Daily variations in effluent water turbidity and diarrhoeal illness in a Russian city.

To assess an association between temporal variations in drinking water quality and gastrointestinal (GI) illness, a cohort study involving 100 randomly selected families (367 individuals) was conducted in the city of Cherepovets, Russia from June through November 1999. Participants maintained daily diaries of gastrointestinal symptoms, water consumption and other behavioural exposure variables, while daily effluent water quality data were provided by the water utility. The cumulative incidence rate of self-reported gastrointestinal diseases, 1.7 cases per person-year, was almost two orders of magnitude higher than that of officially reported GI infections in the city. An interquartile range increase in effluent water turbidity of 0.8 Nephelometric Turbidity Units was associated with a relative risk of self-reported GI illness of 1.47 (95% Confidence Interval 1.16, 1.86) at a lag of 2 days after control for daily rate of consumption of non-boiled tap water, behavioural covariates, day of the week and a seasonally-related linear trend. In the analysis by subsets of study participants stratified by non-boiled tap water consumption, no statistically significant associations between turbidity and GI illness were found for the study participants who always boiled their drinking water. For individuals who drank non-boiled tap water, statistically significant associations between turbidity and GI illness were detected at lags 1, 2 and 7 days.