Measure for measure: the cost of improving quality.

There is an urgent need to ensure that cancer patients receive high-quality care. This is coupled with the concern that the cost of cancer care is rising faster than many other areas of medicine and the hope that improving quality will both curb these rising costs and improve outcomes (1). The quality of cancer care people receive is highly variable. While some patients often do not receive treatments and interventions that are known to be beneficial, over-use or misuse of unnecessary and sometimes harmful treatments are also common (2). Addressing the barriers to achieving quality care and identifying intervention to improve care depend on the ability to define and measure quality. Information on quality of care can be classified into " structure, " " process, " and " outcomes " measures. Structure includes all of the factors that affect the context in which care is delivered, process is the sum of all actions that make up the manor in which care is delivered, and outcome contains all the effects of healthcare on patients or populations (3). In 2008, representatives from the American Society of Clinical Oncology (ASCO) and the National Comprehensive Cancer Network (NCCN) collaborated to create three breast cancer and four colorectal cancer metrics suitable for national performance measurement based on high-level evidence (4). Only evidence-based clinical indicators predict patient outcomes and are true measures of quality, however indicators based on professional consensus without evidence may at times be all that is feasible (5). While more comprehensive lists of quality indicators have been developed, many are not supported by high-quality clinical evidence, and the ability of these indicators to predict outcomes that are meaningful to patients is unknown (6). In the paper that accompanies this article, Dr. Hassett and colleagues (7) have shown that quantifying the relationship between quality process indicators, cost, and outcome is not straight forward. In a population-based sample of elderly women with breast cancer and ductal carcinoma in situ (DCIS), an association between spending and adherence with quality measures was not found. In addition, no association was observed between spending and survival. The results are disturbing. If we take this at face value, we may erroneously conclude that there is no reason to advocate for quality. Either the truth is that adherence to quality guidelines does not reduce cost or improve outcome, or it may be that the measures of quality, cost, or outcome used …