Prostate cancer is characterized by large racial differences in both incidence and post-diagnosis survival. Known disparities in health care access and utilization complicate uncovering underlying differences in the natural history of the disease. Such differences can be caused by racial variations in risk factors such as diet or genetic predisposition. Understanding and quantifying them is important for the development of health policies. We propose to use modeling techniques to estimate the manner and extent of racial differences in the natural history of prostate cancer. Our model uses the dynamics of prostate cancer incidence after the introduction of prostate-specific antigen (PSA) screening to obtain information about the preclinical stage of the disease. We will rely on the SEER-Medicare database for detailed estimates of prostate cancer incidence and screening behavior.
Our specific aims are to estimate age- and race-specific rates of transurethral resection of the prostate and PSA tests in the SEER 9 registries; to develop a model for screening behavior variability; to develop a model for prostate cancer incidence; and, finally, to use the models and the screening dissemination data to assess similarities and differences in the natural history of prostate cancer between blacks and whites.
