In 2008, Legacy Health, a six-hospital system based in Portland, Ore., embarked on an initiative to reduce hospital mortality and preventable harm. Called Big Aims, it zeroed in on Legacy's four most prevalent infections—catheter-associated urinary tract infections, bloodstream infections associated with central-line catheters, surgical-site infections, and ventilator-associated pneumonia—and garnered big results in a fairly short period.

By 2010, Legacy saw a 44.6 percent reduction in hospital-acquired infections and a 13.5 percent reduction in hospital mortality. The estimated savings was $6.8 million over two years.

Jodi Joyce, vice president for quality and patient safety, says key to success was focusing on a manageable set of improvements that would show dramatic results. "One of the things that makes us complicated as an organization is we have six different hospitals and we range in the services we offer," she says. "So there's a part of us saying, 'Would a single approach work across our system?' And it did."

Systems with similar success stories—including Ascension, Geisinger and the Mayo Clinic—also began with "small tests of change," says Jeff Selberg, executive vice president and COO of the Institute for Healthcare Improvement. "And as success is generated and those results are publicized," an organization gradually develops a culture that strives to find ways to perform even better.

Legacy's initiative centered on implementing best-practice bundles for reducing infection, based on IHI guidelines, such as establishing mortality review boards at each hospital and instituting regular multidisciplinary rounds in which clinicians asked scripted questions about safety and whether the patient was receiving the best possible care. Joyce says getting the clinical staff to support the project went "remarkably smoothly," in part because they had a hand in the specifics. While some broad systemwide approaches were developed, clinicians at each hospital tailored solutions based on their needs and patient populations.

Mortality review teams of doctors, nurses and administrators used an IHI approach of reviewing 50 consecutive deaths at a time to determine whether lapses were one-time issues or part of a pattern. "When you look at data points one at a time, you can draw conclusions that you don't link to other things that occurred that were quite similar," says George Cioffi, chief medical officer at Legacy.

Since 2010, Legacy has reduced infection rates even further. "We still probably have some real opportunity around central-line infections, urinary tract infections and surgical-site infections, but it's specific by hospital now," Cioffi says.