PCI process measures may not correlate with 30-day mortality

CHICAGO — Hospital performance measures for PCI processes failed to demonstrate an association with 30-day risk-standardized mortality rates, according to report findings presented at the American Heart Association Scientific Sessions.

Philip W. Chui, MD, and colleagues conducted a study to describe current performance on PCI process and outcomes measures; to examine whether hospitals that perform well on process measures also perform well on 30-day risk-standardized mortality rates; and to calculate the explained variance and correlation between hospital performance on PCI process measures and 30-day risk-standardized mortality rates.

Chui noted a dramatic increase in efforts to measure and report quality of care delivered to PCI patients. The National Quality Forum has endorsed a set of PCI process measures and outcome measures such as 30-day risk-standardized mortality rates. He said process and outcome measures represent important assays of hospital quality.

However, “no study has examined the relationship between hospital performance on PCI process measures and 30-day risk-standardized mortality rates,” said Chui, from the department of medicine at University of California Irvine School of Medicine. Further, “understanding this association can guide future improvements in hospital performance on PCI measures.”

Chui and colleagues analyzed data from the National Cardiovascular Data Registry CathPCI Registry on 1,268,860 PCIs performed from 2010 to 2011 in patients at 1,331 facilities. Patients were stratified with regard to the presence or absence of STEMI or cardiogenic shock.

The following performance measures were used: aspirin, thienopyridines and statins at discharge; proportion of door-to-balloon time within 90 minutes; referral to cardiac rehabilitation; and composite measure of performance on aforementioned process measures.

Chui reported that performance in many of the process measures was close to 100%. Although the researchers observed a correlation between process metrics and discharge medications, they found no association between process measures and 30-day risk-standardized mortality rates.

“Hospitals are doing quite well with discharge medications, almost near 100%,” Chui said. He noted a large range in hospital use of cardiac rehabilitation at discharge.

In terms of mortality rates among patients with or without STEMI or cardiogenic shock, Chui highlighted modest but “perhaps clinically significant variation in both groups.”

The strongest correlation observed was between referral to cardiac rehabilitation and the overall composite of measures of performance, according to Chui.

Other findings indicated that performance at the hospital level in terms of current National Quality Forum-endorsed PCI process measures explain only a small proportion of variation in risk-standardized mortality rates.

The results suggest that processes related to PCI and outcome measures fall in different quality domains.

The analysis highlights the complementary role of process and outcome measures in benchmarking hospital quality, Chui said. He recommended that the clinical community develop process measures that lead to outcomes, particularly in light of new American College of Cardiology/American Heart Association/Society for Cardiovascular Angiography and Interventions guidelines.

For more information:

Chui PW. Abstract #15729. Presented at: American Heart Association Scientific Sessions; Nov. 15-19, 2014; Chicago.

Disclosure: The study was funded by the American College of Cardiology Foundation’s National Cardiovascular Data Registry. Chui reports no relevant financial disclosures.