The number of patients with higher severity-of-illness scores for patients with hospital-acquired sepsis dropped from 82% to 75% between 2009 to 2011, "indicating that we are catching symptoms sooner and preventing patients from becoming severely septic," says Kristine Otterson, Intermountain's process improvement projects coordinator for the Southwest Region. That decreased sepsis length of stay by just over a half a day and readmission rates by 2.7 percentage points, Otterson says, from 13.8% to 11.1%.
Of course making such dramatic changes can be controversial. Some physicians and surgeons don't like being told they have to adapt to a protocol. They call it "cookbook," and say it's not why they went to medical school, VanNorman says.
For example, the push to have patients consider alternatives before they see a surgeon has drawn some fire from specialists, says Jevsevar, an orthopedic surgeon.
"We think the decision should be an interdisciplinary one. We think that surgeons are biased toward intervention to help each specific patient and have a difficult time telling patients no," Jevsevar says.
As surgeons, he continues, "we develop a certain amount of confirmation bias where we've done things for so long we think we can get by with this diabetic patient, too, not understanding that at some point it's going to burn us."
Key to making collaboratives like this work, Jevsevar and VanNorman emphasize, is to have tight, correctly risk-adjusted patient outcome data for each doctor. That way, operators have a tougher time arguing—as many of them will—that their results weren't worse because their patients were sicker from the start.
In dealing with two such physicians, the data made a convincing case that led the surgeons to finally conclude, Jevsevar says, that the procedure "may be something I shouldn't be doing anymore."
Another collaborative unified all eight children's hospitals in Ohio to improve two problems: infection rates in orthopedics, neurosurgery, and cardiovascular surgery, and adverse drug events related to the use of narcotics.
The groups met and found wide variability in the timing and dosage of antibiotics usually given to patients just prior to surgery, says Vera Hupertz, MD, vice chair of quality and safety for Cleveland Clinic Children's Hospital.
A collaborative leader "simplified the process, standardized the dosing of antibiotics during and before surgery so that everybody gets a higher, stronger dose" and in intervals of three hours instead of four. Ever since, surgical site infections have dropped from 8 in 2010 to 7 in 2011 and to 1 in 2012 as of August 8.
The second issue involved the common problem of dangerous constipation in children treated with narcotic painkillers. "Constipation increased patients' morbidity and potentially the length of stay, yet it wasn't being aggressively addressed by the teams," Hupertz says. "We were able to evaluate which laxatives worked best and determine the best doses," and to date, adverse drug events have dropped by 33% since 2010.