OVERVIEW OF IMPORTANT DESIGN ISSUES FOR A NATIONAL HUMAN EXPOSURE ASSESSMENT SURVEY

Exposure issues have important consequences for regulatory decisions. Reliable answers to exposure questions are critical for site cleanup, model validation, and cumulative risk issues, as well as giving perspe ctive on our risk estimates. This paper discusses some of the importan t issues in designing the National Human Exposure Assessment Survey (N HEXAS) and, by implication, other exposure-monitoring-based studies as well. Sampling design issues are discussed in terms useful to exposur e assessors. These issues include simple random sample designs versus more complex multistage designs, design efficiency, how to determine t he sample size for the desired precision of the estimate, and the effe cts of stratification and oversampling on the needed sample size. This paper also discusses several important nonsampling issues such as pop ulation definition, response rates, and several potential sources of e rror in interpreting the monitoring results.