Environmental Protection Agency, National Health Environmental Effects Research Laboratory (MD85), Research Triangle Park, NC 27711-0001

A -- PESTICIDE INTERACTION STUDIES:STATISTICAL MODELING SOL 99-NTD-001 DUE 062299 POC a -- GINGER MOSER(919)541-5075 E-MAIL: PESTICIDE INTERACTION STUDIES:STATISTICAL MODELING, mmann@epa.gov. The Neurotoxicology Division and the Environmental Toxicology Division of the US EPA's National Health and Environmental Effects Research Laboratory (NHEERL) intend to conduct research on the neurotoxicological effects of multiple pesticide exposure in young and adult rats. NHEERL further intends for statistical modeling of the data, an integral part of the research plan, via the establishment of an assistance agreement. The project period will be three years with anticipated funding to be $145,000 total. The studies will involve combinations of up to five pesticides, using a range of neurological tests which yield various forms of data. Interested parties should have the: 1) capablitiy of analyzing chemical combinations of up to five dose-additivity using efficient and innovative experimental designs linked to analytical methods that incorporate the concepts of thresholds for noncancer events and estimates of variability, 2) capability to estimate confidence limits and prediction intervals around the point estimate of the effects of the mixture, 3) ability to use the model with mixtures of up to five compounds, 4) ability to develop statistical models for neurological, biochemical, and behavioral data, including continuous, rank, and quantal data, and 5) experimental designs and statistical methodology employed have either been developed with toxicity data sets or have been applied to mixtures of environmental pollutants, 6) demonstrated ability to collaborate effectively with experimentalists in all phases of experimental design and statistical analysis of data, and 7) documented experience in the use of these methods, including publications in high-quality, peer-reviewed journals. Due to the limited nature of this funding, this agreement will not be used to develop these methods de novo. The first phase of this study will involve collaborations between the cooperator and EPA personnel to determine an experimental design which will fit the assumptions used in the statistical model to be tested, and to design the most efficient model to use. In the second phase, experiments will then be conducted at the NHEERL laboratories using the specific mixtures of these pesticides, and observed responses will be compared to that predicted under a threshold additivity model using prediction intervals. Comparisons will be made between the additivity models for rats of different ages while accounting for biological variability. Timely statistical analysis of the data will be required in order to guide the subsequent experiments. Posted 06/07/99 (W-SN339970). (0158)

Loren Data Corp. http://www.ld.com (SYN# 0003 19990609\A-0003.SOL)