WASH In Schools for Everyone (WISE): a Randomised Evaluation Assessing the Impact of a School-based Water, Sanitation and Hygiene (WASH) Intervention on Child Health and School Attendance in Addis Ababa, Ethiopia

This study uses a cluster-randomised controlled trial design to evaluate the impact of the WASH in Schools for Everyone (WISE) programme. Project WISE, implemented by Splash, aims to improve child health and attendance by 1) improving WASH infrastructure, 2) promoting WASH behaviour change and 3) strengthening menstrual hygiene management (MHM) services targeting girls aged 10 and above, in public schools across Addis Ababa.

60 public schools from planned implementation groups that will receive the WISE intervention will be randomised: 30 schools will be allocated to the intervention group (to receive the WISE intervention during the 2021/2022 academic year) and 30 schools will be allocated to the control group (on a waitlist to receive the WISE intervention after study completion). Randomisation will be stratified by school size (number of students) and presence of a kindergarten. In each selected school, between two and four classes will be randomly selected from grades 2-8 - to obtain a sample size of approximately 100 students in total per school. Students in the selected classes will be enrolled at the start of the academic year (September 2021) and followed up with unannounced visits every 4-6 weeks until the end of the academic year (May 2022). Primary outcomes of pupil-reported illness and absence will be assessed during enrolment and at each follow-up visit. Secondary outcomes are restricted to subsets of students by age and gender, and will be assessed during the final follow-up visit. Students joining the class mid-way through the year will be enrolled in the study, and students who leave the class will be recorded.

Key outcomes of interest are longitudinal prevalence of absence and illness (diarrhoea and respiratory infection) over the course of the academic year, measured as cumulative number of days absent/with illness per 100 school days after adjustment for child- and school-specific covariates. Data collected in intervention schools prior to and/or during the active Splash intervention will not be included in the primary impact evaluation analysis. Sub-population analysis of evaluation outcomes in girls post-menarche will allow assessment of the gendered effects of the intervention. We will also monitor intervention fidelity and compliance, as well as cost-effectiveness of the intervention, through routine data collection with students and school staff.

40 public schools with a kindergarten will be enrolled to assess the impacts of the intervention among kindergarten students (20 intervention, 20 control). Schools enrolled in the intervention and control arms of the main study among older children that have a kindergarten will be also enrolled in the kindergarten evaluation. If there are fewer than 20 such schools in each arm, additional kindergarten schools will be randomised (estimated 10 additional schools per arm). In each selected school, 20 kindergarten students will be randomly selected for participation. Parents of selected students will be invited to participate in weekly telephone surveys over a four-week period in February/March 2022 in order to assess the longitudinal prevalence of illness and absence among kindergarteners.