Public Reporting of PCI Mortality Tied to Fewer High-Risk Cases But Better Outcomes

 

Public reporting of PCI mortality rates results in better procedure quality and lower risk of adverse events, including mortality, than does collaborative quality improvement. Yet the practice also translates into fewer high-risk patients undergoing PCI, according to a registry study comparing 2 states that employ the different strategies published online September 15, 2015, ahead of print in the American Heart Journal.

Public Reporting of PCI Mortality Tied to Fewer High-Risk Cases But Better Outcomes

Drawing on the NCDR CathPCI Registry, researchers led by Thomas F. Boyden, MD, MS, of Spectrum Health Medical Group (Grand Rapids, MI), analyzed data on patients who underwent PCI between January 2011 and September 2012 at participating hospitals in New York, which employs public reporting, and Michigan, which does not. The state of Michigan utilizes a collaborative quality improvement system that provides peer-reviewed analysis and promotes accountability through sharing of information to institutions and providers.

Compared with Michigan patients, those in New York were less likely to be female or white, or to have a history of MI, congestive heart failure, hypertension, dyslipidemia, cerebrovascular disease, PAD, and chronic lung disease. New York patients also were less likely to undergo PCI for STEMI or NSTEMI and had lower rates of cardiogenic shock and cardiac arrest at the time of PCI. The baseline differences resulted in a lower percentage of patients with extremely high predicted mortality risk (> 20%) in New York vs Michigan.

Measures of PCI quality varied between the states. For example, a higher proportion of New York patients had PCI classified as appropriate. However, PCI of uncertain appropriateness was more common in New York, while the proportions of inappropriate procedures were similar between the states.

In addition, New York patients were more likely to have markers of myonecrosis assessed but less likely to undergo pre-PCI evaluation of renal function. At discharge, New York hospitals far less often referred patients to cardiac rehabilitation than did Michigan hospitals (OR 0.15; 95% CI 0.14-0.16), although similar proportions of patients in the 2 states were discharged on optimal medical therapy.

New York Patients Fare Better

Propensity matching resulted in 2 cohorts of 40,916 patients each. Risks of vascular complications, access-site bleeding, post-PCI transfusion, and referral for urgent, emergent, or salvage CABG were lower in New York than in Michigan. Moreover, New York patients were less likely to die in the hospital (table 1).

Table 1. Propensity-Matched Analysis of Procedural/In-Hospital Events: New York vs Michigan


Sensitivity analyses to assess the impact of different estimated levels of unmeasured confounders confirmed the robustness of the mortality findings.

The authors say their observations “confirm prior studies showing that extremely high risk patients are less likely to undergo PCI in states with [public reporting], which may be related to risk avoidance.”

Furthermore, observed Jeffrey W. Moses, MD, of Columbia University Medical Center (New York, NY), in an email with TCTMD, “If an operator has lower mortality in a low-risk population but avoids sicker people for whom the number needed to treat is far lower, mortality for all comers—those treated and untreated—goes up.”

Avoiding Unintended Consequences

However, there are ways to mitigate such “unintended consequences” of public reporting, Dr. Boyden and colleagues say. They cite the “compassionate use criteria” incorporated into the Massachusetts PCI reporting program, which takes into account indicators of extreme risk that may make PCI futile and thus refines risk adjustment. Another strategy is to expand public reporting to “highlight other important related performance measures which may be neglected by too narrow a focus on mortality alone,” the investigators say.  

Considering the propensity-matched results, the lower mortality rate in New York compared with Michigan is unlikely to be due to patient mix alone, the authors contend. In fact, the comparable or better quality of care and lower adverse events rate suggest that public reporting “may be a strategy that still warrants consideration for improving outcomes for PCI,” they say, adding that further study should help understand how cath lab directors respond to public reporting so that useful strategies can be extended to more hospitals.

 “As the [American College of Cardiology] moves towards publicly reporting mortality statistics for PCI, and as states and the federal government increase the use of [public reporting] for more procedures and conditions, it will be important to closely follow use patterns to ensure that PCI is being appropriately offered and performed,” the investigators stress.

The current results should be interpreted with certain caveats, the authors acknowledge. For example, the benefits associated with PCI may be affected by selection bias, as not all hospitals and health systems in New York provide data to the NCDR, while all hospitals in Michigan do, they say. Thus, participating hospitals in New York may represent institutions that have more resources or provide better care than nonparticipating hospitals. In addition, the more rigorous auditing process employed by the Michigan collaborative quality improvement program may be better than the self-reporting strategy used by the CathPCI Registry at accurately identifying adverse events and clinical outcomes, Dr. Boyden and colleagues add.

In the end, the authors say, the best approach to protecting access to care while still optimizing patient outcomes may be to combine public reporting and collaborative quality improvement.


Source: 
Boyden TF, Joynt KE, McCoy L, et al. Collaborative quality improvement versus public reporting for percutaneous coronary intervention: a comparison of PCI in New York versus Michigan. Am Heart J. 2015;Epub ahead of print.

Related Stories:

 

 

Disclosures
  • The study was supported by the American College of Cardiology Foundation’s National Cardiovascular Data Registry.
  • The paper contains no statement regarding potential conflicts of interest for Dr. Boyden.
  • Dr. Moses reports no relevant conflicts of interest.

We Recommend

Comments