Projecting cancer incidence and mortality rates, and relating those projections to the attainment of national cancer control objectives are the goals of this intramural research project. The project includes development and continued refinement of a computer model which projects cancer incidence and mortality, meshing together data from a variety of sources, and adapting quantitative cancer control objectives to fit the modeling framework. The NCI staff has developed and written a large interactive Fortran program used to project cancer figures for a forty year period. The model incorporates different models for survival from cancer, includes data for a number of cancer sites, the ability to examine temporal trends in underlying cancer incidence and mortality from other causes, adjustment of rates to different populations, and production of annual projections of cancer incidence and mortality. The crux of the model is the flexibility to analyze the effect of cancer prevention, screening, and treatment activities (in any combination) on cancer mortality. Work in the current year has involved continued updating of the basic underlying database used by the program, modifications related to implementation on the NIH Convex computer, calculation of rates to different standard populations, and coordination with a similar program developed for the IBM personal computer. One of the major efforts related to the IBM model involved modeling of cancer screening.
