The continuing rapid evolution of U.S. health care delivery has raised expectations that the medical marketplace will compete on how care is delivered (i.e., processes of care) and how patients fare (i.e., outcomes of care), in addition to how much care will cost. We propose to evaluate the building blocks of a feasible, reliable, and valid population-based measurement system for elderly breast cancer patients so that scientifically based knowledge will be available to those undertaking comparative studies of care and outcomes for cancer patients during the next decade. Using patient self-report data at diagnosis and one year after diagnosis and medical record data as supplements to registry data, we will describe how variations in patient characteristics at the time of diagnosis, processes, and outcomes of care can vary as a function of the definition of performance measures, the data source used to assess performance, and the definition of the study cohort. The project s specific aims are. 1. To identify a complete cohort of population-based, newly diagnosed older breast cancer patients using existing Rapid Case Ascertainment (RCA) strategies of the Cancer Surveillance Program in Los Angeles County. 2. To use available patient self-report and medical record data as supplements to registry data, to describe variations in patient characteristics at the time of diagnosis, measures of process of care, and patient outcomes. 3. To assess the degree to which process scores can be explained by baseline patient characteristics such as burden of illness at diagnosis, demographics, and system of care. 4. To assess the degree to which outcomes can be explained by processes of care after accounting for baseline patient characteristics such as burden of illness at diagnosis, demographics, and system of care. 5. To assess the degree to which the reliability and validity of measures of baseline patient characteristics, processes of care, and outcomes vary as a function of data sources for those elements. 6. To assess the degree to which the reliability and validity of measures of baseline patient characteristics, processes of care, and outcomes can be influenced by selection biases. 7. To assess ways in which important comparisons between policy relevant groups such as patients treated with fee-for- service vs. managed care, younger vs. older age cohorts, or more vs. less affluent socioeconomic groups vary as a function of: definition of the measures of baseline patient characteristics, processes, and outcomes; data source; and definition of the cohort.
Showing the most recent 10 out of 11 publications
