Development of Cost-effective Statistical Sampling Strategies and Optimal Design Considerations for Exposure Assessment as Part of the National Children's Study

• Activity Number: 95
• Type: Invited
• Date/Time: Monday, August 9, 2004 : 10:30 AM to 12:20 PM
• Sponsor: General Methodology
• Author(s): Warren J. Strauss*+ and Jeff Lehman and Haluk Ozkaynak
• Companies: Battelle Memorial Institute and Statistics and Data Analysis Systems and U.S. Environmental Protection Agency
• Keywords: optimal design ; measurement error ; designed missingness ; longitudinal ; environmental exposure ; replicate sampling

Abstract

This project focused on innovative statistical study design guidance for the acquisition of exposure data over time in a longitudinal study of children who participate in the National Children's Study (NCS). The NCS is intended to investigate environmental influences on children's health and development, including understanding any environmental exposures that may cause or exacerbate health impacts. Due to the large sample size and longitudinal nature of the NCS, unique statistical issues arise that must be addressed before a cost-effective sampling design can be developed to gather environmental and personal exposure data. A key issue for the NCS relates to obtaining enough samples to provide adequate statistical power to detect health effects attributable to exposure, while being cost-effective, minimizing participant burden, and staying within the study's budget. Relevant specific issues include identifying potential sources of bias and/or uncertainty in the exposure measures (nonresponse, subject burden, attrition, and measurement error); and strategies to address these issues, including sample weighting techniques, and replicate sampling to assess measurement error variance.