Productivity and compensation benchmarks can be useful when negotiating with hospital administrators for increased reimbursements and support resources, when recruiting hospitalists, and when conducting self-evaluations. For many of these processes, hospitalists—and, indeed, hospital administrators—turn to the information contained in the voluminous SHM 2005-2006 Survey, “The Authoritative Source on the State of the Hospital Medicine Movement.” (See “For More Information,” p. 32.)
With a response rate of 26%, the survey represents some 2,550 hospitalists across the nation, and its variables present a more comprehensive aerial view of hospital medicine than did previous surveys. But on the ground and in the trenches, hospital medicine groups must be careful to look at the survey’s metrics with a discerning eye.
When applying the survey metrics to one’s own practice, there can be benefits as well as pitfalls, cautions Joe Miller, SHM senior vice president and principal analyst of the survey data. He emphasizes the great variation among hospital medicine groups and warns against looking at survey medians as representing a “typical” hospital medicine practice.
“When you’ve seen one hospital medicine group, you’ve seen one hospital medicine group,” he quips. In several recent conversations, hospital medicine group leaders and SHM leaders involved in compiling the survey discussed the survey’s strengths and limitations as a benchmarking tool.
Healthy to Negotiate
According to the survey 97% of hospitalist programs receive some type of financial support. “Virtually every program in the country is challenged to defend the amount of money [they receive] or to negotiate for support dollars,” says Miller, who believes that negotiation can be a healthy dynamic. “There is a sense of equality of both sides of the table, a mutual respect between hospitalists and the hospital.” In the process of such negotiations, it will be important not to pin one’s position entirely to the survey metrics.
John Nelson, MD, medical director of the hospitalist practice at Overlake Hospital in Bellevue, Wash., a consultant for hospitalist practices with Nelson/Flores Associates, a columnist for The Hospitalist (“Practice Management”), and a co-founder and past president of SHM, believes that some hospitalists mistakenly view the survey as SHM’s position on what a hospitalist should make. “The survey is the best information we have about what hospitalists do make—there is no better source—but it’s still a survey.”
Using compensation medians as yardsticks for actual salaries and compensation packages is analogous to “learning the average weight of an American and deciding that’s what we all should weigh—and that’s a big mistake,” he says. “If you hold up the survey as the governing document, then each party will use it to their advantage.”
Because the survey is regarded as the most authoritative existing source on hospitalists’ compensation and productivity, it nevertheless ends up being used as a benchmark, says Robin L. Dauterive, assistant director of the clinical hospitalist service at Massachusetts General Hospital in Boston.
“Whenever I’m preparing billings reports or dashboard measures—anything that shows my group’s workload—sooner or later, I always have to include something in there that states, ‘This is what other people are doing,’ ” says Dauterive. “It’s something that you can’t get away from, unfortunately, in medicine.”
She realizes that the survey does not purport to set any national standards, and yet, “all administrations want comparisons.” Dr. Nelson has also noted this phenomenon with the survey. In the absence of additional guidance, hospital executives and hospitalists often find that they’re just arguing about the survey. “And that’s unfortunate,” he says. “It means they’ve lost sight of the unique attributes of a given practice that might support higher or lower incomes and higher or lower workloads.”
View in Context
Hospitalists reading the survey for the first time might first seek to analyze the metrics regarding billings and collections. Here it is especially important not to view the reported numbers in isolation, says Dr. Nelson. For instance, to learn how a hospitalist’s annual gross charges (billings) compare with others across the country (question 12 of the Individual Hospitalist questionnaire—p. 87, Appendix 2), details on pages 251 and 252 supply pertinent variables. For instance, in comparing the four regions of the country, Table 056-A shows that the median annual gross charges for physicians in the south are highest, at $354,000. Hospitalists compensated by a 100% incentive method report higher charges per year ($392,000) than those who are on a 100% salary or a mix of the two methods of payment. Turning to Table 056-B, on page 252 of the published survey, hospitalists can find annual gross charges according to practitioner type, specialty, and employment model. Hospitalists should not stop their reading there, however, as a comparison of others’ annual gross collections might give a more complete picture.
Still, the SHM Survey does not reference all possible explanatory variables. Collections can be influenced by location and payer mix. Hospitalists practicing in a large urban hospital are likely to see more indigent patients for whom the hospital is not reimbursed. A careful reading of the survey should include the questionnaire and the tables supporting chapter conclusions, and the reader must recognize the survey’s limitations.
Apples to Oranges
IPC–The Hospitalist Company participates in the SHM survey and also uses it as a recruitment tool, reports IPC Vice President of Physician Staffing Timothy Lary. “We look at the income averages, and we’re able to demonstrate how our averages are, for the most part, higher than the averages,” he explains. “We also look at the survey from an internal viewpoint, but oftentimes you are comparing apples to oranges.”
Like individual hospitalists, hospital medicine group leaders seek comparisons when they read the survey. For her part, Dr. Dauterive has found the data on starting salaries for new hospitalists useful. For example, page 259, detailed table 060-A on hospitalists’ compensation by category and total, breaks out median yearly income by years as a hospitalist, from less than a year to six or more years. (Many of the detailed “A” tables in Chapter 8 on compensation include the “years as a hospitalist” category.) Dr. Dauterive praises the wealth of data in the survey, pointing to examples of the many variables she was surprised to learn. One of those factors was that 48% of surveyed hospitalist programs were at non-teaching hospitals. (See page 7 of the survey, Executive Summary, “Teaching status of affiliated hospital.”)
Those interviewed for this article agree that productivity data are probably more telling about the day-to-day clinical realities for hospitalists. Productivity metrics figure prominently in Dr. Dauterive’s uses for the survey. Accordingly, the annual number of billable patient encounters seen by the hospitalist (Table 58-B, page 256) and the annual number of work relative value units (RVUs) worked by the hospitalist (Table 59-B, page 258) caught her interest.
Still, Dr. Dauterive found herself wanting more data to shed light on those numbers. In negotiations for resources with hospital administrators, Dr. Dauterive would like to be able to pinpoint the reasons behind reported numbers of clinical encounters seen by the hospitalist. If the median number of billable patient encounters seen by the hospitalist in a teaching service was 1,668 (based on 107 responses; page 256, Detailed Table 058-B), what were some of the influences on this number? What was the acuity level of patients? Did the hospitalist have group resources, such as physician extenders, to help with patient admissions and rounds?
“For groups that have low lengths of stay, it would be important for me to know why,” she says. “Did they have extra supports? Do their [doctors] use Palm Pilots? You don’t always know from looking at the numbers how to apply them, make the connection, and justify the resources you’re trying to achieve,” she says.
No Perfect Measure
The ideal survey for Dr. Dauterive would include specific structured models, providing links between categories so that she could compare characteristics that more closely align with her group’s situation.
“Our program is very mixed, so it would be helpful for me to know how work RVUs were being reported,” she says. Pointing to results showing higher productivity (work RVUs) in practices compensated by 100% incentive (Table 060-A, page 259 of the survey), Dr. Dauterive wonders what factors drive these results. While the 100% incentive might appear to be the most important factor, perhaps these groups also have physician extenders or are located in a geographic location that boosts their productivity.
“I’m in a nonprofit hospital, in a clinical hospitalist service, and I want to be able to approach the administration and say, ‘If you want us to see the most patients, these are the kinds of services that see the most patients,’ ”says Dr. Dauterive. “But, if you are more interested in physician retention and work/life, then these are the characteristics of those successful programs.”
This level of detail can be difficult to interpolate from the survey, agrees Dr. Nelson. Patient acuity, for instance, is not specifically queried in the survey questionnaire. “I agree, in the ideal world, this is all information that you would want to know,” he says. Answers to the following questions could help refine product metrics:
- Does your group have teaching responsibilities for residents?
- Do you take a lot of calls from home, or do you have a separate night shift?
- Do you cover more than one physical hospital on the same day?
- Does your group do more than the typical amount of committee and administrative work?
“All these factors,” notes Dr. Nelson, “would influence productivity. There is no perfect way to know the answer to any of those things.” And, he adds, the survey already comprises 292 pages, including numerous detailed tables of data. To include all pertinent variables would entail a longer questionnaire, which might affect the response rate.
Healthcare Delivery Is Local
In his consultations with hospitalist groups, Dr. Nelson always emphasizes that the survey is “a starting point” and not the goal of what hospitalists should make. He favors adjunctive methods for benchmarking practices: “I think that when you’re benchmarking your practice, it’s as important to gather as much local and regional data as you can—in addition to the SHM survey.” He tries to network with other Seattle hospitalist programs to learn about their patterns of work hours, patient loads, and the like.
Thomas Baudendistel, MD, FACP, associate program director of the Internal Medicine Residency Program at Sutter Health’s California Pacific Medical Center in San Francisco, notes that regional markets differ widely. The healthcare market in the Northern California Bay area is very different from the one in Los Angeles in terms of financial remuneration and incentivization.
“The survey,” he says, “gives a global gestalt of the regional flavor of hospital medicine” and reveals general ballpark medians that can be a good starting point for practice benchmarking. “I think what our administration [at California Pacific Medical Center] wants to see is our data compared to the people across the street and down the road, because that’s a closer comparison in terms of payer mix and insurance reimbursements.”
IPC’s Lary agrees. “When I compete, I don’t compete against people across the country; I compete with people across the street,” he says. “As large as IPC is, we realize that healthcare is delivered locally. What we try to do [with the survey] is take the information and, to the best or our ability, figure out how it applies to our individual settings and [to the] different markets that we are in.”
A Stake in the Sand
“I think the benchmarks we have in the survey are just a piece of information—[the survey] is a context, it’s a stake in the sand,” concludes Miller. “We do have variations by type of program, by size of hospital, by geographic location, by size of program. There are numbers for each one of those, and you can clue in as to what some of the more important variations are. We could list probably 25 to 50 variables that would affect hospitalists’ productivity in one way or the other—and that’s not taking into account the individual styles of hospitalists.”
For instance, some hospitalists want to work and earn as much money as possible, while others are searching for a work/life balance that will allow them time with their families.
The survey, says Lary, supplies a piece of information in a complex puzzle about a highly variable profession. “There are so many different ways this business is being conducted right now,” he says. “One medical community may be willing to subsidize a hospital medicine program, and another may not be willing.”
Hospitalists’ professional goals vary widely as well. As far as Dr. Nelson is concerned, the bottom line for hospitalists is to structure independent practices tailored to fit their goals. This means that hospitalists are connected to the economic consequences of their staffing and workload decisions. In that way, he says, rather than approaching administrators about hiring more physicians, the practice itself can decide whether it is worth the decrease in individual hospitalists’ incomes to hire another doctor.
Because their specialty is still evolving, hospitalists will find themselves educating their clients about the profession’s services and advantages. And for that process, the survey can be a helpful adjunct. Miller agrees that the use of the survey requires a certain amount of interpolation on the part of hospitalist leaders. They should be careful, he emphasizes, not to lose sight of the individuality of their own practices.
“If you hold up the survey as the governing document when you negotiate with your hospital, then each party will use it to their advantage,” says Dr. Nelson. “This can push you towards being ‘average’ when that might not be appropriate for your practice.” TH
Gretchen Henkel is a frequent contributor to The Hospitalist.