The best way to improve a hospital’s surgical quality—and sustain that improvement—is through access to timely clinical data, panelists said at a Tuesday afternoon session of the Hospital of Tomorrow, a U.S. News & World Report conference.
[More Hospital of Tomorrow Forum coverage: usnews.com/hospitaloftomorrow]
The session, called “Making Quality Stick,” highlighted the value of data gathered by the American College of Surgeons, founder of the National Surgical Quality Improvement Program (NSQIP), a registry that collects clinical, risk-adjusted, 30-day outcomes data used by 500 hospitals throughout the United States.
“NSQIP is data surgeons will believe,” said panelist Dr. Carl Boyd of Memorial University Medical Center in Savannah, Ga. “If you share data, you will see change in your hospital.”
The session focused on a challenging aspect of hospital improvement: capturing and identifying the subtle shifts in quality of care that signal trouble. Studies have shown that relying on billing data is insufficient, because at least half of all complications go unrecorded.
Dr. J. Michael Henderson, chief of quality for the Cleveland Clinic, said one study at the clinic showed that billing data revealed just 45 percent of known surgical-site infections. It also yielded a false positive rate of nearly 60 percent, because the clerks who fill out billing forms record suspected infections but not the results of tests that ruled the infections out. “Unless I write somewhere else in the chart that there was no infection, you get wrong information,” Henderson said.
The NSQIP data allows hospitals and quality assurance officers to identify trends in surgeons’ performance or in complications that may reflect broader problems throughout the surgical service. Patient data is gathered by trained reviewers who collect information on randomly selected patients before surgery and 30 days afterward. The data are risk adjusted to account for differences in the patients’ ages, severity of illness and the complexity of the operations performed. The reviewers examine how well patients are faring a month after the operation to catch complications that occur after patients leave the hospital.
Boyd said the data is packaged in scorecards that can be shared with multiple hospital committees charged with quality and safety assurance. The overall scores are included in a monthly newsletter. Surgeons with 10 cases in the database get customized report cards that allow them to evaluate their own performance.
In 2008, when Memorial University Medical Center began examining data, the hospital had a statistically higher rate of urinary-tract infections than other hospitals in the database. The hospital initially responded to the results, Boyd said, with denial. “The nurse doesn’t collect the right data.” Hospital officials had the same reaction a year later, but in 2010 began taking the data seriously. Over the next two years, the hospital prevented nearly two dozen infections, sparing patients additional misery and saving the hospital about $3,000 per infection—about $70,000.
“If applied to all general surgery patients,” he said, “The hospital would have saved $320,000.”
Dr. Clifford Ko, of UCLA Medical Center and director of the NSQIP surgical registry, said that there is no “one-size-fits all” approach to surgical quality improvement. In one study of eight major hospitals, he said, researchers found that there was no common factor associated with surgical site infections in every center. Every center got better, but they got better in different ways. “They get there," he said. "But they get there by different paths.’’