Determinant Characteristics of Stunting in Babies and Toddlers in Sungai Rengit Murni, South Sumatra, Indonesia

. Stunting is a condition where a child fails to optimally grow and develop his/her bodily organs. Due to the high rate of stunting case in Indonesia, particularly in South Sumatra, this study aimed to determine the causes of stunting in children of 0 yr to 5 yr old as the dependent variable. Independent variables of infection, birth weight, gender, and age were studied accordingly. The purposive sampling method selected 100 babies and toddlers in Sungai Rengit Murni, Banyuasin, South Sumatra, Indonesia. The cross-sectional design applied univariate, bivariate (chi-*


Introduction
Fetal phase and the first two years of infancy are the most crucial stages of a child's life [1] for growth and development both physically and mentally.While nutrition is of the highest requirement during these periods, the latter becomes prone to undernourishment [2] due to scarce health access and inappropriate diet [3,4].Such a condition often leads to stuntinga condition when a child fails to optimally grow and develop bodily organs.Stunted children are expected to be shorter than average as grown-ups as they have lost their individual productivity up to 1.4 % [5].Moreover, stunting dwindles one's intelligence between fivepoints and 11 points [6].When occurring in an early age, the impact of stunting is likely more severe (P < 0.000) [7].
While undernutrition is often assumed as the outcome of failing food security, food is proven to be an insignificant causewith famine as an exception.Various studies have discovered parents' knowledge, parenting pattern, health service access, water access, sanitation, and diet [8,9] as some of more sizeable factors.Food-abundant countries have also been recorded to face household nutrition problems [10].Despite high food productivity rates, Arsi, Ethiopia and Iringa, Tanzania have been dealing with stunting as high as 62 % and 66 % respectively [11].
The prevalence of stunting in Indonesia, particularly among children of 0 mo to 59 mo, is as high as 28.4 % [12,13], and it is detrimental towards the national human resource quality [13,14].This situation had ranked the nation in 124 th out of 187 regarding the 2011 International Human Development Index, which was lower than the neighboring countries Malaysia (61 st ), Thailand (103 rd ), and the Philippines (112 th ) [15].Despite that, the rate of stunting cases in South Sumatra (3° 19' 9.973" S 103° 54' 51.836" E), a province in Indonesia, has been rising in 2017 to 2018 (17.2 %) compared to 2016 (14.58 %) [16].Realizing the urgency of the matter, Indonesia government has responded by issuing Presidential Regulation No.72 of 2021applying in synergy with National Population and Family Planning Board Regulation No. 12 of 2021to accelerate the decline of stunting prevalence rate to the targeted 14 % by arranging a multisectoral collaboration.

Material and methods
Performed in Sungai Rengit Murnia village in Banyuasin Residence, South Sumatra Province (38' 46.267" E, 50' 31.497"S)with aimed to map out determinant characteristics of stunting occurrence in children of 0 mo to 59 mo old in South Sumatra due to its highand-rising rate of stunting cases.This research was conducted with the description of Ethical Approval No. E.  Amount of 100 children of 12 mo to 59 mo in the area who were not under any medical treatments, were free of infection within 1 m prior, and were of Z-scores ≤ 2SD were taken as samples [17,18], thus purposive sampling.With stunting as the dependent variable, four independent variables of infection, birth weight, age, and age entails.Data were obtained by applying WHO-standard Anthropometry [19], interview, and 2010 National Basic Health Research.
Involving editing, coding, and scoring processes, the results were run through crosssectional methods.All data were analyzed by employing Microsoft Excel 2007 and Statistical Packages for the Social Science (SPSS) ver.24.T-test quantified the average Z-score of length by age [20].Univariate analysis calculated distribution and proportion of all categories entailing the variable.Bivariate analysis through chi-square test evaluated the correlation between free variable and stunting variable.Multivariate analysis through Multiple Logistic Regression test determined the role of free variable towards stunting.Regression analysis of backward method was also included to tackle any sign of multicolinearity [21].

Univariate analysis
The categories that indicate potentials of stunting in young children were low birth weight, low height growth, and low weight growth.Comparing observation data on each category based to standard, a stunting child should be distinguishable from a normal one.

Birth weight
By computing birth weight, birth length, weight at measurement, and length at measurement, a child can be verified to have a normal or low one as detailed in Table 1.A child is considered to have a normal birth weight when he/she is born with 2.5 kg to 3.8 kg of it.Although a birth length of 45 cm may be of either normal or low birth heights, below the number means low birth length that more likely goes along with low birth weight.
Weight and height growths vary widely considering the range of ages between 1 yr old and 5 yr old.Regarding the samples in this study, the weight rate of 12.10 kg and height rate of 89.52 cm are referred to as normal while the weight rate of 10.25 kg and height rate of 85.91 cm are low.

Height growth
The child characteristics of infection, birth weight, sex, and age were then cross-compared with height-at-measurement to conclude whether a child is very short, short, normal, or tall.The result is presented in Table 2. Out of 64 children with no infection, more than half of them were of normal height and some of them were even tall; yet 20 (31.3 %) were short.Meanwhile, only 8 (38.1 %) out of 21 children with worms were normalthe rest were either short or very short.One child with diarrhea was of normal height, while the other eight were short or very short.Out of six children with acute respiratory disease, half of them were normal and tall while the other half were short and very short.
Regarding age, most children of 1 yr old, 2 yr old, 3 yr old, and 5 yr old were short to very short -rated50 %, 62.1 %, 57.1 %, and 85.7 %, respectively.Four years old children were in the best condition as most of them were normal and tall, leaving 28.6 % to be short or very short.

Weight growth
Low weight growth is the most factor in stunting.Contrasting birth weight and birth length to measured weight and length, the result should be able to indicate the parameters of normal children and stunting one.The data can be seen in table 3. Normal children were 2 970.21 g weight and 50.15 cm height when born and grew up to 12.88 kg weight and 95.08 cm height when measured for this study.Meanwhile, children with stunting were generally born with 2.8 kg weight and 49.42 cm length, and when measured they grew only up to 10.99 kg weight and 83.62 cm height.Studies [22][23][24] have explained how children with low birth weights and birth lengths are prone to diseases and fatality, particularly when undernourished during their growing period.When such children survive, they have a tendency to suffer from stunting.

Bivariate analysis
A bivariate analysis employing Chi-square test was established to find out whether demographywas affective towards stunting cases among children in the study site.The result is as listed in Table 4 While most children with no infection grew up normally in height, almost half of them (43.8 %) were stunted.Stunting also appeared in most children with worms (61.8 %) and with diarrhea (88.9 %) as well as in exactly half of the children with acute respiratory infection.Seeing the P value of 0.058 (P > 0.05), it is conclusive that infection is insignificant in triggering stunting among babies and toddlers in the study area.This result differs from a number of research stating that infectious diseasesproportionate to hygiene is substantial in causing stunting [25][26][27][28].
Nine out of eleven children born with low weight and nearly half of the ones born with normal height (48.3 %) were indicated for stunting.With the P value of 0.036 (P < 0.05), low birth weight has higher risk of stunting.
From 55 boys observed, 33 of them (60 %) showed signs of stunting in their height growths.The same thing appeared in 19 out of 45 girls (42.2 %).Since the P value came out with 0.077 (P > 0.05), gender is therefore insignificant in instigating a stunting case.
Referring to ages, indications of stunting occurred by 50.0 %, 62.1 %, 57.1 %, 28.6 %, and 85.7 % in children of 1 yr old, 2 yr old, 3 yr old, 4 yr old, and 5 yr old, respectively.The P value of 0.029 (P < 0.05) affirms the significance of age in the presence of stunting, thus in agreement with research articulating that the risk of stunting heightens when a child is older [29,30].Despite the vacillation, this study has proven that 5 yr old children with stunting are of the highest rate.

Multivariate analysis
Run to ascertain characteristic factor towards stunting cases in the study area, the multivariate analysis was simultaneously performed with the logistic regression test to discover the most influential independent variable towards stunting as the dependent one.Independent variables with P < 0.25 as per bivariate analysislisted in Table 5 were made candidates for the purpose.With P value of 0.649much higher than the limit of 0.25the independent variable of age was pronounced as insignificant and therefore eliminated.The second stage calculated the three remaining independent variables, and the answers are presented in Table 7.As all independent variables are of P < 0.25, no further removal was required.With a negative regressive coefficient rate (-0.459) and P value of 0.077 (P > 0.05), infection is less significant in stunting towards babies and toddlers in the study area.However, the highrisk rate (CI 95 %) indicates that children with infectionsparticularly helminthiasis, diarrhea, and acute respiratoryare 0.632 times more prone to stunting than those with no infection.
The positive regressive coefficient rate (1.708) and P value of 0.044 (P < 0.05) conclude that birth weight is significant in stunting.Often found in children with low birth weight (< 2500 g), they have tendency to stunting 5.519 times more than children with normal birth weight.
Gender also has a positive regressive coefficient rate (0.872), with P value of 0.045 (P < 0.05).In the study area, signs of stunting are shown in boys as they are 2.392 times more liable to it than are girls.
An accuracy raterepresented by overall percentageof 61.0 % is established on infection, low birth weight, and gender as significant causes of stunting, with low birth weighthaving the highest odds ratioas the most significant independent variable.However, the three independent variables appointed cover only 15.1 % of stunting foundation in the study area.

Conclusion and recommendation
This study has discovered that in Sungai Rengit Murni Village, Banyuasin Residence, South Sumatea Province, Indonesia, boys of 5 yr old and younger who were born with below-normal weight and have experienced infection are most likely impaired in their height growth.An early response to the conditions should be advantageous to diminish the rate of stunting in the area.While relatively accurate, the coverage of the study needs to be broadened in the future researches on the matter of stunting in babies and toddlers in the hope of keeping the next generation healthy and productive.

Table 1 .
Data t-test on birth weight.

Table 2 .
Child characteristics compared to height.
very short (14.6 %) was notable.As of 11 children with low birth weight, only 2 (18.2 %) were of normal height while the rest were short (36.4 %) or very short (45.5%).

Table 3 .
Stunting determination based on birth weight and birth length compared to weight and length at measurement.

Table 5 .
Significant independent variablesin stunting among children in Sungai Rengit Murni, Banyuasin, South Sumatra, Indonesia.The logistic regression test operated LR Backward method, where the aforementioned independent variables were taken account on before removed in accordance with certain statistical significance.The outcome of the first stage is as detailed in Table6.