The Fractional Poisson: a Simple Dose-Response Model

• Activity Number: 566
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Risk Analysis
• Author(s): Michael Messner*+
• Companies: USEPA
• Keywords: Norovirus ; dose-response ; beta-Poisson ; fractional Poisson

Abstract

The beta-Poisson and exponential dose-response models are commonly used to describe infection probability as a function of exposure to microbial contaminants. The fractional Poisson is a limiting case of the beta-Poisson, and is approached as the beta-Poisson's parameters (alpha and beta) become small. The fractional Poisson allows for maximum host-to-host variability (each subject is either perfectly susceptible or perfectly protected against infection).

Norovirus human challenge data are used to illustrate the fractional Poisson model. Compared with the beta-Poisson, the fractional Poisson has equally good fit, improved error structure and fewer parameters. It may also have broader applicability, as when doses are not Poisson and when particles (in this case, virions) are aggregated, but the aggregation distribution is unknown. New Norovirus dose-response data would be of value to further clarify the NoV dose-response relationship and support improved hazard predictions at environmental exposure levels.