Drinking Water Turbidity and Emergency Department Visits for Gastrointestinal Illness in New York City, 2002-2009

Background Studies have examined whether there is a relationship between drinking water turbidity and gastrointestinal (GI) illness indicators, and results have varied possibly due to differences in methods and study settings. Objectives As part of a water security improvement project we conducted a retrospective analysis of the relationship between drinking water turbidity and GI illness in New York City (NYC) based on emergency department chief complaint syndromic data that are available in near-real-time. Methods We used a Poisson time-series model to estimate the relationship of turbidity measured at distribution system and source water sites to diarrhea emergency department (ED) visits in NYC during 2002-2009. The analysis assessed age groups and was stratified by season and adjusted for sub-seasonal temporal trends, year-to-year variation, ambient temperature, day-of-week, and holidays. Results Seasonal variation unrelated to turbidity dominated (similar to 90% deviance) the variation of daily diarrhea ED visits, with an additional 0.4% deviance explained with turbidity. Small yet significant multi-day lagged associations were found between NYC turbidity and diarrhea ED visits in the spring only, with approximately 5% excess risk per inter-quartile-range of NYC turbidity peaking at a 6 day lag. This association was strongest among those aged 0-4 years and was explained by the variation in source water turbidity. Conclusions Integrated analysis of turbidity and syndromic surveillance data, as part of overall drinking water surveillance, may be useful for enhanced situational awareness of possible risk factors that can contribute to GI illness. Elucidating the causes of turbidity-GI illness associations including seasonal and regional variations would be necessary to further inform surveillance needs.