Challenges for Evaluating Individual and Community Level Data in a Children’s Pesticide Intervention Study. W. C. Griffith and E. M. Faustman, IRARC, University of Washington
Estimating the amount by which child pesticide exposures may be lowered through a community education program in Washington’s Yakima valley has presented a number of challenges because it involves combining statistical methods for estimating urinary metabolites of pesticides with methods for estimating effects in randomized community intervention trials. Multiple metabolites are measured in urine and then combined to estimate the total uptake of dimethyl and diethyl organophosphate pesticides. The assays for some metabolites reported a high percentage of samples below detection levels and required development of statistical methods appropriate for combining the individual metabolite measurements into the combined diemtyl and diethyl uptakes. Comparisons of uptakes, before and after the education program, were used to estimate the effectiveness of the community intervention where the unit of analysis is the community rather than the individual. Adjustments were made to the original protocol because it was difficult to locate children in some communities. At present baseline information is available and indicates there are no pre-intervention differences between communities randomly assigned to the education program and those not assigned. Approaches to evaluate the final results are described.
This work was performed with joint support of U.S. Environmental Protection Agency grant EPA-R826886, and National Institute of Environmental Health Science grant 1 PO1 ES09601.
Go to . . .
2001 SRA Annual Meeting Table of Contents
2001 SRA Annual Meeting Author Index
Main Abstracts Menu Page
RiskWorld Home Page