NY State to Exclude Emergency PCI Deaths From Physician-Level Public Reports
“This is a great standard that others should follow,” Ajay Kirtane says, although the impact on risk-averse behavior is unknown.
The mortality outcomes of emergency PCI cases will not be included on future physician-level reports in New York State, according to a policy change announced last week.
In a letter dated December 28, 2022, officials from the state’s health department said data collection would remain the same but beginning with the 2018-2021 outcomes report set to be published at the end of 2023, physician-level results would no longer include these high-risk cases, defined as patients with acute MI or cardiac arrest in the 24 hours leading up to PCI or those in nonrefractory cardiogenic shock at the time of the procedure. Public reports in New York already exclude cases involving refractory cardiogenic shock or pre-PCI anoxic brain injury (since 2006 and 2010, respectively).
The outcomes of emergency PCI still will be included on the hospital-level reports.
The policy change, based on a recommendation from the New York State Cardiac Advisory Committee, “is the result of careful deliberation weighing the importance of transparency in outcome assessment against the potential for increased access to PCI for critically ill patients who may benefit from the procedure,” the letter reads.
We need to think about case complexity and technical expertise, case selection, and other outcomes other than mortality, and all of those things are what quality assessment ought to be. Not just a body count. Lloyd Klein
The impact of the policy will be assessed annually over the next 4 years to watch for unintended consequences.
Since the advent of public reporting of PCI outcomes, which varies by state, there has been debate about how to make sure it is done fairly while providing a true reflection of PCI quality. The major concern with reporting mortality outcomes, specifically, is that it will lead to operators and hospitals avoiding the treatment of the highest-risk patients; a previous survey showed that most interventional cardiologists believe public reporting of PCI outcomes leads to avoidance of high-risk patients and upcoding of patient comorbidities.
Interventional cardiologists interviewed by TCTMD say the change announced in New York State is a step toward making the reports fairer, although whether it will reduce risk-averse behavior among operators in the state will not be known for several years.
“This is a great standard that others should follow,” Ajay Kirtane, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), told TCTMD. Not all states require public reporting of PCI outcomes, he noted, but for those that do, it’s important to make “intelligent changes” like excluding cases involving shock, anoxic brain injury, and emergency PCI on the operator level, as done here.
Lloyd Klein, MD (University of California, San Francisco), also applauded the move, saying “the change is long overdue.” Klein is the lead author of a 2016 statement on public reporting from the Society for Cardiovascular Angiography and Interventions (SCAI).
“If we’re going to use 30-day mortality after PCI as an indicator of operator and programmatic quality, then acute cases anticipated to have a high intrinsic risk have to be excluded because when they’re included, the resulting event rate is higher inherently, which reflects only case acuity and not quality of service,” Klein said. Factors like the size of the infarct, the presence of shock, the severity and extent of underlying CAD, premorbid conditions, and delays in the patient getting to the hospital all affect procedure-related mortality but can’t be controlled by the operator or hospital, he added.
By including emergency cases in public reports, “all you’re doing is including, knowingly, as a quality indicator, deaths which aren’t really related to quality,” Klein said, adding, “That leads to risk avoidance, not better quality.”
Impact of the Change
Klein said the exclusion of emergency cases from the physician-level reports is a reasonable way to improve the public reporting process. “What will happen is that operators will be happy that they’re not being unfairly graded,” he said.
However, because mortality associated with these high-risk cases will remain on the hospital-level reports, administrators may still encourage more-conservative case selection to keep their rates down. “Given the circumstances, I think they’re doing it the best way that they could come up with, and I can’t think of a better way to do it, but there are still going to be negative consequences to the way they’re doing it.”
SCAI President Sunil Rao, MD (NYU Langone Medical Center, New York, NY), who recently moved to New York from North Carolina, a state that does not publicly report PCI outcomes, said he doesn’t expect the policy change, though “potentially a step in the right direction,” to change physician behavior too much.
“Physicians are always trying to do what’s best for the patients. I don’t think that they’re letting the public reporting drive their decision-making,” Rao told TCTMD, although he acknowledged that operators do pay attention to these public reports when they come out. He added that there’s a possibility that including mortality data on the hospital-level but not the physician-level reports could exacerbate the tension between physicians and hospital systems.
“On a personal level, I’d like to say that I don’t think about those things when I do cases. I do them for the clinical indications,” Kirtane said. “But if you look across populations of practitioners . . . you basically do see that there’s risk avoidance happening, and that’s not in anyone’s interest.”
In that context, this iterative change in New York’s policy ultimately is going to be a good thing for patients, he predicted. Plus, he added, “I think that there’s no question that at the very least it will give operators relief.”
Looking Beyond Mortality for Quality Assessment
Taking a broader view, Klein said the best way to report PCI quality might be to look at outcomes other than mortality, including quality of life, persistence of angina, rehospitalization, revascularization, and recurrent MI. He, Rao, and H. Vernon Anderson, MD (University of Texas Health Science Center at Houston), previously released a proposed framework for assessing PCI quality in JAMA Cardiology.
Changes to public reporting policies like those being taken in New York are “just a way to kind of clean up after the parade,” Klein said. “It’s not the best way to do it. We need to think about case complexity and technical expertise, case selection, and other outcomes other than mortality, and all of those things are what quality assessment ought to be. Not just a body count.”
I think that there’s no question that at the very least it will give operators relief. Ajay Kirtane
For Rao, the ideal quality-improvement process would involve a random but systematic review of angiograms, something that is being implemented within the Veterans Health Administration. That’s what’s needed to move the needle, he said, although he acknowledged that it would be difficult to scale up that type of process nationwide.
Kirtane agreed that mortality is not the best metric to reflect PCI quality. Disease-specific registries that also capture patients who do not undergo an intervention will play an important role, he said, in assessing how well operators and hospitals are doing when it comes to the care of all patients, and not just those treated with an invasive procedure.
New York State has both a PCI registry and an MI registry. Looking at information from both will allow centers to get a handle on how well they’re doing and on whether there is risk avoidance occurring, Kirtane said. “My feeling is that the progression towards these disease-based registries is going to be a good way of addressing PCI quality.”
Kirtane added that multiple research groups have been digging into the best ways to assess PCI quality over the past several years, and he credited those investigators for influencing policy changes like the one in New York. “I do think that it was the cumulative weight of those types of publications that has led to the awareness of these types of issues” and resulted in change, he said.