Welch, Shari J. MD
Decide today. Choose a patient flow problem in your emergency department that needs solving and commit to working through the problem using continuous quality improvement tools. This will be the first step on the road to a perfectly functioning department. Know, however, that the end of the road is always up ahead. You will never arrive at your destination.
There are a number of means to improve an ED process or operation: Six Sigma, Lean, and Rapid Cycle Testing, to name a few used successfully by business and industry. Within health improvement organizations and the corporate crowd, there are groupies for every school of thought. Take away the slogans and the T-shirts, and they are simply methodologies for solving process or operational problems. You need tools like these to help you through an improvement initiative, but one approach used regularly and well will likely be enough for an emergency department to make desired changes.
Six Sigma offers two pathways to improvement depending on whether the process is dysfunctional or doesn't exist in a coherent, recognizable way. One approach is used for tinkering — define-measure-analyze-improve-control (DMAIC) — and the other to devise a new process for dealing with a problem — define-measure-analyze-devise-verify (DMADV).
Define, the first step in both, is often more difficult than it sounds. A subjective grievance, “We have problems with the lab,” must be turned into specific actionable events. What exactly is the problem? A flowchart often helps the team see the inefficiency in a process. The REACT (Rapid Entry and Accelerated Care at Triage) study examined the many steps involved in entering a patient into the system in the University of California at San Diego emergency department and getting the lab operations started. In particular, the UCSD team headed by Theodore Chan, MD, and James Killeen, MD, had difficulty in getting labs properly labeled and sent off in a timely fashion. There were delays, lost specimens, and mislabeled specimens. (Ann Emerg Med 2005; 46:491.)
All quality improvement projects must involve step two, a measurement for tracking the improvement. In the ED, there are only two measurable quantities: those we can count (mislabeled labs) and those we can measure in time (time-to-lab results). Baseline data are usually obtained through a data-run off a tracking system or a manual chart review. Table 2 shows lab turnaround times at UCSD before and after the quality improvement initiative began. The REACT project used bar codes, expedited triage protocols, and an integrated tracking system to accelerate emergency care.
Small teams of stakeholders look at problems and processes, breaking them down into parts, looking for redundancy and inefficiency, and identifying faulty processes to fulfill step three, analyze.
At this time, the team works to find a way to improve the old process or devise a new one. At UCSD, the team implemented an entirely new process to expedite registration and lab collection in triage. Instituting a bar code system and an integrated tracking system improved lab processes and overall departmental efficiency.
The final step, verify or control, ensures that the change has been effective by measuring the metric again and comparing it with the baseline. If a minor change has been implemented, then occasional spot measurements are performed to ensure that gains have been maintained after the improvement initiative. The graphs showed improvement in efficiency and decreased lab errors with the new innovations. This also had the unexpected side effect of reducing the number of patients who left the ED before being seen.
Six Sigma techniques have been employed by Motorola and General Electric for decades with great success. The training in Six Sigma has its own vocabulary and culture, and practitioners can achieve various certifications (green belt, black belt, etc.).
The Lean Method
The Lean method began in manufacturing, and has been used successfully for decades, particularly in Japan. These principles are now being applied successfully to the delivery of health care. The approach is driven by the concept of value stream mapping, and requires less training and infrastructure. Practitioners learn by doing, and projects are short in length (one to three weeks) and are based on best practices. The basic premise is that anything that does not add value to the patient's care should be eliminated. Lean methodology seeks to weed out waste in health care processes.
In instituting value stream mapping, the team looks at every operation from the perspective that each step should add value to the patient's care. If it does not, ways are sought to eliminate that step from the process. The goal is to drive out waste so that all work adds value and serves the patient's needs. Toyota has put these principles and practices to use with great success in its plants. Table 1 compares and contrasts the two methodologies, called Lean Six Sigma, which increasingly are being used together. Virginia Mason Medical Center in Seattle has been using Lean methodology to redesign processes to eliminate waste, and has saved more than $1 million. (Womack J, et al. “Going Lean in Healthcare.” Institute for Healthcare Improvement, Innovation Series 2005.)
Rapid Cycle Testing
A newer tool for your QI toolkit is Rapid Cycle Testing (RCT), which has great potential in health care settings. This is an easy-to-learn, easy-to-teach, results-oriented model for change. With RCT, changes in operations occur on a small scale, and the results inform subsequent changes. It is perhaps the nimblest of the tools presented here. Not only is staff resistance less likely (because only small changes are being made to existing processes), but staff members are encouraged to become involved in these changes to their protocols, procedures, and processes. The work team asks three questions:
* What are we trying to accomplish?
* How will we know that a change is an improvement?
* What change can we make that will result in an improvement?
David Hnatow, MD, of the University of Texas Health Science Center, implemented a number of small changes in ED processes to achieve greater efficiency. One RCT project was low-tech, inexpensive, and the brainchild of front-line staffers. It involved using colored slips of paper in jars to cue housekeeping staff when rooms needed cleaning and in turn to cue clerks when rooms were cleaned. The turnaround time on inpatient beds was reduced by more than 100 minutes in just three weeks! (Urgent Matters e-newsletter, 2004; 1; www.urgentmatters.org/enewsletter/vol1_issue4/BP_rapid_cycle.asp; cited Oct. 19, 2005.)
The Institute for Healthcare Improvement offers a close variation of RCT called the PDSA cycle: Plan-Do-Study-Act in its toolkit. (Table 3.) The PDSA model is shorthand for testing a change in a real work setting. (How to Improve. Institute for Healthcare Improvement; www.ihi.org/IHI/Topics/Flow/PatientFlow; cited Nov. 4, 2005.)
The PDSA cycle and RCT involve this small, ongoing, and incremental change methodology that is very effective for quality improvement projects, but completely at odds with traditional medical research methodology! Usually more than one cycle of PDSA is needed to adopt a change. Monitoring whether the change is being implemented as planned is important. Often, more than one change may be carried out at once. Plot the data over time to judge progress toward the desired goal.
At the recent Urgent Matters Conference in Las Vegas, Kirk Jensen, MD, MBA, an expert on patient flow at the Institute for Healthcare Improvement, offered tips to help implement change within the emergency department. He noted that change is uncomfortable for everyone so preparation before and highly visible buy-in from the top down are key. Medical directors, nurse managers, and administrators need to support any initiative in a way that staff can see.
High-impact teams are the newer, nimbler model for the traditional work group. These smaller groups are made of three to six people representing the stakeholders in the process, and they are empowered to make changes. The members of the high-impact teams do most of the groundwork before so meetings are brief, focused, and effective. The meetings are where the solutions get hammered out. Three to six meetings per project are optimum, though frequent informal huddles or phone calls to monitor progress are utilized and are effective.
Data married to powerful anecdotes are compelling and should never be underestimated in the process of change at the organizational level. Crude data are better than no data, and he who owns the data wins!
QI experts can attest to the observation that some of the best solutions come from the front line, from those who intimately know the operational details of a system. They also have the advantage of having innate buy-in from staff. This is preferable to having administrative mandates imposed from the top down, which may be resisted on principle.
Two web sites are particularly useful for ideas about improvement, success stories, and information on quality improvement tools. The web site for the Institute for Healthcare Improvement (www.ihi.org) allows users to download its monthly newsletter and reports for free. The site offers several white papers on ED QI. “Optimizing Patient Flow” and “Forming the Team” are worthwhile for emergency physicians trying to launch QI efforts.
A web site sponsored by the Robert Woods Johnson Foundation (www.urgentmatters.org) is dedicated to health care improvement. It offers a free newsletter describing quality improvement initiatives from around the country. The newsletter, for example, recently discussed Rapid Cycle Testing, patient flow coordinators, and the dissolution of triage.
Both organizations have conferences dedicated to ED operations and flow. The Emergency Department Benchmarking Alliance sponsors a conference each year in Florida, and more information about this year's meeting, which was held in early March, “ED Benchmarks 2006: Discover Real World Solutions to Common ED Problems,” is available at www.emlrc.org/edbenchmarks2006.htm.
© 2006 Lippincott Williams & Wilkins, Inc.