UR study sheds light on decision-making in era of health 'report cards'
Many interventional cardiologists opt out of treating patients if chance of death is high
Tuesday, January 11, 2005
Nearly 80 percent of interventional cardiologists in New York state admit they have avoided performing a risky but potentially life-saving angioplasty on a patient, out of fear that if the patient dies it skews the doctor’s personal mortality “report card,” according to a University of Rochester survey.
The anonymous poll was designed to measure whether the state Department of Health’s system of tracking doctor’s cardiac death rates has an impact on how doctors make treatment decisions – and what types of patients they choose to accept.
The groundbreaking study, its results published in the Jan. 10 issue of the Archives of Internal Medicine, found that 79 percent of New York interventional cardiologists polled responded that they have made decisions about accepting critical cases into their cardiac catheterization labs that hinged on how the outcome would affect their state mortality data report card.
New York is one of only a handful of states to publish mortality data for consumers in an effort to raise awareness about physicians’ and hospitals’ records, and more states are considering adopting the practice. Many medical professionals, though, have long suspected that these report cards may influence some physicians to avoid taking critically ill patients to the catheterization lab, even though the patient might benefit from angioplasty, because of the possibility of adding a potential death to their files, says principal investigator Craig Narins, M.D., an interventional cardiologist.
“While these reports attempt to provide the public with objective information about physician quality, they can in some instances create a conflict for the physician that may actually worsen patient care,” he says.
For the study, an anonymous, one-page questionnaire was sent to all interventional cardiologists in New York – 186 physicians – who were included in the angioplasty report released in January 2003. They were asked to respond to nine statements regarding the state mortality reports, indicating whether they strongly agreed, agreed, disagreed or strongly disagreed with each statement.
Conducted by Narins with Frederick Ling, M.D., Wojciech Zareba, M.D., Ph.D., and Ann Dozier, R.N., Ph.D., the UR survey received an extraordinary 65 percent response rate. Investigators speculate that doctors responded in such large numbers because they want the public to know that a physician’s mortality rate, the only statistic measured on the state’s report card, is by itself an inaccurate indicator of the physician’s skill level or quality of care.
Of the 120 physicians who responded to the survey, the vast majority agreed or strongly agreed that the publication of mortality statistics has, in certain instances, influenced their decision regarding whether to perform angioplasty on individual patients. Physicians expressed an increased reluctance to intervene upon critically ill patients with high-expected mortality rates, even though these patients may have the most to gain from angioplasty.
Among the respondents, 83 percent agreed or strongly agreed that patients who might benefit from angioplasty may not receive the procedure as a result of public reporting of physician-specific mortality rates. The scoring system attempts to not penalize physicians as much when a more severely ill patient dies following a procedure; yet 85 percent of those surveyed believed that the risk adjustment model used in New York is not sufficient to avoid punishing physicians who perform higher risk interventions.
The state reporting system may, as intended, lead some interventional cardiologists with low volumes or poor outcomes to improve their performance or stop performing the procedure. However, the unintended effects of the scorecard system on patient care decisions, as described in the paper, may adversely affect outcomes for patients who might benefit from angioplasty but are denied the procedure.
“Take, for instance, a case of a very ill patient who presents to the emergency room with a large heart attack that is complicated by shock. Attempting to unplug the patient’s blocked artery with an angioplasty procedure has been shown to save heart muscle and reduce the patient’s chance of dying, but even if the cardiologist decides to perform angioplasty, some patients will not survive,” Narins says. “So the cardiologist must decide whether to perform a procedure that is possibly life-saving for the patient at the risk of making their own statistics will look worse.”
If the patient receives necessary angioplasty, but ultimately dies from their illness, that death still shows up on the cardiologist’s record and counts against them as well as their medical institution when the state Department of Health releases cardiac mortality data. However, if the cardiologist decides not to perform the angioplasty, which may increase the patient’s chance of dying, their statistics will not be affected. In this way, the presence of the report card may in some instances influence a doctor not to perform a procedure that may be helpful for the patient.
“There is a fear among health care professionals that as more states adopt mortality data reports, insurers may direct patients to hospitals with better scores, even though better scores probably do not correlate with better patient care,” Narins says.
At the University of Rochester Medical Center, protocol dictates that all patients, regardless of their expected outcome, be treated in the cardiac catheterization lab if there is a chance they may benefit from angioplasty.
“Our numbers, although at times above the state average, reflect the deaths of patients who came to us critically ill,” says Ling, director of the cardiac catheterization lab. “We’re aware of the potential outcome, but it is our duty as physicians to treat patients to the best of our ability.”