Public Reporting: Policy Changes in New York May Have Boosted PCI Rates of MI Complicated by Cardiogenic Shock
Highlighting the potential shortcomings of publicly releasing PCI outcomes, two new studies show that exclusion of patients with acute MI complicated by cardiogenic shock from New York’s reporting requirements was associated with an uptick in invasive management and a decline in in-hospital mortality for these high-risk patients.
There is accumulating evidence that public reporting of PCI mortality data—which is meant to improve quality of care, increase transparency, and allow patients to make more informed choices—can have unintended effects, including a reluctance on the part of physicians to treat the highest-risk patients, including those with cardiogenic shock, “for fear of being singled out either institutionally or individually,” according to Frederic Resnic, MD (Lahey Hospital & Medical Center, Burlington, MA), who was not involved in either analysis.
“If in certain states—New York, Massachusetts—the physicians are incorporating something other than the patient’s condition into their decision as to whether to proceed or not, it is creating an ethical dilemma for the physicians, even if they don’t realize it,” he told TCTMD. “I, as an interventionalist in Massachusetts, try very hard to not allow that to affect my decision-making, but I don’t know that it doesn’t affect my decision-making.”
He noted that the impact of the policy change in New York on treatment rates is blurred by the fact that gains were seen over time in all states, even those that do not require the release of PCI mortality data, but said the current findings support that idea that making public reporting less punitive increases treatment rates.
“With guideline modifications, with quality improvement efforts, in the absence of the pressure of public reporting, I think there’s an ever increasing interest and willingness on the part of interventionalists to take a higher and higher level of instability to the lab to try, knowing that this is the best—even if still a small—hope of survival for a patient,” Resnic told TCTMD. “I just think it’s much easier for nonreporting states to have the buy-in of physicians to accept that logic.”
Both studies were published online July 27, 2016, ahead of print in JAMA Cardiology.
Immediate Increase in PCI Rates
New York State implemented public reporting of operator-specific mortality for CABG and PCI in the late 1980s. Based on studies showing potential negative effects of reporting outcomes of high-risk patients, the state excluded patients with cardiogenic shock from reporting requirements, first on a trial basis in 2006 and 2007 and then permanently starting in 2008.
In the first analysis evaluating the impact of that change, Sripal Bangalore, MD (NYU Langone Medical Center, New York, NY), and colleagues identified patients with acute MI complicated by cardiogenic shock who were treated between 2002 and 2011 and included in the National Inpatient Sample. The investigators used propensity matching to compare 1,063 patients from New York with an equal number from Michigan, a nonreporting state.
The study included three time periods:
- 2002 to 2005, when cardiogenic shock was still included in publicly reported outcomes
- 2006 to 2007, when it was excluded on a trial basis
- 2008 and later, when cardiogenic shock was excluded permanently
From the beginning of the study to the end, the proportion of patients with cardiogenic shock who underwent PCI significantly increased from 31.1% to 40.7%, with similar upturns in any invasive management and any revascularization.
Rates of PCI, invasive management, and revascularization also increased in Michigan, however, such that rates of all types of treatment remained lower in New York throughout the study period. The results were generally similar in analyses comparing New York with two other nonreporting states, New Jersey and California.
In-hospital mortality declined in both New York and Michigan over time, with no differences between states at any time point.
Results from the second analysis—reported by James McCabe, MD (University of Washington, Seattle), and colleagues—were generally consistent with the findings of Bangalore et al. McCabe and colleagues used several statewide hospitalization databases to compare patients treated in New York between 2002 and 2012 with those treated in three nonreporting states (Michigan, New Jersey, and California) and Massachusetts, which does not exclude patients with cardiogenic shock but has a risk-adjustment scheme to account for compassionate use.
The study included 45,977 patients (24.6% from New York). Overall, the rate of PCI increased from 44.9% before the policy change to 49.2% after exclusion of patients with cardiogenic shock, with a greater increase seen in New York compared with the other states combined (P < 0.001 for interaction). As in the analysis from Bangalore et al, the rate lagged behind in New York.
Although the rate of coronary angiography did not change over time across all states, a greater percentage of patients in New York underwent testing after the reporting policy change (67.9% vs 63.6%; P < 0.001).
The in-hospital mortality rate fell overall from 44.7% to 37.9%, but there was a greater decline seen in New York than elsewhere (P < 0.001 for interaction).
The study “suggests that the censoring of adjudicated, extreme-risk cases may have been effective at facilitating guideline-directed revascularization and improving outcomes,” McCabe and colleagues write. “Further research is required to better understand how to balance the desires for healthcare transparency with a system that encourages appropriate care for the highest-risk patients.”
Why Is New York Still Playing Catch-up?
To TCTMD, Bangalore acknowledged that the improvements observed in New York could be related to factors other than the change in policy, considering the fact that similar trends were observed in nonreporting states. Over time, he noted, there have been modifications in the guidelines to increase the emphasis on revascularization in the setting of cardiogenic shock—based on the results of the SHOCK trial—and better dissemination of that information, both of which could have contributed to greater use of invasive management.
Bangalore said rates of treatment in New York likely have not caught up to those in other states yet because of the strict definition of cardiogenic shock and recording requirements mandated by the state. If all of the criteria are not met or are not documented correctly—which can occur in such an emergent situation—the state can refuse to exclude a patient from the public reporting requirements, he said.
Those challenges inevitably lead to the outcomes of some patients with cardiogenic shock being reported, which might explain the lingering reluctance of some New York physicians to take them to the cath lab, he said.
Bangalore suggested that easing the stringency of the definition and perhaps starting to track processes of care rather than patient outcomes could resolve the issue. “It needs much more thought and discussion as to what to do with this, and creating more flexibility in the definition of shock and also taking into account various other aspects might be helpful,” he said.
In an accompanying editorial, Ajay Kirtane, MD (Columbia University Medical Center/NewYork-Presbyterian Hospital, New York, NY), and colleagues say the persistently lagging treatment rates among patients with cardiogenic shock in New York “remains sobering.”
“Even with the recognition that risk adjustment was not enough to mitigate risk aversion, which led to the New York shock exclusion in 2006, this policy change—although perhaps helpful—still did not do enough to encourage physicians to care for the sickest patients who had the most to lose,” they argue.
“These findings should therefore give pause for those who advocate more widespread and indiscriminate public reporting of PCI mortality without careful consideration of the consequences,” they continue. “It may simply be time to recognize that mortality following PCI is the wrong metric with which to arbitrate its quality across heterogeneous patient scenarios, despite attempts to separate these scenarios into discrete entities, such as shock and nonshock.”
Note: Two of the editorialists are faculty members of the Cardiovascular Research Foundation, the publisher of TCTMD.
- Bangalore S, Guo Y, Xu J, et al. Rates of invasive management of cardiogenic shock in New York before and after exclusion from public reporting. JAMA Cardiol. 2016;Epub ahead of print.
- McCabe JM, Waldo SW, Kennedy KF, Yeh RW. Treatment and outcomes of acute myocardial infarction complicated by shock after public reporting policy changes in New York. JAMA Cardiol. 2016;Epub ahead of print.
- Kirtane AJ, Nallamothu BK, Moses JW. The complicated calculus of publicly reporting mortality after percutaneous coronary intervention. JAMA Cardiol. 2016;Epub ahead of print.
- The funding for the study by McCabe et al came from Massachusetts General Hospital’s Hassenfield Scholar Award and the Richard and Susan Smith Center for Outcomes Research.
- Bangalore and McCabe report no relevant conflicts of interest.
- Resnic reports serving as the senior medical advisor for interventional cardiology to the Massachusetts Data Analysis Center (Mass-DAC).
- Kirtane reports receiving institutional research grants (to Columbia University) from Medtronic Cardiovascular, Boston Scientific, Abiomed, Abbott Vascular, St. Jude Medical, Eli Lilly, and GlaxoSmithKline.