Efficiency, economics might not translate from small to big HM groups
by John Nelson, MD, FACP
Editor’s note: This is the second of a two-part series addressing issues at large HM groups.
Last month (“The Bigger They Are…,” May 2009, p. 58), I discussed the difficulty large HM groups (more than 20 full-time equivalent hospitalists) face in trying to ensure that nurses and other staff always know which hospitalist is attending a patient daily, as well as issues raised by the common practice of separating admitter and rounder shifts. This month, I want to address patient distribution and economic issues faced by large groups.
Large and small groups often work diligently to ensure all rounding doctors start the day with a nearly identical patient load. Such “load leveling” might take only a few minutes in groups of two or three rounding doctors, but it may take up to an hour a day if there are eight to 10 rounding doctors. Think about what this costs a large group. If a group has eight rounders spending the first 30 minutes of each day distributing patients, the practice is devoting 1,460 hours annually to this function. Those 1,460 hours equate to 0.7 FTE, and if each FTE costs the practice $220,000 annually in salary and benefits, then the practice is spending $154,000 per year to distribute patients each morning.
Is that the best way to use $154,000?
An alternative is to establish a system that allows the evening and night admitters to know in advance which rounding doctor will assume each patient’s care the next morning. The night/evening admitters would then write “admit to Dr. Satriani” for the first new admission, and “admit to Dr. Johnson” for the second, and so on. The hospital would never list evening/night admitters as a patient’s attending on the chart or in the computer. And each rounding doctor could arrive in the morning to find a list of new patients from overnight, eliminating the need for a meeting of all rounding doctors just to distribute the patients. There may be other reasons to meet each morning, such as case management rounds, but eliminating the need to spend time distributing patients will make the meetings shorter and get everyone out to the floors to see patients more quickly.
There are two simple ways the evening/night admitters can know how to assign new patients to the rounding doctors for the next morning. If the group wants to have each rounding doctor start with a nearly identical patient load, then the rounding doctors could indicate their load at the end of each day and the evening/night admitters would follow an algorithm of assigning the first admissions to the doctor who finished the prior day with a lighter load. The other option is for evening/night admitters to assign new patients through the night according to a fixed protocol, which wouldn’t vary based on the current patient load of each rounder. Of course, this will mean daily patient load could vary significantly from one rounder to the next, but over any long period, the workloads will tend to even out.
Hospitalists and hospital executives tend to have a different view of the benefits of practice growth to require more hospitalist FTEs. The hospitalists themselves often are convinced that when the group has more doctors, there will be more scheduling flexibility for each individual doctor, and perhaps each doctor will have to work fewer weekends. Sadly, neither is true to any significant degree in most practices. Some aspects of scheduling are easier when there are more doctors. For example, it usually is easier to find someone to fill in for an unexpected absence in larger groups. However, each doctor’s schedule usually doesn’t get much better or more flexible.
Hospital executives, or whoever is responsible for coming up with funding to support the practice, often look at a larger practice as one that can take advantage of economies of scale. For example, executives may project that when the practice is larger, the hospital’s contribution to the practice on a per-FTE or per-encounter basis will become smaller. But just like the elusive scheduling benefit of larger groups, few practices realize any economies of scale. The vast majority of the costs in most programs are provider labor costs, which scale with program volume. So in most cases, the larger a practice becomes, the larger the overall hospital financial support will be on a roughly linear basis.
More often than not, night-shift work does become more cost-effective as practices grow. For example, an in-house night shift for a practice of eight FTEs might generate $200 to $400 in collected professional fee revenue each night, leaving the hospital to pay the remaining $700 to $900 each night. (These numbers are for illustration and aren’t intended to represent benchmarks or realistic targets for any practice.) But as the practice grows to support 20 FTE hospitalists in total, nights get busier. The night doctor might average three or four admissions per night in an eight-FTE practice but could average 10 or more in a 20-FTE practice. Those 10 admissions might generate around $1,200 in professional fee revenue, leaving the hospital to contribute only a small fraction (about $200 per night) of the cost of each night shift. So night shifts typically require diminishing dollars of hospital support as the practice grows.
Unfortunately, any economy of scale for night-shift coverage usually is offset by inefficiencies and costs that larger practices incur—and small practices often don’t—such as a significant amount of management infrastructure, professional administration, dedicated physician leader time, and an information technology infrastructure. These resources often are necessary to manage the complexity of a large practice, but every practice should challenge itself to demonstrate that these things actually improve the practice’s efficiency and performance enough to justify the money spent on them.
The larger the practice, the more likely there have been attempts to implement a triage pager system in which all new admissions and consults are routed to one pager (the triage, or “hot,” pager), which is held by one hospitalist at a time. In large practices, the “triage hospitalist” is usually so busy answering pages that they can do little else. And in many cases, ED doctors may describe a new admission to the triage hospitalist in detail only to have the triage hospitalist pass the information along to a colleague who will actually see the patient. This is an inefficient chain of communication, and I think most groups could do away with the triage pager. I described this issue in detail in my December 2008 column (“Technological Advance or Workplace Setback,” p. 69).
Large hospitalist groups work in large hospitals and end up doing a lot of inefficient walking between nursing units during the day. They may have patients on 10 or more nursing units and end up spending only a little time on each unit, which probably diminishes the “constant availability” that most see as key to the hospitalist model. So many groups decide to have each hospitalist cover only a small number of nursing units. This really has become a hot topic in the past couple of years, and I discussed it in detail in my September 2007 column (“Unit-Based Hospitalist Practice,” p. 84).
Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson/Flores Associates, a national hospitalist practice management consulting firm. He also is part of the faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program” course. This column represents his views and is not intended to reflect an official position of SHM.
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.