EXECUTIVE SUMMARY
The ED at the University of Colorado Hospital (UCH) has undergone a dramatic transformation in recent years, doubling in size while also using process improvement methods to dramatically reduce wait times, eliminate ambulance diversion, and boost patient satisfaction. Throughout this period, volume has continued to increase while the cost per patient and avoidable hospital admissions have experienced steady declines. Guiding the effort has been a series of core principles, with a particular focus on making sure that all processes are patient-centered.
- To begin the improvement effort, ED leaders established a leadership team, and hired a process improvement chief with no previous experience in healthcare to provide fresh, outside perspective on processes.
- In addition to mandating that all processes be patient-centered, the other guiding principles included a commitment to use and track data, to speak with one voice, to value everyone’s perspective, to deliver high-quality care to all patients, and to set a standard for other academic medical centers.
- To get points on the board early and win approval from staff, one of the first changes administrators implemented was to hire scribes for every physician so they wouldn’t be bogged down with data input. The approach has essentially paid for itself.
- Among the biggest changes was the elimination of triage, a process that improvement teams found no longer added value or quality to the patient experience.
- Leadership also has moved to equilibrate the size and staff of the various zones in the ED so that they are more generic and less specialized. The move has facilitated patient flow, enabling patients in zones with resuscitation bays to connect with providers quickly.
EDs are busy and complicated, so changing the way things are done always is a monumental task. Physicians and nurses generally are adept at guarding their own turf — as are all the other key staffing groups, from the techs and transporters to environmental services staff. Thus, even simple adjustments can prove devilishly difficult.
Despite these obstacles, the ED at the University of Colorado Hospital (UCH) in Aurora, CO, offers compelling evidence that large-scale change is not only possible, but it also can deliver dramatic improvements in many of the key metrics that ED administrators track, including wait times, leave-without-being-seen (LWBS) rates, patient satisfaction, and costs.
Certainly, a transformation of this magnitude requires commitment and resources, but administrators note that another key weapon that continues to be leveraged in the ED at UCH is a fresh set of eyes.
Put the Patient First
Richard Zane, MD, chief innovation officer at University of Colorado Health and chair of emergency medicine at the University of Colorado School of Medicine, took charge of the ED’s change process when he became the hospital’s emergency medicine chief in 2012.
At the time, he relates that the department logged a sky-high LWBS rate, it was on diversion roughly one-third of the time, and patient satisfaction scores were dismal. A major expansion of the ED was planned to serve the roughly 230 patients who were presenting to the ED daily, but the hospital’s senior leadership team concluded that structural changes would not solve all the ED’s deficiencies.
To chart a course toward improvement, Zane assembled a senior leadership team, which then agreed on six core, guiding principles. The most important one was what Zane refers to as patient centricity.
“Everything we do is for our patient; that is why we exist and it is how we design and run the department,” he says. “If we are going to implement change, it has to be around that, not around what is most convenient for the provider, and not around what is best for the financial enterprise, but what is best for the patient.”
In practice, every proposed change or improvement was scrutinized according to that primary core principle.
“Nothing is considered unless you show how it is going to improve patient care. Nothing. And we were emphatic about that,” Zane recalls.
The other guiding principles included a commitment to use and track data, to speak with one voice, to value everyone’s perspective, to deliver high-quality care to all patients, and to set a standard for other academic medical centers, but the patient centricity piece was the hallmark of the change effort.
A big believer in industrial engineering methods, such as Lean and Six Sigma, the first person Zane hired when he arrived at UCH was Derek Birznieks, an industrial engineer who had worked for Caterpillar, the tractor and heavy machinery manufacturer.
“I very much wanted an outsider’s view of process engineering — not just an outsider who had never been in this institution, but an outsider who had never been in healthcare,” Zane explains.
Zane thought this fresh perspective was necessary to eliminate steps or processes that may have made sense at one time, but were no longer ideal or efficient.
“A third of the things that we do that don’t seem sensical can be explained because of medical necessity, but two-thirds of them are truly non-sensical,” he says. “Like many processes in healthcare, there is a lot of dogma around why we do them without any evidence.”
Perhaps the biggest example of a process that was no longer adding value from the patient’s perspective was triage, Birznieks observes.
“The nurses thought it would be unsafe [to remove triage]. They didn’t understand why we would take that away,” he says.
However, Birznieks notes that when an improvement team walked through what triage meant for the patient experience in terms of added steps and delays, views began to change.
“When we mapped it out and showed them what it was actually doing, it was clear that [triage] wasn’t a patient-centered process,” he says. “From big things like triage to little things like standardization ... it wasn’t that anyone was pushing back; they just didn’t realize that what they were doing every day was adding waste and time to the process, which affects the patient.”
Birznieks, who is now the senior director of process improvement for the entire University of Colorado Health system, says that when he was working in the ED, his role was to facilitate process improvement events or team meetings, and make sure participants stayed focused on the issue at hand and stuck to the plan.
“We wanted to make sure that we were patient-centered and that we were always driving decisions with data,” he says. “When [I and my process improvement team members] would meet with people new to the process ... they tended to have a more traditional mindset of how they were used to caring for patients, but then the more we were exposed to them, the more events we had and the more data capture we did ... the less that mindset was prevalent.”
Birznieks acknowledges that not coming from a healthcare background was an asset initially because it gave him the freedom to ask why things were done a certain way, inviting new thinking about traditional processes, such as triage, that were no longer delivering optimal value.
“I could play the ignorance card,” he says. “But I gained credibility by showing that I knew what I was doing around process improvement subject matter. I didn’t need to be an expert in emergency medicine, but I had to know what I was doing from a Six Sigma/Lean perspective.”
As time went on, and Birznieks became more knowledgeable about medicine and healthcare delivery, he still could relate industrial management techniques to healthcare delivery.
“You don’t ever want to compare humans to machines, especially in healthcare, but being able to bring in analogies like retail to healthcare helps,” he says. “It gets people out of their normal thinking and out of that comfort zone of how they have always done things.”
Chart Early Wins
Convincing veteran staff to consider alternative approaches is a tall order, especially in healthcare, and the ED at UCH was no exception.
“There was a lot of discomfort ... because healthcare providers dislike change intensely,” Zane acknowledges.
To get around this obstacle, Zane and Birznieks prioritized communication with staff about what the guiding principles of their improvement effort were going to be, and what the change management process would involve. Further, to get points on the board early, one of the first changes Zane implemented was to hire scribes for every physician so that they wouldn’t be bogged down with data input. Of course, the move was a hit with providers, and it made sense from a financial and efficiency standpoint.
“From a straight activity-based costing perspective, [scribes] pay for themselves,” observes Zane, explaining that a physician with a scribe can see 2.2 patients per hour, compared with 1.9 patients without a scribe. What this means is that providing a scribe to a physician just requires the physician to see one more patient in that week, Zane notes.
“Change management [requires] celebrating some early wins ... so that was an early win that we celebrated,” Zane stresses. “And we specifically [implemented scribes] early so that people would know that we were making their lives better, not harder.”
Making use of scribes is just one way Zane endeavored to optimize the use of human capital.
“We had an industrial engineer follow every provider around and actually calculate how much time they were spending doing work that didn’t require their level of expertise,” he says.
For example, Zane notes that one of the smaller principles driving the improvement effort was that physicians and nurses should not spend time trying to find things they need to do their jobs.
“We had to define what things need to be right at the bedside, what needs to be in a central location, how you identify when you need those things in a central location, and who is going to get them,” he says. “This is so that when a doctor or a nurse goes to the bedside, everything is ready for them instead of going in and seeing a patient, and then realizing that they are going to need nose-packing for a bloody nose.”
By eliminating some tasks and streamlining others, clinicians now spend a significant amount of their time performing work that only they can, Zane notes.
“We have more transporters, more people doing stocking, and more people doing environmental services so that doctors and nurses and advance practice providers can spend the majority of their time practicing medicine,” he says. “In the end, that leads to more people, but at a lower cost per patient encounter.”
Select Leaders with Vision
Another way that Zane and Birznieks battled against staff pushback was by enlisting frontline staff to design new and better ways of doing things. In typical Lean fashion, they did this through small teams that were led by a physician and a nurse, often with the addition of an administrator or an industrial engineer.
“You take the key stakeholders and leaders who are involved in a process, and then you put them in charge of it,” Zane explains.
One must pick the right people to be in charge, Zane adds.
“You need people who can embrace change and do not have issues of job protectionism,” he says. “They can’t be worried about how redefining a care process or a care team will impact them individually, so you need visionary leaders on the team; it can’t just be one person. And then you have to lead by example.”
Birznieks stresses that part of his job was making sure that the organization’s core guiding principles were reinforced constantly — not just in process improvement meetings, but also in staff meetings and leadership meetings. For maximum effect, it is important to stress that this is not just a one-time or temporary push, but rather an effort to turn the ED into a continually improving department, he says.
“We spent a huge amount of time on the change management and cultural transformation piece [on the front end] so that when it came time to change specific processes, we weren’t doing it just for the sake of changing processes; we were doing it because we were changing the culture of the department,” Birznieks shares.
As new processes were implemented, Birznieks and ED leaders would round on the floor to make sure staff were adhering to them.
“More importantly, if staff were drifting [away from the improvements], we would ask why, and we would try to figure out why what the process improvement team came up with wasn’t working,” he says. In some cases, improvement teams would have to return to the drawing board to correct deficiencies in the original plan, he says.
Importantly, frontline staff had an outlet for discussing why a process was working or not working, and they had a stake in figuring out how to make it work better, Birznieks adds.
“[The process] is owned by the department or by the working area that we are engaged in. It is not owned by process improvement,” he says. “We are there to facilitate the group through a change, and we don’t necessarily have any skin in the game from an outcomes perspective. We are there to do the best for our patients and drive to a solution, but we don’t know what that solution will be because we don’t work in that environment every day.”
For maximum effect, an improvement effort must be more of a cultural transformation than a process transformation, Birznieks advises.
“What we did from day one [at UCH] was stick to our guiding principles around leadership, speaking with one voice, always being data-driven, and always being patient-centered,” he says.
Standardize Practices
In the past couple of years, Zane has focused on reducing practice variability in the ED.
“We have embedded more than 60 care pathways into the electronic medical record (EMR), which means that instead of simply having pathways that people can look at, now the pathways are embedded in the EMR, so if you follow a pathway, it populates the order sets in the EMR, decreasing variability,” he explains.
Early in the improvement effort, the ED investigators could document statistically and scientifically significant decreases in the variability and use of high-cost imaging through this approach, Zane explains.
“Now, we have done that with 60 different diagnoses,” he adds. “We have also partnered with a company to put clinical decision support around pharmacy ordering so that with the push of a button, a provider can get recommended medicines based on the information in the EMR instead of just free-forming it, so we have decreased the variability specifically on outpatient antibiotic utilization.”
Another big focus this year has been on improving the patient experience. There is a white board in every room with the names of the physician and nurse, Zane explains.
“We have also scripted certain encounters, such as how we discharge a patient, re-emphasizing what [his or her] instructions are, and what medicines they will be taking,” he says. “This is routine stuff, but now we have made it much more specific, deliberate, and uniform.”
Leadership also has moved to equilibrate the size and staff of the various zones in the ED so that they are more generic and less specialized, Zane says. This followed an internal evaluation that found that putting all the resuscitation bays in one zone was problematic.
“If there were multiple resuscitations at the same time, the physician and team [manning that zone] would do nothing but resuscitation, and the patients who were in the zone that did not need resuscitation would not get as much timely attention,” Zane explains.
“So, we rezoned the department so that the resuscitation bays are now split into two zones ... and we have equalized all the zones so that they have equal numbers of providers and beds, and there is a more consistent and constant flow,” he adds.
There have been multiple changes, and more are on the way as Zane and colleagues continually use data to drive improvements, and results are impressive thus far. Both volume and patient satisfaction continue to rise, and patients generally see a provider within a few minutes of arrival. Patients no longer leave the ED without seeing someone, and the ED never goes on ambulance diversion. The cost per patient and avoidable hospital admissions are also way down, according to Zane.
Make a Full Commitment
Achieving improvement on key data points is laudable, but early gains are notoriously difficult to sustain. That is one of the reasons why Zane has employed what he calls “active management.” This involves directing leaders to round in the ED periodically throughout the day to check with providers in all the different zones. The practice enables leaders to identify and assist providers with any issues or difficulties they may be experiencing, but it also provides them with an opportunity to see how the department is operating, Zane notes.
In addition to the rounding, leadership is updated three times a day on key metrics, such as how many patients are visiting the ED, what the average length-of-stay is, what the chief challenges are in the department, and what is going well.
Another must for sustaining early gains is to be all in, all the time.
“A lot of institutions in healthcare dabble in [process improvement]. They say this could fix some problems for them and then go do a Lean event in their ED, figuring they will see how it goes, and if it sticks, they might try it in some other places,” Birznieks explains. “If you dabble in it, it will not work. It has got to be full commitment from your leaders all the way down to the frontline staff. And if you don’t build that commitment level, [the improvements] will not be sustained.”
SOURCES
- Derek Birznieks, Senior Director, Process Improvement, University of Colorado Health, Aurora, CO. Email: [email protected].
- Richard Zane, MD, Chief Innovation Officer, University of Colorado Health, and Chair, Emergency Medicine, University of Colorado School of Medicine, Aurora, CO. Email: [email protected].
Administrators note the importance of establishing a leadership team structure and guiding principles to focus staff, drive continuous improvement.
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