Prevalence of Anemia and Associated Factors among Infants and Young Children Aged 6–23 Months in Debre Berhan Town, North Shewa, Ethiopia

Background Anemia is a problem of both the developed and developing world, which occurs in all age groups of the population. Half of the anemia cases are due to iron deficiency and affects physical growth and mental development. Nevertheless, there is a scarcity of information about anemia and associated factors among infants and young children aged 6 to 23 months in low-income countries like Ethiopia. Objective The aim of this study was to assess the prevalence of anemia and associated factors among infants and young children aged 6–23 months. Methods A community-based cross-sectional study design was used among 531 mothers/caregivers-children pairs in Debre Berhan Town, North Shewa, Ethiopia, from February 1 to March 2, 2018. The cluster sampling technique was used to select the study participants. Sociodemographic data were collected from mothers/caregivers using pretested structured questionnaires. Hemoglobin levels were measured using a HemoCue analyzer machine (HemoCue® Hb 301, Ängelholm, Sweden). All relevant data were described using descriptive statistics such as frequencies, proportions, mean, and standard deviation. Odds ratio and 95% CI were estimated using binary logistic regression to measure the strength of the association between anemia and explanatory variables. The level of statistical significance was declared at P < 0.05. Results The overall prevalence of anemia was 47.5% (95% CI: 43.1–51.4%) of which 18.3% were mildly anemic, 25% were moderately anemic, and 4.1% were severely anemic. In multivariable logistic regression analysis, household food insecurity (AOR = 2.7, 95% CI: 1.6–4.5), unmet minimum dietary diversity (AOR = 2.5, 95% CI: 1.4–4.3), stunting (AOR = 2.3, 95% CI: 1.2–4.3), and underweight (AOR = 2.7, 95% CI: 1.4–5.4) positively associated with anemia while having ≥4 antenatal care visits (AOR = 0.5, 95% CI: 0.3–0.9) and met minimum meal frequency (AOR = 0.25, 95% CI: 0.14–0.45) had a protective effect against anemia. Conclusion Generally, the study showed that anemia was a severe public health problem among infants and young children in the study setting. Antenatal care visit, meal frequency, dietary diversity, underweight, stunting, and food insecurity significantly associated with anemia. Therefore, efforts should be made to strengthen infant and young child feeding practices and antenatal care utilization and ensure household food security, thereby improving the nutritional status of children.


Introduction
Nutrition during the first 1000 days is very critical for child health, growth, cognitive development, and productivity in later life [1]. Children have rapid physical growth, mental development, and the highest nutritional demand in the first two years of life. is period of life is coupled with the introduction of complementary foods, which are low in their iron content and can result in anemia. us, inadequate nutrition leaves short-term and long-term irreversible effects on physical growth and mental development [2,3].

Study Design, Setting, and Period.
A community-based cross-sectional study was conducted from 1 February to 2 March, 2018, in Debre Berhan Town, North Shewa Zone, Amhara, Regional State, Ethiopia. Emperor Zara Yaqob founded the town and served as the capital of the North Shewa Zone. e town is located 130 km from Addis Ababa, the capital city of Ethiopia, and 690 km from Bahir Dar, the capital of the Amhara Region. e town has an altitude of 2840 meters above sea level. e area practices a mixed farming system: crop production with animal husbandry. e main crop production in the area is barley, wheat, peas, lentils, and linseed. Cattle, sheep, horses, donkeys, and mules are the main live stocks.
According to the 2017 Town Health Administrative Office report, the town has a total population of 88369, of whom 14011 are under-five children and 6707 are children under two years. e town has nine kebeles (the smallest administrative unit) in Ethiopia, one referral hospital, three health centers, and 14 health posts.

Study Participants and Sampling Procedures.
e source population of the study was all infants and young children aged 6-23 months and their mothers or caregivers in Debre Berhan Town. e study population was all infants and young children aged 6-23 months and their mothers/caretakers living in three randomly selected kebeles/clusters. All mother/ caregiver-child pairs living for at least six months in the study were included in the study. Mothers or caregivers who were unable to respond to the interview due to their child's or their own illness and infants and children who had taken iron or vitamin A supplements or subjected to deworming or blood loss due to injury in the past three months were excluded from the study. e cluster sampling technique was used to select mother/caregiver-child pairs. Debre Berhan Town has 9 kebeles and three randomly selected kebeles are considered as clusters. e total number of children in each selected cluster was obtained from health extension workers (HEWs) family folder documentation. Based on the records of HEWs, 577 infants and young children aged 6-23 months were found in the selected clusters. e sample size was determined using a single population proportion formula with the following assumptions: prevalence of anemia among children aged 6-23 months to be 66.6% (22), 5% margin of error, 95% confidence level, design effect of 1.5, and 10% for nonresponse, which gave rise to 564 samples. In the case of more than one child being available in a given household, both children were included in the study. Due to the nature of cluster sampling, 577 infants and young children-mothers pairs living in selected clusters were included in the study.

Data Collection Methods and Instruments.
Socioeconomic and demographic data of mothers or caregivers and their children were collected through home-tohome visits using a pretested structured interviewing-administered questionnaire which was adapted from similar studies [21,22]. e birth date of the children was recorded based on mothers' or caregivers' verbal reports. e child's dietary diversity score was assessed using the dietary diversity assessment tool adapted from the WHO standardized questionnaire for infant and young child feeding practices. It was based on the mother's or caregiver's recall of all foods given to her child in the past twenty-four hours prior to the survey. e dietary diversity score was based on seven food groups consumed by the child: grains, roots and tubers, legumes and nuts, dairy products, flesh foods, eggs, vitamin A-rich fruits and vegetables, and other fruits and vegetables [18]. Household food security status was measured using the Household Food Insecurity Access Scale (HFIAS), a structured, standardized, and validated tool developed by the Food and Nutrition Technical Assistance (FANTA), which has nine occurrences and frequency of occurrence questions based on the previous four weeks or one-month recall method [27].
Anthropometric data, such as the child's height and weight, were also collected. e child's length was measured to the nearest 0.1 centimeters using the United Nation Children's Fund (UNICEF) horizontal wooden length board with a movable headpiece on a flat surface. Children were kept in a recumbent position and the five contact points, including the head, shoulders, buttocks, calves, and heels, were maintained against the length of the board in a straight direction. e child's weight was measured to the nearest 0.1 kg. e weight of a child was estimated by subtracting the mother's/caregiver's weight record from the weight record of both mother and child obtained together. Each anthropometric measurement was measured after removing shoes, heavy clothes, and capes. Each participant was measured twice and the average value was taken when there were variations between two consecutive measurements. e weight scale was adjusted to zero level and calibrated using a standard 2 kg weight object before weighing each study participant.
Hemoglobin level was measured with a HemoCue analyzer machine (HemoCue ® Hb 301,Ängelholm, Sweden). e HemoCue HB 301 analyzer has internal quality control, the self-test. Every time the analyzer is turned on, the analyzer automatically verifies the measurement performance.
is test is performed at regular intervals if the analyzer remains switched on. Upon passing the self-test, the display will show the HemoCue system and three dashes showing that the analyzer is ready to perform the measurement [28]. When an error code was displayed due to self-test failure, a quality control measure was performed according to the recommended guideline.
A separate lancet was used for each child's finger pricking. After wiping off the first two drops of the blood sample, a third drop was collected and completely filled to a cuvette in one continuous motion. Hemoglobin data were adjusted during analysis at an altitude of 2840 meters above sea level and hemoglobin adjustment was done according to the WHO 2011 recommendation [29]. e hemoglobin cutoff point is based on the WHO's classification of under-five anemia, defined as hemoglobin level <11 g/dL [5]. A child with a hemoglobin value <11 g/dL was confirmed as anemic.

Measurements
Anemia: In this study, anemia among children aged 6-23 months was the outcome variable, and it was understood as hemoglobin level <11 g/dL after adjustment for individual hemoglobin value by subtracting 1.9 g/dL for an altitude of 2840 meters above sea level [29]. Food-secure household: If the household head responds no to all questions items 1-9 or responds yes to question item 1 and experiences rarely in the past four weeks [27]. Food-insecure household: If the household head responds at least yes to question item 1 and experiences sometimes in the past four weeks [27]. Minimum dietary diversity was achieved: e proportion of children aged 6-23 months who received 4 or more food items among the seven food groups [18]. Minimum meal frequency was achieved: e minimum number of times a child aged 6-23 months received solid, semisolid, or soft foods. If a child is breastfed, aged 6-8 months, s/he should receive two times and if aged 9-23 months, s/he should receive three times while a nonbreastfed child aged 6-23 months should receive four times during the previous day [18]. Stunting: Length for age <−2SD of the WHO growth standard [30]. Underweight: Weight for age less than −2SD of the WHO growth standard [30]. Wasting: Weight for length <−2SD of the WHO growth standard [30]. Household wealth index: e proxy measure of household living standards based on the sum of available assets of the owner like productive assets, durable assets, domestic animals, and housing characteristics [10]. Diarrhea: Children with a history of ≥3 times loose stool or watery diarrhea per day for two weeks prior to the study [31]. Timely introduction of complementary feeding: e child starts to receive solid, semisolid, or soft foods at the age of six months in addition to continued breastfeeding [18]. Early initiation of breastfeeding: e proportion of children born in the last 24 months who were put to the breast within one hour of birth [18].

Data Quality Control.
Data collection tools were prepared in English and translated into Amharic and then translated back into English to check for its consistency. Pretest was done on 5% of the study sample in the nonselected kebele. Two days of training was given to data collectors and supervisors on the objectives and context of the study, content of the questionnaire, how to fill the questionnaire in the field, interview technique, household selection procedure, respondent approaching technique, hemoglobin, and anthropometric measurement. e relative technical error of measurement (%TEM) was calculated to minimize intra-and interobserver variability [32]. Data collection was supervised by two BSc nurses, and the principal investigator supervised the overall data collection process. Data were double-entered by two independent data clerks for cross-validation.

Statistical Analysis.
First, data were checked for completeness, accuracy, and consistency before entering the computer. Data were then coded and entered into Epi-Data version 3.1 and exported to IBM-SPSS version 22 statistical software for analysis. e household wealth index was computed using principal component analysis (PCA) with all its assumptions, after which it was categorized into five quintiles: lowest, second, middle, fourth, and highest.
Bivariate logistic regression was done to see the association between each independent variable and the outcome variable, anemia. Covariates with P value <0.25 during bivariate logistic regression analyses such as maternal education, wealth index, child's sex and age, food security status, antenatal care follow-up, birth interval, introduction of complementary foods, dietary diversity, meal frequency, undernutrition (stunting, underweight), and presence of fever and diarrhea were retained for multivariable logistic regression analysis to control for all possible confounders and to identify predictors of anemia. Multicollinearity between independent variables was checked using the value of standard error (SE), whereby all variables with SE less than 2 were considered. Model fitness was checked with the Hosmer-Lemeshow test and its P value was greater than 0.05. In a multivariable analysis, adjusted odds ratio (AOR) and 95% confidence interval were estimated to measure the strength of association between the dependent variable and covariates. e level of statistical significance was declared at P value <0.05.

Ethical Approval.
Before the commencement of data collection, Haramaya University Institutional Health Research Ethics Review Committee (IHRERC) reviewed and approved the study with reference number C/Ac/R/D/01/ 878/18. Each study participant was informed, and voluntary, written, and signed consent was secured. Children who were found to be anemic during data collection were linked to the nearest health facility for treatment.

Sociodemographic Characteristics.
Of five hundred seventy-seven mothers/caregivers of infants and young children aged 6-23 months, a total of 531 participated in the study and making a response rate of 92.0%. Seven motherchild pairs were excluded according to the exclusion criteria and 39 were nonrespondents. e mean (±SD) age of the mother was 27 (±4) years. e majority, 467 (87.9%) mothers, had formal education. e mean (±SD) age of children was (14.7 ± 5.1) months. Slightly more than half, 271 (51%), of them were females. More than two-thirds, 356 (67%), of children were in the age range of 12-23 months ( Table 1).

Health-Care-Related Characteristics of Mothers.
e majority, 519 (97.7%), mothers had at least one antenatal care (ANC) visit for the index child. Nearly all, 522 (98.3%), mothers delivered the index child at a health institution. All households had access to safe drinking water or pipe water (Table 3).

Factors Associated with Anemia.
In bivariable logistic regression analyses, maternal education, household's wealth index, child's sex and age, food security status, minimum dietary diversity, met minimum meal frequency, antenatal follow-up, history of diarrhea, fever two weeks prior to data collection, stunting, underweight, birth interval, and age at the commencement of complementary feeding were significantly associated with anemia. In multivariable logistic regression analysis, food insecurity, ANC follow-up, minimum dietary diversity, met minimum meal frequency, underweight, and stunting were significantly associated with anemia.
Children from food-insecure households had 2.7 times (AOR � 2.7, 95% CI: 1.6-4.5) higher odds of being anemic than those children from food-secure households. e odds of anemia were reduced by 50% (AOR � 0.5, 95% CI: 0.3-0.9) among children born to mothers who had four or more ANC visits compared with children born to mothers who had fewer than four ANC visits. e odds of anemia were reduced by 75% among children who met their minimum meal frequency (AOR � 0.25, 95% CI: 0.14-0. 45

Discussion
e present study assessed the prevalence of anemia and associated factors among infant young children aged 6-23 months.
e World Health Assembly set a fifty percent reduction of anemia among pregnant women by the year   2025 as a global nutrition target. But the magnitude of anemia remains a significant public health concern across the globe [33]. Anemia is one of the severe public health issues, where 57% under-five children are anemic in Ethiopia [11]. e study findings showed that 47.5% of infants and young children aged 6-23 months were anemic. e magnitude is unacceptably high, which needs a strong effort to reduce where the level is not more a public health problem.
is finding is similar to the studies reported in China (51.3%) [34] and Romania (46%) [35]. However, it is higher than the results of studies done in China (18.7%-36.6%) [36] and Brazil (29%) [37]. e possible reason for such variation could be the differences in socioeconomic and study period. On the other hand, the result of this study is lower than studies conducted in India (71.9%) [38], Egypt (66%) [39], Cote Devore (78.1%) [40], Kenya (73.2%) [41], and Ethiopia (66%) (53.7%) [21,22], respectively. e possible differences between the current study and the aforementioned studies might be variations in the study setting, period, and geographical location. e variation might also be due to a difference in socioeconomic status, urbanization, place of residence, and child feeding practices of the participants.
In this study, there was a significant association between ANC follow-up and anemia. is is in line with a study done in Ethiopia [22]. Children born to mothers who had focused ANC follow-up had 50% fewer odds of being anemic than children of mothers who had fewer than four antenatal care visits. is is due to the fact that mothers who attend ANC may receive nutrition education and counseling about maternal and infant and young child feeding practices and care during ANC contacts [42]. One should recognize that an investment in maternal health services could also improve the health of their offspring, in which case anemia is one of the public health problems among under-five children and requires such an intervention.
Children from food-insecure households were 2.7 times more likely to be anemic compared to children in foodsecure households. is is consistent with a study done in the US [43]. is could be explained by the fact that food-insecure households are less likely to access adequate, diversified, nutrient, and energy-dense diets. is finding implies that prevention of anemia among the target children requires giving full attention to the household's food security status.
Dietary diversity and meal frequency were also significantly associated with anemia. is is consistent with studies done in China [44] and Ethiopia [21,22]. Children with less minimum dietary diversity were 2.5 times more likely to be anemic than children who used to eat ≥4 food groups.
is might be due to the effect of cereal-based undiversified or monotonous complementary foods that lack essential minerals and vitamins (iron, vitamin A, vitamin B12, and folate) and thus is a major risk factor for anemia among infants and young children [45]. Similarly, children who met minimum meal frequency were 75% less likely to be anemic than those children with unmet minimum meal frequency. is might be due to the fact that children with high meal frequency are more likely to get enough amount of macronutrient, which ensures the child's optimal nutrition through the prevention of deficiency of essential nutrients. e findings related to dietary practices imply that there is a need to improve children's feeding practices to reduce the burden of anemia among the study population.
Moreover, undernutrition was found to be a significant predictor of anemia.
is finding is consistent with the studies done in Rwanda [46] and northeast Ethiopia [21]. Stunted children were 2.3 times more likely to be anemic than their counterparts, and underweight children had 2.7 times higher odds of being anemic than nonunderweight children. is is due to the fact that undernourishment leads to both macronutrient and micronutrient deficiencies, such as protein, iron, and vitamin A, which are responsible for iron deficiency.
is implies that prevention of anemia among infants and young children should take into account designing interventions that target the prevention of macronutrient undernutrition.     is study has the following limitations. First, due to the nature of the design, the study might not show causal relationships between predictors and anemia. Second, the study did not assess other risk factors for anemia such as intestinal parasites, HIV, genetic disorders, folate, vitamin B12, and vitamin A deficiencies. ird, the study used only hemoglobin level to define anemia, which does not reveal the body's level of iron stores. Fourth, household food security was assessed for the past four week's period. is could subject the study to recall and social desirability bias. Lastly, there could be possibilities for anthropometric and hemoglobin measurement errors, but all possible efforts were made to minimize all forms of errors. Despite all the aforementioned limitations, the study tried to assess the prevalence of anemia and associated factors among infants and young children aged 6-23 months at the population level. Nevertheless, the interpretation of the findings of the study needs to be carefully made as measurement error in anthropometry and hemoglobin level determination can either underestimate or overestimate the results of the study.

Conclusion
Based on the WHO cutoff point, the prevalence of anemia observed among children aged 6-23 months in the study area is a severe public health problem. Household food insecurity, less frequent antenatal care follow-up, dietary diversity, meal frequency, and undernutrition were significantly associated with anemia. erefore, integrated efforts need to be made to improve infant and young appropriate feeding practices and maternal ANC service utilization. Similarly, multisectorial work needs to be emphasized to ensure household food security in the study community.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declared that they have no conflicts of interest.