Inadequacies in the area of water, sanitation and hygiene (known collectively as WASH), are responsible for a huge number of health problems throughout the developing world, including diarrhoea, the leading cause of childhood mortality. As outlined in our WASH web page, research in the WASH sector has often produced conflicting and ambiguous results as to which interventions are effective. However, two world-class research institutions, Innovations for Poverty Action (IPA) and Poverty Action Lab (JPAL), believing that progress could be made in this area, carried out a series of randomized control trials (RCTs) to try and find a solution. The results of these trials provided strong evidence that installing chlorine dispensers next to water sources is a highly effective method of ensuring the chlorination of unsafe water. Convinced of the effectiveness of chlorine by existing research, Dispensers for Safe Water (DSW) was then launched. Giving What We Can has been researching DSW since September, and believes that on many measures it looks like a highly cost-effective charity with promising opportunity for leverage. Despite this, we hesitate to include DSW among our top charities at this time, because we have lingering concerns about the strength of the existing evidence for the efficacy of water quality interventions.

What makes chlorine dispensers good - four problems, DSW's solutions:

Problem #1:

783 million people in the world, or 11% of the global population lack access to safe drinking water. As a result, people are drinking water contaminated with bacteria that lead to diseases such as diarrhoea.

Solution #1:

Water can be treated with disinfectants, and some research has shown chlorine to be particularly effective here, suggesting a 29-35% reduction in diarrhoea. This is due in part to its significant residual effect: it keeps water safe from recontamination for the period over which we expect water to be retained (usually only 24 hours).

Problem #2:

Even when the health benefits are well-known, chlorine isn't used if people have to pay for it.

Solution #2:

The chlorine from dispensers is free. On top of this, instead of the distribution of individual bottles of chlorine as some charities and trials have done, DSW's system is much more cost-effective - only one dispenser is needed for the whole community.

Problem #3:

People forget to use chlorine.

Solution #3:

The public nature of the dispensers, and their location next to water sources serve as a reminder. On top of this, DSW uses community education and local promoters to encourage usage. In keeping with their evidence-based practice, they have carried out RCTs in order to ensure that that such promotion has the widest possible impact, and is continued over a long period of time. The outcome of this is a programme that has seen an average adoption rate of 43% in the scaled up programme. This compares to adoption rates in other WASH interventions of 5 -10%.

Problem #4:

WASH interventions are often neglected, and end up in disrepair because maintenance is costly.

Solution #4:

DSW has been set up with the idea that it provides a service - not just the initial infrastructure. Maintenance costs have not only been factored into the organisation's budget, but they have once again carried out RCTs in order to ensure that they are using the most effective methods for this part of the programme too.

Scaling up - Dispensers for Safe Water and Evidence Action

Given their early successes, DSW is in the process of scaling up their programme with the aim of reaching 25 million people in Africa by the end of 2018. DSW is now part of the larger organisation Evidence Action that also runs another of our recommended charities, Deworm the World, though you can continue to donate directly to both projects.

There are several key elements of their scaling up model that give us reason to believe that DSW will continue to be effective:

  • Continuous evaluation and monitoring

DSW carries out research and pilots in regions into which it is considering expanding, in order to determine whether it is a suitable destination for the chlorine dispenser system. In addition, it carefully monitors any partner organizations to ensure that best practice continues to be used.

  • Carbon credits

The Kyoto protocol allows organizations to gain carbon credits if they can show that the activity they are carrying out reduces carbon usage. Since using chlorine means that people no longer have to boil their water in attempts to purify it, DSW earns carbon credits which it can then sell in carbon markets such as the European Union Emissions Trading System. Because of this, they are able to cover a large amount of the fixed costs of running the organization, allowing more of the donated money to be spent on chlorine and dispensers.

  • Economies of Scale

The DSW scaling model is such that as the programme is expanded, average costs will increasingly fall. This is because chlorine supply chain costs, and maintenance costs will be lower if several can be attended to in one round. This means that even if DSW expands into areas with lower rates of diarrhoea such that potentially fewer lives would be saved for a given amount of money, the reduced costs overall offset this worry, and permit them to expand into more regions.

Room for funding?

Because of the carbon credits system they are using, DSW hopes to be financially sustainable by 2018. That is, once they reach a certain scale (the projected 25 million people) the carbon credits system should be raising enough money to sustain the project without need for further donations. However, to reach that point they have a substantial funding gap to make up. Since a large part of their model is based on economies of scale, donations will be increasingly valuable in making it to this goal.

Cost-effectiveness estimate

Current internal estimates supplied to us by DSW put the net cost of one DALY averted at $67.5. DSW estimate that one dispenser with a life-time of 5 years averts 3.97 DALYs, relying on estimates of diarrheal reduction due to Fewtrell et al. (2005) and Kremer et al. (2009). The marginal gross cost for a dispenser of this kind installed in 2014 is $671, but the net cost of $268 is significantly lower due to sale of carbon credits.

The internal cost-effectiveness estimate is indicative of a highly cost-effective charity. In addition, two things should be noted. The figure given is a conservative estimate based on the minimum number of diarrhoea-related deaths averted. This figure doesn't take into account the benefits of the prevention of other water-borne diseases that chlorine dispensers help to prevent, and so DSW's programs may well have a larger potential benefit than this estimate indicates. In addition, costs are expected to decline over time due to economies of scale.

Our reservations

In spite of DSW's many positive aspects, we remain cautious about recommending them at this time. This is due to our uncertainty about the current evidence for water treatment interventions. As GiveWell explore in comprehensive detail in theirpage on this issue, despite many randomized control trials that demonstrate efficacy for water treatment interventions in reducing diarrhoeal disease, the small number of blinded trials to date have failed to find evidence of an effect. While the blinded trials are smaller in number, they also represent a higher standard of evidence. Like GiveWell, we don't believe that we have encountered an adequate explanation for this discrepancy that would allow us to entirely discount the null results reported in blinded trials. As a result, whilst DSW looks highly promising, at the moment we cannot recommend them as a top charity. However, we believe that they are likely to be among the most effective WASH charities, and quite possibly the most effective in this area. Furthermore, future research on water interventions may well resolve our lingering concerns regarding the evidence for water treatment interventions.