I have been using MGMA data on physician compensation for many years now, and I generally like it. I have found that there are occasional apparent glitches in the data, but I still think that MGMA data on physician compensation is the gold standard.
What I like about MGMA data on physician compensation
It isn’t perfect, but there is a lot to like about MGMA data on physician compensation.
Citing subjective benchmarks is more persuasive
Most physicians, when analyzing a physician employment agreement, ask for more money or vacation or a higher wRVU conversion factor. I assume other law firms also try for better compensation terms when doing physician contract review. However, without subjective benchmarking, I think it is difficult to persuade a potential employer that more should be paid.
Analyzing physician productivity compensation can be tricky for people without a lot of experience. Trying to parse through these provisions and obtain a fair result is much harder if you don’t have a subjective basis to argue for more.
Similarly, attempting to obtain appropriate compensation for call coverage requirements is obviously more effective when backed up by objective benchmarks.
MGMA data on physician compensation separates academic compensation
Unlike SullivanCotter, MGMA uses a separate database for academic compensation. Academic physicians generally have less clinical duties and more research and teaching duties. As a result, academic physicians are generally paid significantly less than non-academic physicians.
SullivanCotter has a bigger sample size, but I think that is at least partly the result of including academic physicians in the database. I have not personally attempted it, but I understand that SullivanCotter does not give you the ability to filter out academic physicians. In my mind, that makes the entire survey suspect.
AMGA also has a survey that supposedly allows you to filter out academic physicians. Whenever I do an MGMA compensation analysis and the employer counters with results of an AMGA survey, I am always left wondering if academic compensation was filtered out. Here again, I’m always a bit suspicious that the employer is using their entire database, without filtering.
MGMA data on physician compensation can be very granular
Depending on the specialty, I can generally narrow the results down to the state level. When looking at a salary offer, I believe it is very persuasive when you can argue that physicians in that specialty employed by a hospital in that state have median compensation of $X.
Similarly, I can generally provide benchmarks for PTO, wRVU production, collections, and the ratio of wRVU production to compensation (helpful in analyzing a proposed bonus) down to the state level and type of entity that is employing the physician. Here again, arguing that the median wRVU to compensation ratio for a physician employed by a physician practice in that state is $X can be very persuasive when analyzing a lowball offer.
I have seen supposed experts use regional data in negotiating physician employment agreements. I think that is better than nothing, but obviously taking the time to find out state compensation in the particular practice setting will be more persuasive.
What I don't like about MGMA data on physician compensation
As I said before, MGMA data on physician compensation is not perfect. I have seen two issues that strike me as suspicious in the data.
CME data is sometimes questionable
I have seen many specialties where the database indicates that the median amount paid for a starting physician in that specialty for CME is $16,200. Having reviewed thousands of physician employment agreements, I rarely see a CME allowance that generous. I am sometimes concerned about the employer reaction when I cite these MGMA benchmarks.
Occasionally I see agreements that pay for CME, professional dues, staff dues, medical licensure, and other physician benefits with an annual limit for those expenses. However, the definition of CME in the MGMA data on physician compensation indicates that the benchmark is just for CME, and not these other professional expenses.
Physicians earning more than total collections
A second issue I have seen in MGMA data on physician compensation is benchmarks which indicate that the median total collections are less than the median total compensation for a given specialty in a given state or region. While it is not impossible that a physician may be hired at a loss, my experience is that this doesn’t happen often.
Perhaps this apparent anomaly can be explained by the exclusion of the technical component of income from the benchmark. The most common benchmark shows collections where technical component income is excluded from compensation. The technical component represents income for non-professional services, such as imaging, labs, etc.
Hospital employed physicians and non-owner physician practice physicians typically do not share in the technical component, while owners in physician practices typically do share in the technical component. I suspect one reason for this apparent anomaly is the inclusion of private practice owners in the benchmark for total compensation. These physicians do receive a portion of their income from the technical component. However, this would not explain the discrepancy when the report is limited to hospital employees only.
In spite of these relatively minor issues in MGMA data on physician compensation, I still believe that it is a critical component of a thorough physician contract review.
You may also be interested in my book on physician employment agreements, or my blogs on physician compensation, and wRVU compensation traps for physicians.
If you would like a personalized MGMA analysis of your current position, click here. If you have an agreement you would like us to review, you can start your review here. We can also provide a free consultation to talk about how we can help.
