Heart attack patients in states that require health care providers to report the outcomes of procedures to open blocked arteries are less likely to receive those live-saving treatments than similar patients in states without public reporting mandates, according to a new study.
The disparities in care, however, appeared to have little effect on patient survival rates.
The analysis, based on nearly 100,000 Medicare patients in 10 states, comes amid a nationwide push for greater transparency in how doctors and hospitals measure up against one another.
Most health policy experts are in favor of publishing such performance data so that medical providers can be held accountable for their results and patients can make informed choices about where they receive care. But the study, published Wednesday in the Journal of the American Medical Association, shows that public reporting does not automatically lead to better outcomes - and potentially can backfire.
Dr. Hitinder Gurm, a cardiologist at the University of Michigan who was not involved in the study, said the analysis raised concerns that some extremely sick patients were being denied care because doctors were under pressure to keep their success rates high.
"The study highlights the unintended consequences of public reporting," he said.
But considering the comparable patient survival rates, he said there could be another reason for the disparities: Doctors were managing to avoid the procedures in hopeless cases.
Both explanations were probably at work, said Dr. Karen Joynt, a cardiologist at Brigham and Women's Hospital in Boston, who led the study.
In Massachusetts, which has published outcome data since 2005, the "buzz among cardiologists" was that some doctors were not operating on patients with the lowest chance of survival, Joynt said. She was skeptical but decided to investigate.
Her team looked at nine years of Medicare data, limiting the study to patients 65 and older who had suffered acute heart attacks.
In 2010, the most recent year in the analysis, 38 percent of patients in the three states that require public reporting - Massachusetts, New York and Pennsylvania - received an artery-opening procedure. That compared with 43 percent of patients in seven states without the requirement - Maine, Vermont, New Hampshire, Connecticut, Rhode Island, Maryland and Delaware.
The differences were more pronounced in patients with full blockages: 62 percent versus 68 percent.
Many acute heart attack patients arrive at the hospital in dire condition and have little chance of survival regardless of treatment. The most common artery-clearing treatments involve threading a catheter to the site of the blockage and inflating a tiny balloon to expand the vessel or installing a stent to prop it open. But deciding whether intervention would be futile can be more of an art than a science.
"It's a really hard decision," Joynt said.
The researchers found the overall death rates after 30 days were about 12 percent regardless of reporting requirements. Among patients with fully occluded arteries, the death rate in reporting states was 13.5 percent, compared with 11 percent in nonreporting states - a slight but statistically significant difference.
Public reporting has long been the norm in transplant surgery, where it is viewed as a safeguard against wasting scarce donor organs. In recent years, similar requirements have spread to other areas of medicine. Medicare now publishes several types of hospital performance data at www.hospitalcompare.hhs.gov.
Many doctors have been opposed to public reporting and may try to use the new study as ammunition for their cause. But several experts said the results - and the questions they raise - underscored the need for sophisticated performance measures that take into account the differences between patients and thus allow for valid comparisons.
Dr. Harlan Krumholz, a Yale cardiologist who helped develop the outcome measures used by Medicare, said the next steps were to figure out why some doctors and hospitals did better than others, and then to spread the best practices throughout the health care system.
"Just putting the data online is not enough," he said.