A system for the early detection of outbreaks of water-related gastroenteritis

Prof Christopher Fairley (Melbourne University)
Dr Jim Black

Funded by: Cooperative Research Centre for Water Quality and Treatment

Current methods for the surveillance of waterborne diseases, and gastrointestinal diseases in general, are acknowledged to be relatively insensitive and slow since they rely mainly on laboratory identification of pathogens and subsequent notification to health authorities. Such methods detect only the "tip of the iceberg" in terms of illness in the community, as less than 10% of people with gastroenteritis attend a doctor, and perhaps only 10% of these have a faecal specimen tested for pathogens.

Relatively large outbreaks of disease could be missed by current surveillance systems, and the usual time lag of two weeks or more between the causative event and recognition of an outbreak by health authorities limits the usefulness of response measures.

Retrospective analysis of outbreaks, including the Milwaukee Cryptosporidium outbreak in 1993, have pointed to several options for increasing the speed and sensitivity of surveillance. A Melbourne Water/ CRC funded feasibility study by Dr Alex Padiglione confirmed that a number of existing data sources had the potential to form the basis of a greatly improved surveillance system for waterborne diseases in Australia(1).

Work in this area was continued by PhD student Dr Jim Black who developed artifical neural networks to analyse health related and water quality data to detect and predict outbreaks. The project also involved the development of other computing tools to analyse and present data in an easily accessible form to water authorities and health agencies. In the future, automatically acquired health data and water quality figures could be linked to Geographic Information Systems to create an integrated set of programs capable of detecting outbreaks at the earliest possible stage, and determining whether they are related to drinking water supplies.

1) Padiglione, A. and C. K. Fairley (1998). "Early detection of outbreaks of waterborne gastroenteritis." Water 25(6): 11-15.