Monitoring of 82 surface water treatment plants revealed that 1.4% of 1,690 100‐L finished water samples tested positive for infectious Cryptosporidium using the cell culture‐polymerase chain reaction). Infectious oocysts were detected in finished water samples from 22 water treatment plants (26.8%). Genotype analysis identified 23 isolates as Cryptosporidium parvum and one isolate as Cryptosporidium hominis. All isolates subgenotyped were shown to be genetically distinct environmental isolates. Analysis of the water quality and treatment plant characteristics showed no differences between the positive and negative sites. More than 70% of the positive samples occurred in filtered water samples of <0.1 ntu, and 20% of the positive samples were in water of <0.05 ntu. There was no association among Cryptosporidium occurrence and source water type, microbial indicators, or plant operation and treatment. It was concluded that given sufficient testing, nearly all conventional treatment plants would be at risk for passing infectious oocysts. Based on this study, the overall risk of Cryptosporidium infection for conventionally treated drinking water was 52 infections/10,000 people/year, with an 80% credible range of 9–119 infections/10,000 people/year. It was also concluded based on these studies that conventional treatment requires an additional treatment barrier, such as ultraviolet light disinfection, to meet the US Environmental Protection Agency risk goals.

A new strategy for the detection of infectiousCryptosporidium parvum oocysts in water samples, which combines immunomagnetic separation (IMS) for recovery of oocysts with in vitro cell culturing and PCR (CC-PCR), was field tested with a total of 122 raw source water samples and 121 filter backwash water grab samples obtained from 25 sites in the United States. In addition, samples were processed by Percoll-sucrose flotation and oocysts were detected by an immunofluorescence assay (IFA) as a baseline method. Samples of different water quality were seeded with viable C. parvum to evaluate oocyst recovery efficiencies and the performance of the CC-PCR protocol. Mean method oocyst recoveries, including concentration of seeded 10-liter samples, from raw water were 26.1% for IMS and 16.6% for flotation, while recoveries from seeded filter backwash water were 9.1 and 5.8%, respectively. There was full agreement between IFA oocyst counts of IMS-purified seeded samples and CC-PCR results. In natural samples, CC-PCR detected infectious C. parvum in 4.9% (6) of the raw water samples and 7.4% (9) of the filter backwash water samples, while IFA detected oocysts in 13.1% (16) of the raw water samples and 5.8% (7) of the filter backwash water samples. All CC-PCR products were confirmed by cloning and DNA sequence analysis and were greater than 98% homologous to the C. parvum KSU-1hsp70 gene product. DNA sequence analysis also revealed reproducible nucleotide substitutions among the hsp70fragments, suggesting that several different strains of infectiousC. parvum were detected.

Ultraviolet light is now recognised to be very effective for inactivation of Cryptosporidium parvum oocysts; however, its application for disinfection of finished water necessitates validation of UV reactors prior to their installation. Although reactor performance will likely be assessed using non-pathogenic microorganisms as biodosimetry surrogates, it would be prudent for the water industry to simultaneously measure Cryptosporidium oocysts inactivation in controlled bench-scale studies using the water matrix intended for disinfection. The likelihood of that occurring is dependent upon the availability of infectivity measurement procedures that are more user-friendly than the mouse infectivity assays currently used. This study describes a modified cell culture procedure that would enable reliable measurement of changes in oocysts' infectivity following their UV treatment. Also, a number of different biodosimetry surrogates were examined and one selected for comparing the UV doses delivered between bench-scale and full-scale biodosimetry studies. Impacts of UV disinfection on production of disinfection byproducts, effects of lamp ageing on effectiveness of disinfection and the costs associated with employing this technology were also examined.