The challenge
Unsafe drinking water causes diarrhoea in young children, but individual trials were too small to show whether treating water saves lives.
The case
Michael Kremer, Stephen Luby and colleagues gathered child mortality data from randomised trials of drinking water treatment in low- and middle-income countries that included children under five. Of 52 trials identified, 18 had collected death data. Pooling them, they estimated that water treatment reduced the odds of a child dying before age five from any cause by 24–27%, and that a new study would be expected to find a 23% reduction. They then estimated the cost of three approaches: chlorine dispensers at water collection points, chlorination built into piped water systems, and free chlorine solution through mother and child health services. Each cost between US$28 and US$72 per year of healthy life gained.
Why it worked
The authors conclude that treating drinking water is one of the most cost-effective child health programs available. Pooling trials answered a question no single trial was large enough to answer.
Sources
Spot an error or know a better source? Tell us