Abstract of Meeting Paper

Society for Risk Analysis 1996 Annual Meeting

An Animal Model for Assessing Individual Differences in Susceptibility to Environmental Pollutants. R. C. MacPhail, Neurotoxicology Division, National Health and Environmental Effects Research Laboratory, U.S. EPA, RTP, NC 27711; and J. R. Glowa, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD 20892

Current approaches to risk assessment for systemic (non-cancer) toxicity typically include an uncertainty factor of 10 to account for individual differences in susceptibility to toxicants. Variability in a heterogeneous human population could easily exceed a factor of 10 due to differences in gender, age and health status. A behavioral approach was developed to quantitatively assess individual differences in the sensitivity of laboratory animals to toxic substances that might reflect a lower-bound estimate of the range of sensitivity in the human population. A repeated-measures (baseline) design was used to establish complete acute dose-effect functions in individual male outbred mice and rats exposed to solvents and pesticides respectively. Dose-effect functions for each animal were quantitatively described, from which a dose producing a 10% decrement in behavioral function was calculated (ED10). Distributions of ED10s were then established, from which doses producing 10% decrements in successively smaller proportions of the population were estimated (i.e., a probabilistic dose-tolerance model). Ratios of the highest-to-lowest ED10s (X " 3SD) varied by 1 1/2-4 orders of magnitude. These data suggest an uncertainty factor of 10 may greatly underestimate intraspecies differences in susceptibility to environmental pollutants.