Rapid model requires innovative thinkersEven an accelerated improvement model won’t make wait times and delays automatically disappear. No one knows better thanMarilyn Rudolph, director of Performance Innovation Collaboratives at VHA Inc., headquartered in Dallas. Two years ago, when Rudolph was nurse manager of outpatient surgery at Sewickley (PA) Valley Hospital, she experienced firsthand how to overcome barriers to change by combining the right team with the right concepts.
For example, in January 1996, outpatient surgery was delayed 83% of the time. Six months later, only 33% of cases were delayed. How did they do it?
Rudolph shared with QI/TQM the lessons she learned when the 225-bed community hospital participated in the Boston-based Institute for Healthcare Improvement’s Collaborative on reducing delays and waiting times as follows.
Give it your best — and focused — shot.
The team’s goal was to have all outpatients ready for incision within 90 minutes after arriving for surgery.
"Set your aim high and make it clear; let it be known that maintaining the status quo is not an option," she says. "If you’re clear about this, people will start to look for ideas [to overcome barriers]. Chances are, they’ve had them for years, and now you’re giving them an opportunity to try them."
Thomas Nolan, PhD, developer of the IHI plan-do-study-act model, which features testing big changes on a small scale before implementing them systemwide, calls this a "stretch goal."
"One role of leadership is to serve notice that the goal can’t be met by tweaking the existing system," Nolan says.
Rudolph also advises team leaders to make sure members are focusing on the same goal. "They may pretend they are, but if you sense they are veering, then put the goal up on the wall and say This is where we are going,’" she says.
Nolan terms this phenomena "aim drift."
"Be careful not to drift away from what was originally a stretch goal," he says. "Because people are so cautious about change, it’s easy for a goal of 50% reduction to slip almost imperceptibly to 40% or even 30%."
If that happens, remember Rudolph’s advice, he says. "Repeat the aim constantly by starting each meeting with an explicit statement of aim," she says.
Make sure a frontline worker is on the team.
"Without a staff member who understands firsthand the process to be improved, the team will be wasting its time," Rudolph says.
For example, previously the scheduling was done according to the time frames allotted in the computer. "We realized that we were scheduling cases with what amounted to be guesswork. One surgeon would say he could do a case in 45 minutes, and another would estimate 20 minutes," she explains.
So the team relied on "good old-fashioned common sense," as Rudolph puts it. "We asked the operating room nurse on the team to help us schedule cases based on reality, rather than estimation," she says.
Make data collection a means rather than an end.
For example, instead of measuring delays in all surgical cases, the team used sampling to help them understand why their system was not working.
"We measured delays at 11 a.m. and at 2 p.m. because we knew those were prime times for activity," Rudolph says.
And instead of measuring delays every day, the team measured each of the seven operating rooms one day per week. "By computing the median of those times and then plotting them on a chart, we had an ongoing data collection that was accurate but not cumbersome," she says. "It’s important to keep data collection simple, so you can put energies into testing change rather than data management."
By varying design parameters such as number of patients, doctors, units, or beds, points out Nolan, you can accelerate testing.
"Try sampling the next 10 patients instead of a getting a sampling of 200," he says.
Nolan reassures those who worry that small-scale data lack "scientific" validity. "Randomize clinical trials are needed to establish standards of practice, not to test best methods for putting those standards into practice," he explains. Also, teams can use sampling to test change on a small scale, and once they agree on improvements, they can collect additional data.
Understand and identify change concepts.
Change concepts, defined by Nolan as general scientifically grounded ideas for change, can also reduce the need to collect too much data. (Out of more than 70 such concepts Nolan identified for industry and manufacturing, 27 are pertinent to reducing wait and delays in health care.)
For example, at Sewickley by answering the question, "What change can we make that will result in an improvement," the team selected the change concept — do tasks in parallel.
"Instead of setting up instruments in the operating room and then preparing the instruments for surgery, we now perform these two tasks simultaneously," says Rudolph. (For Nolan’s list of change concepts, see chart, p. 103.)
Test change concepts in a small, friendly environment.
"We encourage people to take action based on the rapid cycles of testing and learning," explains Marie Schall, MA, IHI collaborative director. "If you were learning to ride a bike, you wouldn’t spend months and months plotting how to do a randomized clinical trial of 10 flat streets and 10 hills, you’d just get out there and try it."
Instead, you’d probably rely on the help of someone you could count on to provide support as well as learn to ride on familiar streets.
"We started in one operating room because we realized if we tried to tackle everything at once, we would fail," says Rudolph. "We packed that room with staff we knew were open to the idea of change. In that environment, we set ourselves up for success because we weren’t hesitant to try any idea — we knew no one would say I told you so’ if it didn’t work."
Don’t worry about buy-in during initial stages.
On your first attempt to ride a bike when you were six, you also wouldn’t worry about getting your parents’ permission for a cross-country trip you may take at age 18.
"By small scale testing of a big idea, getting consensus early on is not a fruitful strategy," advises Nolan. "Later on, yes, when you’re experienced and ready to implement on a larger scale. By then you’ll have evidence that you are capable of making the idea work."
Sometimes buy-in will happen as the evidence of change speaks for itself.
"Surgeons begin to notice that the test room had less delays than the others," says Rudolph. "They began to ask, What’s going on in room three?’ Then they wanted to know, When is it going to be our turn?’"