USE OF PROBABILISTIC EXPERT JUDGMENT IN UNCERTAINTY ANALYSIS OF CARCINOGENIC POTENCY

A new approach to characterizing the state of knowledge about carcinog enic potency is described. In this approach, the carcinogenic risk pos ed by a specific dose is characterized by a probability distribution, indicating the relative likelihood of different risk estimates. The ap proach utilizes expert judgment and a probability tree and is illustra ted in a case study of chloroform exposure. Experts in cancer biology/ toxicology, pharmacokinetics, and dose-response modeling were identifi ed by a panel of science-policy specialists. In a workshop, experts re viewed the chloroform data, received training in probability elicitati on, and constructed a consensual probability tree based on biological theories of cancer causation. Distributions of carcinogenic risk were developed based on the probability tree, chloroform data, judgmental p robabilities provided by the experts, and classical statistical techni ques. Risk distributions varied considerably between experts, with som e predicting essentially no risk from 100 ppb chloroform in drinking w ater while others have at least some probability on risks generally co nsidered of regulatory significance. Estimated human risk was much low er when extrapolating from liver tumors in animals than from kidney tu mors. Issues of scientific disagreement leading to different risk dist ributions between experts are discussed. The resulting risk distributi ons are compared to standard EPA risk calculations for the same exposu re scenario as well as to the expert judgments of epidemiologists abou t cancer risks of chlorinated drinking water. Issues in combining expe rt judgments are discussed, and several alternative methods are presen ted. Strengths and weaknesses of the distributional approach are discu ssed. (C) 1994 Academic Press, Inc.