul variate analysis only caste, number of Children Ever Born (CEB) and mother's educa on were found significant determinants of height for age, mother's educa on was found significant determinant for weight for height and caste, age of children and mother's educa on were found significant determinants of weight for age of children.


ORA 18 INTRODUCTION
Adequate nutri on is prerequisite for all the individuals to maintain their physical as well as mental health; however, more focus has to be given par cularly for women and children who are especially vulnerable to it.Childhood is a stage of ac ve growth in all round aspects of a child.Children need varie es of nutrients in this stage, if the nutrient requirements of children are not met, the consequences can be serious for their physical, psychological as well as social development.
The poor nutri onal status of under five children has been considered a serious public health issue in Nepal for many years.The most common forms of malnutri on in the country are protein energy malnutri on (PEM), iodine deficiency disorders (IDD), vitamin A deficiency (VAD), iron deficiency anaemia (IDA) and other micronutrient 1 deficiencies (DoHS/MOPE).Some studies have shown that the nutri onal condi on of Dalit children is very poor in comparison to general children, in Asian countries; however, there is lack of sufficient evidences in Nepalese context.A study conducted in Kapilvastu district of Nepal reported more than 60% under five children with any kind of malnutri on while measuring by anthropometric measurements.Out of them nearly one-2 fourth children were in cri cal condi on (below -3SD).The socio-demographic factors like socio-economic status, mother's age, birth order, exclusive breast feeding, and supplementary foods were associated with child health.Another Study conducted in Rupandehi district reported that 65% children aged between 36 to 59 months were stunted and 46% were underweight and the figure was more serious 3 among the children of uneducated mothers.Similarly, the study conducted by Coffey D et al.in India found that the height gap between Scheduled Tribe children and children from general caste, or historically dominant groups were en rely associated with economic wealth and even a er accoun ng for economic differences, an important height 4 shor all remained for Scheduled Caste.
There is essen al to explore the nutri onal status of Dalit children along with socio-economic and demographic status.Therefore, the present study has aimed to explore the nutri onal status of Dalit children in eastern Terai ecological belt of Nepal along with its comparison to Non-dalit children and also to aims to explore its socio-demographic determinants.

METHODOLOGY
This study is based on cross-sec onal study design.Predesigned, pretested semi-structured interview schedule was used to collect the informa on.The nutri onal status was assessed by the anthropometric measurements: height, weight and Mid-Upper-Arm Circumference (MUAC).
To calculate the sample size a pilot study was conducted among 50 subjects and in an average 30% under five children were found malnourished.Based on the data of pilot study the sample size was obtained by using following formulae: The height for age of children has been given in table 1.In total, 41.4 percent children were found stunted (<-2SD) and among them 18.1 percent were severely malnourished (<-3SD).The table also reveals whether any socio-demographic factor is associated with the Height for Age.Among the seven Kafle TK et al variables considered, caste, age of children, numbers of CEB to mother, wealth index and mothers' educa on were found significantly associated with height for age of children in bi-variate analysis (p<0.05).According to the age of children, 50.0 percent children aged 12-23 months were found below -2SD followed by the children 24-47 months (43.6%).Number of children ever born (CEB) was nega vely associated with the height for age of children.As the number of ever born children increased the percentage of stun ng also increased.
Higher propor on (55%) of children were found stunted Educa on also had be er impact on achieving the height of children.Higher percent (46.6%) children with noneducated mothers were found stunted in comparison to the children of primary and high educated mothers (p<.05).Mean Z score (SD) was also severe among the children of non-educated mother (Table 1).

Weight for Height
Weight for height parameter is close to the body mass index (BMI) for the children.Among the total study children, about 25 percent were found wasted (<-2SD) including 9.5 percent severely wasted (<-3SD).In mul variate analysis only mother's educa on was found significant determinant of weight for height.The odds of children with mothers' educa on as primary and some secondary were 2.36 mes higher (CI: 1.39 -4.10) and 1.79 mes higher (CI: 1.08 -2.98) as compared to the children of non-educated mothers (table 4).

Weight for Age
Weight for age is one of the most common indicator to measure nutri onal status of children has given in table no. 5.Among 705 children measured successfully, 38.6 percent were found underweight (<-2SD) and 13.6 were severely underweight (<-3SD).
Educa on, wealth Index, caste, age of children and numbers of children ever born were significantly associated with the weight for age (p<.05).About one and half mes higher propor on of Dalit children (46.4%) were underweight as compared to Non-dalit (30.7%) and children of aged 2 years and above were more vulnerable to be underweight than those below 2 years.
Higher propor on of first (64.3%)and second (66.4%) order live births had be er weight than the forth order and more (46.8%).Mother's educa on was highly significant in maintaining the normal weight; underweighted children without mother's educa on were nearly two mes higher than the children of educated mothers'.Similarly, the varia on is also found in mean Z score (SD) of weight for age according to the socio-demographic status.Children of age below 6 months had be er (-1.02) mean Z score than the children aged 24-47(-2.03)and 48-59 months (-2.10), (Table 5).The result of mul variate analysis confirmed that the caste, age of children, numbers of ever born children and educa on of mothers were significant determinants of weight for age of the children.The remarkable varia on is that the odds of having normal weight in children less than 6 months was 5.08 (CI: 2.63 -9.81) mes higher than the children of 48 months & above.Similarly, the odds of Non-dalit children gaining normal weight was 1.55 (CI: 1.09 -2.21) mes higher than the Dalit children.Educa on was also another significant determinant of weight for age.Children of educated mothers had higher chances of ge ng normal weight as compared to the children of not-educated mothers.The children with mothers' primary-secondary educa on had 1.87 (CI: 1.16 -3.02) mes higher and with mothers' educa on grade 10 & above had 1.62 (CI: 1.07 -2.45) mes higher odds than the children of not-educated mothers to gain be er weight for age.The decision-making capacity of mothers also had some influence for gaining normal weight of children (table 6).

Mid-Upper-Arm Circumference (MUAC)
Mid-Upper-Arm circumference of children determines the muscle volume in the body.In the present study, in total, 21.

DISCUSSION
Adequate nutri on is prerequisite for all the individuals to maintain their health; however, more focus has to be given par cularly women and children who are especially vulnerable to it.Problems related to malnutri on affect the popula on of all ages and areas; women and children are especially vulnerable because of their unique physiology and 5 socioeconomic backgrounds.Women's nutrient needs increase during pregnancy and lacta on.If the requirements are not met, the consequences can be serious par cularly for 6 their infants.The poor nutri onal status of children has been considered a serious issue in Nepal for many years.The malnutri on status has been trying to assess by anthropometric measurements in this study to understand the real situa on among Dalit children along with its comparison with Non-dalit.The height for age, weight for height, weight for age, mid-upper-arm circumference (MUAC) and mean Z score (SD) were obtained and compared between groups according to the WHO standard.
In total, 41 percent children were found stunted (<2SD), 24.7 percent wasted and 38.6 percent underweight.Moreover, the prevalence of severe malnutri on was also high in children.Eighteen percent children were severely (<-3SD) Kafle TK et al stunted, 9.5 percent were severely wasted and 13.6 percent were severely underweight.Mean Z score was also calculated to find out the severity of malnutri on.The average Z score was - The finding of NDHS 2011 is similar with present finding except was ng, they indicated that 39 percent children were stunted, 29 percent underweight and 11 percent were wasted.Was ng found higher in present study.They also found that the problem was deeper among the Dalit, low

6 income
households and non-educated mothers' children.Later Panday JP et al (2013) further analysed NDHS 2011 data and found almost similar results.They found that caste/ethnicity, age of child and wealth index were significant predictors of weight of age of the children along with delivery 7 by an SBA.Another study by Shah N (2004) also reported similar finding as the present study.They analyzed the final survey of Child Nutri on Program 2003 data and found that older children (above 36 month) and children whose mothers were illiterate were more likely to be underweight and

Table 1 : Percentage Distribu on of Children according to Height for Age of Children Aged 6-59 Months and Socio- demographic Characteris cs.
Mul stage cluster sampling technique was used to select study districts and VDCs.Three districts: Jhapa, Sunsari and Saptari were selected randomly in the first phase among total 5 districts.Again three VDCs: Surunga from Jhapa, Chhitaha from Sunsar and Inarwa Phulbariya from Saptari were also selected randomly.In third stage, equal numbers (n=240) of study subjects from each VDC were selected using simple random sampling technique from the list of mothers with under-five children.The sampling frame was prepared based on the newest record available from the local health unit (Health Post or Sub-health Post).Face to face interview was taken from the mothers by researchers themselves.Bi-variate analysis was done using Pearson's chi-square test and mul variate analysis was done in binary logis c regression.Significance of the test was considered at 95% confidence interval.The ethical approval was taken from Nepal Health Research Council (NHRC), Kathmandu.*Test of significance taking 3 columns: normal, undernourished and severely undernourished ** Also includes the children below -3 SD

Table 2 : Logis c Regression Analysis on Socio-demographic Determinants of Height for Age of Children Aged 6 -59 Months
SD) if mother had four and more live births.The mean Z score (SD) was cri cal among the children with mothers of four and more live births (-2.20).Birth spacing of children had also influenced the height of children; majority of children (61.8%) of the first birth order had be er height than others.Wealth index of family had also significant impact on achieving the height of the children.Children with poor wealth index had more chances to be stunted; 50.3 st found cri cal among families with 1 quar le wealth index.

Table 2
describes the result of logis c regression analysis on the determinants of height for age among the children.Age, number of children ever born and mothers' educa on were found as significant determinants of height for age.The odds of 6-11 month children were two mes higher to have normal height than the children of 48 month and above age (AOR:2.08;CI:1.07 -4.04).The children of mothers having two live birth had nearly two mes normal height than the children of mothers with four and more live births (AOR :2.02, CI : 1.17 -3.50).Similarly, the odds of children of mothers with grade 10 & above educa on having normal height were more than one and half mes higher than the children of non-educated mothers (AOR :1.64, CI : 1.03 -2.61).

Table 3 : Percentage Distribu on of Weight for Height of Children Aged 6-59 Months by Socio-demographic Characteris cs
*Test of significance taking 3 columns: normal, undernourished and severely undernourished ** Also includes the children below -3 SD.

Table 4 : Logis c Regression analysis on Socio-demographic Determinants of Weight for Height of Children Aged 6 -59 Months Table 5: Percentage Distribu on of Children according to Weight for Age and Socio-demographic Characteris cs Determinants of Height for Age N Logis c Regression Analysis AOR CI
*significant at 95% CI, ** significant at 99%CI

Table 6 : Logis c Regression analysis on Socio-demographic Determinants of Weight for Age of Children Determinants of Height for Age N Weight for Age AOR CI Caste
*Test of significance taking 3 columns: normal, undernourished and severely undernourished ** Also includes the children below -3 SD. *significant at 95% CI, ** significant at 99%CI, ***significant at 99.9% CI

Table 8 : Logis c Regression analysis on Socio-demographic Determinants of MUAC of Children
SDBirth spacing of children was also significantly associated with MUAC (p<.001).The first child had be er MUAC status as compared to others.Higher percentage of female children had MUAC thinner as compared to male children.Educa on of mother again has significant impact on the MUAC status of children.Twenty seven (27.4%) percent children with noneducated mother had MUAC below -2SD and 4.7 percent had below -3SD, however the same figure was be er among the children of educated mothers (p<.001).
*Also includes the children below -3 *significant at 95% CI, ** significant at 99%CI, ***significant at 99.9% CI 1.76 SD for height for age, -1.72 SD for weight for age and -1.23 SD for weight for height.Similarly, 21 percent children had MUAC below -2SD and the Mean MUAC was found 13.7 cm.While observing according to the socio-demographic characteris cs caste, mother's educa on and wealth index of family were strongly associated with all three indicators i.e. height for age, weight for height and weight for age.In addi on, age, number of children ever born (CEB) and birth spacing also posi vely associated with height for age and age and number of CEB with weight for height.While adjus ng with caste, age, sex, birth spacing, number of CEB, wealth index, mother's educa on and mother's decision making capacity in mul variate analysis, only caste, number of CEB and mother's educa on were found as significant determinants of height for age.Mother's educa on was found significant predictor (determinant) for weight for height.Similarly, Caste, age of children and mother's educa on were found as significant determinants of weight for age.