Public Reporting Linked to Less PCI, Higher Mortality in Acute MI Patients
Acute MI patients treated in states with mandatory public reporting of revascularization outcomes are less likely to receive PCI and, without intervention, more likely to die in the hospital than those treated in states without such requirements, according to a study published in the March 24, 2015, issue of the Journal of the American College of Cardiology.
Over the past 2 decades, public reporting has been adopted by several US states with the aim of improving quality of care. But previous evidence has suggested that physicians may be reluctant to revascularize critically ill patients when faced with the prospect of adding to adverse outcomes, Robert W. Yeh, MD, MSc, of Massachusetts General Hospital (Boston, MA), and colleagues observe. As a result, they say, “Better reporting methods are needed to adequately balance transparency and accountability with the potential influence of risk aversion.”
Using the Nationwide Inpatient Sample, the investigators compared procedural management and in-hospital outcomes between acute MI patients treated in Massachusetts and New York (n = 57,629), which require public reporting, and those treated in 5 states that have comparable regional populations but no public reporting (n = 26,492). The study period lasted from 2005 to 2011. Patients hospitalized at facilities not offering PCI or transferred to another hospital were excluded from the analysis.
The groups were similar with respect to age (about 68 years) and sex (about 40% female), but patients from public-reporting states were less likely to have private insurance, anemia, peripheral vascular disease, or concomitant cardiac arrest or cardiogenic shock. In addition, length of hospital stay was longer in reporting states (median 4 vs 3 days; P < .001).
After multivariate adjustment, PCI was performed less often in reporting than nonreporting states (OR 0.81; 95% CI 0.67-0.96). The discrepancy was especially marked among older patients, Medicare beneficiaries, and those with STEMI or concomitant cardiac arrest/cardiogenic shock (P for interaction < .001 for all). There was no difference between Massachusetts and New York in the likelihood of undergoing PCI.
In contrast, adjusted rates of surgical revascularization were similar for patients in reporting and nonreporting states. The odds of any revascularization remained lower for those with STEMI or cardiac arrest/cardiogenic shock treated in reporting states.
Overall in-hospital mortality was 6%. The adjusted risk was higher for patients in reporting states than for those in nonreporting states (OR 1.21; 95% CI 1.06-1.37). The finding was consistent across differences in age and Medicare status, although the mortality increase was more pronounced among those with NSTEMI (P for interaction = .035).
Mortality Lower in Patients Who Receive PCI
When patients were stratified by whether or not they received PCI, the adjusted mortality risk in public-reporting states was lower among those who underwent intervention (OR 0.71; 95% CI 0.62-0.83) and higher among those who did not (OR 1.30; 95% CI 1.13-1.50; P for interaction < .001).
According to the authors, “these data suggest that public reporting may improve PCI-related outcomes but may also have the unintended consequence of increasing risk aversion to the detriment of overall outcomes for patients with [MI].”
Reduced mortality in patients selected for percutaneous revascularization “may reflect the intended effect of improvement in PCI quality in [public-reporting] states or a greater avoidance of futile cases,” they observe. “However, those who were not selected for revascularization… had a marked increase in mortality, perhaps reflecting the avoidance of very high-risk cases that may have, in fact, benefited from revascularization.”
Better Reporting Strategies Needed
“There is a significant need to develop strategies to combat risk aversion for percutaneous revascularization among critically ill patients with [MI] while maintaining the important goals of transparency, accountability, and quality improvement supported by public reporting,” Dr. Yeh and colleagues say.
In an accompanying editorial, Mauro Moscucci, MD, of Sinai Hospital of Baltimore (Baltimore, MD), writes that “the answer [to the unintended consequences of public reporting] is not to avoid participation in national or regional registries that provide valuable benchmarks and comparative data. Rather, we should refine our risk adjustment methodologies, educate the public on how to interpret available data, and continue to engage the interventional cardiology community in regional or national continuous quality assurance and quality improvement programs.”
Dr. Moscucci points to the approach used by Massachusetts. “Data on exceptional risk admissions are collected, adjudicated, and analyzed, but they are eliminated from the models and not reported to the public; data on elective PCI and PCI for STEMI and cardiogenic shock are reported separately,” he explains.
More Complex Than Risk Aversion
Although risk aversion on the part of physicians contributes to lower rates of PCI in reporting states, “it’s a little more complex than that,” Sorin J. Brener, MD, of New York Methodist Hospital (New York, NY), told TCTMD in a telephone interview. It may also relate to more judicious application of treatment, he suggested, because in some cases intervention is futile.
“If you think that [in a given case] primary PCI is an absolute must, public reporting means nothing,” he said. “But if you’re on the fence [about the benefit of PCI], then public reporting could push you toward not doing it.”
In addition, administrative data have major limitations in terms of risk adjustment, he said, and that can lead to “a tremendous amount of gaming of the system in reporting states because everybody wants to look good.”
Dr. Brener applauded efforts to modify reporting so that cath labs are not penalized for high-risk cases. He noted that in New York, for example, if a patient with cardiac arrest in the field undergoes PCI and is hemodynamically stable but then dies of ischemic brain injury, the death is not counted.
But most states are unlikely to adopt public reporting anytime soon, he said, because of the inadequacies of risk stratification and how that might affect decision making. “The only cure for gaming of the system is independent auditing,” Dr. Brener said, although he added that auditing is not feasible from a resource standpoint.
“In theory, public reporting is a very good idea,” Dr. Brener concluded, “but in practice it is complex, because we cannot capture the things that really matter. It remains a work in progress.”
1. Waldo SW, McCabe JM, O’Brien C, et al. Association between public reporting of outcomes with procedural management and mortality for patients with acute myocardial infarction. J Am Coll Cardiol. 2015;65:1119-1126.
2. Moscucci M. Public reporting of percutaneous coronary intervention outcomes: harm or benefit [editorial]? J Am Coll Cardiol. 2015;65:1127-1129.
- Dr. Yeh reports being supported by a career development award from the National Heart, Lung, and Blood Institute.
- Dr. Moscucci reports receiving book royalties from Lippincott Williams & Wilkins.
- Dr. Brener reports no relevant conflicts of interest.