Stunting and the Associated Factors among Under-ve Children in Shire Endaslassie Town, Tigray, North West Ethiopia

Background: Globally, 158 million under ve children suffer from childhood stunting. On the continent, 87 million stunted were in Asia, 59 million were in Africa and 6 million were in Latin America and Caribbean. In Ethiopian the number of stunting children declines from 6.13 million in 2012 to 5.85 million in 2015 with the prevalence of 38% as per 2016 EDHS. Objective: To determine the prevalence and associated factors of stunting among 6–59 months of age in Shire Endaslassie Town, northwest Ethiopia. Methods: A community-based cross-sectional study was conducted in Shire Endaslassie, northwestern Ethiopia, from April 7 to 20, 2017. Multistage sampling technique was used to select 356 study subjects. Child and family status were assessed using the structured questionnaire and anthropometric measurement was conducted to measure nutritional status of the children. Data were cleaned by EPI-Info version 3.5.1, and height for age was converted to Z-score with ENA-SMART software. A multivariable logistic regression analysis was used to investigate factors associated with stunting. An adjusted odd (AOR) with 95% condence interval (CI) is used to show the strength of the association, and variables with P values <0.05 are considered statistically signicant. Results: The prevalence of stunting was 35% (95% CI: 30 %–40%) among study participants. Unprotected water source [AOR = 3.20, 95%CI: 5.73)], Monthly income< 3000 ETB [AOR = 2.59, 95%CI: 4.66)], providing breast feeding when cry 2.22, 95%CI: 3.99) positively associated with stunting, while, prepare meal [AOR 0.50, 95%CI: usage of family planning [AOR 0.26, 95%CI: were found to be preventive factor.

. Stunting has negative consequence both on individual health as well as on economic growth of the country. Adult with short stature has outcomes lower earing and poor productivity in labor market with an estimated 20% with adult non-stature [9,10]. It is also associated with higher morbidity and mortality [3,4].
Stunting is associated with higher risk of choric illness like elevated blood pressure, renal dysfunction, alerted glucose concentration and increase harmful lipid pro le [6][7][8]. Generally, the impact will be expressed in terms of delayed mental development, poor educational performance, reduced intellectual capacity, and low economic productivity. Generally, it is a strong predictor of human capital and social progress.
Globally, 158 million children of under ve age group suffer from childhood stunting, in 2016. Continent wise 87 million stunted were in Asia, 59 million were in Africa and 6 million were in Latin America and Caribbean. Five sub-regions have child stunting rates that exceed 30%: western Africa (31.4%), middle Africa (32.5%), eastern Africa (36.7%), southern Asia (34.1%) and Oceania (38.3%; excluding Australia and New Zealand). Asia and Oceania have experienced slow or minimal progress in reducing child stunting. However, stunting has declined twice in Latin America and the Caribbean, when compared with Africa from 2000 to 2016 [11].
In Ethiopian, stunting declines from 6.13 million in 2012 to 5.85 million in 2015. According to 2016, Ethiopian demographic health survey stunting is reduced to 38% from 40.4% in 2014. Region-wise, it ranges from 15 % in Addis Ababa city administration to that of 46% in the Amhara region [12].
Ethiopia has endorsed major global and national initiatives to see children free from under nutrition including stunting. The Seqota Declaration to end stunting in children under two by 2030 and the Health Sector Transformation Plan to reduce childhood stunting in under-ve years from 40% to 26% by the end of the year 2020 are part of the national initiatives [14]. Therefore, the study was conducted with the aim of identifying the prevalence and determinate factor of stunting in Shire Endaslassie town, Tigray, Northern West Ethiopia.

Study area and design
A community-based cross-sectional study was conducted in Shire Endaslassie, North Western Ethiopia, from April 7 to 20, 2017. The town is 1074 km away from the capital Addis Ababa and located at 14°6′N 38°17′E with an altitude of 1953 meters above sea level [15].

Sample size and sampling procedures
The study population included children aged 6-59 months in the 3 randomly selected kebeles (local smallest administrative unit) in Shire Endesillasie Town. Seriously ill Children during the whole data collection season and children with spinal curvature (Kiphosis, scoliosis and kiphoscoliosis) were excluded from the study. Three out of ve total kebeles in the town were selected using the lottery method. The total sample size (n = 356) was distributed to each selected kebele proportionally using probability proportional to size of population. The total number of each kebele was obtained from the local kebele administrative o ce. The sample interval (k) for each kebeles was calculated for each kebeles, and the rst household in each kebeles was identi ed using a random number from k number of households. Finally a systematic random sampling technique was used to select participants from each household. For households with more than one eligible child, random sampling was used to select one child for the study. Mothers or caregivers were interviewed on socio demography, housing condition, child variables (eating habits and history of illness) with a pre tested structured questionnaires. Interview questions were revised, edited, and those found to be unclear were modi ed after pretest. Trained and experienced data collectors were used to conduct a face to face interview. Anthropometric measurement was conducted on target children to measure the nutritional status. A horizontal wooden length board and a vertical wooden height were used to measure the length of children 6-23 months of age and height of children aged 24-59 months respectively. Both length and height are measured nearest to the 0.1 cm.

Data Processing and Analysis procedures
The collected data using the interview was coded, entered and cleaned for its completeness and errors, and analyzed using SPSS version 24 statistical software packages. Quantitative variables were expressed as mean (± Standard Deviation), or median (interquartile range); qualitative variables were expressed as a percentage. The anthropometric measurement of height for age (HAZ) was calculated by ENA SMART software, and children less than − 2 SD were classi ed as stunted. Children with HFA between − 2 and − 3 SD were classi ed as moderately stunted and < -3 SD were classi ed as severely stunted. The monthly income of the household was categorized based on Ethiopian Birr and the baseline was 3000 ETB. The stunted versus non-stunted groups were compared using Chi 2 . Bivariate and multivariable logistic regression were done. Factors potentially associated with stunting in univariate analysis with a p-value of < 0.25 were included in a backward logistic regression. The strength of association was presented using the odds ratio and 95% con dence intervals and variables with P-values of < 0.05 were considered statistically signi cant.

Socio demography characteristics of the respondent
The total of 337 respondents participated in this study with respondents' rate of 90.7 % (337/356). The majority of the respondents 288(85.5%) orthodox region follower and 301(89.3%) were married. regarding maternal educational status, 287(85.2%) were attained primary school and above. nearly one third of the household 101(30%) earned more than 3000 birrs per month. The detail is displayed in Table 1.  Housing and environmental condition The majority of the households 213(63.2%) used piped water for drinking. The common waste disposal system among the study population is pit 316(93.8%) and 313(92.8%) households have latrine as displayed in Table 3.  Table 4.

Discussion
The intention of this study was to measure the prevalence and associated factors of stunting among preschool children of shire Endassilasie. Stunting is a cumulative process that can begin in utero and continues to about three years after birth. In the study area, the prevalence of stunting was 35 %. There was a slight age difference in stunting between the study group as the prevalence was 18.1% and 16.9 among male and female respectively. Signi cant difference among age group was also detected marking higher prevalence among 24-47 months of age group (15.1%), followed by 48-59 months (10.4%) then the least among study participant were age group of less than 24 months with the prevalence of 9.5%.
The 35% stunting prevalence testi es that the population in this area is affected by high risk of malnutrition according to the WHO classi cation of severity of malnutrition [16]. The prevalence of stunting among the study population was almost the same when compared with the study conducted in other areas with reasonably high malnutrition burden. For instance the study conducted in Bangladesh, Indonesia and Pakistan showed the stunting prevalence of 36.1%, 33.7% and 55% respectively [18, 19,24]. The same nding was also recorded in the study conducted at Jigjiga town Eastern Ethiopia showing 34.9% of under ve were stunted [17]. However, the study conducted in northern part of Ethiopia showed the lower result when compared to the result obtained from this study. For example the detected prevalence in Libo Kemekem, Lalibela town, Wukro town and Medabay Zena were 49.4%, 47.3%, 49.2% and 56.7% respectively [20][21][22][23]. The difference could be attributed to the difference in the study period, variation in the age category of target populations, Geographic location and the recently initiated nutrition-sensitive intervention activities in the study area.
The nding our study was higher when compared to the study conducted in Southern part of Ethiopia.
with this, the stunting prevalence of the study nding in Kembata, Hawassa and Wolaita Sodo shows 18.8%, 26.6% and 27% respectively [25][26][27]. In Ethiopia there is the difference in socioeconomic, culture, feeding habits, environmental factors, and public service utilization of the community in the study area between Southern and Northern part. Therefore, it's believed that the difference in stunting prevalence might have been contributed by such variation in socio cultural differences between southern and northern part of the country. Additionally, the nding of this study showed that the stunting prevalence increases with the age of the child. This association is inline with other studies conducted in the North and Northwest Ethiopia [29,30]. This might be due to the nutritional status of the mother since stunting has a chronic and cyclic nature. Additionally, poor dietary practice, weaning, lower and inappropriate breast and complementary feeding practices have an effect on child growth as age increases. The other possible explanation for the increased risk of stunting might be due to environmental factors like poor sanitation and hygiene practice which increase the risk of choric malnutrition.
There is also variation in stunting prevalence between male and female. This result in line with studies conducted in Ethiopia [31], Nigeria [32] and the sub-Saharan countries [33]. This could be because of social factors like favoritism towards daughters [33]. Environmental factors such as parasitic infections like intestinal helminths could be worse the existing malnutrition. Additionally, biological factors including the common child illness predominately affect males than females and this in turns contribute for the higher prevalence among males [31].
Choric malnutrition arises from multifaceted and interrelated circumstances in which food security is the core factor. The present study showed that monthly income was signi cantly associated with stunting. The odd of being stunted is substantially higher among the low income group. This nding suggests that a child's health status depends upon the socio-economic status of their families. Similar association was seen in studies conducted in Holeta Town [34], Wolayta sodo [27], Lasta district [35]. This could be due to the fact that rich households have greater purchasing power of food to maintain the health of their children. However, being rich is not guaranteed according to the WHO report, stunting may also arise due to inadequate knowledge of food, feeding practices, inappropriate food allocation, and poor hygiene practices [36].
Additionally, this study identi ed that water source is one the factor associated with child stunting. Children from households using unprotected water sources are more likely to be stunted than children from households that are using protected sources. This is also supported by other similar studies conducted in Arba Mich [37], Medebay Zena [23]. Other study conducted out of Ethiopia showed similar nding; for instance study conducted in Tanzania [38], Zambia and other 59 countries [39] showed the same results. Unsafe drinking water causes diarrheal diseases thereby inhibiting nutrient absorption, which can lead to under nutrition and stunting.
Our results show that having enough time to prepare a meal is inversely associated with the child stunting. The odd of stunting among busy mothers to prepare a meal for their child is 3.8 times compared to children's from mothers that have enough time to prepare meal. Consistent nding has been observed in studies conducted in Southern [25,40] and Eastern [41] part of Ethiopia and Bangladesh [42].
This commonly related to the mother's occupational status that impacts the intimacy with their child.
Additionally, mothers who spent most of their time out of a home cannot provide su cient care for their child. Therefore, busy mothers have short period to breastfeed their child, enforced to cease breastfeeding early that increase exposure to bottle feeding and improper complementary feeding practice.
This study also identi ed a signi cant association between stunting and time of breast feeding. Children who breast feed as per the feeding frequency have less chance of being stunted compared to children that feed in response to crying. A imilar association was seen in the studies conducted in Guto Gida of East Wollega [43], Sidama [44] and Lasta district [35]. Proper breastfeeding has an impact on averting early infant death as well as reducing child stunting. However, breast feeding practice could be affected by socioeconomic factors like low parental education especially mother's education, mothers' employment and another poor socio-economic status.
Another identi ed association was family planning usage and child stunting. Mothers who use family planning have lower chance of having stunted children when compared to nonusers of family planning. This nding was in line with studies conducted in the Sodo Zuria District of Ethiopia [45] and India [46].
The observed association might be due to loss of macro and micronutrient that occurs due to repeated pregnancy, delivery and breast feeding. Moreover, unplanned childbearing could elevate the risk of intrauterine growth restriction (IUGR), low birth weight (LBW), premature birth, and small birth size [47].
Even though this study shows reliable nding when compared with domestic and international journals, it exhibits the following limitations. The study cannot declare a temporal relationship between stunting and other independent variables because cross-sectional design was used for this study. Even though standard procedures were used for the measurement of height/length, measurement errors are inevitable especially within evaluators. Moreover, there might be a recall bias in reporting the age of children.

Conclusion
Our ndings demonstrate a higher prevalence of stunting in Shire Endaslassie's district and this justify that malnutrition is the major public health concern in this area. This study also revealed that a monthly income, water source, frequency of feeding, time of breastfeeding and usage of family planning are the major predictor of stunting. Therefore, strong nutrition-speci c and sensitive intervention should be implemented in the study area to avert the long effect of malnutrition.
Abbreviations EDHS: Ethiopian Demographic Health Survey HAZ: Height for age z score, IUGR: Intra Uterine growth restriction, LBW: low birth weight, WHO: World Health Organization Declarations