Drinking Water Quality and Emergency Visits for Gastroenteritis in AtlantaEPA Grant Number: R831629
Title: Drinking Water Quality and Emergency Visits for Gastroenteritis in Atlanta
Investigators: Tolbert, Paige , Singer, Philip C. , Moe, Christine L. , Amirtharajah, A. , Flanders, Dana , Uber, Jim , Klein, Mitchel , Tinker, Sarah , Hooper, Stuart
Institution: University of North Carolina at Chapel Hill , Georgia Institute of Technology , University of Cincinnati , Emory University
Current Institution: Emory University , Georgia Institute of Technology , University of Cincinnati , University of North Carolina at Chapel Hill
EPA Project Officer: Page, Angela
Project Period: September 1, 2004 through August 31, 2007 (Extended to August 31, 2008)
Project Amount: $1,223,366
RFA: Microbial Risk in Drinking Water (2003) RFA Text | Recipients Lists
Research Category: Drinking Water , Water , Health Effects
The investigators propose to conduct a population-level study of the relationship between measured or modeled parameters of a metropolitan drinking water system and incidence of gastrointestinal illness. The impact of distributional system factors and characteristics of finished water from the plants will be assessed separately and in combination. The study site is the five-county metropolitan Atlanta area, served by six water utilities, all in general compliance with water standards. Incidence of gastrointestinal illness in the general population and in susceptible subgroups will be compared across water providers, within providers by hydraulic residence time and other distributional measures, and over time in relation to indicators of likelihood of microbial contamination at the plant and in the distributional system. The study will provide an estimate of the proportion of endemic gastrointestinal illness attributable to distributional factors in a well-functioning drinking water system in a major US city. The study will be the most comprehensive assessment of the relationship of distributional and other water system factors and gastrointestinal illness to date, complementing the existing randomized controlled trials and making advances over earlier time-series studies.
The study will take advantage of a unique database on approximately six million Atlanta emergency department visits for the period 1993-2004 compiled by the investigators under existing USEPA and NIH grants. Emergency department visits for gastrointestinal illness (N=~583,000) will be examined in relation to a variety of water quality parameters collected from participating utilities and the Georgia Department of Natural Resources. Distributional system hydraulic network modeling will be used to estimate residence time as a function of network topology, characteristics of pipes, location of storage tanks and other available information. A selected plant will be the focus of more intensive modeling to estimate pathogen intrusion vulnerability, making use of information on pipe breaks, power loss and other factors. Spatio-temporal analyses will be performed to assess the relationship of plant and distributional factors with rates of emergency department visits for gastrointestinal illness. Studies such as this are vulnerable to a high degree of measurement error, adding noise and a general bias toward null findings, and therefore a thorough assessment of the impact of measurement error is planned for our models.
Because of the extensive emergency department visit database, the study has outstanding power to discern subtle associations, even in the presence of the bias to the null introduced by measurement error. For interquartile comparisons of any given exposure parameter, the study has over 99% power to detect risk ratios as low as 1.01. The study will therefore provide a robust assessment of the questions of interest, even for susceptible subgroups.