Patients entering hospitals with heart attacks in states with mandatory public reporting are less likely to receive angioplasties to fix heart blockages than patients in states without public reporting, according to a new study from Harvard School of Public Health. The patients most affected were those considered "high-risk," that is, those who were extremely sick when they arrived at the hospital. It is the first study to look at public reporting for these interventions on a national level. The study appears in the October 10, 2012 edition of the Journal of the American Medical Association.
"Making performance data available to patients is very important. It helps them choose the best health care," said Karen Joynt, instructor in the Department of Health Policy and Management at HSPH and a physician at Brigham and Women's Hospital in Boston. However, the results of our study make us wonder if we're doing public reporting as well as we could be."
Public reporting is an increasingly popular strategy for improving health care. While currently only three states require public reporting of patient outcomes after percutaneous coronary intervention (PCI, more commonly referred to as stenting or angioplasty, a procedure in which balloons and stents are inserted into heart arteries to fix blockages), many states have voluntary reporting programs, and many more are considering implementing such programs. The idea is simple -- if hospital performance is made public, patients can have information to choose the best hospital and clinicians will be motivated to improve performance. However, a worry is that if hospitals are concerned about receiving negative ratings in public reports, they will be less inclined to offer these procedures to very sick patients.
The researchers, led by Joynt and Ashish Jha, associate professor of health policy and management at HSPH and a staff physician at the VA Boston Healthcare System, analyzed data from patients older than 65 years of age who had suffered heart attacks in Massachusetts, New York, and Pennsylvania, three public reporting states, from 2002-2010. They compared the patients in those three states with patients in seven non-reporting states in the same region.
The results showed that overall, Medicare patients with heart attacks in those three states had 18% lower odds of receiving PCI than patients in states without public reporting. The results were even more striking for extremely sick patients -- those arriving at a hospital with a massive heart attack had about 27% lower odds and those arriving in cardiac arrest or in shock had about 21% lower odds of receiving PCI than comparable patients in other states. In Massachusetts, which implemented public reporting relatively recently, the odds of receiving PCI, relative to other states, dropped 19% after the public reporting program started.
The researchers note that there are at least two possible explanations for the study results. One is that physicians were concerned about the risk of poor outcomes if they performed PCIs on patients who were very ill. However, it is also possible that physicians did not perform PCIs because the procedures were futile or unnecessary. "It will be important for follow-up work to determine why this is going on, and more importantly, how we can address it," said Joynt.
This study raises a caution flag, said Jha. Many states are considering implementing reporting programs and many have put voluntary programs in place. Policies may need to be evaluated with the same rigor that new drugs, devices, or therapies are evaluated, he said.
"Improving transparency of the health care system is critical," Jha said. "However, we have to make sure we do it right. Otherwise we risk creating incentives to deny care to the sickest patients."
- Karen E. Joynt, Daniel M. Blumenthal, E. John Orav, Frederic S. Resnic, Ashish K. Jha. Association of Public Reporting for Percutaneous Coronary Intervention With Utilization and Outcomes Among Medicare Beneficiaries With Acute Myocardial Infarction. JAMA, October 10, 2012, Vol. 308, No. 14
Cite This Page: