I visited during a hot Florida summer in the mid 1990s and could readily see that the practice was great in most respects. The large multispecialty group had recruited talented hospitalists and had put in place effective operational practices. All seemed to be going well, but inappropriate overhead allocation was undermining the success of their efforts.
The multispecialty group employing the hospitalists used the same formula to allocate overhead to the hospitalists that was in place for other specialties. And compensation was essentially each doctor’s collections minus overhead, leaving the hospitalists with annual compensation much lower than they could reasonably expect. With the group deducting from hospitalist collections the same overhead expenses charged to other specialties, including a share of outpatient buildings, staff, and supplies, the hospitalists were paying a lot for services they weren’t using. This group corrected the errors but not until some talented doctors had resigned because of the compensation formula.
This was a common mistake made by multispecialty groups that employed hospitalists years ago. Today, nearly all such groups assess an appropriately smaller portion of overhead to hospitalists than office-based doctors.
Typical Hospitalist Overhead
It is still tricky to correctly assess and allocate hospitalist overhead. This meaningfully influences the apparent total cost of the program and hence the amount of support paid by the hospital or other entity. (This support is often referred to as a “subsidy,” though I don’t care for that term because of its negative connotation.)
For example, costs for billing and collections services, malpractice insurance, temporary staffing (locums), and an overhead allocation that pays for things like the salaries of medical group administrators and clerical staff may or may not be attributed to the hospitalist budget or “cost center.” This is one of several factors that make it awfully tricky to compare the total costs and/or hospital financial support between different hospitalist groups.
SHM’s State of Hospital Medicine report includes detailed instructions regarding which expenses the survey respondents should include as overhead costs, but I think it’s safe to assume that not all responses are fully compliant. I’m confident there is a meaningful amount of “noise” in these figures. Numbers like the median financial support per FTE hospitalist per year ($156,063 in the 2014 report) should only be used as a guideline and not a precise number that might apply in your setting. My reasoning is that the collections rate and compensation amount can vary tremendously from one practice to another and will typically have a far larger influence on the amount of financial support provided by the hospital than which expenses are or aren’t included as overhead. But I am confining this discussion to the latter.
APC Costs: One Factor Driving Increased Support
SHM has been surveying the financial support per physician FTE for about 15 years, and it has shown a steady increase. It was about $60,000 per FTE annually when first surveyed in the late 1990s; it has gone up every survey since. The best explanation for this seems to an increase in hospitalist compensation while production and revenue have remained relatively flat.
There likely are many other factors in play. One important one is physician assistant and nurse practitioner costs. The survey divides the total annual support provided to the whole hospitalist practice by the total number of physician FTEs. NPs and PAs are becoming more common in hospitalist groups; 65% of groups included them in 2014, up from 54% in 2012. Yet the cost of employing them, primarily salary and benefits, appears in the numerator but not the denominator of the support per physician FTE figure.
This means a group that adds NP/PA staffing, which typically requires an accompanying increase in hospital financial support, while maintaining the same number of physician FTEs will show an increase in hospital support per physician FTE. But this fails to capture that the practice’s work product (i.e., patients seen) has increased as a result of increasing its clinical staff.
This is a tricky issue to fix. SHM’s Practice Analysis Committee, which manages the survey, is aware of the issue and may make future adjustments to account for it. The best method might be to convert total staffing by physicians and NP/PAs into physician-equivalent FTEs (I described one method for doing this in my August 2009 column titled “Volume Variables”) or some other method that clearly accounts for both physician and NP/PA staffing levels. Other alternatives would be to divide the annual support by the number of billed encounters or some other measure of “work output” or to report percent of the total practice revenue that comes from hospital support versus professional fee collections and other sources.
Why Allocation of NP/PA Costs and FTEs Matter
Another way to think of this issue is that including NP/PA costs but not their work (FTEs) in the financial support per FTE figure overlooks the important work they can do for a hospitalist practice. And it can lead one to conclude hospitals’ costs per clinician FTE are rising faster than is actually the case.
This is only one of the tricky issues in accurately understanding hospitalist overhead and costs to the hospital they serve. TH