In the past, Jencks says, HCFA has sought to improve care by hunting for "bad apples" among physicians through audits of patients' charts. The new report represents an effort to instead use population-based performance measures to gauge the quality of care across the entire Medicare system.
"We think that a great deal of the gap between what could be done and what is happening is attributable to systems failures," Jencks tells WebMD. "We think the health care system could be engineered so that these things happen as a matter of course, rather than depending on very busy individual physicians to make them happen."
For example, Jencks says, screening for patients' immunization status would improve if doctors' office systems were designed -- through computerization and other mechanisms -- to ensure patients were screened routinely. Now, many offices simply rely on the doctor to ask during checkups. "It is likely the immunization rates under these two scenarios are going to be profoundly different," he says.
In an independent assessment of the study, Elliott Fisher, MD, MPH, professor of medicine at Dartmouth Medical School and a researcher involved in the Dartmouth Atlas Study Series, says the study underscores the inconsistency found in American medical practice. The Dartmouth Atlas series has found wide variation in the utilization of services in the Medicare program.
"I think this drives home the message that for a variety of very well-measured attributes of medical care, there are substantial differences across states," says Fisher, who also is co-director of the VA Outcomes Group, a research group based at the Veterans Affairs Hospital in White River Junction, Vt., "Even the best states could do much better. There is lots of room for improvement."
The challenge, Fisher cautions, will be to use the information to improve overall systems of care, rather than focusing on specific measures on which states fared poorly. "There will be a temptation to look at these results in the same way that physicians look at report cards," he says. "We need take what we learn from these measures of variation and modify the systems to achieve the results we know we can achieve."
But some medical societies in states that did not score well suggest that the study does not necessarily tell the entire story. Leroy Sprang, MD, president of the Illinois State Medical Society, acknowledged there are areas of care that can be improved, but said the report's focus on performance measures may not indicate how the patients actually fared in treatment.
Jencks says that doctors and others who react defensively to the report, and patients who react with fright, risk missing the potential inherent in its findings.
"The point is that there is an opportunity to improve care, and the opportunity is best realized through systems change," Jencks says. "These figures are not about what has gone wrong. We have a system that is getting better."