Onsite Nonpotable Water Systems Pathogen Treatment Targets: A Comparison of Infection and Disability-Adjusted Life Years (DALYs) Risk Benchmark Approaches.

Abstract

Pathogen log10 reduction targets for onsite nonpotable water systems were calculated using both annual infection (LRTINF) and disability-adjusted life year (LRTDALY) benchmarks. The DALY is a measure of the health burden of a disease, accounting for both the severity and duration of illness. Results were evaluated to identify if treatment requirements change when accounting for the likelihood, duration, and severity of illness in addition to the likelihood of infection. The benchmarks of 10-4 infections per person per year (ppy) and 10-6 DALYs ppy were adopted along with multilevel dose-response models for Norovirus and Campylobacter jejuni, which characterize the probability of illness given infection (Pill|inf) as dose-dependent using challenge or outbreak data. We found differences between treatment requirements, LRTINF - LRTDALY, for some pathogens, driven by the likelihood of illness, rather than the severity of illness. For pathogens with dose-independent Pill|inf characterizations, such as Cryptosporidium spp., Giardia, and Salmonella enterica, the difference, LRTINF - LRTDALY, was identical across reuse scenarios (<than 1.0). The differences varied across source waters and uses for C. jejuni and Norovirus and widened when the dose-dependent Pill|inf was characterized using challenge data (i.e., when there was evidence of a small probability of illness at low doses). Norovirus LRTs were highest across pathogens, despite low severity and dose-dependent Pill|inf, given the high infection risks predicted by the multilevel framework. This work highlights updated Norovirus dose-response best practices, the quantitative impact of risk endpoint in determining risk-based treatment targets, and the discrepancy in best available science for illness and infection responses across pathogens.