At Parkland Memorial Hospital in Dallas, emergency department patients once had to wait a median of three hours for inpatient orders once doctors decided to admit them. Closing this gap wasn't easy: Three separate independent consultants tackled the problem and failed. But after Parkland's physicians, hospitalists and ED staff got involved in an effort to streamline processes, the median time between the admission decision and inpatient placement decreased to just an hour. The new system, which went live on July 1, saved more than 2,000 hours of patient care in the first two months alone.
Parkland, a 725-bed hospital that is always at capacity, discovered the root of its ED-to-admission problems after an evaluation of throughput efficiency, which suggested that the interval between the ED's decision to admit a patient and the internal medicine department writing admission orders was too long. When outside consultants had no luck solving the problem, in part because they were unfamiliar with the hospital and its culture, administrators decided to look within.
A group of Parkland physicians and staff spent nine months analyzing every aspect of the ED/internal medicine interface, including transportation of patients, bed turnaround times, discharge predictions, the timing of orders, staffing levels and the role of the admitting officer on duty (AOD). The committee was led by Ruben Amarasingham, MD, MBA, Parkland's medical director for medicine, who is also a hospitalist and an assistant professor of medicine at University of Texas Southwestern Medical School.
“We concentrated on the phase that begins when emergency medicine decides to admit and ends when the ED receives written orders. Some months this process actually exceeded four hours,” said Dr. Amarasingham. “The hospital and the medical school administrators recognized that this was something that was worth tackling.”
Brent Treichler, MD, assistant professor at University of Texas Southwestern and medical director of emergency services at Parkland Health and Hospital Systems, was also integral to the streamlining process.
“Like Dr. Amarasingham, I was also a resident here, and we had always been thinking about ways to more rapidly get the patients upstairs so that they could receive the care and the treatment they needed,” Dr. Treichler said. “[Partnering] with Dr. Amarasingham seemed like a natural fit; we're both relatively young in our careers, so we were willing to try new ways to fix this age-old problem.”
Getting support and buy-in from every level, from nursing through to top-tier administrators, was critical to the project's success, the physicians said. “We could have made all the plans we wanted, but unless we had significant buy-in from high levels in the hospital as well as our own medical school department, it would have gone nowhere,” Dr. Treichler said.
Gary Reed, FACP, Parkland's vice chair of internal medicine, chief of general internal medicine at University of Texas Southwestern Medical School and director of the hospitalist program, agreed that holding discussions with all of the stakeholders was a key part of the planning process. “I don't think anyone would have agreed to the changes otherwise, since it was such a major change from the way we did things in the past,” he said.
Drs. Amarasingham and Treichler recognized early on that they had to prepare everyone for significant cultural changes that might require some transparency and scrutiny.
“One of the ways we dealt with the unease was to say, ‘I practice in this institution, too, and we are all going to be scrutinized equally,’” Dr. Treichler said. “Everyone eventually understood that these were really systems-based problems and not the result of any individual clinician; over time, this generated a tide of support.”
The group began to track simultaneous actions and behaviors throughout all departments in order to identify what was causing the bottleneck in the ED.
“We built an electronic database which fed from multiple hospital information systems; this allowed us to map discrete clinical processes in real time. In addition to patient times and throughput, the database included diagnosis, risk, comorbidities, timing of orders and destination unit, for example. The data helped us pinpoint obstacles to both quality and efficiency. We also spent a lot of time deconstructing the [hospital] ‘mythology’ that had developed over time and was limiting cultural change,” Dr. Amarasingham said. “Longstanding ideas about what, or who, constituted the problem turned out to be false.”
The team created a 10-foot-by-6-foot map of all of the processes that occurred at the hospital from ED arrival to admission. “We walked through the entire process from patient arrival by ambulance to arrival into a bed, and found that in complex cases the process could take up to 500 to 700 individual steps, sometimes parallel, sometimes sequential,” Dr. Amarasingham said. Once the map was created, the group overlaid data from their electronic database, revealing the steps that created the greatest bottlenecks. Finally, they considered cultural aspects—both good and bad—that were also in play, Dr. Amarasingham explained.
Once the map was mocked up, the committee solicited feedback from clinical leaders throughout the institution—a process that alone took two and a half months. Next, the committee created individual groups of internal medicine and ED physicians who brainstormed solutions for very specific aspects of care.
“With each group, we talked about what could go wrong, what wouldn't go wrong, [and] what was preventable, and out of this process bubbled some of the best solutions,” Dr. Amarasingham said.
Give and take
Negotiating those solutions was challenging, the group found, because at that point sacrifices needed to be made. Departments had to give up control of things that had been in their domain or take on new responsibilities that could easily be perceived as getting extra work dumped on their already overflowing plates.
The committee members then realized that July 1, when the new interns began work, would be the ideal time to implement the new process. “We began a mad rush to try to get everyone to agree to the final plan,” Dr. Amarasingham said, meeting with all of the department chairs, chief residents and faculty members. “In the three months leading up to July 1, when we were trying to secure agreement from all of the departments, we went from four major drafts to 52 major drafts. It took an enormous amount of negotiations, and there were times when we wondered if it was going to happen at all. But we were able to get everyone to agree on the final process at 2 a.m. on July 1. We went live at 7 a.m.”
To Jennifer Sharpe, RN, Parkland's director of emergency services and co-chair of the committee for capacity management, the project's success was even more impressive given that the flawed system had about 100 years of culture behind it.
Like most hospitals, Ms. Sharpe said, Parkland's ED and internal medicine staffs had very different perspectives about patient treatment, which contributed to the protracted ED-to-admission wait time. ED personnel thought internal medicine staff asked for tests that should have been done once patients were on the floor, and internal medicine thought that if the ED ordered the right tests in the first place, there wouldn't be a holdup. Meanwhile, the AOD was seen as someone who duplicated the ED's efforts, which compounded both the waiting time and the animosity between the two departments, she explained.
“Before, it felt like our patients were in the ER because we couldn't get them admitted efficiently, and that created dissatisfaction. We provide excellent care, but the ER is not a comfortable, stress-free environment,” Ms. Sharpe said. “Knowing that we were able to make this change to moving patients to where we feel they are more comfortable in a more controlled environment was a big plus for us. From the nursing perspective, I think that was our biggest win.”
Ms. Sharpe sees the success of the process as a product of the open, honest dialogue between departments. “The process gave us a forum to legitimately review and discuss cases as colleagues. For the ER nursing staff, this change has engendered a feeling of teamwork with internal medicine. There used to be great animosity between the staffs, especially aimed at the AOD from the ER staff, and that's no longer true,” she said. “The AOD is more of a colleague now. We're working with that person trying to solve the patient's crisis, doing what we need to do but hurrying it along at the same time. That aligning of goals puts us on the same page.”
Now that Parkland has addressed its ED-to-admission problems, its next challenge is tweaking the processes and keeping them going. “To say that this is all 100% easy from the day it was implemented would be wrong,” said Dr. Reed. “We're still working out the kinks.”
Establishing and maintaining a plan to prevent reversion to long-standing, counterproductive habits is an integral piece of the puzzle, according to Dr. Treichler.
“There can be a lot of slippage, so it's important to have a very active process improvement committee that will meet to address problems that crop up,” he said. “It ensures that the lines of communication are kept open.”
Rochelle Nataloni is a freelance writer in Sewell, N.J.
Have a success story?
If your hospitalist program has helped solve a problem or improved processes, let us know about it at acphospitalist[EACHAT]acponline.org. The best ideas may be featured in upcoming issues.