High Prevalence of Undernutrition among Children in Gondar Town, Northwest Ethiopia: A Community-Based Cross-Sectional Study

Objective To assess undernutrition and associated factors among children aged 6–59 months in Gondar Town, northwest Ethiopia. Methods A community-based cross-sectional study was conducted in 2014. Multistage sampling method was used to select study participants. Structured interviewer administered questionnaire and anthropometric measurements were used. Binary logistic regression was fitted to identify associated factors. Results The prevalences of wasting and stunting were 6.8% and 45.7%, respectively. Higher odds of wasting were observed among children whose fathers were daily laborers (AOR = 2.63), children who had eating problem (AOR = 2.96), and those who were not exclusively breast-fed for the first six months (AOR = 5.63). Similarly, higher odds of stunting were found among female children (AOR = 1.65), children who lived in households having four to six families (AOR = 2.14), and children who did not start breast-feeding within one hour of birth (AOR = 0.67). Conclusion Childhood undernutrition was a significant problem. Child eating problem, paternal occupation, and exclusive breast-feeding were associated with wasting, whereas family size, child sex, and breast-feeding initiation time were associated with stunting. Therefore, strengthening of early initiation and exclusive breast-feeding, promoting healthcare seeking behavior, and designing social support programme for poor family are recommended to reduce undernutrition.


Background
Undernutrition encompasses all stunting, wasting, underweight, and micronutrient deficiencies [1,2]. According to World Health Organization (WHO), undernutrition is defined as -scores less than −2 standard deviations, irrespective of the indicators used [3]. Child undernutrition (stunting and wasting) has been a serious global public health problem for the past many decades, especially in the developing countries including Ethiopia [4,5]. It is considered as a very critical public health problem when the prevalence of stunting and that of wasting among children are higher than 40 and 15%, respectively [6]. The consequences of undernutrition are numerous, such as increased mortality, 2 International Journal of Pediatrics irreversible brain damage, and negative effect on cognitive ability, which have a cumulative effect on social development [7][8][9]. In addition, it has been associated with adverse functional consequences, such as overweight, obesity, insulin resistance, hypertension, dyslipidemia, a reduced capacity for manual work, and other chronic noncommunicable diseases during adolescence and adulthood [8,10,11]. Furthermore, it has a negative effect on future reproductive outcomes [12].
Globally, around 8 and 26% of children below five years of age are wasted and stunted, respectively. The highest prevalence of childhood undernutrition is found in Africa (38.2%) followed by Southeast Asia (27.6%), while the lowest magnitude is found in Latin America and the Caribbean countries (13.5%) [13,14].
In Ethiopia, the magnitude of childhood undernutrition has decreased from 58% in 2000 to 40% in 2014 with regional differences ranging from 52% in Amhara to 22% in Addis Ababa [15,16], but it is sustained as the major public health problem. For instance, 28% of all child mortality, 16% of all repetitions in primary school, and 8% of the workforce lost are associated with undernutrition. In addition, it is also associated with an estimated annual cost loss of 55.5 billion Ethiopian Birr (ETB), which is equivalent to 16.5% of the country's gross domestic product (GDP) [17].
Previous researchers from different corners of the globe confirmed that undernutrition is a complex interaction between different factors. Accordingly, poor child feeding practice [18], being male [19], poor household wealth status [19], prelacteal feeding and large household size [4,5], and poor paternal educational status and an eating problem [5] were factors associated with undernutrition.
The Government of Ethiopia has planned to reach the zero-level undernutrition by 2030 [20]. Accordingly, it endorsed a National Nutrition Program, prepared infant and young child feeding manual, implemented monthly child growth and monitoring program, and has been working in collaboration with different nongovernmental organizations, such as Micronutrient Initiatives (MI), United Nations Children's Fund (UNICEF), and Save the Children. However, about 40 and 9% of children below five years of age were stunted and wasted, respectively [16].
In addition, level of undernutrition depends on the ecological setting of the country, and there is a need to investigate the magnitude and risk factors to generate comprehensive data for policy makers and governmental and nongovernmental organizations working on child health and nutrition. Therefore, the aim of this study was to assess the magnitude and associated factors of undernutrition among children aged 6-59 months in Gondar Town, northwest Ethiopia.

Study Period and Setting.
The study was conducted in Gondar Town from 15 October to 30 November 2014. The town is found in North Gondar Zone, Amhara Regional State of Ethiopia, and is located at 750 km from Addis Ababa to the northwest. According to population projection rate, a population of around 300,000 had been estimated to have resided in Gondar Town in 2014. Administratively, the town is divided into 12 administrative areas (subcities) which consist of 21 kebeles (the smallest administrative units in Ethiopia).

Study Participants, Sample Size, and Sampling Procedure.
All children who are aged 6-59 months and had lived in Gondar Town for at least six months were eligible for the study. The sample size of the study was calculated using the OpenEpi software, version 2.3, by assuming 46% of stunting prevalence [4], a 95% Confidence Interval, and a 5% margin of error. Finally, the sample size of 764 was obtained by considering the design effect of 2. A multistage sampling followed by the simple random sampling technique was employed to reach the study participants. Initially, four of the 12 subcities were selected by the lottery method. Then, four kebeles (one kebele from each subcity) were selected. A total number of children in the selected kebeles were obtained from health extension workers, and then the total number of children included in the study was proportionally allocated to each kebele. Finally, a simple random sampling method was used to select the study subjects from the list of children in each kebele.

Data Collection Instrument and Procedure.
A structured interviewer-administered questionnaire was used to collect data. The questionnaire was adopted from EDHS (Ethiopian Demographic and Health survey) 2011 and other similar studies with some modifications to fit the local context. The questionnaire consisted of the sociodemographic characteristics of the children and their parents, child health, dietary practice related factors, and maternal health service utilization.
The questionnaire was initially prepared in English and translated to Amharic and, finally, retranslated to English to maintain consistency. A pretest was done on 5% of the sample outside the study area. Two-day training was given to data collectors and supervisors. The data were collected by clinical nurses and public health experts.

Assessment of Comorbidity Symptoms and Eating Problem.
Assessment of comorbidity symptoms was begun by asking the mothers using a two-week recall period. Finally, the prevalence of undernutrition was compared among children who had the symptoms and those who did not. Similarly, eating problem was assessed by asking the mothers whether or not the child faced difficulty of swallowing, eating things that are not really food, and eating only certain types of food.

Assessment of Exclusive
Breast-Feeding. Children were considered exclusively breast-fed for the first six month when the child had only received breast milk (including breastfeeding by a wet nurse and feeding expressed breast milk and children received ORS, drops, and syrups) [21].

Anthropometric Measurement.
Birth date was determined by asking mothers and by cross-checking against immunization status certificates of the children. The weight of children aged 24-59 months was taken in light clothes and no shoe to the nearest 0.1 kg by a Seca beam balance International Journal of Pediatrics 3 (Germany, serial number 5755107131646 with graduation of 0.1 kg and measuring up to 160 kg). Similarly, the weight of children aged 6-23 months was measured by the salter scale to the nearest 0.1 kg (Germany, serial number 3541317009 with graduation of 0.1 kg and measuring up to 20 kg). The height of children aged 24-59 months was measured in Frankfurt (the child's head, shoulders, buttocks, knees, and heels touch the vertical board) position using the Seca vertical height scale (Germany, model number 213) to the nearest 0.1 cm, whereas the length of children aged 6-23 months was taken by a measuring board to the nearest 0.1 cm in recumbent position. Instrument calibration was done before measuring the next child. The weight and height of mothers were also measured in standing position using Seca beam balance and Seca vertical height scale, respectively.
The anthropometric data of the children were entered into the ENA/SMART software, version 2011. -scores of nutritional indices, such as Weight-for-Age (WAZ), Weight-for-Height (WHZ), and Height-for-Age (HAZ), were calculated using the WHO Multicenter Growth Reference Standard. Finally, children were classified as stunted, underweight, and wasted when the HAZ, WAZ, and WHZ scores were less than −2 standard deviations (SD), respectively [3]. Maternal body mass index (BMI) was calculated.

Data Processing and Analysis.
The collected data were manually checked for completeness and consistency of responses. Then, the data were entered into Epi Info, version 7, and exported to SPSS, version 20, for further analysis. Descriptive statistics were used to summarize variables. Both bivariable and multivariable binary logistic regression analyses were used to identify variables associated with undernutrition. Variables with a value less than 0.2 in the bivariable analysis were fitted into the multivariable binary logistic regression analysis to control the possible effect of confounding. Both Crude Odds Ratio (COR) and Adjusted Odds Ratio (AOR) with the corresponding 95% Confidence Interval (CI) were calculated to show the strength of association. Finally, in the multivariable analysis, variables with a value less than 0.05 were considered statistically significant.

Sociodemographic and Socioeconomic Characteristics of
Parents. A total of 707 child-mother pairs were included in the study. The median age of the mothers was 27 years (Interquartile Range (IQR): 25-31 years). About 9.6% of the respondents had a family size of greater than seven. Around one-third (31.1%) of the mothers were unable to read and write. Most (70.7%) and nearly one-third (32.9%) of the mothers and their husbands were housewives and government employees, respectively (Table 1).

Child Feeding and Health-Related Characteristics.
Of the total children, 378 (53.5%) were males. The mean (±SD) age of the children was 30 ± 14.39 months. One-fifth (21.5%) of the children were sick two weeks prior to the date of survey. Four (0.6%) of the children had vitamin A deficiency (Bitot's spot) ( Table 2). Fifty-eight (8.2%) of the children were given prelacteal feeding, and 5.8% of the mothers used bottles for complementary feeding. More than two-thirds (81.6%) of children were exclusively breast-fed for the first six months (Table 3).

Discussion
Childhood undernutrition was a significant public health problem in the study area. Eating problem, exclusive breastfeeding status for the first six months, family size, and breastfeeding initiation time were associated with undernutrition. The prevalence of stunting was 45.7% (95% CI: 41.9, 49.5%), which is comparable to the finding of a study done in Dembiya District, Ethiopia (46%) [4], but it is lower than other districts of Ethiopia: Shire (Inda Selassie) (56.6%) [5] and Belesa District (57.7%) [2]. This may be due to Ethiopian recent nutrition programme that gives special attention to child nutrition and the promotion campaign on infant and young child feeding options. In addition, in this study, most of the mothers were educated: attended secondary and above school. Educated mothers can easily understand the effect of undernutrition on child health and survival. Furthermore, educated mothers are more likely to provide nutrient dense, adequate, and frequent complementary feeding to their children. Previous reports have also documented that boosting maternal knowledge is one of the fundamental instruments to break the intergenerational cycle of malnutrition [22].
Furthermore, the observed prevalence of stunting is also higher than what is reported in Indonesia (28.4%) [19], Bhutan (34.9%) [23], and EDHS (40%) [16]. This could be explained by cultural and food security status differences which may have an effect on the infant and young child feeding practices in the three settings: Ethiopia, Indonesia, and Bhutan.
The prevalence of wasting in the study area is in line with the finding from Ethiopia (9%) [24]. However, the prevalence is higher than the reports in Kenya (2.6%) [25] and Nigeria (3.7%) [26]. This is due to low family income in Ethiopia compared to Kenya and Nigeria. Supporting this argument, paternal occupation was one of the determinant factors of infant and young child wasting, according to the finding of this study [27].
In agreement with another study [28], higher odds of stunting were observed among females compared to males. In contrast, stunting was higher among males compared to females in studies done in Indonesia [19] and China [29]. This might be due to cultures in which parents give special attention and feed male children better than females.
Similar to other studies [30][31][32], the odds of stunting were higher among children from larger families compared to children from a family with less than three members. This is due to the fact that when families are large and their resources are limited, the available food is shared by all members, reducing the amount individuals get. In addition, low-quality foods and inappropriate feeding practice may contribute to the high prevalence of undernutrition [33][34][35].
Exclusive breast-feeding for the first six months is one of the strategies employed to decrease the burden of undernutrition [30]. Similarly, the higher odds of wasting were observed among children who were not exclusively breast-fed for the first six months compared to their counterparts. This is due to an early and late introduction of complementary foods. Early initiation of complementary feeding has been strongly associated with infection, which leads to an increased energy demand and loss of appetite and nutrients [36]. In addition, after six months, the energy requirements of children are not satisfied with breast milk alone [37].
In this study, the odds of being wasted were nearly 3 times (AOR = 2.96; 95% CI: 1.13, 7.78) higher among children who had eating problems for the last two weeks prior to the survey compared to children who did not have eating problems. This finding is in line with another finding from Ethiopia [38]. It is evident that recurrent infection is one of the determining factors for the high prevalence of undernutrition among under-five children [39].
Higher odds of wasting (AOR = 2.63; 95% CI: 1.10, 6.27) were found among children whose fathers were daily laborers compared to children whose fathers were government employees. This finding is supported by finding from Bangladesh [40]. In Ethiopia, men are the main source of family income. Family income depends on the type of paternal occupation, as the food purchasing power of the family is dependent on family income. Studies show that poor household wealth status has been strongly associated with inadequate intake of nutrient and poor personal and environmental hygiene, which are likely to increase the frequency of infection and undernutrition [29,41]. In addition, government employees have a fixed monthly income, and most of them are educated. Educated individuals are more likely to follow the recommended child feeding practices, which improve the nutritional status.

Limitation of the Study
Even though adequate training was given to data collectors and supervisors, instruments were calibrated frequently, and respondents were clearly informed about the objectives of the study; still, there might be some limitations. First, there might be some intraobserver bias during weighing and recording. Second, the study may not show the causal relationship between undernutrition (stunting and wasting) and selected variables due to its cross-sectional nature.

Conclusion
Despite the fact that Ethiopia has been planning to reach zero level of childhood undernutrition, the study showed a significant public health problem in the study area. The finding alerts the public authorities about the implementation of undernutrition reduction initiatives. Therefore, strengthening early initiation and exclusive breast-feeding, promoting healthcare seeking behavior, and designing social support programme for poor family are recommended to reduce the burden of childhood undernutrition. Weight-for-Age WHO: World Health Organization WHZ: Weight-for-Height.

Data Access
Data will be available upon request from the corresponding author.

Disclosure
Mulugeta Melku and Zegeye Abebe are co-first authors of the manuscript.

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Authors' Contributions
Mulugeta Melku conceived the study, developed the tool, coordinated the data collection activity, and carried out the statistical analysis. Degefaye Zelalem Anlay participated in the design of the study, tool development, and drafting the manuscript. Belete Biadgo, Asemarie Kebede, Bamlaku Enawgaw, and Tsedalu Melku were involved in data collection supervision and drafting the manuscript. Zegeye Abebe participated in the design of the study and tool development, performing statistical analysis, and reviewing and editing the manuscript. All authors read and approved the final manuscript. Zegeye Abebe and Mulugeta Melku contributed equally to this work.