A Comparison of Rapid and Conventional Measures of Indicator Bacteria As Predictors of Waterborne Protozoan Pathogen Presence and Density

Samuel Dorevitch,Mary Doi,Fu-Chih Hsu,King-Teh Lin,Jennifer D. Roberts,Li C. Liu,Ross Gladding,Ember Vannoy,Hong Li,Margit Javor,Peter A. Scheff
DOI: https://doi.org/10.1039/c1em10379b
2011-01-01
Journal of Environmental Monitoring
Abstract:E. coli and enterococci in recreational waters are monitored as indicators of fecal contamination, pathogen presence, and health risk. Quantitative polymerase chain reaction (qPCR) tests for fecal indicator bacteria can provide beach managers with same-day information about water quality, unlike culture methods which provide that information the following day. The abilities of qPCR measurements of indicator bacteria, as compared to culture measurements of indicator bacteria, as predictors of pathogen presence or density in surface waters are not well understood. The purpose of this study was to make such comparisons between water samples collected from Chicago area surface waters, including rivers, inland lakes, Lake Michigan, and the Chicago Area Waterways System, which is dominated by wastewater effluent. A total of 294 twenty-litre samples were collected and analyzed for Giardia and Cryptosporidium. qPCR and membrane filtration methods were used to quantify E. coli and enterococci. Correlation, logistic regression, and zero-inflated Poisson modeling were utilized to evaluate associations between indicators and parasites. qPCR and culture measures of the indicator bacteria were similar in their ability to predict parasite presence and density. Correlations between parasites and indicators were generally stronger at waters not dominated by effluent. Associations between indicator density and Giarida presence were observed more consistently than between indicator density and Cryptosporidium presence. Associations between enterococci and parasites were generally stronger than associations between E. coli and parasites. The use of qPCR monitoring in our setting would generate more timely results without compromising the ability to predict parasite presence or density.
What problem does this paper attempt to address?