Health system sets 'zero errors' as its goal for patient safety, quality

Instead of compliance percentages, errors are used as benchmarks

Winners of quality and safety awards often have much in common, but there is something about Eastern Maine Healthcare System in Brewer, recent recipient of The VHA Foundation and National Business Group on Health's Patient Safety Leadership Award, that sets it apart: the way it measures success.

"We've done a few things that have changed our approach, specifically relating to the culture of safety, and also a different way of thinking about our goals for specific projects," says Erik Steele, DO, chief medical officer for the system. "We said that as an organization, our goal was to achieve zero errors in key safety and quality initiatives, rather than having a goal of being better than other organizations or hitting a certain percentage. We have now set getting to zero errors as our default for quality and safety initiatives." If in certain instances it is decided that it is not reasonable to aim for zero errors, he adds, the goal can be changed.

Steele is responsible for the management of systemwide quality and safety initiatives. "We have seven hospitals, each of which has individual programs, but I am responsible for the ones we do in concert," he explains. The name of the systemwide quality and safety project is the Zero Defect Project. "So up front we are stating our goal," says Steele. "We have stopped reporting by percentage, such as the percentage of patients who have been given aspirin or a beta blocker for acute myocardial infarction, and instead just report the number of patients who have not. So everyone registers, first, that our goal is zero and, second, the number of patients for whom we did not achieve this goal."

Creating a plan

The system created a strategic plan in November 2007, with a goal of becoming "the best rural health system in America." "We asked ourselves what such a system would look like in terms of quality and safety, and I said I think you'd have zero errors," Steele recalls.

The system leaders identified about 40 initiatives based on standards of the Centers for Medicare & Medicaid Services (CMS), the National Quality Forum (NQF), The Leapfrog Group, The Joint Commission, and the Institute for Healthcare Improvement. "These were expert-vetted, bang-for-your buck kinds of things; we evaluated them with a systemwide committee of doctors and nurses and determined the most important ones," says Steele. "So, to be specific, NQF says every patient admitted should be screened for risk of blood clot, and high-risk patients should get prophylaxis. Our goal is everyone, so we count how many we fail to do."

When reports are made to the board, a composite number of errors is shown. "Instead of a blizzard of data, the board can follow one big-picture number over time," Steele explains. "We were pleasantly surprised at the impact it had on our reporting and how it changed how everyone thinks. Setting zero as a goal changes how you think and has a profound impact on what you're willing to do.

"So, for example, if zero is the goal, as you talk about how to get to the goal, you have a very specific target, instead of something nebulous like 'We're trying to be better than,'" Steele continues. "Second, you are much less likely to have discussions where you decide you want to keep educating people forever; you're much more willing to get into discussions about hard wiring the right thing."

Hard wiring success

Eastern Maine is installing systemwide computerized physician order entry. "We want 100% of our patients to get a screening for blood clot risk when admitted, so we are hard wiring into admission status that you must assign a risk category," Steele explains. "If you don't do that, you can't complete the admission process."

To help reinforce the "zero defect" attitude among the staff, says Steele, the "Your Mamma" rule is used. "Anything you'd be willing to do to protect your mother from harm, you have to be willing to do for all patients," he explains. "When we have discussions about quality and safety, we challenge ourselves; would any of us accept the possibility of this error for a parent, daughter, or spouse? If the answer is no, then it's not acceptable for any of our patients. That's really hard, but it's a very effective check on our tolerance or systems of care that allow errors, and works hand in hand with zero defects."

What's more, he continues, percentages can be deceiving. "As soon as you move out of percentages to errors, you find you're making a lot more than you thought because you had such a small percentage," he notes. "If you admit 20,000 patients and screen 95% for risk, you are at the top of the curve, missing on only 5%. But 5% is 1,000 patients not screened for the most common cause of an unanticipated death in U.S. hospitals — a blot clot that travels to the lungs. Any time you do something a lot, you can have a very good percentage of doing the right thing — but would it be OK if your mother was one of those 1,000?"

Hard wiring, Steele adds, also can be used to engender rapid change. "So, for example, CMS has said recently that if you're going to prescribe vancomycin to a hospital patient, you must provide documentation as to why you use it, since it is one of the few remaining antibiotics that is effective against MRSA [methicillin-resistant Staphylococcus aureus]; if resistance were to develop, that would be devastating. So, we said it was very important to achieve change quickly."

In a more traditional model, he notes, you would educate staff and then feed data back. "It takes about 10-15 years to really achieve a high level of performance that way," he observes. "We simply said that as of a certain day you can't order this medication without providing a reason. It's easier to do this in an EMR [electronic medical record] environment, but you can do it in a paper environment as well."

Change on the hospital level

Vicki Loughery, RN, BSN, MS, chief nursing officer at Sebasticook Valley Hospital, part of the Eastern Maine system, explains how these initiatives translate into change on the hospital level. "Erik will bring forth at a CCC [the systemwide clinical coordinating committee] meeting a new best practice or initiative and ask everyone if this is something we want be involved in; we all vote as a group," she shares. "He generally provides some of the beginning information, we discuss it as a group, come up with a plan of where we want to go, and set the benchmark. Then, each hospital has a quality team within its own organization. We bring it back there and come up with a plan on how to roll it out and get education to everybody."

Every hospital, she continues, has a quality web page, and each is in the same format. "This way, we all look at and do the same things, and we try to measure the same way," Loughery explains." Being an affiliate, we are a small community hospital, but patients will go back and forth between us and the tertiary hospital, and they need to see we provide the same quality care, speak the same language, and that the same things are important to all of us."

She describes her hospital's participation in the venous thromboembolism initiative. "The very first thing we asked was how we could hard wire this," says Loughery. "The problem is that at present we have all different levels of connectivity to the EMR in the system; we don't have it yet. So, for example, at Eastern Maine they can set things up so a physician can't go to the next question unless he takes a certain action, but we can't do that." So her facility developed an order set to achieve the same goal.

"We have our own zero defect team here, which includes the director of quality, vice president of medical affairs, nursing directors for various departments, and a clinical nurse consultant, and we meet every Tuesday morning to look at every initiative," Loughery continues. "We look at any gaps that exist and discuss how we can fill them. A lot of our doctors were ordering prophylactic Coumadin, Heparin, or Lovenox but not assessing the patient's risk, so we were not sure if they were getting the proper med at the proper risk level."

That's where the order set came in. "We re-developed our order set so you had to do an assessment first and assign a high-, medium-, or low-risk level, and we developed appropriate dosing for each level. This made it easy for the doctor to follow the flow."

Unfortunately, this was not a perfect system, because some doctors still skipped over the step. "This is where we had to use people as gatekeepers," says Loughery. "As soon as the doctor turned in the chart, they were either chased down the hall or, after we gave them time get to the office, we would call and note that they had not completed the form. We'd tell them, 'I'll fax it to you; I can't give the patient meds until you fill it out.'"

Every week Loughery runs a report telling her which patients did not get a risk assessment done. "If it is an end-of-life situation, or if they are on a therapeutic dosage, you can put those two on to indicate a reason why it was not done," she notes. This initiative was started in October 2009; in that month, there were 19 charts on which the assessment was not done; in November there were 18, in December 16, and by January it was down to 11. "I hope in February we'll be at five, and by March, which will be the six-month point, we should be at zero," says Loughery.

Loughery says that for some indicators her facility is already at zero, while some others will never be. "For some, if we said we were at zero we'd be lying," she notes, "But we do all that we can. For example, hand washing has got to be the hardest. For January we were at 95.8% with 504 observations monitored. But, we had started at about 50%."

All medical data are reported "unblinded," notes Loughery, so that it can be determined which staff member is not following a given process. "We immediately pull them aside — myself or another nurse or physician," she notes. "We have made it a comfortable culture for everyone; we're not punitive, but educative. I would say, 'Did you realize that just now you didn't wash your hands? What can we do to make this entrance to the room safer? Do we need to move the Purell dispenser, or couldn't you see the sign?'"

"We try to set expectations at very specific levels at a certain period of time, moving up the quality 'pyramid' in progressive hard wiring," adds Steele. "We try to spend less and less time educating and more time using the tools and methods we know will work."

[For more information, contact:

Erik Steele, D.O., Chief Medical Officer, Eastern Maine Healthcare, Phone: (207) 973-7979.

Vicki Loughery, RN, BSN, MS, Chief Nursing Officer, Sebasticook Valley Hospital, Phone: (207) 487-4057.]