Collaborative uses new process to cut waits, delays in half within a year
Who says TQM takes months or years to implement?
In maturing managed care markets, competition among health care systems is shifting from price to service and quality. Being responsive to customers’ expectations of prompt treatment will become a real competitive advantage, says Thomas W. Nolan, PhD, co-founder of Associates in Process Improvement, a consulting firm in Silver Spring, MD.Some managed care companies are already setting targets such as scheduling appointments within a week or limiting waiting times to no more than a few minutes, he says. And if they aren’t now, companies should consider that the hidden costs of delays are driving up already spiraling costs.
"A patient waiting to be transferred from the intensive care unit (ICU) is not just a service issue because the ICU is a very expensive place to wait," Nolan points out. "Or if your organization must meet the target of offering appointments within seven days, then you have two choices: Hire more physicians and staff, or redesign your system."
But redesign doesn’t have to be traumatic or paralyzing for your organization — if you improve your improvement process, insists Nolan, co-creator of the Model for Improvement for the Breakthrough Series Collaborative of the Institute for Healthcare Improvement (IHI), a nonprofit organization in Boston.
Nor is this model meant to replace your current one, he adds. "The same concepts can be incorporated into your models to accelerate their improvements," he says.
One of the essential parts of the model is based on what Nolan calls "change concepts," or scientifically grounded ideas for change. Nolan and his colleagues identified and wrote about 70 such change concepts applicable to many business settings in their book, The Improvement Guide: A Practical Approach to Enhancing Organizational Performance (For a list of change concepts applicable to health care, see box, p. 103.)
"By understanding the various concepts of change, teams don’t have to start from scratch, so they can begin to test a change in hours or days, rather than weeks or months," he explains.
The second part of the model calls for answering three fundamental questions about the actual process change and testing the process repeatedly on a small scale before implementing it on a larger one, he says. Teams from 27 health care organizations now agree with him.
In 1995, IHI invited organizations to work together for a year to reduce by 50% the delays and waiting times in surgery, emergency departments, and clinics and physicians offices, as well as delays in scheduling appointments.
"The collaborative was designed to help committed organizations make major rapid changes that produced breakthrough results," says Marie Schall, MA, the IHI collaborative director. (For more information on this and other breakthrough series topics, see related story, p. 101.)
Many organizations met and exceeded their initial goals, she says.
• At Sewickley (PA) Valley Hospital, the median delay for patients scheduled for surgery was 55 minutes. Twelve months later the delay had been reduced to 25 minutes.
• At MetroHealth in Indianapolis, only 42% of patients were offered a routine pediatric appointment within seven days. In one year, the chance of getting the appointment in that time frame rose to nearly 100%.
Even those organizations that fell short of their goals made substantial progress, says Nolan. For example, instead of waiting for the anticipated scheduling system to solve Deborah Heart and Lung Center’s 105-minute wait time in ambulatory care services, one department concentrated on using the IHI model to identify the real bottleneck, cutting off 43 minutes of the average wait time.
"Although this time is divided up into several shorter periods, patients still have to wait 62 minutes and that’s too long. But now we’ve laid the foundation so that the scheduling system will finish correcting the problem," says Charles Dennis, MD, chairman of the department of cardiology in the Browns Mills, NJ, hospital.
Just get on with it!
"Many QI teams will talk for months and months, planning an ambitious change, trying to figure out ahead of time what the downside will be," says Nolan. "And while I realize that tests need to be planned, the learning comes when the actual change is made and you can observe the results," he says.This trial-and-learning approach is inherent in the Model for Improvement. Repeatedly testing change on a small scale — then observing the consequences and learning from them — allows a team to quickly see when a change is actually working. "Small scale refers to the size of the test — one physician, one operating room, for example — not the size of the change. The change itself may be quite big or innovative," he says. (See illustration of cycle’s components, below.)
Typical tests last for a week or two, he notes. After the first cycle is over, the next immediately begins, building on the experience of the first. "Ask yourself, What worked and what didn’t? What should we keep or throw away?’" says Nolan.
After several rapid cycles that improve on the previous ones, the change is then ready for broader implementation. "This approach really reduces fear and decreases the natural resistance to change," says Schall. "Also, people are more willing to go along with the change if they don’t think it’s going to be permanent. As they engage in the process and begin to get results, they’ll be more open to spreading the change." (For more tips on this rapid improvement model, see related story, p. 102.)
But before you start running a cycle to make a change, you must know exactly what you’re trying to fix! The model stresses that three fundamental questions about aim, measure, and change must be answered upfront. "Although they can be addressed in any order, it’s imperative that you do so," says Nolan.
1. What are we trying to accomplish?
State the aim clearly and specifically. A numerical goal may be helpful. For example, the emergency department at York (PA) Health System was becoming a holding area for patients who needed beds. So the aim was to reduce delays in transferring patients to inpatient beds by 50%.
"This succinct statement not only clarifies the aim but also suggests the level of support that will be needed to reach it," Nolan says. "And it serves notice that the goal cannot be met by tweaking the existing system."
2. How will we know that a change is an improvement?
"You need to identify measurements that will indicate whether a change is actually an improvement," explains Nolan.
For example, York established this measure: A change is an improvement if the time it takes to transfer a patient from the emergency department decreases. The team measured the time from when a decision was made to admit a patient from the emergency department until the time the patient was actually taken to the patient care unit. The team also monitored patients and staff satisfaction to ensure that change didn’t result in unexpected consequences.
3. What changes can we make that will result in improvement?
"All improvement requires making a change, but not all change is an improvement, so you have to be able to identify the most promising changes," Nolan says.
For example, York’s delays in transfer of patients were occurring because no beds were available in the ICU. In order to reduce delays, changes would have to be made to free up beds in the ICU.
Form and support the team
Schall suggests that members of senior leadership carefully explore those questions about their particular wait time or delay dilemma before they create a team. "It’s important to define the aim thoroughly so you can fully construct the team," she explains.Next, consider the system processes that will be affected by the improvement. Although the model recognizes that teams vary in size and composition depending upon the aim, it must contain — or be supported by — the right mixture of physicians, nurses, managers, and administrators from the following areas:
• System leadership.
"This person is someone with enough clout in the organization to get things done, so it’s important that he or she have the authority in all of the areas affected by the change," she explains. He or she doesn’t necessarily have to be a member of the team, Nolan says.
• Technical expertise.
"You also need a subject matter specialist who understands the entire process of care being improved," she says. QI professionals can also provide technical support in areas such as data measurement and analysis.
• Day-to-day leadership.
"Look for someone who understands the process thoroughly and has the desire and ability to drive the project on a daily basis," Schall says.
For example, the QI team at Deborah Heart and Lung Center in Browns Mills, NJ, consisted of a physician who was chairperson of the department of cardiology, a physician who was chairperson of the QA/UR committee, a nurse who was assistant director of nursing for clinical practice, and a scheduling clerk.
"Our job was to cut waiting time in half for ambulatory care patients, so we were very task oriented," says Sandra Morelli, RD, BSN, who was part of Deborah’s team. "We met every two weeks, and each agenda was driven off tasks assigned from the previous meeting."
But even after the team is formed and functioning, aims may need to be refined, warns Schall. "It’s a constant cycle of [aim] clarification that may involve adding or dropping team members," she explains.
For example, after some initial changes, Deborah’s team members realized they would be more effective to concentrate on changes in the cardiology area, rather than throughout the entire ambulatory care system.
"We stumbled along for three to four months and then realized that we didn’t have the system leadership [on the team] to take it all on," says Dennis. "But we did have control of how cardiology was controlled and designed. Once we narrowed the scope of the aim, the project became manageable."
Measure twice; cut once
After setting the aim and forming the team, the next step is to establish measures that will indicate if a change leads to an improvement. But unlike many other QI initiatives that get bogged down in collecting data or waiting for them to be analyzed, measurement under the IHI model is designed to speed up the process."Remember that improvement is the goal, not measurement, so you want to collect just enough data to know whether changes are leading to improvement and then move to the next step," Nolan says.
Nolan cautions that teams need to use balancing measures to make sure changes to improve one part of the system don’t cause problems in another.
Consider MetroHealth’s collaborative to improve access to routine primary care appointments. "You also have to consider related measurements such as wait times," says Eric Bindewald, MD, former medical director at MetroHealth. "What would happen if you do succeed in getting patients into the office quicker but then they have to wait twice as long? You may be reaching your aim but not necessarily improving patient satisfaction."
[To order The Improvement Guide: A Practical Approach to Enhancing Organizational Performance ($29.95), contact: Jossey-Bass Publishers, 350 Sansome St., San Francisco, CA 94104. Telephone: (888) 378-2537.]
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content