3150.0: Monday, October 22, 2001: 2:30 PM-4:00 PM | ||||
Oral Session | ||||
| ||||
| Environmental health studies pose particularly difficult problems of statistical analysis for several reasons. Realistic data are rarely obtainable under controlled experimental conditions, so that most data are observational and subject to numerous confounding factors. Sampling of environmental conditions is rarely uniform over both space and time, is often sporadic, and may be incomplete or perturbed by instrument failure. Environmental exposure is often only vaguely conceptualized (e.g., cigarette smoke exposure), and is always measured with error arising from extrapolation of spot measures to immediate personal (self-specific) environments, extrapolation of short-term measures over time, and/or reliance on memory and fallible record systems. By definition, many environmental factors are regional, calling the primary independence assumption of classical statistical methods into question for proximal observational units. In this session, leaders in the statistical analysis of air quality, dietary, and toxicology studies address relevant statistical issues and approaches to dealing with them, in the context of substantial applications. | ||||
| See individual abstracts for presenting author's disclosure statement. | ||||
| Learning Objectives: At the conclusion of the session, the participant should be able to: 1. Describe uses of loglinear models with natural cubic splines, and of spatial regression, in studying association between mortality and particulate pollution, and list conditions that determine the appropriateness of these statistical methods for analyses of other environmental problems. 2. To be added. | ||||
| Andrew A. White, PhD | ||||
| Peter B. Imrey, PhD | ||||
| Introductory Remarks | ||||
| Assessing the prevalence of nutrient inadequacy Alicia L. Carriquiry, PhD | ||||
| The National Morbidity, Mortality, and Air Pollution Study: Estimating regional and national dose-response relationships Francesca Dominici | ||||
| Challenges to statistical analyses of toxicological data Christopher Portier, PhD | ||||
| Discussion | ||||
| Sponsor: | Statistics | |||
| Cosponsors: | Environment; Socialist Caucus | |||
| CE Credits: | CME, Environmental Health, Health Education (CHES), Nursing, Pharmacy, Social Work | |||