The long-range objectives of this proposal are to develop and evaluate statistical techniques useful in the design and analysis of human studies arising in cancer research. Specific topics for the current funding period include: (1) methods for the design and analysis of clinical studies involving recurring events which can be either symptomatic or asymptomatic, (2) methods for the design and analysis of studies to assess agreement between reviewers when response-stratified sampling is employed; and (3) methods for estimating the number of disease cases attributable to an exposure.