Drinking Water Chlorination and Child Survival in Rural Kenya

The purpose of the study is to estimate the effect of community-wide provision of water treatment (chlorine) solution on all-cause child mortality and on infectious disease related child mortality. We will also examine effects on the following secondary outcomes: 7-day diarrhea prevalence, all-cause under-2 mortality, diarrheal disease related child mortality, school attendance, and school enrollment. In addition, and for a subsample of children, we will examine effects on motor development, emergent language and literacy, emergent math/numeracy, and socio-emotional development.

The World Health Organization (WHO) estimates that in 2015 over 2 billion people consumed drinking water contaminated with feces and that approximately half a million people died from diarrheal disease associated with fecal contamination of water (1). Dilute chlorine solution is widely used to treat water as it is effective, safe, and low-cost. While nonexperimental studies suggest strong impacts of chlorine water treatment on child survival (2-4), there has been no previous randomized controlled trial of a water treatment intervention powered to evaluate effects on child survival. We will examine the effect of community-wide provision of chlorine solution on child survival in rural Kenya, where a long-term village-wide chlorination randomized evaluation was implemented. The WASH Benefits Kenya trial was a randomized controlled trial of water treatment, sanitation, handwashing, and nutrition interventions in western Kenya. WASH Benefits Kenya enrolled pregnant women in their second or third gestational trimester between November 2012 and May 2014 and followed children for their first 2 years of life. In communities that were randomized to water treatment, chlorine solution dispensers were installed and refilled as needed. After the WASH Benefits Kenya trial ended, the NGO Evidence Action continued to refill most of the dispensers in the treatment villages. We intend to re-visit all water treatment and control clusters in the trial approximately 6 years after the chlorination intervention was initiated to enroll all women 50 or younger, who had a child since January 1, 2008. We note that our target population includes both women (and children) who were and who were not "enrolled" in the original WASH Benefits Kenya study. The latter group is made up of women (and their children) who were pregnant before the study, women who were in their first trimester at the time of enrollment, and women who got pregnant after study enrollment. We expect to find approximately 22,000 such women. We will briefly survey these women to identify those that gave birth to a child who later died, and then conduct verbal autopsies to ascertain the cause of death. We will also perform free and total chlorine residual testing at households to assess current usage rates and collect GPS data on the location of houses and dispensers. We will estimate the intent-to-treat effect of the community-wide provision of chlorine solution on child survival by 1) comparing post-intervention mortality rates between water treatment and control areas; and 2) comparing changes in mortality rates (before and after the intervention) across treatment and control areas (a difference-in-difference analysis). The primary outcomes are all-cause child mortality and infectious disease related child mortality; secondary outcomes include: 7-day diarrhea prevalence, all-cause under-2 mortality, diarrheal disease related child mortality, school attendance, and school enrollment. In addition, for the subsample of children who were enrolled in the original WASH Benefits Kenya study, we will examine effects on motor development, emergent language and literacy, emergent math/numeracy, and socio-emotional development. Our findings will provide evidence on whether community-wide provision of chlorine reduces all-cause child mortality and infectious disease related child mortality.

Outcome collectors were not informed of intervention status, but could have inferred status from observing intervention hardware during household visits.

This arm includes all households in villages randomized to the active control arm (double-sized) or passive control arm of the original trial. Village-level promoter visited households enrolled in the WASH Benefits Kenya study active control arm and strictly engaged in recording the child's MUAC and referring children identified as malnourished (MUAC<11.5 cm) to health clinics, for two years. These visits were also conducted in all active comparator arms. Households in active control and active comparator villages which were not enrolled in the original study did not receive such visits.

This arm includes all households in villages randomized in the original WASH Benefits trial to the water treatment arm, combined water treatment with handwashing and sanitation (WASH) arm, and combined WASH + nutrition arm. Village-level promoter visited households enrolled in the original trial to promote the interventions for approximately two years.

Hardware: Chlorine dispensers provided for free at communal water sources, available to households in intervention arm who were and who were not enrolled in the WASH Benefits Kenya study. Promotion: Local promoters visited compounds enrolled in the WASH Benefits Kenya study at least monthly during the first year and bi-monthly during the second (and last) year to deliver behavior change messages that focused on the treatment of drinking water for all children living in the household. Compounds in intervention villages that were not enrolled in the WASH Benefits Kenya study did not receive such visits. After the completion of the WASH Benefits Kenya study, the NGO Evidence Action conducted educational campaigns to promote the use of dispensers.

Only households enrolled in the original WASH Benefits study and assigned to the combined WASH and combined WASH + nutrition arms received this intervention. Hardware: Free child potties, sani-scoop hoes to remove feces from household environments, and new or upgraded pit latrine for each household enrolled in the WASH Benefits Kenya study. Upgrades included structural improvements, plastic slabs, and superstructure improvements. Households in intervention villages not enrolled in the original trial did not receive any hardware. Promotion: Local promoters visited compounds enrolled in the WASH Benefits Kenya study at least monthly during the first year and bi-monthly during the second (and last) year to deliver behavior change messages that focused on handwashing with soap at critical times around food preparation, defecation, and contact with feces. Compounds in intervention villages that were not enrolled in the original trial did not receive such visits.

Only households enrolled in the original WASH Benefits study and assigned to the combined WASH and combined WASH + nutrition arms received this intervention. Hardware: Handwashing "dual tippy tap" stations, including jugs for clean and for soapy water, for each compound. Handwashing stations were stocked with soap for the duration of the WASH Benefits Kenya study. Compounds in intervention villages who were not enrolled in the WASH Benefits Kenya study did not receive any hardware. Promotion: Local promoters visited compounds enrolled in the WASH Benefits Kenya study at least monthly during the first year and bi-monthly during the second (and last) year to deliver behavior change messages that focused on the use of latrines for defecation and the removal of human and animal feces from the compound. Compounds in intervention villages that were not enrolled in the WASH Benefits Kenya study did not receive such visits.

Supplement: Lipid-based Nutrient Supplement (LNS) twice daily from ages 6 to 24 months, among children enrolled in the WASH Benefits Kenya study and for the duration of that study. Children in intervention villages who were not enrolled in the WASH Benefits Kenya study did not receive any supplements. Promotion: Local promoters visited compounds enrolled in the WASH Benefits Kenya study at least monthly during the first year and bi-monthly during the second (and last) year to deliver the following behavior change messages: (1) practice exclusive breastfeeding from birth to 6 months of age; (2) continue breast feeding with the introduction of LNS; (3) provide your child micronutrient-rich foods and vitamin A rich fruits and vegetables; and (4) feed your child at least 2-3 times per day when 6-8 months old and 3-4 times per day when 9-24 months old. Compounds in intervention villages that were not enrolled in the WASH Benefits Kenya study did not receive such visits.

Measured 6 years after start of intervention (recall period from start of intervention to day of survey)

Mortality for children who died under the age of 5 years from infectious or parasitic diseases. Based on the 2016 WHO Verbal Autopsy instrument, the category of deaths by "infectious and parasitic diseases" include the following causes (ICD-10 codes in parenthesis): Sepsis (A41) Acute respiratory infection, including pneumonia (J22/J18) HIV/AIDS related death (B24) Diarrheal diseases (A09) Malaria (B54) Measles (B05) Meningitis and encephalitis (G03; G04) Tetanus, excluding neonatal tetanus (A35) Pulmonary tuberculosis (A16) Pertussis (A37) Hemorrhagic fever (A99) Dengue fever (A90; A91) Unspecified infectious disease (B99)

Measured 6 years after start of intervention (recall period from start of intervention to day of survey)

Diarrhea is defined as 3+ loose or watery stools in a 24 hour period. Data will be measured in interviews using mother-reported symptoms with a 7-day recall among children under 5 years.

Emergent language and literacy measured with the International Development and Early Learning Assessment tool.

Emergent math/numeracy measured with the International Development and Early Learning Assessment tool.

Socio-emotional development measured with the International Development and Early Learning Assessment tool.

Mortality for children who died under the age of 2 years from infectious or parasitic diseases. Based on the 2016 WHO Verbal Autopsy instrument, the category of deaths by "infectious and parasitic diseases" include the following causes (ICD-10 codes in parenthesis):

Measured 6 years after start of intervention (recall period from start of intervention to day of survey)

Inclusion Criteria: Had one or more live births since January 1, 2008. Live in a village which was randomized to water treatment or control arms during the WASH Benefits Kenya trial.

World Health Organization. Drinking-water. Available at: https://www.who.int/news-room/fact-sheets/detail/drinking-water. (Accessed: 20th February 2019)

Bhalotra, S. R., Díaz-Cayeros, A., Miller, G., Miranda, A. & Venkataramani, A. S. Urban Water Disinfection and Mortality Decline in Developing Countries.

Galiani, S., Gertler, P. & Schargrodsky, E. Water for Life: The Impact of the Privatization of Water Services on Child Mortality. J. Polit. Econ. 113, 83-120 (2005).

Cutler D, Miller G. The role of public health improvements in health advances: the twentieth-century United States. Demography. 2005 Feb;42(1):1-22. doi: 10.1353/dem.2005.0002.