Public disclosure about surgeons' open-heart outcomes has been controversial ever since New York released the nation's first state mortality scorecard for coronary artery bypass graft surgery (CABG) in 1991.
Providers in New York and elsewhere have complained that it's impossible to adjust such reports to fairly reflect factors that put their patients at higher risk of dying. To score surgeons like that, some say, provokes hospitals and doctors to cherrypick their cases, refusing patients they think might be more likely to die within 30 days.
Indeed, cardiac surgeons in New York and Pennsylvania, states where mortality of patients undergoing CABG is regularly reported, have answered surveys saying they were subsequently less likely to accept higher risk patients, according to a report earlier this year by Patrick Romano MD, Clinical Lead for the Agency for Healthcare Research and Quality's national Quality Indicators program. Similar reporting is done in three other states, California, New Jersey and Massachusetts.
If this is true, it would mean some patients would be left to die without care, or would have to travel out of state to get it. I recall physicians insisting to me that they couldn't afford to have such patients pull their reputations down—though they felt bad about sending such patients away.
But Romano, professor of general medicine and pediatrics at the University of California at Davis, an expert on quality outcome reporting metrics, says there's little evidence in the data to back up their claims.
"Despite all of this concern about discrimination against high-risk patients, empirical evidence of this phenomenon is very limited," he says.
In other words, patients are not being turned away in droves. And surgeons who claim that they are being turned away may be full of bluster. Maybe they just don't like the transparency.
There was one report from the Cleveland Clinic in Ohio that noted a higher than normal number of cardiac surgical patients came to their care from New York. These patients did have risk factors that placed them in a higher risk category than the Clinic's patients from Ohio and higher than patients who received surgery in New York.
But that increase was only from 61 to 96 patients between 1989 and 1993. "Other data suggest that out-referral of high-risk patients is not a serious problem," Romano says.
The Office of Statewide Health Planning and Development (OSHPD) in California, which has had mandatory public reporting of CABG mortality in place since 2003 and voluntary reporting before that, was concerned by sufficient numbers of reports of physician discrimination that it asked Romano to look at the issue.
"I've heard this from surgeons myself," says Joe Parker, OSHPD's director of healthcare outcomes. "Surgeons have expressed fears that their associates are choosing not to perform surgery on the sickest cases, in the belief that the risk modeling doesn't adequately adjust for that."
Romano's report was released this week, and he and his colleagues at the University of California at Davis, put much of that concern to rest.
Romano's report documented two significant findings.
The first is that overall relative CABG mortality—by hospital and by physician ? has consistently dropped since March, 2006 when the first of three hospital outcomes reports revealed mandatory mortality outcomes for 2003. Most noteworthy was a 27% drop from the average number of deaths counted during the first seven quarters and the average in the seven quarters thereafter. The latest report was issued on Monday and included hospital variation in CABG-related stroke.
"The most dramatic decrease occurred between the fourth quarter of 2005 (3.18%) and the first quarter of 2006 (2.01%)," Romano wrote.
The reports cover all operative mortality during the hospitalization when the CABG was performed and within 30 days after the surgery, regardless of where the patient died.
The second finding comes from two reports that evaluated surgeon specific mortality rates, with specific surgeons identified by name, in July 2007 for 2003-04 and April 2009 for 2005, 06.
Those surgeons identified as having high mortality rates seemed to be performing fewer CABGs than the surgeons who were labeled as having normal patient mortality outcomes.
What seems to be happening is that within hospitals, medical staffs are policing their own, repurposing surgeons with bad outcomes into other types of care.
In fact, in the California report, Romano notes that among the 12 surgeons who had been labeled as having high risk-adjusted mortality, four discontinued performing those surgeries. What's more, the change from the reduction of those surgeons' volume did not account for the dramatic decline in CABG patient mortality.
Disclosure of the information to the public has, in Romano's opinion, fostered a major soul searching among hospitals and providers to do the right thing to get their practices in shape for themselves and their patients.
"The surgeons in the hospital are responding to the information because it's in the public domain more vigorously than they would respond if it weren't in the public domain. It's pride," he says. "There's now a diffusion of best practices. People are learning from each other. People have perhaps gained more experience over time."
"I speak as a physician here," Romano continues. "The culture in medical practice, and in healthcare in general, is to try to do a good job and to strive for continual improvement. It's embarrassing when you're marked as a bad outlier. It leaves most of us to question: What did we do wrong? What could we have done better?"
The take home message is that providing transparency and accountability with public release of this information "appears to be associated with improved performance across the system statewide," Romano says. "It does not appear to have had an obvious adverse effect in terms of avoidance of high risk patients."
Transparency is not going away. In fact, the Affordable Care Act will soon make dozens of hospital outcomes reportable and comparable. I expect many hospitals and physicians who do well on any metric will shout their success to the world.
The hospitals that don't do well should be prepared to deal with it and improve.
So perhaps the blustering should stop.