![[logo] US EPA](http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
2007 CompTox Forum
Abstract - Development and Use of Predictive Hazard Modeling in the Categorization and Screening of Existing Substances at Health Canada
M.E. (Bette) Meek, J. Paterson and K. Hughes
M.E. (Bette) Meek (presenting author)
Manager, Existing Substances Division
Health Canada
Safe Environments Programme
269 Laurier Avenue West, Ottawa, Ontario, K1A 0K9, Canada
Phone: 613-957-3129
E-mail: bette.meek@hc-sc.gc.ca
Canada was the first country to introduce a legislative requirement for systematic priority setting of all chemicals in commerce (Existing Substances). The Canadian Environmental Protection Act, CEPA '99, required the Ministers of Health and Environment to complete "categorization" (priority setting) of all of the approximately 23,000 substances on the Domestic Substances List by September 2006.
Substances identified as priorities following categorization are to be considered subsequently in screening assessments. This is in addition to the continuing mandate to more fully assess compounds designated as "Priority Substances."
The approach to priority setting for the health components involved the development of simple and complex exposure and hazard tools, as a basis to draw maximally on the often limited information available for many of the 23,000 substances. The program entails a completely iterative approach to priority setting and assessments of increasing complexity. This enables efficient and inclusive identification of highest priority environmental contaminants for subsequent introduction of control measures to reduce exposure.
This presentation will address the novel predictive methodology developed to set priorities on the basis of potential hazard to human health from among the many thousands of Existing Chemicals. This has included development of a hierarchical approach to consideration of data and the output of predictive modeling for hazard.
Specifically, a transparent framework to address weight of evidence for cancer and genotoxicity, based on often limited data and the considered output of a range of expert and statistically-based modeling systems, has been proposed and considered in external peer consultations. Advances in the development of more robust predictive methodology for other health-related endpoints will also be described.