Publication of mortality data affects angioplasty rates
■ A study highlights the active influence of physician report cards.
By Andis Robeznieks — Posted Jan. 31, 2005
Public reporting of physician mortality outcomes in New York has an effect on medical decision-making, according to a report published in the Jan. 10 Archives of Internal Medicine.
Almost 80% of interventional cardiologists responding to a survey agreed that publication of mortality statistics has influenced their decision to perform angioplasties on high-risk patients. The researchers noted that this provides evidence that "physician scorecards are not passive devices simply monitoring quality" but have an active influence on clinical decision-making.
Physicians have long wondered what effect the state's scorecards may have on medical decision-making, said the report's lead author, Craig R. Narins, MD, an assistant professor of medicine/cardiology at the University of Rochester School of Medicine and Dentistry in New York.
"I was surprised to see how widespread it was," Dr. Narins said. "The scorecards provide useful information, but our study shows that they have unintended consequences."
The New York State Dept. of Health has been collecting mortality data since 1994, and Dr. Narins and his colleagues received completed surveys from 120 (65%) of the 186 interventional cardiologists included in the state's report on angioplasty outcomes between 1998 and 2000.
More than 83% of the responding physicians agreed that high-risk patients who might benefit from angioplasty might not receive the procedure because of the state's publication of physician-specific mortality rates.
One New York City heart surgeon, however, suggested that if a physician is distracted by what could appear on his or her scorecard, the patient might be better off if that doctor doesn't perform the procedure.
"Those who don't feel comfortable should refer [high-risk patients] to those who do," said O. Wayne Isom, MD, chair of the Dept. of Cardiothoracic Surgery at New York Presbyterian-Weill Cornell Medical Center.
The American Medical Association did not comment specifically on the study, but AMA Trustee Cyril M. Hetsko, MD, said the Association has concerns about public report card systems for individual physicians.
"There's always a difficulty with public reporting and physician report cards of any kind -- and that has to do with the validity of the report cards to begin with," said Dr. Hetsko, a clinical professor of medicine at the University of Wisconsin Medical School.
Dr. Hetsko said there is a concern that report cards could give physicians incentives to avoid sicker patients or patients who might be viewed as noncompliant to recommendations that they quit smoking, change their diet or start exercising. "You can try to adjust for these differences, but it's very difficult."
In general, Dr. Hetsko said, AMA concerns include: having an adequate sample size to ensure an appropriate analysis, appropriate risk adjustment, uniform implementation of data collection, opportunities for physicians to review data before publication, and fairness.