Stage 5: Scaling
Dispensers for Safe Water grew out of a research study that tested several ideas to address the problem of diarrheal disease. Researchers at Harvard University and the University of California, Berkeley -- in collaboration with Innovations for Poverty Action (IPA) -- identified chlorine dispensers as the most effective, low-cost intervention. The researchers found that 50-61% of people adopted the chlorine dispenser system, which was a six-fold increase over the control group.
Inspired by the sustained adoption of this low-cost intervention, a program was developed to scale up chlorine dispensers as a service. Between 2009 and 2011, various operational models for service delivery were tested, and additional research was conducted on how best to engage local promoters to boost adoption. The program installed 2,500 dispensers in Kenya during this period, serving approximately half a million people. Having refined our model in Kenya, we successfully replicated and scaled the program in Uganda and Malawi, demonstrating successful implementation of the program at scale.
Today, we have over 26,000 dispensers currently installed across Kenya, Uganda and Malawi serving 4.5 million people with an average program wide adoption rate of 60%. Specifically, we have over 17,000 dispensers serving over 2.2 million people in Kenya, over 5,000 dispensers serving 1.6 million people in Uganda and over 3,000 dispensers serving around 600,000 people in Malawi. Such expansion has proved the cost effectiveness of our model at scale, which is currently at $1.24 per person per year.
We have already developed and proved our capacity to create financial returns from the program by successfully validating, monitoring, auditing, issuing, and selling carbon credits for dispensers with our carbon development partner, South Pole Group. Evidence Action recovers part of its operating costs as it earns carbon revenue, since people with access to a chlorine dispenser are able to treat their water instead of boiling it by burning firewood, resulting in lower greenhouse gas emissions. The program has also secured a futures contract worth $13 million for purchase of credits generated from 2017-2021. Our buyers have paid a higher price for these credits, recognizing twin benefits of reducing carbon emissions as well as the provision of safe rural water services.At this juncture, we seek funding partners to invest in our current projects as well as our growth plan. Investing in our current projects will allow us to continue serving 4.5 million people at less than $1.24 per person per year and will help us continue on a path to sustainability. On the other hand, in order to continue to grow the program, we need to invest in lowering costs across the chlorine supply chain, expand our capacity to advocate to governments for support of chlorine dispensers, and grow our strategic partnerships to support global expansion. Investing in this growth plan will help us further refine our business model and grow our program effectively.
We currently serve nearly 4.5 million beneficiaries in rural households across Kenya, Uganda and Malawi. Our beneficiaries are people who are not serviced by centrally-piped water systems and are thus vulnerable to diarrheal diseases caused by contaminated water. Globally, the potential user base could be more than 1 billion people. Our demographics include poor rural households in Kenya, Malawi and Uganda, with an average of 5.16 persons per household, including 2.82 females, 2.53 males and 0.81 children under five. Our beneficiaries also have an average of 1 mobile phone 0.75 radios,1 sheep and 1.31 cattle per household.
Water treatment by the household can significantly improve water quality and health, and several appropriate technologies exist such as filters, dilute chlorine solution and solar disinfection. However, for all of these technologies, adoption remains low.
Randomized controlled trials have suggested several behavioral barriers to sustained high adoption. Inconvenience of acquiring these technologies may be one such barrier. In one study, for example, mothers of young children were given coupons that could be exchanged for free chlorine at stores. Adoption was 60 percent initially, but dropped to only 20 percent by six months suggesting that even the small inconvenience of going to a store and redeeming a coupon could be a barrier. When chlorine was distributed directly to homes, take-up rates remained high even 6 months after distribution, however this was a costly solution.
A study of spring protection also suggests that convenience is an important determinant of demand for safer water; it shows that people will walk only 3.5 additional minutes, on average, to collect water from a protected versus an unprotected spring, despite substantial health gains from the cleaner source.
Charging for products typically dramatically reduces adoption, both for water treatment as well as for other preventative health technologies, as shown by several randomized controlled trials performed in the last 10 years. More qualitative and anecdotal results suggest that remembering to treat water regularly, learning how to use chlorine with appropriate dosage and dislike of the taste of chlorine (potentially due to overdosing) may also be barriers.
Randomized controlled trials suggest that education and marketing messages may be able to increase adoption incrementally, but have not yet been shown to have large or continued effects. Making health information salient and relying on local promoters to reinforce messages about health and water treatment may hold more promise.
The source-based chlorine dispenser solution makes drinking water treatment convenient, salient, and public. Using the dispenser is convenient because the dispenser valve delivers an accurate dose of chlorine to treat a standard transport container of water and the required agitation and wait time for chlorine treated water are at least partially accomplished automatically during the walk home from the source. Accurate dosing also reduces the likelihood that treated water will have a chemical after-taste. The dispenser hardware provides a frequent visual reminder to households to treat their water at the moment when it is most salient – at water collection. This visual reminder combined with chlorination encourages habit formation. Emerging evidence also suggests that social effects may be important in adoption and the public nature of the dispenser system maximizes the potential for learning, habit formation, and social norm effects.
Randomized controlled trials have shown that the dispenser system dramatically increased water treatment compared to the traditional retail model. In pilot programs during unannounced visits 30 months after dispenser installation (and two years after payments to the local promoter terminated), 55-60% of household drinking water samples in these communities tested positive for chlorine. These test results represent a conservative estimate for the adoption of dispensers, since self-reports of use were closer to 75 percent. In comparison, only 5-10% of samples with access to the traditional retail model of chlorine distribution tested positive. Bacteriological testing showed that those reporting chlorine use had less contaminated water than those who said they had not used chlorine. Evidence Action has maintained 60% adoption rate on average across Kenya, Uganda and Malawi (as measured by the presence of residual chlorine in stored household drinking water) at scale.