You are here:
VARIATIONS IN SEASONAL PATTERNS OF GASTROINTESTINAL INFECTIONS ALONG A RIVER
Citation:
JAGAI, J. S., J. K. Griffiths, P. K. Kirshen, P. Webb, AND E. N. Naumova. VARIATIONS IN SEASONAL PATTERNS OF GASTROINTESTINAL INFECTIONS ALONG A RIVER. Presented at Water and Health: Where Science Meets Policy Conference, Chapel Hill, NC, October 25 - 26, 2010.
Impact/Purpose:
This preliminary analysis explores seasonal patterns of waterborne diseases along the Ohio River watershed in the U.S.
Description:
Epidemiologic analysis of waterborne diseases typically considers socio-economic, demographic, and pathogen-specific characteristics. However, hydrological parameters may need to be considered as well. Fate and transport models of watersheds have demonstrated impairment due to livestock agriculture and have predicted pathogen concentrations at intakes to water treatment plants. The affects of these watershed impairments on human health has not been adequately assessed. Increases in rates of waterborne diseases have been associated with water quality parameters such as turbidity and river flow at a single location however a watershed level analysis has not been considered. Each watershed has unique characteristics, such as streamflow, which may affect the rates of waterborne diseases in populations residing within the watershed. We hypothesized that watershed level attributes predict the rates of waterborne disease in these populations. This preliminary analysis explores seasonal patterns of waterborne diseases along the Ohio River watershed in the U.S. We selected six counties along the river which use surface water as the primary source of water. For each county we gathered streamflow data from the U.S. Geological Survey monitoring network and meteorological data (temperature and precipitation) from the National Climatic Data Center. For health outcome data, we utilized hospitalization records from the Center for Medicare and Medicaid Services (CMS) from 1991 to 2004. We considered three diagnoses of gastrointestinal infections and cases were aggregated according to diagnosis codes, location, and date of admission. We assessed seasonal patterns for disease rates using an annual harmonic regression controlling for streamflow and meteorological parameters. We found that streamflow demonstrated strong seasonal patterns for all counties selected along the Ohio River with peak timing ranging from the 4th week of February to the 4th week of March. All three disease outcomes considered demonstrated strong seasonal patterns with peaks in late winter however, none of the disease outcomes demonstrated a trend in peak timings along the watershed. Controlling for streamflow and meteorological parameters did not significantly affect the estimate of peak timing of outcome rates; suggesting that there is an interaction between the seasonal pattern of GI illness and streamflow. We also found statistically significant, though weak correlations between the time series of streamflow and disease outcome rates. This is a first attempt at modeling rates of waterborne diseases associated with river characteristics within a watershed. Further, interdisciplinary research is needed to properly assess environmental associations with seasonal patterns of waterborne disease. An understanding of the relationship between watershed characteristics to health outcomes will allow water treatment authorities to tailor treatment practices in order to reduce these seasonal risks and for appropriate public health protection to be implemented. (This abstract does not necessarily reflect EPA policy)