PCI Rates for Acute MI Lower in States with Public Reporting

In states that publicly report percutaneous coronary intervention (PCI) outcomes, rates of PCI for acute myocardial infarction (MI) are lower than in states without public reporting, according to a large observational study published in the October 10, 2012, issue of the Journal of the American Medical Association. However, reporting status had no effect on short-term mortality, according to the findings.

Karen E. Joynt, MD, MPH, of Brigham and Women’s Hospital (Boston, MA), and colleagues, retrospectively analyzed Medicare data of patients treated for acute MI from 2002 to 2010 in 3 states with public reporting of PCI outcomes (n = 49,660; Massachusetts, Pennsylvania, and New York) and in 7 states in the same region without such reporting (n = 48,142; Maine, Vermont, New Hampshire, Connecticut, Rhode Island, Maryland, and Delaware).

Less PCI With Public Reporting

In 2010, patients hospitalized in public reporting states were less likely to receive PCI than those in nonreporting states. The difference was most pronounced in the STEMI (n = 6,708) and cardiogenic shock or cardiac arrest subgroups (n = 2,194; P < 0.001 for interaction), but was not significant in the NSTEMI cohort (table 1).

Table 1. Rates of PCI for AMI: Reporting vs. Nonreporting States

Similar patterns were seen for the likelihood of catheterization. On the other hand, after adjustment there was no difference in rates of CABG between reporting and nonreporting states (adjusted OR 1.01; 95% CI 0.80-1.26; P = 0.95).

Overall, 30-day mortality rates did not differ between reporting and nonreporting states (adjusted OR 1.08; 95% CI 0.96-1.20; P = 0.20). However, mortality was higher in the STEMI subgroup in reporting states (adjusted OR 1.35; 95% CI 1.10-1.66; P = 0.004).

Since Massachusetts was the only state to initiate public reporting in the recent era, the investigators analyzed these patients separately. Before the state began reporting in 2005, Massachusetts patients had rates of PCI similar to those in other nonreporting states. However, the pattern began to diverge in 2005, and by 2010 Massachusetts patients were less likely to receive PCI than their counterparts in nonreporting states (adjusted OR 0.81; 95% CI 0.47-1.38; P = 0.03).

Tough Decisions

A self-declared proponent of public reporting, Dr. Joynt told TCTMD in a telephone interview that treating acute MI patients puts clinicians “in a tough spot” because “it really requires the whole hospital to take care of that person, . . . but it’s the one interventionalist who has to make the up-front decision about what we’re going to do with this patient. Sometimes you can hurt people more by doing these procedures and sometimes we overuse procedures.” 

Based on this perspective, she said the fact that the PCI rates decreased after the introduction of public reporting may be positive. “It probably made people look a little more closely at what they were doing,” she suggested, adding that it in some cases not performing procedures was probably “not the right decision,” as well.

Going forward, states should keep better track of PCI appropriateness, she suggested. Also, it might be wise to consider tracking the rate of PCI for acute MI patients and the rate at which hospitals accept very sick patients. In addition, Dr. Joynt added, perhaps the risk adjustment model should be amended so that hospitals receive “credit” for treating such patients.

“There’s just got to be some place that will take on the sickest of the sick, and we really don’t want to make [it] harder for a hospital to want to be the hospital of last resort,” Dr. Joynt said.

‘Unintended Consequences’ Confirmed

“The major value of this study is that through access to Medicare data, the authors have been able to include all of the patients presenting with acute MI, thus including in the denominator all patients eligible for PCI and thereby assessing the association between public reporting and 30-day mortality,” Mauro Moscucci, MD, MBA, of the University of Miami Miller School of Medicine (Miami, FL), writes in an accompanying editorial.

The similar CABG rates are “reassuring,” he continued, because they “suggest that the lower PCI rates in public reporting states were not attributable to a switch from PCI to CABG, and they do not support an inappropriate higher use of PCI as an alternative to CABG in nonreporting states.”

Because PCI provides great benefit in acute MI patients with cardiogenic shock, the potential for denial of care for these patients could have a “chilling effect,” Dr. Moscucci writes, though he adds that the similar 30-day mortality rates “mitigate this concern.”

He suggests several possible explanations for the similar mortality rates between patients with and without PCI. These include assessment of the futility of intervention in patients with multiple comorbidities, improved care resulting in better outcomes that offset the adverse effect of not performing PCI in high-risk patients, and the practice of ‘upcoding,’ or delineating complications to their fullest extent, in states with public reporting.

The study “confirms the possible unintended consequences of public reporting, while highlighting its association with (or lack of association with) clinical outcomes,” Dr. Moscucci concludes. “In addition, these findings may help spearhead a new focus on procedures that, while perceived [as] appropriate based on current use criteria, might not result in added benefit in selected patients.”

Upcoding Muddies the Waters

Morton J. Kern, MD, of the University of California, Irvine (Irvine, CA), told TCTMD in a telephone interview that the study confirms what people already think—that hospitals are very conservative about who can receive PCI in an effort to keep their mortality rates low. He said he observes this practice at his own institution. Public reporting “influences how the physicians treat the patients, and I think that’s what the bottom line for this study was,” he said.

Still, Dr. Kern said, “just because you report doesn’t mean you get better outcomes, and just because you don’t report doesn’t mean you get worse outcomes. Probably the issue of public reporting needs to be refocused on the type of procedure for the type of patient and then compare apples to apples.”

In a telephone interview with TCTMD, Jeffrey W. Moses, MD, of Columbia University Medical Center/Weill Cornell Medical Center (New York, NY), made a similar point.

“You can’t compare acute STEMI mortality with PCI to [that of] NSTEMI with PCI. The time courses and magnitude are different,” he said. “So there’s a little bit of a specious argument here in that the outcomes are different.”

In addition, Dr. Moses said, the issue of upcoding is “critically important.” While the practice is “not dishonest,” he continued, it emphasizes capturing more comorbidities to help the risk-adjustment models “because obviously you want to present the sickest population you can to make your outcomes relatively better.” That is much less likely to occur in a nonreporting state, he added.

Dr. Joynt agreed that there are incentives for hospitals in public reporting states to upcode, but said the issue is difficult to study. However, public reporting poses “a little bit of a catch 22, because we want hospitals to code aggressively enough that we capture differences in sickness but we don’t want them to overcode because then we are taking away their ability to assess who’s sick and who’s not.”

 

---

Sources:
1. Joynt KE, Blumenthal DM, Orav EJ, et al. Association of public reporting for percutaneous coronary intervention with utilization and outcomes among Medicare beneficiaries with acute myocardial infarction. JAMA. 2012;308:1460-1468.

2. Moscucci M. Public reporting of PCI outcomes and quality of care: One step forward and new questions raised. JAMA. 2012;308:1478-1479.

 

 

Related Stories:

• New York State Public Reporting on PCI Mortality Does Little to Change Behavior
• Public Reporting of State PCI Results May Backfire