Developments in microbiological risk assessment models for drinking water--a short review.

Microbiological risk assessment (MRA) is emerging method to predict the risks of infection from waterborne pathogens (e.g. rotavirus and Cryptosporidium parvum) in the drinking water supply. The objectives of this paper are to review the appropriateness of current models, with emphasis on pathogen exposures through drinking water, and to consider the information necessary to further their development. Calculating pathogen exposures in MRA is currently limited by the fact that pathogen density data for drinking water supplies are only available for very large volume samples--much larger than imbibed daily by any consumer. To develop MRA, information is needed on how pathogens are dispersed within those volumes at the resolution of volumes typically consumed daily by individuals. Available evidence suggests that micro-organisms, including pathogens, are clustered to some degree, even within small volumes, exposing some drinking water consumers to much higher doses than others. By assuming pathogens are randomly dispersed, current models overestimate the risk from the more infectious agents (e.g. rotaviruses) but underestimate the risk from less infectious pathogens (e.g. C. parvum). Approaches to modelling pathogen densities in drinking water from source water data and treatment removal efficiencies require additional information on the degree to which treatment processes (e.g. filtration and coagulation) increase pathogen clustering. The missing information could be obtained from large-scale pilot plant studies.