The Value of Biomarker Information in Aflatoxin Risk Management. R. C. Lee, University of Washington Dept. Environmental Health, Seattle, WA and Golder Associates Ltd, Calgary, AB; S. M. Bartell, R. A. Ponce, W. H. Griffith, A. C. Cullen, and D. L. Eaton, University of Washington Department of Environmental Health, Seattle, WA
Aflatoxins, which are likely liver carcinogens, are controlled in the food supply using an action level (AL) approach. We represent cancer risk associated with aflatoxins as a function of exposure and carcinogenic hazard. Population variability in exposure and hazard is accounted for by stratified scenarios. We examine uncertain measures of exposure information within these scenarios; including survey/questionnaires, urinary DNA adducts, and albumin adducts, as well as uncertain hazard estimates. We estimate the expected value of information (EVI), a measure of the value of uncertainty reduction in making optimal decisions regarding alternative ALs, as the difference in expected opportunity loss under different levels of exposure uncertainty. Additionally, we estimate the expected value of perfect (no uncertainty) information for exposure and carcinogenic hazard. Cost-effectiveness (CE) ratios are estimated for a range of ALs. The EVI of different exposure and hazard measures and the CE of ALs are strongly dependent on the risk scenario evaluated. The EVI is greater and CE ratios are smaller for high risk compared to low risk scenarios. The EVI of albumin adducts exceeds that of urinary adducts. There is no value of improved exposure information for restrictive a priori ALs. The value of improved hazard information in most cases outweighs the value of improved exposure information. This framework is applicable to other environmental contaminants.
Partially supported by CRESP, Dept. of Energy DE-FCO1-95EW55084.
Go to . . .