Evaluation of Safe-Water Programs in Baja California Sur, Mexico

A Stepped Wedge, Cluster-randomized Impact Evaluation of a Household UV-treatment and Safe Storage Drinking Water Intervention in Rural Baja California Sur, Mexico

Fundacion Cantaro Azul (FCA) is a non-profit organization in Baja California Sur, Mexico (BCS). Since 2006 FCA has piloted a safe drinking-water program in rural regions of BCS. The premise of their safe drinking-water programs has been the installation of household drinking-water disinfection systems which utilize an ultra-violet technology (UV) developed at the University of California, Berkeley. While the systems have been tested for safety and effectiveness at inactivating waterborne pathogens, FCA is interested in rigorously evaluating the impact of their safe drinking-water program at the population level. FCA is looking to expand their safe-water program during 2009 and 2010 to newly identified communities that lack safe-drinking water. In order to evaluate the community level effectiveness of their program FCA has agreed to randomize the timing of this expansion which will allow the lead investigators and key personal in this protocol to conduct a meaningful, scientific evaluation of the impact of their program through a randomized stepped wedge design. The research described in this protocol has four (4) primary objectives: To evaluate the impact the implementation of the safe drinking-water programs has on rates of gastrointestinal events in rural BCS communities; To evaluate the impact of the safe drinking-water programs on concentrations of fecal contamination in household drinking-water in rural BCS communities; To evaluate other-health and non-water impacts on communities where the safe-water programs are implemented, including school and work absenteeism, and health care costs; To identify household, program and system design characteristics that affect user compliance with the disinfection strategies. The investigators hypothesize that households that receive an UV based drinking-water disinfection system through the safe-water program will have reduced prevalence of gastrointestinal illness, and reductions in fecal contamination of household drinking-water, measured as concentrations of Escherichia Coli per 100 ml of water. Similarly, the investigators hypothesize that these communities will also have reduced health care costs, and school and work absenteeism due to the implementation of the safe drinking-water programs. The investigators further hypothesize that household level characteristics and specific program characteristics will differentially impact user compliance, measured as the sustained use of the systems over the course of the study. In order to evaluate the last hypothesis (Objective 4) two program variations will be rolled out to inform future programmatic decisions. A priori the investigators do not anticipate that these program variations will impact population measures for Objectives 1 and 2, but the investigators will explore these assumptions during analysis.

Twenty-five to thirty communities (400-600 households) have been identified by Fundacion Cantaro Azul (FCA) as communities that are appropriate for their safe water program. These communities were identified based on their sources of water (chemical and taste characteristics) and resource availability (access to electricity and lack of access to piped water). Fundacion Cantaro Azul is planning to expand their safe-drinking water programs to all 600 households in these communities. In order to allow for a meaningful evaluation of the impact of their program the organization has agreed to randomize the timing of implementation of the safe drinking-water program to eligible communities. This will allow independent researchers (the lead investigator and key personal in this protocol) to make household visits during the implementation of the programs and will further allow for a valid evaluation of the programs' impact using a stepped wedge study design. Through the stepped-wedge study design all communities, over a 12-15 month, period will be approached by Fundacion Cantaro Azul for promotional meetings; since no community will have received the intervention they will be acting as their own controls until the time they are approached by the organization. Randomizing the time at which a community is approached reduces bias and allows for meaningful evaluation of differences in outcomes due to the safe drinking-water program. As part of this study design a randomly assigned, predetermined number of rural BCS communities will be approached every two-months for enrollment in the safe water program by Fundacion Cantaro Azul, and thus 'cross-over' from the comparison group to the safe-water group. Within each of these crossover groups half of the clusters will receive a "Basic" program variation, and half will receive an "Enhance" program variation. This process will continue until all consenting households in the 25-30 communities have received a UV disinfection system, and related support through the safe drinking-water program. Field teams, trained and supervised by the lead investigator and key personal listed in this protocol, will conduct independent (from Fundacion Cantaro Azul) household visits to collect data on the outcomes of interest. During household visits structured interviews and water-quality sampling will be used to measure these outcomes (prevalence of gastrointestinal events, concentrations of E.coli in household drinking-water, school and work absenteeism and compliance with the disinfection strategy.) The stepped-wedge design will allow for evaluation of the impacts of the safe drinking-water programs while allowing the relatively small organization to manage the logistics of installing 600 household disinfection systems, in a manner consistent with their mission, commitments to regional governmental collaborators, and fiscal restraints.

All communities (independent units/clusters) start in the control and have no intervention. All control communities cross-over to the intervention in a randomized sequence over the course of 12-15 month study.

The Basic program represents a variation in the implementation of the overall intervention: a community level safe-water program that is initiated with promotional meetings and followed by the offer to install Mesita Azul disinfection systems at the household level. Communities (independent units/cluster) cross-over from the control group to the intervention group in a randomized stepped-wedge fashion -- within each crossover group of 4 clusters, 2 are randomized to the Basic program.

The Enhanced program represents a variation in the implementation of the overall intervention: a community level safe-water program that is initiated with promotional meetings and followed by the offer to install Mesita Azul disinfection systems at the household level. Communities (independent units/cluster) cross-over from the control group to the intervention group in a randomized stepped-wedge fashion -- within each crossover group of 4 clusters, 2 are randomized to the Enhanced program.

The intervention is community promotional meetings followed by household installations of a UV-based drinking water disinfection system (Mesita Azul). Households can opt-in at the meetings, or when the organization returns for installations in that community (~2 weeks later). Mesita Azul utilizes a 15-watt UV lamp to disinfect water and a closed mouth collection container for safe storage. The system utlizes UV-Tube technology developed as a method to disinfect water in low- and middle-income communities. The Basic Safe Water Program receives: the community meeting; installation of a Mesita Azul; a six month technical guarantee; one safe-storage container; training of one household member in operation and basic maintenance of the technology; training of at least one local resident in repair and system maintenance; and follow up after four to six months of the installation date.

The intervention is community promotional meetings followed by household installations of a UV-based drinking water disinfection system (Mesita Azul). Households can opt-in at the meetings, or when the organization returns for installations in that community (~2 weeks later). Mesita Azul utilizes a 15-watt UV lamp to disinfect water and a closed mouth collection container for safe storage. The system utlizes UV-Tube technology developed as a method to disinfect water in low- and middle-income communities. The Enhanced program complements the Basic version with: a water quality analysis and presentation of results to households at the community meeting; a six-month money back guarantee; the provision of two safe-storage containers to facilitate access to disinfected water at different locations within the household; training of two household members in operation and basic maintenance of the technology; an additional follow up visit about two weeks after the installation date.

We measure household drinking water quality using Escherichia Coli concentrations measured through standard techniques. Each household is visited a maximum of 7 times over 15 months. Water is collected from household drinking water storage containers by a household respondent as though they were going to take a drink, and a 100 ml sample is taken and processed. We use multiple definitions of water contamination risk based World Health Organization classifications (>1 E.coli (low), >10 E.coli (medium), >100 E.coli (high)

We measure the 7-day prevalence of diarrhea symptoms (3 or more loose/watery stools in a 24 hour period, or one with blood or mucus) through household interviews. Each household is visited a maximum of 7 times (every 2 months) over the 15 month study period for a total possible of 49 days of recall.

Inclusion Criteria: Communities: are within the La Paz and Los Cabos counties of Baja California Sur, Mexico; lack access to centrally treated drinking water distribution; access drinking water sources (i.e. wells and springs) year round for household purposes Exclusion Criteria: Community source water: contains levels arsenic that are below Mexican limits; has taste characteristics (e.g. salinity) that are acceptable for drinking among community members.

Brownell SA, Chakrabarti AR, Kaser FM, Connelly LG, Peletz RL, Reygadas F, Lang MJ, Kammen DM, Nelson KL. Assessment of a low-cost, point-of-use, ultraviolet water disinfection technology. J Water Health. 2008 Mar;6(1):53-65. doi: 10.2166/wh.2007.015.