Half of U.S. valve hospitals misclassified as low- or high-performing

Nearly half of all valve hospitals in the U.S. are misclassified as either low- or high-performing, according to research published in JAMA Cardiology—but that problem might have a simple fix.

Identifying high-performing, positive-outcome surgical valve centers in the U.S. has been challenging for some time, first author Rohan Khera, MD, and colleagues wrote in the study, because payers and policymakers have for years graded hospitals based on surgical volume. These endorsements largely ignore surgical outcome statistics, favoring volume as the metric of choice—even the Centers for Medicare and Medicaid Services requires hospitals to exceed specific minimum annual surgical volumes if they want to be reimbursed for transcatheter aortic valve replacement (TAVR) and MitraClip costs.

Volume can be a helpful measuring tool, Khera and co-authors said, due to its simple nature and the fact that it’s easy data to collect, interpret and monitor.

“However, there are several downsides to this approach,” they wrote. “Most important, using solely volume as a surrogate may be less relevant when carefully collected and risk-adjusted outcomes are directly available.”

In their analysis, Khera et al. used an all-payer data set of 682 American hospitals that performed a total of 70,295 surgical aortic valve replacement (SAVR) surgeries, 19,913 mitral valve (MV) replacements and 17,037 MV repair procedures between 2007 and 2011. The researchers used in-hospital risk-standardized mortality rate (RSMR) numbers to calculate more accurate hospital grades in addition to the number of patients seen per institution.

The surgical centers had a mean annual volume of 43 SAVR surgeries, 13 MV replacements and nine  MV repair procedures, according to the research. Khera and colleagues found that of 225 SAVR hospitals in the study’s highest-volume tertile, 34.7 percent and 36 percent were in the highest-RSMR tertile for SAVR and coronary artery bypass grafting (CABG), respectively, while 21.5 percent and 17.5 percent of SAVR hospitals in the lowest-volume tertile were in the lowest respective RSMR tertile. The other procedures recorded similar results; 36.8 percent and 43.5 percent of hospitals in the highest volume tertile for MV replacement and repair were in the corresponding highest RSMR tertile.

There was a limited correlation between SAVR and MV procedural outcomes in hospitals, the research team reported. There did appear to be a strong link between SAVR and MV procedure volumes at individual centers, but there was little correlation between RSMRs for aortic and MV surgical procedures within the same hospital.

If surgical center quality rankings continue to rely on volume stats alone, Khera and co-authors wrote, 44.7 percent of all valve hospitals would be misclassified as either high- or low-performing. In the study, the authors found 305 of 682 institutions were misclassified.

“Our findings have potentially important policy implications,” the researchers wrote. “Hospital variation in surgical outcomes has been central to advocating regionalization of high-risk procedures to well-performing centers. However, our findings suggest that an approach relying heavily on institutional surgical volume might not adequately categorize low-performing or high-performing institutions for aortic or MV procedures in the current era.”