Page 22 - IJPS-5-2

Risk factors of child undernutrition in India

anthropometric indicators of unit-level data of 209,377 (after excluding missing and flagged cases) children aged
0-59 months.
2.1. Dependent Variable

Three markers of height-for-age, weight-for-age, and weight-for-height z-scores provide a complete picture of children’s
severe undernourishment; subsequently, we utilized anthropometric information on these three-pointers to study the child
undernutrition. To estimate all three markers, we embraced another reference population of the WHO 2006 (Multicenter
Growth Reference Study Group, 2006). As indicated by the WHO rules, if height-for-age Z-score of a child is <2 standard
deviations (SDs), weight-for-age Z-score of a child <2 SDs, and weight-for-height Z-score of a child <2 SDs then the
child is classified as stunted, underweight, and wasted, respectively. Thus, these three nutritional indices are treated as
dependent variables and dichotomized. While those that were malnourished (Z- score below −2 SD) were coded as one
and nourished children (Z- score above −2 SD) were coded as 0.
2.2. Independent Variable

Three classes of elements were evaluated as independent factors; First, socio-economic and demographic variables such
as wealth index (poorest, poorer, middle, richer, and richest), and religion (Scheduled Caste, Scheduled Tribe, Other
Backward Class, and Other caste) are taken into consideration. Second, child characteristics such as child age in months
(0-6, 7-12, 13-24, and 25-59), birth order (1, 2, 3, 4, and above), and size at birth (average, small, and large) are taken.
Third, maternal characteristics considered (Maternal age 15-49 years), maternal nutritional status (underweight: Body
mass index [BMI] <18.5, normal/healthy weight: 18.5 <BMI <25, and overweight/obese BMI >25.0), and mother’s
education (No education, <5 years education, 5-7 years, 8-9 years, 10-11 years, 12 years or more) are considered. For
the independent variable child age, 0-6 months were taken as the reference category for stunting and underweight (Khan,
Zaheer and Safdar, 2019). For wasting, the reference category was chosen as 25-59 months since children aged 0-6, 7-12,
and 13-24 months age group are the most affected by wasting and severe wasting (Akombi, Agho, Merom, et al., 2017).
2.3. Statistical Analysis

2.3.1. Logistic regression
The binary logistic regression analysis was carried out to identify the factors associated with child malnutrition. Odds
ratio with 95% confidence interval was used to study the effect of independent variables (namely, child age in months,
mother’s educational status, mother’s nutritional status, type of caste, wealth index, birth order, and size of a child at
birth) on dependent variables being the three indicators of nutritional status (stunting, wasting, and underweight) (Bewick,
Cheek and Ball, 2005). This analysis was performed utilizing the Statistical Package for the Social Sciences (SPSS),
version 19.0 software.

2.3.2. CI
It quantifies socio-economic inequality and CI for nutritional status. It can be described utilizing the concentration curve,
(Y-axis) which determines the cumulative percentage of undernourished children, whereas (X-axis) determines the total
percentage of children ranked by household wealth status, and it begins with the poorest wealth quintile to the richest
wealth quintile. When “y” takes on higher quality for poorer people, the concentration curve lies above the equality
line. If the “y” takes lower confidence, the reverse is true, the health concentration curve lies below the equality line.
The CI is termed as positive when the concentration curve lies below the diagonal and negative when it lies above the
diagonal. Thus, the minimum and maximum value that CI assumes is −1 and +1 (Wagstaff, Paci and van Doorslaer, 1991;
Giashuddin, Rahman, Rahman, et al., 2009; Kumar, Kumari and Singh et al., 2015). We calculated the corresponding
indices and their CI, by the following formula for the grouped data case.

C = ( p L - pL ) ( pL+ 2 3 - p L 2 ) ... ( p L++ T -1 T - p L T -1 )
1 2
3
2 1
T
Where p is the cumulative percent of the sample ranked by economic status, L(p) is the corresponding concentration
curve ordinate, and T is the number of socio-economic groups (O’Donnell, van Doorslaer, Wagstaff, et al., 2008; Zere,
Tumusiime, Walker, et al., 2010).
16 International Journal of Population Studies | 2019, Volume 5, Issue 2