The objectives of this ongoing research are to develop quantitative methods for assessing the combined effects on chronic disease of one or more environmental exposures and lifestyle factors, and to supply these methods to epidemiologic data. Improved statistical techniques will be sought in two problem areas. The first concerns methods for dealing with incomplete covariates in multiple regressions of disease-free survival. Examples of incomplete covariates include missing covariates, grouped and censored covariates, and covariate measurement errors. New methods will be tested by application to cohort mortality data among occupationally exposed groups and among former college students, and to case-control data for various site-specific cancers. The second problem area concerns methods for detecting and evaluating spatial, temporal and occupational clusters of disease. An expanded statewide cancer reporting system in California has been motivated by increased public awareness of the potential carcinogenic effects of chronic exposures to environmental toxicants. The proposed research will develop new methods for analyzing tumor registry data in search of unusual patterns of cancer occurrence. These methods will be tested by application to data from the San Francisco Bay Area Surveillance, Epidemiology and End Results (SEER) Program, and from the new California Tumor Reporting System. Attention will focus on detecting incidence patterns among histologic subtypes of site- specific cancers and among ethnic subgroups of the population.
Showing the most recent 10 out of 11 publications
