Next QI hurdle, rapid-cycle change: Most programs have elements in place
Accelerated improvement: Industry's most underused strategy
As the cycles of merger, acquisition, and restructuring tear through health care organizations as often as once a year, industry insiders warn that quality improvement cycles must keep pace. Regula-tory agencies such as the Joint Commission on Accreditation of Healthcare Organizations are demanding more than promises and good intentions. And who could blame them, asks Tim Hallock, MS, CQI facilitator at the Madison, WI-based St. Mary's Hospital and Medical Center. As an industry, Hallock asserts, "We have wait times that would make people in airports revolt."
Though barraged by all these pressures for swift change, less than 20% of QI professionals are truly accelerating their change cycles, according to quality consultant Chip Caldwell, senior vice president of Premier Performance Services in Charlotte, NC. (See interview with Caldwell, QI/TQM, December 1997, p. 160.) And merely going faster isn't enough, says Rosemary Keeley, director of Improvement Services at VHA Central Atlantic, also in Charlotte. "We have to double and triple our change rates if we're going to survive."
If comments like Hallock's and Keeley's tempt you to write off this story, don't put it down yet. "It's possible to speed change rates," Keeley says, "and you can do it without cutting corners."
She has the experience to prove it. When she introduced rapid improvement concepts at VHA facilities in New Jersey and Pennsylvania, she produced an 89% time improvement compared with their earlier methods. (For comparisons between the implementation times of traditional teams and rapid change teams, see comparison table, above. For the start-to-implementation time span for team processes, see summary table, above right.)
Strip old models for seeds of quick change
To improve your change rates, you don't necessarily have to scrap your current efforts, but you'll have to strip them down to their most functional elements. Below are three guidelines for extracting the best nuggets and building upon them:
1. Prepare teams to hit the ground running.
In traditional QI models like the one at St. Joseph's Hospital and Medical Center in Paterson, NJ, a quality council could deliberate for two hours a week for two years and achieve a problem identification, mission statement, and define the problem parameters.
It's quite a different story today, says Barbara Niedz, RN, PhD, Director of Quality Management at St. Joseph's. In accelerated change, as she uses it, teams don't even meet until the above-mentioned preliminaries are complete. Using the Wilton, CT-based Juran Institute's Blitz Team method, Niedz works with a varied cast of characters, depending on the nature of the QI project. Typically, they would include the team leader, a trained facilitator like Niedz, a data analyst, and one or two key managers with the clout to secure buy-in from the project team.
A good managerial choice, Niedz explains, would be a supportive vice president possessing the power to free up staff and monetary resources. Background tasks often include identifying the right players. With a high-level manager involved, Niedz says, getting participation from department directors is no problem.
2. Outfit your team with ready-made tools.
Another background task is creating the needed change process tools:
- Document the problem through graphic presentations of root cause analysis data.
- Assemble the literature reviews.
- Design a flowchart isolating the parameters of the problem, and write a glossary of working definitions of key terms.
That way, when your process team meets for the first time, you can hand them a set of tools and get down to solutions.
3. Create data collection tools on the run.
When one of Niedz's project teams gathers for its first three-hour meeting, it typically comes up with a pilot plan for a process change. While the team works, Niedz and her associate work with their laptops, sketching a data collection tool.
As the meeting closes, they tell the team to be at their desks in two hours for a flash review of the initial draft.
Why is a high-level manager's participation key?
Niedz hammers out the draft, distributes it, and incorporates the flash review feedback. By the end of the day, each team member has multiple copies of the revised instrument. So data collection can start as early as that night if the issue involves a clinical situation.
Keeley notes that it's not always realistic to expect a pilot plan out of the first project meeting, especially on very complicated tasks. It can take three to four meetings before a pilot is ready to launch. If the processes involved need more than 25% improvement, it usually requires an initial design plus refinement. (For a capsule of her meeting agendas, see time model, above right.)
Rapid change in action
Here are three examples of accelerated change from inpatient, operational and ambulatory care settings.
Don't medicate problems if you can eliminate them.
St. Mary's took change cycles that yielded minimum results and accelerated them into clinical improvements of 50% within four months. They model their process after the plan/do/study/act approach designed by the Boston-based Institute for Healthcare Improvement's Breakthrough Series.
Here are the questions Hallock asked to accelerate clinical improvements:
- What are you trying to accomplish?
- What changes are necessary to effect improvement?
- How are you going to measure your change?
In one project, a team of cardiac surgeons, nurses, pharmacists, and anesthesiologists were searching for appropriate anti-anxiety meds to calm post-cardiac surgery patients so they could come off of respirators sooner. Hallock explains that the literature shows the sooner patients come off respirators, the sooner their recovery begins.
But, Hallock continues, "Someone asked the obvious question, 'Instead of finding anti-anxiety drugs, why we couldn't just take them off the respirators sooner?' Because obviously having those tubes down their throats when they woke up from the anesthetic was making them anxious."
That was the key Hallock's team needed to pilot earlier extubations instead of administering anti-anxiety meds in the first place. "Many of the changes we wanted to use were founded in science." he says. "But there's a lag between research and practice in the clinical setting. We tried to apply what was already out there."
Four months into the project, an average of 50% of patients were extubated within 16 hours, 90% within 32 hours. The benchmark they chose was a six- to eight-hour average intubation time for 50% of post-cardiac patients. St. Mary's aimed for eight hours. Now they average 50% extubation within six hours, and 90% of the patients are off intubation within 16 hours. Hallock points out that 16 hours used to be the average intubation time for the hospital's cardiac surgery patients.
How did St. Mary's accelerate the change cycle? Once people answer the three questions stated earlier, they start to pilot changes, Hallock explains. "Some would say we leave out root cause analysis, but in the accelerated model it's done, we just don't bury ourselves in proving the root causes."
With post-cardiac surgery, for instance, the literature review persuaded the clinicians that finding the root cause of post-surgical anxiety would yield less positive outcomes than just getting them off the respirators as soon as possible. Hallock adds, "Now, we go for solutions that get us the improvement we want."
'Backlog is bad.'
Three and a half years ago, staff threw away the procedures manual in the health information management department at the University of Texas Medical Branch in Galveston. Carl Hula, associate director of records management, says the book was impeding progress in reducing turnaround times and continuous improvement.
In place of departmental sections, staff created 18 self-empowered teams with team leaders instead of supervisors. They trimmed middle management to four positions below the director. "The staff's charge was to concentrate on outcomes and design their own procedures," explains Hula. "They figure out how to do it."
The results testify to the teams' effectiveness. Before teams existed, it used to take up to 32 weeks to file loose reports into medical records. In 1994, that task was pared down to 24 to 36 hours, "and we haven't looked back since," he says.
To foster success, staff combined technology and improvements in the physical environment with training in teamwork. Among the technical upgrades, they installed e-mail as their main communication tool. Now, the departmental climate is one of trust and confidence in the teams. For instance, teams hire new co-workers; they also recommend removal of those who don't work out.
Cross-training needed for staffing shifts
To enable staffing shifts according to workload cycles, records, and data management, staff cross-train each other. That practice has more than a single value, observes Hula, "It's where a lot of continuous improvement comes from." Newcomers are more likely to question poor procedures, while the veterans might adapt and put up with them.
Instead of productivity standards, the teams establish job-specific criteria. In filing patient records: no unfiled backlog at the end of each 24-hour period. To meet that goal, the seven members of the loose report filing staff tuck 10,000 to 11,000 loose pieces of paper in their proper places every day. What are the quality sacrifices for all that speed? Almost none - accuracy consistently runs 99% as determined by intensive quality checks.
Further improvements: Of the 2,800 charts that go to the clinics for patient appointments every day, 99.9% arrive on time, compared to 70% in 1995. That same year, turnaround for release of information forms was 15 to 20 days; now it's three days. In 1995, 14 million unbilled accounts awaited billing; now it's less than three million.
How did Health Information Management accelerate change?
Simple. They threw out productivity standards. Hula contends that people calibrate their production to the standards when it could actually be much higher. Instead, he empowers his people with adequate training and resources, doesn't worry about the absence of written procedures, and feels no particular need to know the process details. He lets the teams know he expects them to find the best ways to do their jobs.
Good medicine by phone.
Eugene Nelson, DSc, MPH, director of Quality Education, Measurement and Research at Dart-mouth-Hitchcock Medical Center in Lebanon, NH, recalls a rapid-cycle change in urinary tract infection (UTI) treatment protocols. At one of the Dartmouth-Hitchcock satellite clinics, an internist and his associates learned from a literature review that most women with UTI have good clinical outcomes when they describe their symptoms by phone to a nurse practitioner who then calls in the appropriate prescription to the pharmacy.
The internist adopted the phone intervention with his patients, and they were delighted to skip the standard doctor's office visit. Symptom resolution rates improved, Nelson reports, and patient satisfaction scores are well over 90%.
How did Dartmouth-Hitchcock accelerate change?
This was another case of "nonacademic detailing" of the change process, says Nelson - not bogging down in the laborious documentation of change. Instead, the internist took the upfront work represented by the existing clinical studies and transplanted it to a clinical practice.
Since the technique worked so well, Nelson continues, he asked the internist and nurse practitioner to visit eight other practices for informal lunchtime inservices describing the streamlined protocol and sharing the literature. The other physicians were asked to try it for their next 15 UTI patients.
They also received a simple checklist to collect data on the outcomes. That was last June. In September, representatives from the eight practices attended a review session. Six of the eight had collected the data and either adapted or duplicated the telephone protocol in their practices.
(For more information on implementing accelerated change cycles in your own facility, see related stories, at left, and p. 46.)
[For further information, contact:
· Tim Hallock, MS, CQI facilitator St. Mary's Hospital and Medical Center, 707 Mills St., Madison, WI 53715. Telephone: (608) 251-6100.
· Barbara Niedz, RN, PhD, Director of Quality Management, St. Joseph's Hospital and Medical Center, 703 Main St., Paterson, NJ 07503. Telephone: (973) 754-3146. Fax: (973) 754-3105.
· Rosemary Keeley, director, Improvement Services for VHA Central Atlantic, 521 E. Morehead St., Suite 300, Charlotte, NC 28202. Telephone: (704) 378-2444. E-mail: [email protected].
· Carl Hula, University of Texas Medical Branch at Galveston, Records Management Division, 3.320 McCullough Building, Galveston, TX 77555-0782. Telephone: (409) 747-0699. E-mail: [email protected].]
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