October 9th, 2012

PCI Utilization Lower in States with Public Reporting of Outcomes

In patients with acute MI, utilization of percutaneous coronary intervention (PCI) is lower in states that publicly report outcomes data, according to a new study published in JAMA. Despite the difference in utilization, however, there was no difference in mortality between reporting and nonreporting states.

Karen Joynt and colleagues used Medicare data to analyze PCI utilization and mortality in acute MI patients in three states with public reporting of PCI outcomes (New York, Massachusetts, and Pennsylvania) and other states in the same region without public reporting. The differences in utilization were greatest in patients at highest risk, who presented with ST-segment elevation MI (STEMI), cardiogenic shock, or cardiac arrest.

  • Overall unadjusted PCI rate: 37.7% for reporting states versus 42.7% for nonreporting states
  • Risk-adjust odds ratio: 0.82, CI 0.71-0.93, p=0.003)

Overall mortality did not differ between the reporting and nonreporting states (12.8% and 12.1%, respectively; adjusted OR 1.08 (CI 0.96-1.20], p=0.20), although there was a significant mortality difference in the STEMI subgroup (13.5% vs. 11.0%; OR 1.35, CI 1.10-1.66, p=0.004).

In Massachusetts, where outcomes reporting was initiated during the course of the study period, PCI utilization was at first no different from the other nonreporting states, but was significantly lower than nonreporting states after the change.

The authors offer two potential explanations for the findings:

…the foregone procedures were futile or unnecessary, and public reporting focused clinicians on ensuring that only the most appropriate procedures were performed. Alternatively, public reporting may have led clinicians to avoid PCI in eligible patients because of concern over the risk of poor outcomes.

The mortality findings, they write, suggest “that the foregone procedures might have been a mix of appropriate and inappropriate PCIs.”

In an accompanying editorial, Mauro Mosucci writes that the mortality finding may be due to “a conscious or unconscious ‘futility assessment'” in states with public reporting,” leading to “avoidance of PCI for patients who are less likely to benefit.” Alternatively, “public reporting might have resulted in a drive toward improved quality of care and improved outcomes in patients receiving PCI, offsetting the adverse effect of not performing PCI in high-risk patients.” Mosucci also points out that the data may be skewed because public reporting might result in “gaming” the coding of cases.

In his Voices blog, Joe Ross wonders what’s really behind the lower rates of PCI use.  Read his theories, and share your comments or questions, here.

3 Responses to “PCI Utilization Lower in States with Public Reporting of Outcomes”

  1. Gosh, theheart.org headline really misrepresents the findings of this article. It states: Public reporting reduces PCI rates in AMI patients – but this article is hardly in a position to prove that rates dropped because outcomes were reported – and, if it did, you would think that mortality would have dropped because the presumption would be that practitioners were avoiding the highest risk patients with concerns that the risk-adjustment would not account adequately for them. In the end it is really hard to evaluate the effect of public reporting. In any case, I think it is likely that reporting alone is not enough to produce true improvement – measurement and transparency is just a first step.

  2. Am I reading this correctly? Increased PCI in non-reporting states results in a significant reduction in AMI death and a trend toward reduction in all cause mortality despite a relatively small difference in PCI rates.

    This sounds like an indictment of reporting. It has resulted in increased coronary death, the opposite of what we want with health care reform.

  3. Robin Motz, M.D., Ph.D. says:

    Isn’t it true that once NY State started reporting open heart surgical death rates that some cardiac surgeons would not operate on the most hopeless cases? I seem to recall an op-ed piece in the NY Times several years ago about this.