Quality-of-Care Indicators for Early-Stage Prostate Cancer

PUBLIC ABSTRACT

With over 230,000 new cases and 29,000 deaths estimated in 2004, prostate cancer is the most common cancer and the second leading cause of cancer death in American men. Much of the scientific literature has focused on the choice among treatments for early-stage prostate cancer, including watchful waiting, surgery, and radiation, and which one of these treatments leads to greatest survival. Decisions regarding treatment are frustrated not only by inadequate evidence favoring one treatment modality, but also by the absence of data comparing quality among providers. In fact, the choice of provider may be as important as the choice of treatment in the ultimate outcomes of treatment.

Significant variation in the use of screening tests and standard primary treatment for prostate cancer has been demonstrated over time, across geographic regions of the United States, and among racial/ethnic and socioeconomic groups. More importantly, large racial differences in mortality also have been shown. The sources of these variations are not well understood. One potential source that has never been investigated is variations in provider practice patterns. We hypothesize that significant variation exists nationally in the treatment of early-stage prostate cancer by urologists and radiation oncologists.

Quality-of-care research has been applied to many surgical diseases and procedures, most notably cardiac bypass surgery and carotid endarterectomy. However, little progress has been made in communicating the results of these studies to the public in the form of consumer reporting. At RAND (Research and Development Corporation), we previously developed candidate quality-of-care indicators as a starting point for the evaluation of variations in quality of care for men with early-stage prostate cancer. These indicators have been successfully pilot tested; formal validation of the indicators is the next step.

This project will evaluate the RAND candidate indicators by applying them to a national sample of patients. We will compare quality of care among hospitals and among urologists and radiation oncologists in order to develop a tool for patients to select where to go for their treatment. We will determine whether indicator compliance rates differ by patient characteristics, particularly race, to investigate whether disparities in outcomes of prostate cancer care can be partially explained by quality-of-care differences. We will perform statistical analyses to determine which components of quality of care are most strongly associated with better clinical outcomes. Finally, throughout each stage of the project, we will report back our results to the hospitals in order to identify areas for improvement.

The end product of this study will be the development of a system to measure the quality of care for early-stage prostate cancer. Identifying hospitals practicing higher quality care in the treatment of prostate cancer will yield an immediately valuable resource to other hospitals by focusing on the factors that lead to better outcomes. Patients will benefit by using the results to assist them in making treatment decisions as well as by the possible by-product that hospitals will have greater incentive to compete on quality rather than costs alone. This study will allow us to begin to develop quality improvement programs aimed at narrowing variations in care and consumer-oriented resources such as report cards comparing individual providers and institutions.