Elective PCI Appropriateness Not Linked to Outcomes in the VA System

The appropriateness of elective PCI based on published criteria is not related to the publicly reported outcomes of those procedures, at least within the US Veterans Affairs system, researchers found.

After accounting for potential confounders, the risk of 90-day readmission or death was no different when comparing “appropriate” and “rarely appropriate” procedures (HR 0.85; 95% CI 0.60-1.21), Vinay Kini, MD (University of Colorado Anschutz Medical Campus, Aurora), and colleagues report.

Kini told TCTMD he was not surprised to see the lack of correlation between appropriateness and outcomes because “they do measure two fundamentally different aspects of PCI quality. Appropriateness has to do more with patient selection and whether or not patients are going to benefit from the procedure. It’s more patient-centered. Whereas outcomes really kind of focus on the procedure itself—so making sure that complications are minimal and that patients do well after the procedure.”

This could have implications for the assessment of PCI quality, Kini said. He brought up the debate around the potential for “risk-averse behavior” among operators stemming from the public reporting of PCI outcomes—such that higher-risk patients are not taken to the cath lab to avoid a higher mortality rate. “Most of the patients who are at higher risk of readmission and mortality are likely to have highly appropriate indications for PCI, and so it’s possible that by including appropriateness in public reports of PCI quality that that could mitigate some of the negative consequences of public reporting of outcomes alone,” he said.

The findings were published as a research letter in Circulation: Cardiovascular Quality and Outcomes and presented recently as a poster at the virtual American Heart Association 2020 Scientific Sessions.

Two Aspects of PCI Quality

For the study, the investigators turned to the VA Clinical Assessment, Reporting, and Tracking (CART) Program. Within the VA system, there are no financial incentives to perform PCI, a strength of using this database to study the correlation between appropriateness and outcomes, Kini said.

The analysis focused on 2,561 patients (mean age 66; 99% men; 86% white) who underwent elective PCI for stable CAD across 59 VA facilities between 2013 and 2015. Because of the time period, appropriateness was based on the 2012 US criteria rather than the more recent 2017 criteria. The procedures were classified as “appropriate” in 29.6% of cases, “may be appropriate” in 60.0%, and “rarely appropriate” in 10.4%. Ninety-day rates of mortality across those three categories were 0.7%, 0.4%, and zero, respectively, with corresponding readmission rates of 14.9%, 15.3%, and 16.4%, respectively.

On multivariate adjustment, there were no significant differences in 90-day outcomes across appropriateness groups, a finding consistent when looking at facility-level data. That suggests “that each metric alone does not fully characterize PCI quality,” Kini et al write.

“Since PCI appropriateness is based on the potential for patient benefit,” they say, “additional studies should test whether concurrent reporting of appropriateness could 1) provide valuable information to insurance payers and patients and 2) empower providers to perform PCI in patients at high risk of death or readmission.”

A Holistic View of Quality

Commenting for TCTMD, Karen Joynt Maddox, MD (Washington University School of Medicine in St. Louis, MO), noted that appropriateness and outcomes are really at opposing ends of a spectrum of quality, with appropriateness focused more on reducing overuse in lower-risk patients and outcomes driven primarily by higher-risk patients. This study suggests, then, “that these are two separate realms of quality and that we probably need to be thinking about all of these things as we try to optimize the care that centers deliver.”

Asked about the possible approach of detailing both appropriateness and clinical outcomes on a single public report, Joynt Maddox said that “would give a fuller picture. I have argued for many years that only looking at the outcomes among people that get PCI is not a good idea.”

She alluded to risk-averse behavior among operators that keeps high-risk patients out of cath labs. “My personal opinion would be that the quality of care that an institution provides is not only about those people who get the procedure but it’s also about the people who come in who should have gotten the procedure but didn’t,” she said.

The quality of care centers are providing can be distilled down to three questions, according to Joynt Maddox: “Do they get the care to the people that need it? Do they not do things when people don’t need it? And do they keep people living longer?”

Whether inclusion of appropriateness on public reports would fully mitigate risk aversion among operators is not clear, she said, noting that “the fear, the concern, with the high-risk patients is pretty strong. And I’m not sure that appropriateness would really offset that completely to be honest.”

The overarching message, Joynt Maddox said, is that the field should be thinking more holistically about the measures used to assess PCI quality and about how that information can best be used to influence positive change.

“I think most clinicians want to be better, most places want to do better, but most of the quality measures haven’t been built such that they really engender that,” she said. “So I think this paper’s really interesting because it sort of expands the thinking about why it’s important to have multiple measures. Because they don’t necessarily run together. Measuring one thing won’t tell you all you need to know.”