Published in the August 2013 issue of Today’s Hospitalist
Editor’s note: This is the third in a series of articles on how one hospitalist program successfully launched unit-based care. The full series covers four themes:
IN MY PREVIOUS TWO ARTICLES, I discussed the problems we faced at Presbyterian Hospital in Albuquerque that led to the inception of unit-based care with whiteboard multidisciplinary rounds, as well as the political and financial challenges we had getting people on board to launch the model.
But we did launch our first geographic unit “called Unit Base One “in April 2010. That “unit” consisted of both a telemetry and nontelemetry floor with approximately 52 beds and four rounding teams. (Each team consisted of one hospitalist and one care manager.) We were committed to a daily census for each team of 12 or 13.
During the first few weeks, I as the hospitalist medical director, along with the hospitalist lead physician, case manager supervisors, nursing director and nursing supervisors all attended the morning whiteboard meetings to set expectations regarding hospitalist and nursing attendance.
Skepticism and suspicion
We started with two groups of nonphysicians solidly in our camp: The physical therapists were already geographically isolated and welcomed the idea of whiteboard rounds, while the care managers had their patient assignments drawn for the unit-based model.
Meanwhile, the hospitalists assigned to Unit Base One were all volunteers with wide-ranging motives. Some were truly progressive and loved the idea of being part of a major quality improvement process. Others were more mercenary and drawn to the prospect of a much lower census.
And the rest of the hospitalist group? Most were skeptics who thought we wouldn’t be able to realize the promise of maintaining a lower starting census and greater efficiencies through geographic isolation. They had running bets on how long it would take before the project imploded.
Many of the nurses shared that skepticism. Initially, their attendance was very sporadic because many suspected the hospitalists would not bother to show up at the whiteboard rounds. They too expected the new model to collapse within a few weeks.
Turning things around
The key factor that turned that suspicion around was that hospitalist attendance at whiteboard rounds was 100% from the start. Once nurses realized that the hospitalists were committed to discussing the care of their patients with nurses during whiteboard rounds, their attendance began to improve.
It also helped that a round of nursing turnover coincided with the implementation of Unit Base One. Some nurses were so set in their routines that they did not want to communicate or interact with the hospitalists in this systematic process. These nurses resigned or asked to transfer to different floors, which nursing supervisors saw as a necessary turnover for the floor.
Once Unit Base One became fully operational, throughput started to improve due to shorter average length of stay (aLOS). But one big downside for the nurses was that shorter aLOS meant more beds turning over “and more work. Making matters worse, nurses didn’t get credit for that extra effort because it was not captured by the overnight census used to gauge nursing productivity.
Even worse, nurses on non-unit based floors had easier workdays because patients’ average length of stay was higher. While this was a huge dissatisfier for nurses, we weren’t able to resolve it until the rest of the floors converted to unit-based care. At that point, nursing workloads were evened out across the units.
The unit-based leadership team met with the hospitalists once a week to get feedback and discuss issues, while the nursing director and charge nurses met with the nurses. A unit-based steering committee with unit secretaries, nurses, hospitalists, physical therapists and care managers met once a month.
The feedback from the nurses, physical therapists, care managers and hospitalists was very positive. There was a general consensus that unit-based care improved communication and established a sense of community and cooperation “which, everyone agreed, didn’t exist before.
By then, the staff on non-unit-based floors began hearing about that change in culture, and word quickly spread among the nurses about the huge benefits of having the hospitalists available not only during whiteboard rounds, but throughout the day. That’s when nurses from non-unit-based floors started asking when the model was coming to their floor!
Within four months, the data started to show a significant trend in improved aLOS, which translated into real financial improvements. Armed with that data, we approached the administration to spread unit-based care to two additional medical floors, to be named Unit Base Two. We quickly came up with a business proposal for that second unit, and Unit Base Two was launched in September 2010.
The hospitalists who were not part of Unit Base One had kept a very close eye on the patient census there, and they quickly realized that the census was much lower for unit-based rounders. (Non-unit-based rounders, by comparison, had no natural cap.) That brought about a noticeable change in their attitude toward unit-based care, and we had no shortage of hospitalist volunteers for Unit Base Two.
And because the model at that point was no longer a pilot, nursing directors decreed that nursing attendance at whiteboard rounds on Unit Base Two would be mandatory. With nurses on Unit Base Two wanting to participate, none of them pushed back against that mandatory policy.
The financial case
With Unit Base Two, the finance department wanted to approach the financial modeling differently. With the first unit, they looked at how decreasing aLOS would add to the contribution margin.
For Unit Base Two, however, they wanted to see how reducing aLOS would decrease variable expenditures. After 21 months, Unit Base One had added $2.2 million to the contribution margin. After 15 months with Unit Base Two, we reduced variable expenditures by $1.3 million, which was due to patients’ spending less time in the hospital.
The adoption of unit-based care also allowed the hospitalists to increase their number of admissions from 12,503 in 2010 to 14,411 in 2011, a 15% increase. We achieved that by capturing all the patients we had been losing through our ED who left without being seen. As a result, the percentage of patients in the ED who left without being seen dropped from 10% in 2010 to 2.7% as of May of this year. At the same time, ED diversion during these last two years has been unheard of.
And because of the dramatically increased throughput, the hospital was able to close the nine-bed temporary holding unit that we had used to clear out our ED every morning. That unit, which was started in 2009, closed in summer 2011, eliminating $800,000 a year in fixed costs.
With the success of Unit Bases One and Two, we were able to start Unit Base Three in May 2011. Our final Unit Base Four on the cardiac floor opened in January 2013, and we now admit for our entire cardiology department.
Currently, unit-based care is in place on all of the medicine floors at Presbyterian. Our YTD aLOS as of the end of June this year is 4.14 days. This represents a decrease of 0.92 days since 2009, when aLOS was 5.06. Our group is projected to perform well over 15,500 admissions this year.
Beyond the throughput and financial metrics, unit-based care has transformed our entire group and hospital. In my next and final article, I will discuss the transformative changes that we did not anticipate.
David J. Yu, MD, MBA, is the medical director of adult inpatient medicine services with Presbyterian Medical Group at Presbyterian Hospital in Albuquerque, N.M. Send him questions about starting or maintaining unit-based interdisciplinary rounds at email@example.com.