Study shows Keystone program achieved significant savings

Authors say results make a "business case" for quality initiatives

One of the common challenges faced by healthcare quality professionals is being able to make a business case to administrators — that is, to demonstrate their QI programs will save the hospital money.

"Although many people talk about the business case for quality, there has actually been little evidence that shows this to be true," says Peter J. Pronovost, MD, PhD, FCCM, professor, departments of anesthesiology/critical care medicine and surgery at The Johns Hopkins University School of Medicine; professor, department of health policy & management, The Bloomberg School of Public Health; professor, school of nursing; and medical director, Center for Innovation in Quality Patient Care, and director, Quality and Safety Research Group.

Now, however, that may all be starting to change with the release of a study in the American Journal of Medical Quality. The study, which set out to evaluate the financial achievements of the well-known Michigan Keystone ICU Patient Safety Program, focused on several ICUs in facilities that had participated in the program; the facilities were selected to represent a broad variety, including large, medium, small, academic, urban, and rural hospitals. The researchers found that on average the initiative averted 29.9 catheter-related bloodstream infections and 18 cases of ventilator-associated pneumonia each year. "The average cost of the intervention is $3,375 per infection averted, measured in 2007 dollars," wrote the authors. "The cost of the intervention is substantially less than estimates of the additional health care costs associated with these infections, which range from $12,208 to $56,167 per infection episode."1 These savings, they added, did not take into account any additional benefits that might have been achieved through the reduction in cases of sepsis, the prevention of mortality, improving teamwork, or reducing nurse turnover.

"We knew that the project we did in Michigan reduced infections and saved lives, but we did not know if it saved money; we weren't sure if there was a business case on it," says Pronovost, whose "checklist" served as the foundation of the Keystone project and who was a co-author of the paper. "If this kind of program was going to spread, we had to show that."

Pronovost says that when he spoke to hospital CEOs in Michigan, they were generally split on whether they thought they were saving money. "They all supported reducing infections, but some said they lost money for doing it, while others said they saved money," he says. "We wanted to gain a deeper understanding of that issue."

When the results of the study were extrapolated, he continues, it was found that the average hospital saved a million dollars. "Another way of looking at it is this: There was some cost in preventing each infection, but the return was basically 10 times that cost," Pronovost says. "What other investment today can give you that kind of a return?"

Quality, safety "not enough"

Part of the authors' rationale for the study clearly was that the safety and quality improvements of the program were not enough in and of themselves to convince all hospital CEOs to duplicate the program, and lead author Hugh Waters, deputy director of the healthcare outcomes and quality program at RTI International in Research Triangle Park, NC, bears that out.

"This was a scientific approach to intervention — with 10 different pieces," he notes in describing Keystone. "Some just call it a checklist, but the underlying philosophy is to make sure everything is done correctly every single time — for example, nurses check that tubes are properly inserted in the mouth and nose, check for bruising, and a whole series of steps — each common sense, but when done all together you get a dramatic effect."

That's all well and good, he continues, "but because of the way the U.S. healthcare system is organized, it has to be obvious to the people that implement such programs that they will benefit financially." This, he notes, could be the hospital, or it could be the insurance company — or both. "As a result you get people saying, 'Sure, this helps, but show me where the money is,'" Waters says.

Pronovost couldn't agree more. "We were able to reduce these infections dramatically in Michigan, but often to do so you have to spend money creating infrastructure," he explains. "And people will not make those investments if they do not believe there is a return. And if you just do the program and expect people to take on more work and try harder, it will not work."

"You're removing any economic arguments," Waters adds. "We set out to measure the costs and benefits." The study, he says, involved "very intensive data collection on the financial side," including initial education and training; capital purchases; ongoing time spent on the intervention; the average annual salary for each type of position involved in the project; and product purchases. "Training is a big cost, but the biggest is adding some staff," says Hughes. "Some of these activities require additional nurse time and some additional physician time."

The biggest cost in any hospital, Waters continues, is human capital. "For me, it was most difficult to calculate cost because time makes up most of cost, so we had to know how much time was involved by doctors and nurses," he says. "In this study we recognized the most accurate way was 'activity-based accounting.' In a nutshell, we measure what people spend time on and use it to allocate industry-recognized costs of overhead, hiring staff, and so on. And we got into the nitty-gritty of how procedures are done." Calculations also included supervision follow-up, staff meetings, and so on, as well as physical purchases such as line insertion carts.

Using the results

Waters asserts there are a number of ways quality professionals can use the findings of the study to make their own case with administration. "One, literally show them the results of this study and how it demonstrates that attempts to control infection can be financially rewarding," he says. "But maybe a bigger point is they can use this kind of approach to figure out what can make sense for their hospital. There are a lot of QI approaches, and you can use this to show that they make sense for hospitals to implement. And third, that maybe we need more of this kind of research to show these types of interventions make sense to do."

"This is really a huge 'little' study, and hopefully this one shows them that there really is a business case for quality," adds Pronovost, who has reason to be encouraged. The Agency for Healthcare Research and Quality (AHRQ) has backed the expansion of the Keystone concept into 22 states, and so far that initiative has cut infections in half. "This is really dramatic — it's one of the first, if not the only, nationally scalable QI program where we have measurable results," Pronovost says, adding that he would "love" to do a business case for this larger initiative, but AHRQ has not yet funded it. "To their credit, however, they believe the results Hugh and I just published," he says.


  1. Waters HR, Korn R Jr, Colanuoni E. et al. The business case for quality: economic analysis of the Michigan Keystone Patient Safety Program in ICUs. Am J Med Qual.2011;26:333-339.

[For more information, contact:

• Peter J. Pronovost, MD, PhD, FCCM, Professor, Departments of Anesthesiology/Critical Care Medicine and Surgery, The Johns Hopkins University School of Medicine; Professor, Department of Health Policy & Management, The Bloomberg School of Public Health; Professor, School of Nursing, Medical Director, Center for Innovation in Quality Patient Care; Director, Quality and Safety Research Group. 1909 Thames Street, 2nd Floor, Baltimore, MD 21231. Telephone: 410-502-3231. Fax: 410-502-3235. E-mail:

• Hugh Waters, Deputy Director, Healthcare Outcomes and Quality Program, RTI International, 3040 East Cornwallis Road, Cox 275, Post Office Box 12194, Research Triangle Park, NC 27709-2194. Phone: 919-541-6565]