The investigators evaluated the impact of water, sanitation and hygiene educational Intervention and hand washing with soap on the incidence of diarrhea among under-five children.

METHODS Ethics statement The study protocols have been reviewed and approved by the Jigjiga University ethical review committee. Study consent was obtained from district administration, district health office and community leaders. Control group received books and pens to support the children's education during the study period and health education after the completion of the study. Written consent was taken from the mothers of the children. During the study no one was harmed nor died of the disease.

Study Setting The current study was conducted in the rural areas of Jigjiga district of Ethiopian-Somali Regional State (ESRS) in the Eastern Ethiopia from February 1 to July 30, 2015. The ESRS is one of the nine regional states that constitute the Federal Democratic Republic of Ethiopia. Pastoralism is a way of life whereby the livelihood of the people depends on raising livestock and living on its milk and meat (23). Nearly 90% of the population in ESRS resides in rural areas, leading either a pastoralist or agro-pastoralist lifestyle. The majority of these pastoralists are nomadic and have a pattern of seasonal movement from place to place for cattle grazing. Jigjiga district is one of the 68 districts of the region, part of Fafan zone, with a total population of 277,560 according to 2007 census conducted by the central statiscal agency of Ethiopia of whom 149,292 are men and 128,268 women. A total of 34 Kebele is found in the district of which 4 are urban and 30 are rural Kebele in the district. Finally Jigjiga district was selected for the study because of its notable of high incidence of diarrhea disease and acute watery diarrhea epidemics as reported by the Regional Health Bureau.

Sample size determination The sample was calculated by using methods published by Hayes and Bennett (24) to determine the number of communities required to detect a between-group difference in diarrhea incidence density for the under five years old. Given recent studies conducted in Eastern Ethiopia indicated disease rates for incidence of child diarrhea among home based chlorine treatment intervention and non intervention group of 4.5 episodes/100 person week observations and 10.4 episodes/100 person week observations respectively (13), 80% power, 95% confidence interval, considering a coefficient of between-cluster variation (k) of 0.62, adjusting for loss of follow up of individual households, incomplete longitudinal data and 10% drop out and design effect of three from clustering. The final sample 24 communities was calculated (12 intervention and 12 controls) with 50 children of under-five in each cluster.

Study design and procedures The study subjects was divided into 2 study groups according to clusters of community that received hygiene education and hand washing with soap interventions and those control clusters that did not receive any interventions. Only rural community of the district were selected for the study. A household was considered eligible for the study if the following criteria are met: a) at least one child aged 0-59 months living in the home and b) non model health extension household. A household was excluded if the household graduated model health extension households. The 24 communities were allocated to the intervention and control arms by having a meeting with community leaders, Kebele heads and representative from the district health office.

Intervention

The intervention of this study was an educational intervention of water and sanitation hygiene and hand washing with soap to improve water-sanitation behaviours that shown to be associated with high rates of childhood diarrhea in this region from the baseline studies:

1. Water storage behavior
2. Latrine availability and utilization
3. Hand washing after defecation and before meal preparation and eating by using soap Mothers/caretakers in the intervention group were instructed to keep their water storage container clean and covered, to have a latrine and utilize properly, and to wash their hands and children's hands ideally with soap after defecation, before meal preparation and eating.

Mothers of the control group continued their way disposal of water storage behavior, latrine sanitation and hand washing practices.

Data collection A baseline survey was conducted on mother/caretaker, child and environmental characteristics and pre-intervention diarrhea rates by using questionnaire that was translated from English to local language, Somali. The field workers also arranged neighborhood meetings to illustrate the pamphlets and show health problems resulting from hand and water contamination and to show specific instruction on how to use the intervention assigned to the cluster in every week during the study period. Each participating household received a package of health education messages and soap (white bars). The field workers encouraged mothers/care takers to wet their hands, lather them completely with soap, and rub together for 1 minute after cleaning and child who had defecated, before food preparation, before eating and before feeding children.

Outcome assessment The Primary outcome was longitudinal incidence of diarrhea. Diarrhea is defined as the passage of three or more liquid or semi-liquid stools in a 24-hour period or the passage of at least one liquid or semi-liquid stool with blood or mucus (25). The data collectors visited intervention and control households in every 2 weeks period (12 visits) at the same time to collect outcome data. At each visit, daily episodes of diarrhea over the last previous 2 weeks were recorded for each under-five children based on the report of the mother/caretaker of the child. The secondary outcome was bacteriological quality of drinking water at household level.

Data analysis All data collected was cross checked by the field supervisors at the field sites on a daily basis. Prior to data entry, base line and surveillance data forms were checked for completeness and consistence. Data was double entered on to EPI data Version 3.5.3 and statistical analysis was performed by using STATA Version 13. Intention- to- treat analysis was used to compare the incidence of diarrhea among children under-five years of age between intervention and control arms. During the six months after the intervention, the rate of diarrhea (per 100 person-weeks) in children under-five years of age was measured for the intervention communities and for the control communities. Water quality parameters were also determined and compared among the intervention and control communities.

Inclusion Criteria:

• At least one child aged 0-59 months living in the home
• And non model health extension household.

Exclusion Criteria:

• A household was excluded if the household graduated model health extension households.