Information Value of the Children's Chemical Evaluation Programs

EPA Grant Number: U915561
Title: Information Value of the Children's Chemical Evaluation Programs
Investigators: Yokota, Fumie
Institution: Harvard University
EPA Project Officer: Jones, Brandon
Project Period: September 1, 1999 through August 1, 2002
Project Amount: $96,150
RFA: STAR Graduate Fellowships (1999) RFA Text |  Recipients Lists
Research Category: Fellowship - Risk Management , Academic Fellowships , Ecological Indicators/Assessment/Restoration


The objective of this research project is to use a Bayesian decision-theory framework to characterize the expected value of information gained from the Voluntary Children's Chemical Evaluation Program.


Decision analysis provides a useful framework to evaluate the uncertain consequences of a menu of regulatory actions; it identifies the action that maximizes the likelihood that the appropriate choice is made. Value-of-information (VOI) analysis is an extension of the decision-analytic framework and answers the following question: How much should we be willing to pay for the additional information? VOI is defined as the difference between the value of making a decision with and without additional information. The value of making a decision without additional information (a priori action) is determined, in part, by the prior belief of a pilot chemical’s toxicity. The prior distribution will be modeled using a hierarchical Bayesian approach to combine subsets of toxicological testing results, available for the pilot test chemicals and structurally and/or functionally similar chemicals. Once all relevant a priori information is characterized in the prior distribution, a threshold analysis of the posterior distribution will determine the conditions (e.g., sensitivity and specificity of the test, level of exposure to the chemical, monetary value of the health effects) under which VOI exceeds the cost of the testing program.

Supplemental Keywords:

fellowship, statistical decision theory, value-of-information, VOI, uncertainty analysis, decision analysis, hierarchical Bayesian model, meta analysis, Monte Carlo simulation, risk management, public policy, risk regulation, cost-benefit analysis, risk assessment, human., RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Entomology, wildlife, Molecular Biology/Genetics, Biology, butterfly caterpillars, pheromones, local adaptation, chemical mimicry, ants, bioassays, symbiotic evolution