Real Q.I. Improvements

1 Comment

altImagine you are the CEO of St. Elsewhere Hospital, which is a suburban hospital with a 35,000 yearly visit emergency department (ED). The ED is a mess: average door-to-doctor times of 110 minutes, a left-without-being seen (LWBS) rate of 7%, and poor Press Ganey scores. Most everyone in the hospital, including the patients, refer to it as “That ER.”

Why it takes leadership – plus a good idea – to make and sustain positive change


Imagine you are the CEO of St. Elsewhere Hospital, which is a suburban hospital with a 35,000 yearly visit emergency department (ED). The ED is a mess: average door-to-doctor times of 110 minutes, a left-without-being seen (LWBS) rate of 7%, and poor Press Ganey scores. Most everyone in the hospital, including the patients, refer to it as “That ER.”

One day, you decide to fix the ED. You hire a pin-striped, expert consultant who comes to your ED for two days, surveys the place, and talks to the staff on the day and night shifts. He then, smilingly, recommends that you implement a fast track and put another nurse at triage (and leaves a hefty bill). You’re sure your pricey consultant is correct, so you make an email announcement of the consultant’s recommendations to notify the ED staff. One-week later, Voila, your ED is fixed! Door-to-doctor times are now under 30 minutes, LWBS rates are down to 2%, staff is happy, and the patients are even happier. Even that curmudgeonly cardiologist who always spoke badly of the ED is now singing its praises.

While this may describe a hospital CEO’s fantasy, experienced ED staff know there actually is nothing “real” about this story.


In reality, quality improvement is about leadership from within the ED. And recently, two “real” stories by ED leaders were highlighted in February and July 2011 webinars by Urgent Matters, which is a Robert-Wood Johnson Foundation funded program dedicated to improving ED flow and quality.

The first webinar was titled, “Low-Cost High-Impact Patient Flow Strategies” and delivered by Rodney Borger, MD who is the Chairman of Emergency Medicine at Arrowhead Regional Medical Center, which is a large public hospital in California. Arrowhead Medical Center is a teaching hospital with about 130,000 visits annually. It serves as a major referral and Level 1 Trauma Center and sees primarily an urban, indigent population.




The intervention was a reengineering of the front-end of the ED. The goals were to reduce door-to-doctor times, to reduce elopements, and improve satisfaction along with improving patient safety, staff retention, and teamwork (and hopefully revenue too). Dr. Borger described how he had to sell the plan to hospital administrators, which required $60K in upfront investment , and ensure them that they would recoup the costs. More importantly, it also had to be sold to staff, and many attending staff were initially nay-sayers.

The project team was a multi-disciplinary group of ED staff who first conducted flow-charting of the “ideal” process to re-engineer the waiting area. The group even conducted a real simulation in which 30 volunteers moved through the new process within the existing space to show it could be done. Next, they bought a series of well-equipped office-style cubicles that replaced a large portion of the waiting area. When the process was implemented, Dr. Borger and other leaders spent about “100 hours in the triage area in the first week,” and many wanted to revert to the old way.  

In the new process, after walking into the ED, patients presented directly to a nurse who conducted an assessment and determined whether the patient needed an “immediate bed” or not. Stable patients would go to one of the cubicles where they would be further evaluated by a physician or mid-level provider. In this process, 35% of patients could be immediately discharged from the ED. Some patients required additional testing, or would be moved into the higher acuity area. If a patient required a more detailed evaluation, a second physician would assume care, but for more minor cases, the first physician will complete the entire case.

From 2002 to 2009, the waiting time at Arrowhead decreased from 4 hours to less than 30 minutes in 2009. These improvements have been sustained despite dramatic increases in volume. Over the same time period, the LWBS rate went from 20% to below 0.3%. Patient satisfaction has also increased dramatically.

The second webinar was titled, “Rapid Intake and Patient Segmentation” and delivered by staff at Christiana Care Hospital in Newark, DE including Karen Toulson, RN, MSN, ED Nurse Manager, Heather Farley, MD, Assistant Chair of Emergency Medicine, Jason Deal, ED Technician and Jesse Moncrief, Senior Operational Excellence Consultant.

The group described four initiatives that had been implemented over the previous years.

The first was Supertrack for Emergency Severity Index (ESI) 4 & 5 patients. Prior to Supertrack, there was a 6-room fast-track staffed 8AM to 2AM with an average length of stay of about 2.5 hours. Prior to 2008, several initiatives had been tried and failed, and there was a push by hospital administration to improve. The group used Lean process mapping and spaghetti diagrams to diagnose the problem and determine where there was waste. The new approach used two teams working out of two rooms. Work was standardized and synchronized so that the provider and RN would take the history at the same time, and the tech would make sure that patients were ready to be examined. The implementation of Supertrack resulted in a change in median length of stay from 122 minutes to 62 minutes, and LWBS fell from 4.6% to 2.5%.

The second intervention was SPEED (Synchronized Provider Evaluation and Efficient Disposition) which targeted ESI 3 patients. Similar to Supertrack, one of the major goals was to reduce over-processing through synchronized evaluation of patients, verbalizing the plan of care, using intramuscular or oral medications if possible, ordering imaging studies only as needed, and to have the patients in and out of assessment rooms in less than 40 minutes. Also, when results come back, the goal was to disposition the patient as early as possible. Two nurses worked with each team to either initiate the work-up or complete follow-up orders, and alert the provider when disposition was possible. Techs perform point-of care testing, ensure the rooms are full, and keep the process moving. After SPEED was implemented, overall length of stay for ESI 3 patients fell from 5.5 to 3.5 hours. In addition, LWBS fell from 6.9% to 2.5% for ESI 3 patients. Door-to-doctor times also improved dramatically as did productivity.

The third intervention was SHARP (Synchronized Healthcare Approach and Redesign Process), aimed at ESI 2 & high-level ESI 3 patients. Similar to SPEED, both physicians and nurses evaluate the patient at the same time. The clerks on the team post admission orders, follow-up orders, and actively monitor inpatient bed status and imaging results. They also have implemented a synchronized discharge process that has cut about 30 minutes from the discharge time. In this process, the entire team discharges the patient (nurse & physician together).

The most recent initiative was called TART (Triage Assessment and Redesign Team). Similar to the other process changes, the triage process was coordinated where a tech and nurse would triage patients, give early medications, and do point-of-c
are testing. When beds were available, they would “pull till full” (i.e. immediate bedding). There has been some push back from the staff with this new program. Specifically, there has been resistance by main side nurses who find they need to do more when less is done in triage. In addition, this has put more pressure on the physician to see more patients. Since this was just implemented in April, the team did not have formal data to present, but they reported that the initial data demonstrates this has improved the door-to-doctor time.

Finally, the Christiana team highlighted the importance of monitoring and maintenance in quality improvement. This involves regularly monitoring performance to prevent backsliding, and intermittent high-level analysis to make any additional changes. Daily data is sent to the ED leadership on ED flow so that real-time feedback can be given to the providers. In addition, they have created an overall dashboard that can be used to monitor the ED in real-time (figure 1).

Stepping back, several important distinctions can be made between our “fake CEO” story and these two real stories, highlighted in Urgent Matters webinars.

First, real ED quality improvement is about leadership championing from within, not from above. Both programs were originated and lead by multi-disciplinary teams within the ED. However, there was needed “buy-in” at the highest levels. Second, real ED quality improvement requires consistent communication, and leadership must be hands-on. It is also vital that get buy-in from the entire staff so the inevitable criticisms of the nay-sayers will be drowned out by the cheers of supporters. Third, real quality improvement doesn’t happen overnight, especially in a complex system like an ED. Quality improvement occurs over months to years, as was demonstrated in the stories at these two hospitals. It is a continuous process.

Finally, often the best-laid plans need to be changed as the team finds new important issues to address. Ultimately, success is often driven, not just by a good ED-based leaders OR a good idea, but it really needs both. That seemed to be the real key to success in both of these stories.

Jesse M. Pines, MD, MBA, MSCE is the Director of the Center for Health Care Quality, and an Associate Professor of Emergency Medicine and Health Policy at George Washington University. He also serves as the Principal Investigator for Urgent Matters.


1 Comment

Leave A Reply