This article highlights important challenges for drinking water safety monitoring: A single water quality test may miss periods of faecal contamination and overestimate how safe a water supply really is. Analysis of 524 small water supplies across Bangladesh, Nepal and Tanzania found that E. coli contamination is predominantly intermittent rather than consistent. Isolated cross-sectional surveys can underestimate contamination prevalence by factors of 1.4–5 times. Sampling during periods of highest risk, especially in wet weather, is more effective at identifying at-risk water supplies.
Moving towards risk-based operational monitoring that focuses on the timing of sampling and includes strategies to progressively identify different contamination pathways is recommended.
