At present, available performance measures do not include quality indicators for prostate cancer treatment (e.g., measures of the National Committee on Quality Assurance and Foundation for Accountability). Investigators at RAND are developing candidate quality indicators based on a structured review of the literature, key informant interviews of prostate cancer experts, and focus groups with patients as part of a RAND study funded by the Bing Foundation. The next phase of this research will be to test the reliability and validity of these candidate measures in evaluating the quality of care, as well as exploring potential process-outcomes links.

Good indicators of quality are based on evidence from rigorous research, which is not available for most aspects of cancer care. For those aspects of care that have been evaluated, the quality of health care can be precisely defined and accurately measured. Measures of structure, process, and outcomes can all be used to assess quality. An outcomes indicator that is often used to evaluate cancer care has been five-year survival, but more timely and practical measures are becoming available to more precisely assess factors related to health care that can affect outcomes. Process measures can serve as good quality indicators when research has proved that a given process leads to better outcomes. Examples of good process measures for breast cancer include

• use of screening mammography,
• use of radiation therapy following breast conserving surgery, and
• use of adjuvant therapy among women with local or regional breast cancer.

In other cases, research suggests that one process does not have an advantage over another in terms of outcomes, so patient preferences should dictate the course of care. For many women with breast cancer, for example, optimal care involves presenting information on alternative treatments and supporting an informed choice.

Sometimes, research suggests that providing a service does not have a favorable impact on outcomes, indicating that the service should not be provided. Most elderly men with prostate cancer, for example, would not likely benefit from radical prostatectomy if their life expectancy is less than 10 years. High rates of surgery among very old men could indicate that surgery is being performed too often when there is no expected benefit (and there is potential harm from surgery).

Two national databases are available with which to assess the quality of cancer care, but each has limitations in the context of evaluating quality of care. The SEER cancer registry has been valuable when linked to Medicare and other insurance administrative files to assess quality of care for the elderly and other insured populations. It is also useful in identifying cases for in-depth studies of quality-related issues. The SEER registry, however, covers only 14 percent of the U.S. population and thus may not adequately represent the diversity of systems of care. Finding ways to capture measures of process of care, treatment information, and intermediate outcomes and improving the timeliness of reporting would enhance the registry's use in quality assessment. The National Cancer Data Base now includes information on more than one-half of all newly diagnosed cases of cancer and many of the demographic, clinical, and health system