The major goals of this proposal are to develop and extend statistical models which are important in predicting the future incidence and mortality associated with breast cancer in the United States. The diagnosis, treatment and prevention of female breast cancer has been rapidly changing in the decade and shows every evidence of continuing to change. Our main thrust will be to estimate the impact of interventions on national statistics. The interventions will not only reflect current health behavior practice, but will reflect various scenarios if public health practices were to change. Among the problems which will be considered are: (1) to evaluate the relation between early detection and eventual mortality which considers competing risks; to predict the changing incidence patterns if more women participate in routine screening programs; to compare the trade- offs pertaining to both benefit and cost between optimal screening programs and recommended policies; to develop optimal screening programs for high risk sub-populations; to estimate the change in mortality if modern cancer treatment was routinely given to a higher proportion of the population; to use the Eastern Cooperative Oncology Group (ECOG) and the Harvard Pilgrim Health Plan as a resource to gather data on diagnosis and staging currently not available; to investigate screening policies and their benefit for younger women; to estimate the impact of improved methods of diagnosis on incidence and mortality.
Showing the most recent 10 out of 11 publications