Does your hospital or health system rely on a single benchmark salary survey, such as the Medical Group Management Association (MGMA) Physician Compensation and Production Survey Report, when determining fair market value (FMV) compensation? While many believe that MGMA is one of the most comprehensive salary surveys, relying on only one salary survey may create regulatory compliance issues. Additionally, dependence upon a single survey and the reported data within it can result in significantly different conclusions regarding physician compensation.
Regulatory Issues
First and foremost, regulatory guidance encourages healthcare organizations to use numerous salary survey sources when determining FMV compensation. Specifically, with respect to FMV determination, the Phase III rule of the Stark Law states that reference to multiple, objective, independently published salary surveys remains a prudent practice for evaluating fair market value.1 If your organization’s physician compensation is challenged, documentation of multiple, objective, independently published salary surveys should prove beneficial.
Potential Compensation Consequences
Organizations may also make less-than-optimal compensation decisions when relying on only one salary survey. As an example, assume that a hospital located in the southeastern part of the United States desires to recruit a physician specializing in hematology/oncology from within its own market, and based on a detailed production and compensation due-diligence analysis, has decided to pay the physician at the 90th percentile. According to the MGMA 2014 Physician Compensation and Production Survey Report, compensation at the 90th percentile for a hematology/oncology physician in the South is $864,561. However, if the organization also reviewed data from other salary survey sources, it could see far different values. The 2013 salary survey published by Sullivan, Cotter and Associates, Inc. (SullivanCotter)2 and the 2014 salary survey published by the American Medical Group Association (AMGA) report 90th percentile compensation for the same region in amounts of $596,205 and $693,954, respectively. Therefore, the average of the MGMA, SullivanCotter, and AMGA data is $718,240, a difference of $146,321 when compared to the MGMA data. Depending on the specific facts and circumstances associated with the survey data (e.g., number of respondents, etc.) one might conclude that the MGMA data point is an outlier and choose to average the SullivanCotter and AMGA data only. This average is $645,080, a difference of $219,481 when compared to the MGMA data. In this case, and assuming all other factors are the same, a hospital choosing to utilize only one survey data point at the 90th percentile may decide to pay the physician almost $220,000 more than if it had also considered the additional two surveys.
Other Possible Pitfalls
In addition to compensation varying across surveys, the number and type of survey respondents can also differ significantly. When looking at national benchmark compensation from four major surveys, including 2013 SullivanCotter, 2014 AMGA and MGMA, and the 2014 Hospital & Healthcare Compensation Service (HHCS) Physician Salary & Benefits Report for trauma surgery, a straight average at the 90th percentile yields $600,023. However, this fails to account for the large variance in participant data points across surveys, which includes numbers of respondents ranging from 56 for HHCS to 283 for SullivanCotter. Using a weighted average based on the number of respondents for each survey, the 90th percentile across all four surveys is $567,020, approximately $33,000 lower than using a straight average. One should consider using a weighted average instead of a straight average when sample size and reported compensation data from multiple surveys vary significantly for the same specialty.
Another potential pitfall that one must avoid is selecting the wrong type of survey respondent for benchmarking purposes. As there are many surveys, each with different population characteristics, one must always ask whether or not the appropriate survey has been selected for the target population. Examples of such differences could include academic versus private practice, or hospital-employed versus non-hospital-employed physicians. It is crucial to select the right population—one that is often the closest representation of the population to which the data is being compared.
Additional Factors to Consider
It is important to note that in addition to considering potential variances in compensation from national surveys, there are other factors that should also be considered when determining FMV compensation. Such factors include, but are not limited to: geography (physician pay may vary depending on where the hospital or practice is located), physician experience (a physician’s level of experience could warrant higher or lower compensation), and physician productivity (i.e. personally performed and modifier-adjusted work relative value unit [wRVU] productivity as compared to benchmarks may be a factor). In order to gain a well-rounded perspective of physician compensation, all relevant, potential factors that impact physician compensation should be considered.
As illustrated herein, hospitals and health systems should consider multiple salary surveys and other factors to avoid compliance issues and to arrive at the most appropriate data set to begin assessing FMV compensation. To discuss the use of various physician compensation surveys and other factors that should be considered when determining FMV compensation, please contact the experts listed below at PYA, (800) 270-9629.
1 42 C.F.R. § 411.351 (2011).
2 The 2014 salary survey is not yet available.
