This project will focus on the analysis of data on the transmission and risk factors of virus infections. A major source of data is the Tecumseh study, which was designed to determine the occurrence and etiology of diseases caused by influenzavirus, rhinovirus and rotavirus infections. These infections, which result in respiratory and gastrointestinal symptoms, are among the most frequent reasons for seeking medical care by the U.S. population. Data on recent outbreaks of dengue in several cities in Western Mexico will also be included in this project, as the dengue virus may spread to the Southern U.S. Longini and his co-workers developed and successfully fitted stochastic models describing the transmission of infectious diseases. These models assume different rates of transmission in a household and throughout the community. The proposed research will extend these models by assuming that the transmission probabilities depend upon the levels of demographic and environmental risk factors. A statistical analysis based on these models will enable epidemiologists to assess the net effect of each factor as well as the interactions among the various factors. Knowledge of the associations between risk factors and transmission rates is essential for the effective planning and implementation of disease control strategies such as immunization, education and reduction of mosquito densities (in the case of dengue). Special statistical methods will be developed to analyze infectious disease data, such as those resulting from the Tecumseh and Mexican studies. These methods must take the underlying sampling design into account, as individuals from the same household or neighborhood cannot assumed to be independent with regard to the dependent variable (infected or not) or to the risk factors. Recently published methods for the analysis of categorical data from complex sampling surveys will be used. The statistical analysis will be supplemented by simulations of the spread of an infectious agent under various control strategies. In this way, it will be possible to identify the most effective strategy to control the spread of a specific virus in a given community.
