I work with a number of health systems on determining full-market-value (FMV) compensation related to stipends paid to hospital-based specialists. What is your opinion on how compensation should be determined for the physicians to staff hospitals? Would you say that the busier the location, the higher the compensation? Would you say that the more hours the physician works (regardless of productivity), the more pay they receive? Would you say that the more years of experience that a physician has should result in higher compensation?

Curtis H. Bernstein, CPA/ABV, ASA, CVA, MBA,

director, valuation services,

Sinaiko Healthcare Consulting Inc.,

Los Angeles

Dr. Hospitalist responds:

The topic of physician compensation and workload comes up frequently, but this question frames it a bit differently. Namely, what are the external and internal factors at work in determining compensation? Let’s tackle the response in two parts: 1) How do you account for the variability across sites? and 2) How does physician pay vary within a single site? The crux of the first question comes down to trying to interpret physician workload across disparate locations. It’s not laden with quite the same complexity as Gordian’s knot, but it’s close. One could easily answer this question with a lot of “Yes, but … ” in reference to the all the factors that go into determining compensation. Yes, a busier site would generate more encounters, thus more revenue, and thus more pay. However, that same site might also be so busy as to require more than one physician on at night. A higher-paid, but lower-volume, nocturnist would then skew the workload/pay scale. Same thing with the ICU; if it is fully staffed with intensivists (more likely in a higher-volume setting), then that would remove the single highest paying code for a hospitalist (the 99291: critical care time 35-74 minutes, 4.50 wRVU), and that has the potential to drop reimbursement. Practice management columnist John Nelson, MD, MHM, has written more than a few fantastic columns addressing just these sorts of issues, but let’s concentrate on just the key factors:

• Volume;
• Payor mix/collections;
• Ratio of day shifts to night shifts (optimal is 4:1), assuming that there are dedicated night shifts; and
• Value-added services.

Volume is fairly straightforward, with most hospitalists seeing around 15 patients per day. Now, an average is just that, and practices exist where the daily number is 10 and where the census is 25. Still, it’s an easy number to understand.

Payor mix is a little more complex but should not vary substantially quarter by quarter, though it could vary greatly year to year. (Note: This is independent of collection rate percentage, which is a completely artificial variable.) If Medicare pays $1 for a certain code and the charges are set at $1, then the collection rate is 100%. If the charges are set at $2, then the collection rate will be 50% but bring in the same amount of money. Arcane billing convention aside, from hospital to hospital and region to region, the payor mix and attendant collections will vary. Ratio of day shifts to night shifts posits that there is an optimal ratio of roughly four day shifts for every one night shift. Night shifts are more expensive, in general, so the more day shifts you have to cover your fixed cost of covering the nights, the better. The reason the ratio can’t be 10:1 is that 10 day-rounding hospitalists would generate a daily service of 150+ patients, and there is no way for one nocturnist to safely cross-cover all those folks, much less see new admissions.

Lastly, there are the value-added services that provide the raison d’être for hospitalists. We are fundamentally different from the procedure-based specialties in that our value comes not from increasing revenue (more procedures) but from decreasing costs. Initially, a lot of this focused on length of stay, and now it’s shifting to discussions of core measures, readmission rates, and other quality metrics. What a hospital is willing to pay for this service, which goes above and beyond taking call for unassigned patients, will go a long way toward determining the overall stipend and resultant physician compensation. (For more information on hospitalist compensation and productivity, check out the 2011 State of Hospital Medicine report, www.hospitalmedicine.org/survey)