The overall prevalence of anemia among school-aged children was 23.66%, suggesting that anemia is a public health problem among the school-aged children in the area. No similar study was obtained to compare our finding with; however, 11% prevalence was reported from Northern part of Ethiopia . The larger regional variation might be due to differences in geographical variation and differences in life style.
The prevalence of anemia in our study is higher than those similar studies reported from different areas like, Egyptian children 12% , among school-age children in Kenitra Morocco 12.2%  and among Sanliurfa, South-east Turkish children 5.4% . This variation might be due to low socioeconomic status and lower nutritional status of school-aged children in this study area than those reported from elsewhere. For instance a study in Kenitra Morocco indicated that only 8.9% (26/293) and 12.6% (37/293) were stunted and underweight, respectively which is lower than our study where 117 (32.96%) of school-age children were stunted for their age, and 104 (29.30%) were underweight. In addition, majority of our study participants had 5 or more family members which has an effect on their quality of life.
However, prevalence of anemia in our study is much lower than similar reports conducted in Tanzania (79.6%) , in Kenya (35.3%) , in Abia State, Nigeria (82.6%) . The lower prevalence of anemia in our study might be due to the fact that malaria which is one of the major causes of anemia was less prevalent (3.66%) in this study compared to the previous studies. A study in Nigeria reported that malaria parasite, which, may also contribute to the etiology, and severity of anemia through several mechanisms including destruction of red blood cells, was confirmed in majority of the children (93.2%) with Plasmodium falciparum as the primary cause of severe malaria (39.8%). Moreover, the prevalence of hookworm infection in our study is low and shistosomiasis is not detected, which are major causes of anemia but they were the major causes of anemia in Tanzania and Kenya [19, 20].
The prevalence of anemia changed according to socio-demographic characteristics of children and their families which showed statistically significant differences for some variables and not for others. Children from families with low monthly household income were more likely to be anemic than those children from families with high monthly house hold income (AOR = 9.44, 95% CI: 2.88, 30.99). This might be due to the fact that families with low monthly household income may not get enough iron-rich foods and diets of children living in poor families are usually monotonous. This is similar with the study conducted among students of Ningxia and Qinghai’s poor counties of rural China’s .
Stunted children are 5.50 times more likely to be anemic than non-stunted children (AOR = 5.50, 95% CI: 2.83, 10.72) and children who are underweight are 2.07 times more likely to be anemic than children with normal weight (AOR = 2.07, 95% CI: 1.06, 4.05). This might be due to long term effect of low iron intake and other micronutrient deficiencies. It was supported by studies conducted in Tanzania  reported that under weight was significantly associated with anemia (P < 0.05) by suggesting it could be diet related which is long term effect of low iron intake and Vitamin A deficiency among the children.
School-age children infected with intestinal parasites are more likely to be anemic than those none infected children (AOR = 2.99, 95% C.I: 1.05, 8.49 6.73). This is similar with the studies reported from Tanzania  and the study from Edo state, Nigeria . This might be due to the fact that most identified intestinal parasites have their own contribution on blood loss and/or red cell destruction.
Although literature indicates there is indeed a strong association of malaria with increased prevalence of anemia through several mechanisms including destruction of red blood cells , we did not find an association between anemia and malaria infection probably due to low prevalence of malaria in our study area, because the present study did not consider seasonal variation of malaria.