Cross-Sectional Study of Malnutrition and Associated

Factors among School Aged Children in Rural and Urban

Settings of Fogera and Libo Kemkem Districts, Ethiopia
Zaida Herrador1,2, Luis Sordo3,4,5, Endalamaw Gadisa6, Javier Moreno7, Javier Nieto7, Agustı́n Benito1,2,
Abraham Aseffa6, Carmen Cañavate7, Estefania Custodio1,2*
1 National Centre of Tropical Medicine, Instituto de Salud Carlos III (ISCIII), Madrid, Spain, 2 Tropical Diseases Research Network (RICET in Spanish), Madrid, Spain,
3 National Centre of Epidemiology, ISCIII, Madrid, Spain, 4 Department of Preventive Medicine and Public Health, Faculty of Medicine, Complutense University, Madrid,
Spain, 5 Network Biomedical Research Centers, Epidemiology and Public Health (CIBERESP in Spanish), Madrid, Spain, 6 Armauer Hansen Research Institute, Addis Ababa,
Ethiopia, 7 National Centre of Microbiology, ISCIII, Madrid, Spain

Abstract
Introduction: Little information is available on malnutrition-related factors among school-aged children $5 years in
Ethiopia. This study describes the prevalence of stunting and thinness and their related factors in Libo Kemkem and Fogera,
Amhara Regional State and assesses differences between urban and rural areas.

Methods: In this cross-sectional study, anthropometrics and individual and household characteristics data were collected
from 886 children. Height-for-age z-score for stunting and body-mass-index-for-age z-score for thinness were computed.
Dietary data were collected through a 24-hour recall. Bivariate and backward stepwise multivariable statistical methods
were employed to assess malnutrition-associated factors in rural and urban communities.

Results: The prevalence of stunting among school-aged children was 42.7% in rural areas and 29.2% in urban areas, while
the corresponding figures for thinness were 21.6% and 20.8%. Age differences were significant in both strata. In the rural
setting, fever in the previous 2 weeks (OR: 1.62; 95% CI: 1.23–2.32), consumption of food from animal sources (OR: 0.51; 95%
CI: 0.29–0.91) and consumption of the family’s own cattle products (OR: 0.50; 95% CI: 0.27–0.93), among others factors were
significantly associated with stunting, while in the urban setting, only age (OR: 4.62; 95% CI: 2.09–10.21) and years of
schooling of the person in charge of food preparation were significant (OR: 0.88; 95% CI: 0.79–0.97). Thinness was
statistically associated with number of children living in the house (OR: 1.28; 95% CI: 1.03–1.60) and family rice cultivation
(OR: 0.64; 95% CI: 0.41–0.99) in the rural setting, and with consumption of food from animal sources (OR: 0.26; 95% CI: 0.10–
0.67) and literacy of head of household (OR: 0.24; 95% CI: 0.09–0.65) in the urban setting.

Conclusion: The prevalence of stunting was significantly higher in rural areas, whereas no significant differences were
observed for thinness. Various factors were associated with one or both types of malnutrition, and varied by type of setting.
To effectively tackle malnutrition, nutritional programs should be oriented to local needs.

Citation: Herrador Z, Sordo L, Gadisa E, Moreno J, Nieto J, et al. (2014) Cross-Sectional Study of Malnutrition and Associated Factors among School Aged Children
in Rural and Urban Settings of Fogera and Libo Kemkem Districts, Ethiopia. PLoS ONE 9(9): e105880. doi:10.1371/journal.pone.0105880
Editor: Heather B. Jaspan, University of Cape Town, South Africa
Received November 20, 2013; Accepted July 29, 2014; Published September 29, 2014
Copyright: ß 2014 Herrador et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors gratefully acknowledge the financial support of the UBS-Optimus Foundation in Switzerland, (www.ubs.com/global/en/wealth_
management/optimusfoundation.html), via the Visceral Leishmaniasis and Malnutrition in Amhara State, Ethiopia project, and the Tropical Diseases Research
Network in Spain (www.ricet.es/es/) via the VI PN de I+D+I 2008–2011, ISCIII -Subdirección General de Redes y Centros de Investigación Cooperativa RD12/0018/
0001 and RD12/0018/0003. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* Email: [email protected]

Introduction The two main anthropometric indicators used to define

malnutrition– stunting, and wasting or thinness– represent
Adequate nutrition is essential during childhood to ensure different histories of nutritional insult to the child. Linear growth
healthy growth, proper organ formation and function, a strong retardation (chronic malnutrition or stunting) is frequently
immune system, and neurological and cognitive development [1]. associated with repeated exposure to adverse economic conditions,
Nutritional status has a major impact on children’s survival mainly poor sanitation, and the interactive effects of poor nutrient intakes
due to the synergistic relationships between malnutrition and and infection. Low weight-for-height or low body mass index
diseases [2,3]. In Eastern and Southern Africa, malnutrition is a (BMI) for age (acute malnutrition, wasting or thinness) is generally
major underlying cause of the persistently high child mortality, associated with recent illness and/or food deprivation [5].
contributing to more than a third of all deaths among children The causes of childhood malnutrition are diverse, multidimen-
under age 5 [4]. sional, and interrelated. An analytical framework suggested by the

PLOS ONE | www.plosone.org 1 September 2014 | Volume 9 | Issue 9 | e105880

 Malnutrition in School-Aged Children of Ethiopia

United Nations Children’s Fund (UNICEF) categorizes the causes level. According to the 2009 census, the population was 198,374
into (a) immediate causes: inadequate dietary intake and illness, (b) and 226,595 for Libo Kemkem and Fogera, respectively.
underlying causes: insufficient access to food in a household; These two districts are located in a black cotton clay soil flat
inadequate health services and unhealthy environment; and plain. Temperatures are relatively high, but rainfall is unusually
inadequate care for children and women at the household level, abundant, with a mean of 1173 mm per annum. Agricultural
and (c) basic causes: insufficient current and potential resources at activities are dependent on a single rainy season (from June to
societal level [6]. In Sub-Saharan Africa, various indicators of September). Maize, barley and millet are the main food crops,
social economic status have been associated with children’s while rice, vetch and chickpea are the main cash crops. Livestock
nutritional status, such as maternal and paternal educational holdings in sheep and cattle are relatively modest, but livestock
level, parental income, and family assets [7–9]. In addition, child and butter sales make a substantial complement to the predom-
nutrition outcomes in developing countries have been character- inant crop sales. The major hazards to crop production and
ized by large rural-urban disparities over the last few decades [10]. livestock are pests, occasional flooding, and zoonosis such as
In Ethiopia, child malnutrition continues to be a major public anthrax, trypanosomiasis, pasteurellosis and black leg [16].
health problem. According to the Ethiopian National Demo-
graphic Health Survey (2011), the prevalence of both wasting and Study design
stunting in children under 5 years is very high (10% and 44% This cross-sectional survey was part of a UBS Optimus
respectively) [11], while the situation in older children is not so Foundation funded project called Visceral Leishmaniasis (VL)
well known [12,13]. Furthermore, rural-urban disparities in child and Malnutrition in Amhara State, Ethiopia. Among its specific
nutrition, as well as growing urbanization that results in increasing objectives, the project aimed to characterize nutritional, immu-
inequalities in urban areas, underlines the need to improve our nological, and parasitological factors in school-aged children in the
knowledge of the main drivers of urban-rural differences [14]. districts of Fogera and Libo Kemkem. Other methodological
The Amhara Region is one of the four primary agricultural aspects have previously been published [24–26].
regions in Ethiopia [15], and most households rely upon livestock Sampling was carried out by multistage cluster survey. A total of
and crop sales to generate cash income. This region, and especially 886 children aged 4 to 15 years were recruited. Primary sampling
the Tana Zuria Zone, has a moderate population density, fertile units were sub-districts (kebeles) with a high incidence of VL
soils and good rainfall. For this reason, it is amongst the most food according to the 2008 register of the Addis Zemen VL Treatment
self-sufficient regions in the country [16]. Despite this good Centre: one urban (Addis Zemen) and the rest rural: Bura, Yifag
regional profile, other factors may be determining the high Akababi and Agita from Libo Kemkem, and Sifatra and Rib
prevalence of infant malnutrition in this area [17]. According to Gebriel from Fogera. Secondary sampling units were randomly
the Food and Agriculture Organization of the United Nations selected villages (gotts) in each of the selected sub-districts. Third-
(FAO), the four pillars of food security are food availability, stage sampling units were randomly selected households in each of
stability of the food supply, food access and the utilization of food the villages. All children with reported age between 4 and 15 years
by the body [18]. In our context, availability is strongly affected by living in the household at the time of the survey were recruited.
seasonality; many households are only able to produce sufficient Sample size was calculated according to previous estimates of
food to meet their food requirements for less than six months of malnutrition for children under age 5 in the area and taking into
the year [19]. Food access may be affected by market conditions, account a design effect of 2, corresponding to the complex design.
but also by cultural and religious practices. For example, the high
number of fasting days commemorated by the Ethiopian Data collection
Orthodox Church, the main religion in the country, may have All children were measured and weighed according to standard
repercussions on the nutritional status of the community, WHO procedures [27]. Weight was measured to the nearest
particularly in rural Ethiopia [20]. 0.1 kg on a battery-powered digital scale (SECA 881ß). Standing
Even if people get enough to eat, good nutrition requires access height was measured to the nearest 0.1 cm using a portable adult/
to a sufficient, supply of varied, safe and nutritious food to meet infant measuring unit (PE-AIM-101ß).
daily nutritional requirements [21]. Although diet diversity A pre-tested questionnaire translated into Amharic, the local
questionnaires are extensively used in Ethiopia to investigate language, was administered to the caretaker/head of household
relationships between food intake and nutritional status, there is (HH) of each child in the study by trained medical personnel
limited knowledge of nutrition outcomes, dietary practices and (nurses and health officers).
socioeconomic factors among school-aged children in this specific We asked about individual demographic characteristics, health
context [17]. status and behavior determinants. The following household
The present study aimed to [22] describe the prevalence of variables were also recorded: household socio-demographic
stunting and thinness, and their related factors, including dietary characteristics, person in charge of food preparation (PCFP),
habits, and [23] document the differences in nutritional status house construction material and assets (land and cultivation,
across urban and rural areas accounting for household and domestic animal assets) and community variables. Dietary data
individual characteristics in school-aged children in Libo Kemkem was collected through a 24-hour diet recall.
and Fogera, Amhara regional State, Ethiopia.
Statistical analysis
Material and Methods Stunting and thinness were the main outcomes of interest,
defined as height-for-age z-score (HAZ) ,22 and BMI-for-age z-
Study area and population score (BAZ) ,22 respectively. The z-scores were calculated using
The study was carried out during May–June 2009 in the the WHO 2007 reference (for children $5 years) and the WHO
districts (woredas) of Libo Kemkem and Fogera (Amhara regional Growth Standards (for children ,5 years old), both computed by
state, Ethiopia). Libo Kemkem and Fogera woredas are located in WHO Anthro Plus software.
the Tana Zuria Livelihood Zone, within the Amhara Regional The dietary data collected through the 24-hour diet recall were
State, northwestern Ethiopia at an altitude of 2,000 m above sea computed into 9 food groups (1. Basic staples, 2. Vitamin A rich

PLOS ONE | www.plosone.org 2 September 2014 | Volume 9 | Issue 9 | e105880

 Malnutrition in School-Aged Children of Ethiopia

fruits and vegetables, 3. Other fruits, 4. Other vegetables, 5. Chronic malnutrition

Legumes and pulses, 6. Meat or Fish, 7. Oil, 8. Dairy and 9. Eggs) The prevalence of stunting was 42.7% in the rural setting and
based on the FAO/FANTA Household Dietary Diversity Ques- 29.2% in the urban setting (p = 0.001).
tionnaire and Guidelines [21]. Chronic malnutrition in rural communities. In the
Frequencies and percentages were used to summarize data and bivariate analysis (Table S1), the prevalence of chronic malnutri-
to explore the differences between rural and urban communities. tion was similar in males and females, however it was significantly
These differences were assessed by Student’s t-test and x2 tests for higher inchildren age 10 and over than in those under 10 years of
continuous and categorical variables, respectively. age (OR: 2.28; 95% CI: 1.66–3.11) in the rural group. Stunting
Bivariate analyses for thinness and stunting and their related was significantly more frequent in children with a previous history
factors were performed, with stratification by setting. Age and sex, of splenomegaly (OR: 2.26; 95% CI: 1.11–4.59) or fever (OR:
considered biologically relevant, and all variables associated with 1.59; 95% CI: 1.15–2.18), and less common in those who slept
each of the outcomes at the p,0.10 level were included in the under a bed net (OR: 0.68; 95% CI: 0.50–0.92) or herded the
multivariable analysis. Logistic regression models stratified by cattle (OR: 0.70; 95% CI: 0.52–0.95). Regarding dietary habits in
setting for stunting and thinness were obtained by using a manual rural communities, the consumption of vitamin A rich fruits and
backward stepwise procedure. P-values less than or equal to 0.05 vegetables and the consumption of any meat or fish the day before
were considered statistically significant. the survey were associated with a lower prevalence of chronic
Data analysis was performed using Anthro Plus v1.02 (WHO, malnutrition (OR: 0.34 (95% CI: 0.13–0.93) and OR: 0.59 (95%
Geneva, Switzerland) and SPSS version 18.0 (SPSS Inc., Chicago, CI: 0.36–0.97), respectively).
Illinois, USA). Results from the multivariable logistic analysis are presented in
Table 3. The difference between the two age groups remained
Ethical considerations significant in the final model (OR: 3.12; 95% CI: 2.15–4.51),
The study was approved by the ethical review boards of the showing that the prevalence of chronic malnutrition was higher in
Instituto de Salud Carlos III (ISCIII), the Armauer Hansen children age 10 and over. Children with fever in the two weeks
prior to the survey were 1.62 times more likely to be stunted (95%
Research Institute, and the Ethiopian National Ethical Review
CI: 1.23–2.32). Herding cattle (OR: 0.43; 95% CI: 0.30–0.63) and
Committee. Support letters were obtained from the Amhara
the consumption of any food from animal sources (OR: 0.51; 95%
Regional State and the woreda health bureaus. Parents/guardians
CI: 0.29–0.91) showed a significant negative association with
gave written informed consent before enrollment of their children
stunting. With regard to household socio demographic character-
in the study.
istics, children from rural areas were 2.97 times more likely to have
chronic malnutrition if the HH was female (95%CI: 1.47–5.98).
Results The number of people living in the house was positively associated
with the prevalence of stunting (p = 0.042), while millet production
Description of the sample
(OR: 0.50; 95% CI: 0.27–0.93) and consumption of products from
The study included a total of 886 children aged 4 to 15 years, of
the family’s own cattle (OR: 0.67; 95% CI: 0.46–0.96) were
which 462 (52.0%) were males. Around 80% lived in the rural
negatively associated.
setting and 50.6% presented malnutrition. The prevalence of
Chronic malnutrition in urban communities. In the
malnutrition was significantly higher in rural than in urban
urban setting, the significant age differences (OR: 3.73; 95% CI:
communities (53% and 42.1% respectively; p = 0.006).
1.89–7.39) found in the bivariate analysis occurred in the same
The mean HAZ and BAZ for the overall study population were direction as in the rural area (Table S1). Furthermore, the
below the WHO references [28], 21.62 (SD = 1.34, range 26.12– prevalence of stunting was significantly higher among children
5.74, p#0.05) and 21.28 (SD = 1.00, range 24.95–3.82, p#0.05), living in a house where the HH (OR: 1.58; 95% CI: 1.00–2.53) or
respectively. the PCFP (OR: 1.60; 95% CI: 0.99–2.66) had not attended school.
Individual characteristics, including demographic data, nutri- Table 3 summarizes the results adjusted by logistic regression.
tional status and diet habits are summarized in Table 1, Older children were more prone to stunting than younger children
disaggregated by rural and urban strata. (OR: 4.62; 95% CI: 2.09–10.21). Years of school attendance of the
There were several differences between children living in rural PCFP was negatively associated with chronic malnutrition among
and urban settings regarding child health status, behavior and children living in urban communities (OR: 0.88; 95% CI: 0.79–
dietary habits. Children living in rural areas slept under a bed net 0.97).
less frequently than urban children (41.1% vs. 67.4% respectively,
p,0.001). On the day before the survey, 18.1% of children in the Acute malnutrition
rural setting had consumed food from animal sources vs. 64% of The prevalence of thinness was 21.6% in the rural setting and
those in the urban setting.. The proportion of children who had 20.8% in the urban setting (p.0.05).
consumed food from four or more food groups was 16.3% vs. Acute malnutrition in rural communities. In the bivariate
39.9%, respectively (p,0.001 for each comparison). analysis (Table S2), the prevalence of acute malnutrition was
Household characteristics are summarized in Table 2. The higher in children age 10 and over (OR: 4.74; 95% CI: 3.24–6.94)
proportion of rural households that owned land was 97.6% vs. and among boys (OR: 1.58; 95% CI: 1.18–2.12). Children from
10.1% in urban households, while the corresponding figures for rural communities who herded the cattle were 2.43 times more
ownership of domestic animals or chickens were 96.3% vs. 36.5%, likely to be thin (95%CI: 1.63–3.61) than those who did not. The
respectively (p,0.001 for each). The distribution of the rest of the number of children living in the house (p = 0.007) and teff
socio-demographic and household variables but one (years of cultivation (OR: 1.53; 95% CI: 1.02–2.30) were positively
education of the HH) also differed significantly between urban and associated with higher prevalence of thinness, while living with a
rural areas. HH under 40 years old and rice farming showed an inverse
association with thinness (OR: 0.75; 95% CI: 0.56–0.99 and 0.63;
95% CI: 0.42–0.95, respectively).

PLOS ONE | www.plosone.org 3 September 2014 | Volume 9 | Issue 9 | e105880

 Malnutrition in School-Aged Children of Ethiopia

Table 1. Individual characteristics, behavioral determinants and dietary habits of school-aged children in rural and urban areas of
at Libokemkem and Fogera districts, Ethiopia, May–June 2009.

CHARACTERISTICS RURAL (n = 711) URBAN (n = 178) p value

No. (%) No. (%)

Sex (female) 339 47,68 88 49,44 0,676

Age group ($10 years) 259 36,43 58 32,58 0,193
Stunting 302 42,66 52 29,21 0,001
Wasting 153 21,61 37 20,79 0,450
Splenomegaly 34 4,79 1 0,56 0,004
Fever in the last 15 days 223 31,41 37 20,79 0,003
Sleeps under a bed net 292 41,13 120 67,42 ,0.001
Herds the cattle 413 58,17 7 3,93 ,0.001
Consumption on day before survey of:
Food of animal source 129 18,14 114 64,04 ,0.001
Basic staples 708 99,58 178 100 0,511
VitA rich fruits and vegetables 24 3,38 11 6,18 0,071
Other fruits 0 0 3 1,69 0,008
Other vegetables 53 7,45 22 12,36 0,029
Legumes and pulses 643 90,44 106 59,55 ,0.001
Meat/fish 79 11,11 103 57,87 ,0.001
Oil 644 90,58 166 93,26 0,164
Dairy 53 7,45 15 8,43 0,381
Eggs 5 0,7 5 2,81 0,032
5 or more food groups 28 3,94 16 8,99 0,007
4 or more food groups 116 16,32 71 39,89 ,0.001
Mean s.d. Mean s.d. p value
Sum of food groups 3,12 0,60 3,42 0,77 ,0.001

doi:10.1371/journal.pone.0105880.t001

After adjusting for socio demographic and household charac- but no significant differences were observed for thinness. These
teristics in the model, sex differences lost significance, and the results are similar to those observed in other developing countries
relationship between thinness and age was slightly weakened (OR: [29]. Various intermediate and distal factors like age, consumption
4.11; 95% CI: 2.74–6.16) (Table 4). Children from rural of food from animal sources and family size were associated with
communities were significantly less likely to be thin if the HH both types of malnutrition in one or both settings. Other
was female (OR: 0.40; 95% CI: 0.16–0.70). The number of determinants such as years of school attendance of the PCFP
children living in the house showed a positive relationship with and consumption of the family’s own cattle products were related
thinness in this setting (p = 0.027), while children from households to only one kind of malnutrition. Although malnutrition among
that cultivate rice were less likely to be thin (OR: 0.64; 95% CI: pre-school children has been well documented in Ethiopia
0.41–0.99). [5,30,31], to our knowledge this is the first research to assess
Acute malnutrition in urban communities. In the bivar- factors related to acute and chronic malnutrition stratified by
iate analysis (Table S2), no significant associations with thinness setting in school aged children. These results may assist
were found apart from age group (OR: 2.76; 95% CI: 1.13–5.80). stakeholders in planning and undertaking contextual and evi-
After adjusting the analysis, age group remained significantly dence-based policy initiatives.
related to thinness (OR: 3.67; 95% CI: 1.63–8.30). Food We found that the probability of a child being malnourished
consumption from animal sources on the day before the survey increases with age. Age-group differences were significant in both
was inversely associated with acute malnutrition (OR: 0.26; 95% strata for stunting and thinness. No sex differences were found in
CI: 0.10–0.67) and thinness prevalence was lower among in either strata. As children mature, household socioeconomic
households headed by literate persons (OR: 0.24; 95% CI: 0.09– characteristics may emerge in conjunction with behavioral and
0.65) (Table 4). biological variables as important risk factors [32].

Discussion Chronic malnutrition

The prevalence of stunting in rural areas in our study was
Our study shows that there is a high prevalence of stunting higher (42.7%) than that found in a study conducted in the same
(39.8%) and thinness (21.4%) among school-aged children in Libo age group in rural settings of Fogera in 2012 (30.7%) [33]. Our
Kemkem and Fogera regions of Ethiopia. The prevalence of study was carried out in 2009, which may partially explain this
stunting was significantly higher in rural areas (42.7% vs. 29.2%), difference due to possible improvements in local conditions;

PLOS ONE | www.plosone.org 4 September 2014 | Volume 9 | Issue 9 | e105880

 Malnutrition in School-Aged Children of Ethiopia

Table 2. Parental and household characteristics of school-aged children in rural and urban areas of Libo Kemkem and Fogera
districts, Ethiopia, May–June 2009.

CHARACTERISTICS RURAL (n = 711) URBAN (n = 178) p value

No. (%) No. (%)

Sex head of household (HH)

Female 52 7.31 75 42.13 ,0.001
Age HH
$40 years 351 49.72 65 36.93 0.008
HH literacy (read and write)
Yes 282 39.83 97 54.49 ,0.001
Person in charge of food preparation (PCFP)
Wife or HH(she) 690 97.05 144 80.9 ,0.001
Other 21 2.95 34 19.1
Years of HH education Mean s.d. Mean s.d. p value
2.70 12.6 4.27 4.92 0.106
Years of education of the PCFP
0.22 1 3.30 4.53 ,0.001
Number of people living in the house
6.41 1.71 5.24 1.58 ,0.001
Number children in the house
2.85 1.13 2.17 0.89 ,0.001

Does the household own land? No. (%) No. (%) p value
Yes 694 97.61 18 10.11 ,0.001
Have domestic animals or chickens?
Yes 685 96.34 65 36.52 ,0.001
Does the household….
cultivate teff?
Yes 481 67.65 4 2.25 ,0.001
cultivate rice?
Yes 231 32.49 2 1.12 ,0.001
cultivate millet?
Yes 70 9.85 5 2.81 ,0.001
cultivate beans?
Yes 22 3.09 1 0.56 0.038
consume products from their own cattle?
Do not consume own cattle products 224 31.68 0
Consume own cattle products 419 59.26 11 6.29 ,0.001
Do not have cattle 64 9.05 164 93.71
consume products from their own goats?
Do not consume own goat products 35 4.96 0
Consume own goat products 34 4.82 7 3.93 0.008
Do not have goats 637 90.23 171 96.07
consume products from their own sheep?
Do not consume own sheep products 62 8.81 2 1.12
Consume own sheep products 90 12.78 6 3.37 ,0.001
Do not have sheep 552 78.41 170 95.51
consume products from their own chickens?
Do not consume own chicken products 80 11.28 2 1.12
Consume own chicken products 343 48.38 47 26.40 ,0.001
Do not have chickens 286 40.34 129 72.47

doi:10.1371/journal.pone.0105880.t002

PLOS ONE | www.plosone.org 5 September 2014 | Volume 9 | Issue 9 | e105880

 Table 3. Multivariable logistic regression analysis of stunting in school-aged children stratified by setting in Libo Kemkem and Fogera districts, Ethiopia, May–July 2009.

VARIABLES RURAL (N = 711) URBAN (N = 178)

n (%) AOR 95% CI p value n (%) AOR 95% CI p value

CHILD CHARACTERISTICS
Sex male 166 (44.86) 1 27 (30.00) 1
female 136 (40.24) 0.78 (0.55–1.10) 0.160 25 (28.41) 1.29 (0.56–2.96) 0.554

PLOS ONE | www.plosone.org

Age group ,10 years 159 (35.24) 1 24 (20.01) 1
$10 years 143 (55.26) 3.12 (2.15–4.51) 0.000 28 (48.39) 4.62 (2.09–10.21) 0.000
Fever in the last 15 days? No 112 (50.45) 1 13 (35.14) 1
Yes 190 (39.09) 1.62 (1.23–2.32) 0.009 39 (27.66) 1.80 (0.73–4.47) 0.204
Does the child herd the cattle? No 141 (47.64) 1 39 (29.24) 1
Yes 160 (38.93) 0.43 (0.30–0.63) 0.000 13 (28.57) 0.75 (0.07–7.71) 0.811
DOES THE CHILD CONSUME…
Any food from animal sources* No 254 (43.87) 1 20 (31.25) 1
Yes 48 (37.21) 0.51 (0.29–0.91) 0.022 32 (28.07) 0.72 (0.31–1.71) 0.463
vit A rich fruits and vegetables* No 297 (43.42) 1 48 (28.74) 1
Yes 5 (20.83) 0.29 (0.83–1.04) 0.057 4 (36.36) 1.32 (0.29–5.99) 0.717

6
Other vegetables* No 284 (43.36) 1 42 (26.92) 1
Yes 18 (33.96) 1.08 (0.49–2.38) 0.855 10 (45.45) 3.00 (0.97–9.38) 0.058
HOUSEHOLD AND LAND PRODUCTION
Sex of head of household Male 267 (40.70) 1 29 (28.16) 1
Female 35 (67.31) 2.97 (1.47–5.98) 0.002 23 (30.67) 0.66 (0.27–1.64) 0.370
Age of head of household ,40 years 151 (42.66) 1 37 (28.16) 1
$40 years 147 (42.14) 0.73 (0.50–1.06) 0.097 23 (30.67) 0.43 (0.17–1.06) 0.069
Years of school-person in charge Mean (sd) AOR 95% CI p value Mean (sd) AOR 95% CI p value
of food preparation
0.16 (0.86) 0.84 (0.70–1.01) 0.065 2.19 (4.0) 0.88 (0.79–0.97) 0.014
Number of people living in
the house
6.43 (1.85) 1.12 (1.01–1.25) 0.042 5.08 (1.41) 0.84 (0.63–1.13) 0.260
n (%) AOR 95% CI p value n (%) AOR 95% CI p value
Does the family cultivate millet? No 280 (43.82) 1 50 (28.90) 1
Yes 22 (31.88) 0.50 (0.27–0.93) 0.029 2 (40.00) 2.35 (0.27–20.19) 0.435
Does the family consume Do not consume own products 39 (60.94) 1 47 (28.66) 1
products from their own cattle?
Consume own products 107 (48.20) 0.67 (0.46–0.96) 0.030 0 - - N.A.
Do not have cattle 154 (36.84) 1.36 (0.72–2.56) 0.348 3 (27.27) 1.07 (0.11–9.23) 0.989

September 2014 | Volume 9 | Issue 9 | e105880

Malnutrition in School-Aged Children of Ethiopia
 Malnutrition in School-Aged Children of Ethiopia

another reason may be that children in the Fogera study had to be

p value
enrolled in school in order to participate, which could result in
selection bias. In addition, our sampling was done in sub-districts

N.A.
N.A.
with a high incidence of VL, a characteristic that may be
associated with fewer resources and worse health status in
children. The prevalence of chronic malnutrition in the urban
area (29.2%) could not be compared to previous data due to the
lack of research targeting this particular age group in this setting.

95% CI
In rural communities, the setting with the highest stunting
prevalence, we found several factors associated with chronic

-
-
malnutrition: age group, fever in the previous 15 days, herding the
cattle, consumption of any food from animal sources, sex of the
AOR

1 HH, family size, cultivation of millet and consumption of the

-
-
URBAN (N = 178)

family’s own cattle products. These factors should be considered

when targeting chronic malnutrition in this region.
Children who had fever in the previous 15 days were 62% more
52 (30.41)

likely to be stunted than those who did not. Infection and

n (%)

malnutrition are intricately linked through extensive, synergistic,

0
0

antagonistic, and cyclical interactions [34,35]. Although our study

area is known to be a low endemic area for malaria and
p value

leishmaniasis [25,36], other specific infections associated with

0.090
0.833

malnutrition (such as chronic parasitic infestations) are highly

prevalent [37,38]. Infectious diseases manifested in the form of
fever affect both dietary intake and utilization, which may affect
child growth. Not having empirical data on specific infections is a
(0.12–1.16)
(0.43–1.96)

limitation, but we considered fever in the previous 15 days could

95% CI

act as a proxy for non-specific infection. On the other hand,

stunted growth and related immunosuppression may lead to
intermittent fever [39].
In our research, children who herd the cattle were less likely to
AOR

0.37
0.92

be stunted. This could be explained by better activity levels in non-

1

stunted children, given that stunted children show behavioral

RURAL (N = 711)

differences in early childhood including apathy and reduced

activity, play and exploration [40].
277 (43.69)

16 (45.71)

Finally, children from households with millet farming and from

8 (23.53)
n (%)

families who consume their own cattle products were less likely to
be stunted in rural communities. Although most rural families own
land and animals (97.6% and 96.3% respectively), we observed
that they do not consume their own products as often. Only 11.1%
of the children had eaten any meat or fish over the last 24 hours,
while 99.6% and 90.4%, respectively, had consumed basic staples
Do not consume own products

and legumes and/or pulses respectively. In Ethiopia, child diet is

based mainly based on plant foods like the traditional Ethiopian
Consume own products

staple food called ‘‘injera’’, a yeast-risen flatbread made of a blend

of cereals, usually served with legumes or pulses. This may not
Do not have goats

provide all the nutritional requirements of children [41]. We are

aware that market access in this livelihood zone is poor; moreover
what little trade interaction exists is restricted to cash crops that
are attractive for their high price (i.e. rice and teff), while other
essential crops cultivated mainly for personal consumption face
disincentives [16,42].
In urban communities, age group and years of school of the
PCFP were significantly associated with chronic malnutrition in
doi:10.1371/journal.pone.0105880.t003
products from their own goats?

children. This educational factor may operate indirectly to affect

children’s nutritional status by determining the quality of the
child’s diet, care and physical environment [30]. The level of
Does the family consume

education of the PCFP may have a positive impact on his/her

*day before the survey.

knowledge on food facilities, controlling contamination, time and

Table 3. Cont.

temperature parameters for controlling pathogens, and advice on

good dietary habits [43].
VARIABLES

Acute malnutrition
The prevalence of acute malnutrition in the rural settings of
Fogera and Libo Kemkem was 21.6%. The prevalence of

PLOS ONE | www.plosone.org 7 September 2014 | Volume 9 | Issue 9 | e105880

 Table 4. Multivariable logistic regression analysis of thinness in school-aged children, stratified by setting in Libo Kemkem and Fogera districts, Ethiopia, May–July 2009.

VARIABLES RURAL (N = 711) URBAN (N = 178)

n (%) Adjusted OR 95% CI p value n (%) Adjusted OR 95% CI p value

CHILD CHARACTERISTICS
Sex male 97 (26.22) 1 21 (23.33) 1
female 56 (16.57) 0.69 (0.46–1.03) 0.073 16 (18.18) 0.86 (0.39–1.92) 0.717

PLOS ONE | www.plosone.org

Age group ,10 years 53 (11.83) 1 18 (15.02) 1
$10 years 100 (38.84) 4.11 (2.74–6.16) 0.000 19 (32.81) 3.67 (1.63–8.30) 0.002
Does the child herd the cattle? No 40 (13.51) 1 36 (21.05) 1
Yes 113 (27.49) 1.50 (0.96–2.36) 0.076 1 (14.29) 0.55 (0.06–5.03) 0.598
DOES THE CHILD CONSUME…
Any food from animal sources No 129 (22.28) 1 18 (28.13) 1
Yes 24 (18.60) 0.83 (0.49–1.41) 0.493 19 (16.67) 0.26 (0.10–0.67) 0.005
HOUSEHOLD AND LAND PRODUCTION
Sex of head of household Male 146 (22.26) 1 23 (22.33) 1
Female 7 (13.46) 0.40 (0.16–0.70) 0.043 14 (18.67) 0.67 (0.26–1.72) 0.407
Literacy of head of household (can read and No 93 (21.83) 1 25 (25.77) 1
write)

8
Yes 60 (21.51) 1.39 (0.91–2.11) 0.127 12 (14.81) 0.24 (0.09–0.65) 0.005
Number of people living in the house Mean (sd) AOR 95% CI p value Mean (sd) AOR 95% CI p value
6.45 (1.68) 0.87 (0.75–1.00) 0.054 5.14 (1.72) 0.87 (0.64–1.20) 0.403
Number of children in the house
3.06 (1.14) 1.28 (1.03–1.60) 0.027 2.19 (0.94) 1.02 (0.61–1.70) 0.955
n (%) AOR 95% CI p value n (%) AOR 95% CI p value
Does the family cultivate rice? No 115 (24.01) 1 36 (20.45) 1
Yes 38 (16.59) 0.64 (0.41–0.99) 0.045 1 (50.00) 3.26 (0.15–72.94) 0.456

doi:10.1371/journal.pone.0105880.t004

September 2014 | Volume 9 | Issue 9 | e105880

Malnutrition in School-Aged Children of Ethiopia
 Malnutrition in School-Aged Children of Ethiopia

underweight found in the study conducted in Fogera was 37.2% Rural versus urban: associated factors
[33]. These results cannot be directly compared as different Ethiopia remains one of the least urbanized countries in the
anthropometric indices were used. And again, we did not find any world [42]. Globally, malnutrition is less common in urban areas
data on thinness to compare with our results in the urban [2,31]. We found that malnutrition in rural communities was
population. associated with food habits and the lack of material resources
In rural settings, age group, sex of the HH, number of children whereas in the urban area, it was better predicted by socio
in the house and rice cultivation were factors associated with demographic factors. Inequalities in child health outcomes are
thinness. The number of children in a household and the known to vary between rural and urban areas, and are often due
prevalence of thinness were positively associated. Larger family to unequal allocation of resources [51].
size may put children at higher risk for acute malnutrition, which Significant urban-rural differences remained in the multivari-
could be due to the imbalance between family size and resources able model. This shows that even in the presence of important
[44]. individual factors and socioeconomic variables, area of residence is
Those whose families cultivate rice were less likely to be thin in still a predictor of children’s nutritional status. Our results
rural communities. In this zone, rice production might be acting as highlight the need to stratify data when rural and urban
a proxy for better socioeconomic status, as rice consumption is communities are targeted in nutritional research in this kind of
relatively recent but is one of the main cash crops in the area [16]. context.
The low consumption of animal source foods and its association
with acute malnutrition has been previously identified as a major Limitations
contributing factor to delayed growth in children [45] and The present study was conducted in two single districts in
suboptimal dietary practices among adolescents in Ethiopia [17]. Ethiopia, thus, the findings may not be generalizable to a larger
In urban areas, children with a literate HH were 4 times less population. Additionally, the cross-sectional nature of this data
likely to be thin than those living in houses headed by illiterate does not allow us to examine causality in the relationship between
adults. Some studies have shown that parental education is malnutrition and diverse risk factors. Seasonality should be given
associated with more efficient management of limited household special attention; the season of the year has a significant effect not
resources, improved utilization of available health care services, only on food security and nutritional status, but also on patterns
and better health-promoting behaviors, all of which are associated and trends of infectious disease incidence. Therefore, consecutive
with better child nutrition [46,47]. This result is similar to what we measurements are desirable..
previously observed for stunting and PCFP years of education in This research is part of a project which aimed to characterize
the urban setting. A possible explanation could be the existence of nutritional, immunological, and parasitological aspects in school-
an educational gap in urban but not in rural areas. aged children from urban and rural villages with a high incidence
of VL in 2005–07. However, the VL prevalence found in the study
Stunting versus thinness: associated factors was very low [25], and no association was found between
Clear differences among risk factors for stunting (Table 3) and nutritional status and asymptomatic infection [24]. Therefore,
thinness (Table 4) emerged from this study. The literature on the we are confident that this limitation does not alter our general
causes of stunting is vast, and conventional thinking is summarized conclusions.
in the Lancet series on maternal and child under-nutrition [48].
Recognized causal factors include prenatal and postnatal periods. Conclusions
Stunting is seen as closely tied to poverty and access to services.
Less knowledge is available on risk factors for thinness [1]. In our Our findings suggest that improving food availability is a
research, risk factors for chronic malnutrition encompass a wide necessary but not sufficient condition to improve the nutritional
range of variables.. The relatively consistent pattern of related status of school-aged children in this region. Especially in rural
factors for stunting suggests that continued exposure to adverse areas, the challenge will be for health and development extension
conditions retards children’s linear growth. Conversely, the greater workers to build on this knowledge through educational
diversity observed in the factors associated with thinness is campaigns when advising households about balanced diet, food
consistent with the fact that a relatively short period of risk production and consumption, and hygienic behavior. It is also
exposure can precipitate its onset in children [32]. important to emphasize that nutritional programs should not be
In rural communities, children from male-headed households biased towards rural areas at the cost of excluding the urban poor.
were more likely to be thin than children from female-headed To effectively tackle malnutrition, nutritional programs should
households (p = 0.043), while stunting was significantly more be oriented to the local needs. Our findings can be used to help
frequent in female-headed households (p = 0.002). The disparate policy makers plan and undertake regional initiatives to streamline
sample size in rural and urban areas may have influenced these recommendations.
results. However, the result in the rural area is consistent with the
study conducted in North Ethiopia by Haidar et al. [49]. This Supporting Information
study found a significantly higher proportion of stunted and
Table S1 Factors related to stunting in school-aged children by
underweight pre-school children in female headed-households,
setting in Libokemkem and Fogera districts, Ethiopia, May–June
whereas the prevalence of thinness was similar [48–50]. Women
2009. Bivariate analysis.
who are single HH may be removed from their support structures
(DOCX)
and may face constraints in accessing services, including food, as a
result of insecurity, cultural discrimination and limited mobility Table S2 Factors related to thinness in school-aged children by
[48,50]. This situation may have a long-term impact in child setting in Libokemkem and Fogera districts, Ethiopia, May–June
nutrition. 2009. Bivariate analysis.
(DOCX)

PLOS ONE | www.plosone.org 9 September 2014 | Volume 9 | Issue 9 | e105880

 Malnutrition in School-Aged Children of Ethiopia

Acknowledgments Author Contributions

We thank the study participants for volunteering to participate in the study; Conceived and designed the experiments: LS EG JM JN AA CC EC.
the data collectors for performing field work; the Armauer Hansen Performed the experiments: ZH AB LS EG EC. Analyzed the data: ZH AB
Research Institute/All Africa Leprosy Rehabilitation and Training Center LS EC. Contributed reagents/materials/analysis tools: AB JM AA JN CC.
and the Fundación Española para la Cooperación Internacional, Salud y Wrote the paper: ZH EC LS. Critically reviewed the manuscript: EG JM
Polı́tica Social for providing logistic and technical support; the Amhara JN AB AA CC.
Regional State Laboratory for allowing us to use their laboratory facility
and for creating a conducive environment during the field work.

References
1. United Nations Children’s Fund, World Health Organization, The World Bank http://www.ctahr.hawaii.edu/sm-crsp/phase1/pdf/amhara.pdf. Accessed 2013
(2012) UNICEF-WHO-World Bank Joint Child Malnutrition Estimates. Sep 1.
UNICEF, New York; WHO, Geneva; The World Bank, Washington, DC; 20. International Food Policy Research Institute (2005) An assessment of the causes of
2012. Available: http://www.who.int/nutgrowthdb/jme_unicef_who_wb.pdf. malnutrition in Ethiopia. A contribution to the formulation of a National Nutrition
Accessed: 2013 Aug 19. Strategy for Ethiopia. Washington, DC, USA Available: http://www.ifpri.org/
2. de Onis M, Blossner M, Borghi E (2012) Prevalence and trends of stunting sites/default/files/publications/ethionutrition.pdf. Accessed: 2013 Sep 1.
among pre-school children, 1990–2020. Public Health Nutr 15: 142–148. 21. Food and Agriculture Organization of the United Nations and Food and
3. de Onis M, Onyango A, Borghi E, Siyam A, Blossner M, et al. (2012) Worldwide Nutrition Technical Assistance Project (2008) Guidelines for Measuring
implementation of the WHO Child Growth Standards. Public Health Nutr 15: Household and Individual Dietary Diversity. FAO, Rome, Italy; Reprinted
1603–1610. 2013..Available: http://www.fao.org/docrep/014/i1983e/i1983e00.pdf. Ac-
4. Ezzati M, Hoorn SV, Lopez AD, Danaei G, Rodgers A, et al. (2006) cessed 2013 Sep 2.
Comparative Quantification of Mortality and Burden of Disease Attributable to 22. Lyons S, Veeken H, Long J (2003) Visceral leishmaniasis and HIV in Tigray,
Selected Risk Factors. In: Lopez AD, Mathers CD, Ezzati M, et al., editors. Ethiopia. Trop Med Int Health 8: 733–739.
Global Burden of Disease and Risk Factors. Washington (DC): World Bank; 23. Amati L, Cirimele D, Pugliese V, Covelli V, Resta F, et al. (2003) Nutrition and
2006. Chapter 4. Available: http://www.ncbi.nlm.nih.gov/books/NBK11813/. immunity: laboratory and clinical aspects. Curr Pharm Des 9: 1924–1931.
Accessed: 2013 Sep 1. 24. Custodio E, Gadisa E, Sordo L, Cruz I, Moreno J, et al. (2012) Factors
5. United Nations System Standing Committee on Nutrition (2010) Progress in associated with Leishmania asymptomatic infection: results from a cross-
Nutrition: 6th Report on the World Nutrition Situation (Geneva, Switzerland: sectional survey in highland northern Ethiopia. PLoS Negl Trop Dis 6: e1813.
UNSCN Secretariat, 2010), Available: www.unscn.org/files/Publications/ 25. Sordo L, Gadisa E, Custodio E, Cruz I, Simon F, et al. (2012) Low prevalence of
RWNS6/report/SCN_report.pdf. Accessed: 2013 Aug 19. Leishmania infection in post-epidemic areas of Libo Kemkem, Ethiopia.
6. United Nations Children’s Fund (UNICEF) (1991) Strategy for improved Am J Trop Med Hyg 86: 955–958.
nutrition of children and women in developing countries: a UNICEF policy 26. Gadisa E, Custodio E, Canavate C, Sordo L, Abebe Z, et al. (2012) Usefulness of
review. New York, USA: 1991. Available: http://repository.forcedmigration. the rK39-immunochromatographic test, direct agglutination test, and leishma-
org/show_metadata.jsp?pid=fmo:3066. Accessed: 2013 Aug 19. nin skin test for detecting asymptomatic Leishmania infection in children in a
7. Owusu WB, Lartey A, de Onis M, Onyango AW, Frongillo EA (2004) Factors new visceral leishmaniasis focus in Amhara State, Ethiopia. Am J Trop Med
associated with unconstrained growth among affluent Ghanaian children. Acta Hyg 86: 792–798.
Paediatr 93: 1115–1119. 27. World Health Organization Working Group (1986) Use and interpretation of
8. Abubakar A, Uriyo J, Msuya SE, Swai M, Stray-Pedersen B (2012) Prevalence anthropometric indicators of nutritional status. Bull World Health Organ 64:
and risk factors for poor nutritional status among children in the Kilimanjaro 929–941.
region of Tanzania. Int J Environ Res Public Health 9: 3506–3518. 28. de Onis M, Garza C, Victora CG, Onyango AW, Frongillo EA, et al. (2004) The
WHO Multicentre Growth Reference Study: planning, study design, and
9. Engebretsen IM, Tylleskar T, Wamani H, Karamagi C, Tumwine JK (2008)
methodology. Food Nutr Bull 25: S15–S26.
Determinants of infant growth in Eastern Uganda: a community-based cross-
29. Food and Agriculture Organization of the United Nations (FAO) (2010)
sectional study. BMC Public Health 8: 418.
Assessment of Nutritional Status in Urban Areas. Available: http://www.fao.
10. Van de Poel E, O’Donnell O, Van DE (2007) Are urban children really
org/ag/agn/nutrition/urban_assessment_en.stm. Accessed: 2013 Sep 2.
healthier? Evidence from 47 developing countries. Soc Sci Med 65: 1986–2003.
30. Girma W, Genebo T (2002) Determinants of Nutritional Status of Women and
11. Central Statistical Agency [Ethiopia] and ICF International (2012) Ethiopia
Children in Ethiopia. ORC Macro, Calverton, Maryland, USA. Available: http://
Demographic and Health Survey 2011. Addis Ababa, Ethiopia and Calverton,Mary- dhsprogram.com/pubs/pdf/FA39/02-nutrition.pdf. Accessed: 2013 Sep 2.
land, USA: Central Statistical Agency and ICF International. Available: http://
31. World Health Organization (WHO) (2006) The African Regional Health
www.usaid.gov/sites/default/files/documents/1860/Demographic%20Health% Report. WHO, Geneva, Switzerland; 2007. Available: http://www.who.int/
20Survey%202011%20Ethiopia%20Final%20Report.pdf. Accessed: 2013 Aug mediacentre/news/releases/2006/pr68/en/. Accessed: 2013 Sep 2.
20. 32. Ricci JA, Becker S (1996) Risk factors for wasting and stunting among children
12. Demissie T, Ali A, Mekonnen Y, Haider J, Umeta M (2009) Demographic and in Metro Cebu, Philippines. Am J Clin Nutr 63: 966–975.
health-related risk factors of subclinical vitamin A deficiency in Ethiopia. 33. Mekonnen H, Tadesse T, Kisi T (2013) Malnutrition and its Correlates among
J Health Popul Nutr 27: 666–673. Rural Primary School Children of Fogera District, Northwest Ethiopia.
13. Woldehanna T, Mekonnen A, Alemu T (2008) Young Lives: Ethiopia Round 2 J NutrDisorders Ther S12: 002.
Survey Report. University of Oxford, UK; 2008. Available: http://www. 34. Custodio E, Descalzo MA, Villamor E, Molina L, Sanchez I, et al. (2009)
younglives.org.uk/files/country-reports/country-report-ethiopia-2008. Ac- Nutritional and socio-economic factors associated with Plasmodium falciparum
cessed: 2013 Aug 23. infection in children from Equatorial Guinea: results from a nationally
14. Fotso JC (2006) Child health inequities in developing countries: differences representative survey. Malar J 8: 225.
across urban and rural areas. Int J Equity Health 5: 9. 35. Katona P, Katona-Apte J (2008) The interaction between nutrition and
15. Frank E (1999) Gender, Agricultural Development and Food Security in infection. Clin Infect Dis 46: 1582–1588.
Amhara, Ethiopia: The Contested Identity of Women Farmers in Ethiopia. 36. Ayele DG, Zewotir TT, Mwambi HG (2012) Prevalence and risk factors of
Agency for International Development, Washington, DC. Available: http://pdf. malaria in Ethiopia. Malar J 11: 195.
usaid.gov/pdf_docs/pnacg552.pdf. Accessed: 2013 Aug 23. 37. Amare B, Ali J, Moges B, Yismaw G, Belyhun Y, et al. (2013) Nutritional status,
16. Tools for Food Security and Nutrition Analysis, Household Economy Approach intestinal parasite infection and allergy among school children in Northwest
and Cost of the Diet (2007) Livelihood Profile Amhara Region, Ethiopia. Tana Ethiopia. BMC Pediatr 13: 7.
Zuria Livelihood Zone (TZA). Available: http://www.heawebsite.org/ 38. Mahmud MA, Spigt M, Mulugeta BA, Lopez P, Dinant GJ, et al. (2013) Risk
countries/ethiopia/reports/hea-lz-profile-tana-zuria-livelihood-zone-tza- factors for intestinal parasitosis, anaemia, and malnutrition among school
amhara-region-ethiopia-2007. Accessed: 2013 Aug 23. children in Ethiopia. Pathog Glob Health 107: 58–65.
17. Belachew T, Lindstrom D, Gebremariam A, Hogan D, Lachat C, et al. (2013) 39. Ehrhardt S, Burchard GD, Mantel C, Cramer JP, Kaiser S, et al. (2006)
Food insecurity, food based coping strategies and suboptimal dietary practices of Malaria, anemia, and malnutrition in african children–defining intervention
adolescents in Jimma zone Southwest Ethiopia. PLoS One 8 [2]: e57643. priorities. J Infect Dis 194: 108–114.
18. Food and Agriculture Organization of the United Nations (FAO) (2009) 40. Black RE, Victora CG, Walker SP, Bhutta ZA, Christian P, et al. (2013)
Declaration of the World Summit on Food Security. Rome, Italy; 16–18 Maternal and child undernutrition and overweight in low-income and middle-
November 2009. Available: http://www.fao.org/fileadmin/templates/wsfs/ income countries. Lancet 382: 427–451.
Summit/Docs/Final_Declaration/WSFS09_Declaration.pdf. Accessed: 2013 41. Mekonnen A, Jones N, Tefera B (2005) Tackling child malnutrition in Ethiopia:
Aug 20. do the Sustainable development poverty reduction programme’s underlying
19. USAID Collaborative Research Support Programs Team (2000) Amhara policy assumptions reflect local realities? Young Lives-Save the Children
National Regional State Food Security- Research Assessment Report. Available: Working Paper 19, London, UK. Available: http://www.younglives.org.uk/

PLOS ONE | www.plosone.org 10 September 2014 | Volume 9 | Issue 9 | e105880

 Malnutrition in School-Aged Children of Ethiopia

files/working-papers/wp19-tackling-child-malnutrition-in-ethiopia-do-the- 47. Abuya BA, Onsomu EO, Kimani JK, Moore D (2011) Influence of maternal
sustainable-development-poverty-reduction-programme2019s-underlying- education on child immunization and stunting in Kenya. Matern Child Health J
policy-assumptions-reflect-local-realities. Accessed: 2013 Sept 7. 15: 1389–1399.
42. Dorosh P, Schmidt E (2010) The Rural-Urban Transformation in Ethiopia. 48. Inter-agency Standing Committee (IASC) (2006) Women, Girls, Boys and Men.
Development Strategy and Governance Division, International Food Policy Different Needs. Equal Opportunities. Gender Handbook in Humanitarian
Research Institute.. Working Paper 13. Addis Ababa, Ethiopia; Washington, Action. IASC, December 2006. Available: http://www.humanitarianinfo.org/
DC, USA; 2010. Available: http://www.ifpri.org/sites/default/files/ iasc/gender. Accessed: 2013 Sep 8.
publications/esspwp013.pdf. Accessed 2013 Sep 7. 49. Haidar J, Kogi-Makau W (2009) Gender differences in the household-headship
43. Coates J, Swindale A, Bilinsky P (2007) Household Food Insecurity Access Scale and nutritional status of pre-school children. East Afr Med J 86: 69–73.
(HFIAS) for Measurement of Food Access: Indicator Guide. Food and Nutrition 50. Woldehann T, Tefera B, Jones N, Bayrau A (2005) Child labour, gender
Technical Assistance Project (FANTA). USAID, Washington, DC, USA; August inequality and rural/urban disparities: how can Ethiopia national development
2007. Available: http://www.fao.org/fileadmin/user_upload/eufao-fsi4dm/ strategies best address negative spill-over impacts on child education and well-
doc-training/hfias.pdf. Accessed: 2013 Sep 7. being?. Young Lives-Save the Children Working Paper No 20; Addis Ababa,
44. Filmer D, Friedman J, Schady N (2009) Development, Modernization, and Ethiopia; London, UK. Available: http://r4d.dfid.gov.uk/PDF/Outputs/
Childbearing: The Role of Family Sex Composition. World Bank Econ Rev 23:
YoungLives/wp20-execsummary.pdf. Accessed: 2013 Oct 8.
371–398.
51. Siddiqi A, Irwin LG, Hertzman C (2007) Early child development: a powerful
45. Ayele Z, Peacock C (2003) Improving access to and consumption of animal
equalizer. Final Report for the World Health Organization’s Commission on the
source foods in rural households: the experiences of a women-focused goat
Social Determinants of Health. WHO, Geneva, Switzerland, June 2007.
development program in the highlands of Ethiopia. J Nutr 133: 3981S–3986S.
46. Glewwe P (2005) The impact of child health and nutrition on education in Available: http://www.who.int/social_determinants/resources/ecd_kn_report_
developing countries: theory, econometric issues, and recent empirical evidence. 07_2007.pdf. Accessed: 2013 OCt 8.
Food Nutr Bull 26: S235–S250.

PLOS ONE | www.plosone.org 11 September 2014 | Volume 9 | Issue 9 | e105880