Study design

A cross-sectional study design was used with data collection carried out from June to October 2016. One hundred forty-six children (74 diarrheic and 72 non-diarrheic) in Ambagiorgis and 79 children (38 diarrheic and 41 non-diarrheic) in Gedebge were enrolled. The prevalence of diarrhea was determined from interviews with parents/guardians during house-to-house visits. The source population all came from households in Ambagiorgis and Gedebge towns having at least one under-five child. Nine hundred households from Ambagiorgis and 325 households from Gedebge towns, each having at least one under-five child, were fully registered and participants randomly selected. A systematic random sampling technique was used to identify the participating children. During interviews, parents/guardians of children were asked to visit the health center for diagnosis. Stool samples were collected for clinical diagnosis of the children in the health center.