In one analysis, the mortality rates between highest- and lowest-compliance hospitals were "statistically indistinguishable." In general, compliance on the SCIP measures only accounted for around 3 percent of the variance in mortality, with hospitals not reporting compliance showing mortality rates similar to those of the most compliant hospitals.

Nonetheless, stay length was affected, with patients at highest-compliance hospitals around 12 percent less likely to endure an extended stay, relative to middle-compliance hospitals, the authors said. No difference for stay length was found between middle- and low-compliance institutions.


"These findings, if true, call into serious question the increased time, labor and effort currently expended by hospitals and surgeons across the United States to comply with the SCIP program process measures," Dr. Charles D. Mabry, a surgeon with University of Arkansas for Medical Sciences in Little Rock, wrote in an accompanying editorial.

Still, Mabry noted the study wasn't an open-and-shut case. It's possible, he said, that hospitals that do a bad job of SCIP compliance might also do a similarly bad job of correctly coding complication rates, so they could give the illusion of a more favorable complication rate.

The authors also noted their own limitations. The researchers compared mortality rates of Medicare patients against process measures reported for all eligible admissions. While the authors said Medicare surgical mortality rates are generally correlated with overall surgical mortality rates, it's possible there could be a discrepancy because of insufficient sample size (fewer high-risk procedures are performed at the smaller-volume, low-compliance hospitals).

Still, the authors suggest the research ought to spur investigators to find better measures for patients. They note that organizations like The Leapfrog Group, a coalition of health care purchasers, rely on a composite score of mortality rate and volume for judging hospital quality.

"Despite the limitations of this study, it is an important one that begs the question of whether CMS and the insurance industry should focus on process measures or more on outcomes measures," Mabry said.

Also, some defenders of the SCIP measures note that even if it's not a perfect predictor of hospital quality, the project's goal was more to stimulate the national discussion on important quality measures in order to improve patient safety. For instance, the composite measure of surgical antibiotic prophylaxis for Medicare patients was only around 34 percent in 2002, observed Dale W. Bratzler, with the Oklahoma Foundation for Medical Quality, in a letter to JAMA this week. In 2009, it was 88 percent.

"Although performance on the individual SCIP measures may not predict outcomes at the level of the hospital, the project has stimulated activities related to measurement and reporting of surgical quality that are now routine in many U.S. hospitals," Bratzler wrote.

Back to HCB News

More Industry Headlines