Towards a more accurate quantitative assessment of seasonal Cryptosporidium infection risks in surface waters using species and genotype information.

Many Cryptosporidium species/genotypes are not considered infectious to humans, and more realistic estimations of seasonal infection risks could be made using human infectious species/genotype information to inform quantitative microbial risk assessments (QMRA). Cryptosporidium oocyst concentration and species/genotype data were collected from three surface water surveillance programs in two river basins [South Nation River, SN (2004-09) and Grand River, GR (2005-13)] in Ontario, Canada to evaluate seasonal infection risks. Main river stems, tributaries, agricultural drainage streams, water treatment plant intakes, and waste water treatment plant effluent impacted sites were sampled. The QMRA employed two sets of exposure data to compute risk: one assuming all observed oocysts were infectious to humans, and the other based on the fraction of oocysts that were C. hominis and/or C. parvum (dominant human infectious forms of the parasite). Viability was not considered and relative infection risk was evaluated using a single hypothetical recreational exposure. Many sample site groupings for both river systems, had significant seasonality in Cryptosporidium occurrence and concentrations (p ≤ 0.05); occurrence and concentrations were generally highest in autumn for SN, and autumn and summer for GR. Mean risk values (probability of infection per exposure) for all sites combined, for each river system, were roughly an order of magnitude lower (avg. of SN and GR 5.3 × 10-5) when considering just C. parvum and C. hominis oocysts, in relation to mean infection risk (per exposure) assuming all oocysts were infectious to humans (5.5 × 10-4). Seasonality in mean risk (targeted human infectious oocysts only) was most strongly evident in SN (e.g., 7.9 × 10-6 in spring and 8.1 × 10-5 in summer). Such differences are important if QMRA is used to quantify effects of water safety/quality management practices where inputs from a vast array of fecal pollution sources can readily occur. Cryptosporidium seasonality in water appears to match the seasonality of human infections from Cryptosporidium in the study regions. This study highlights the importance of Cryptosporidium species/genotype data to help determine surface water pollution sources and seasonality, as well as to help more accurately quantify human infection risks by the parasite. Crown