Financial Sustainability

Why benchmarks should not be the sole guide for setting physician compensation

June 1, 2021 2:52 pm

Healthcare organizations that rely on survey benchmark data alone when setting physician compensation may be risking budgets and relationships with physicians, especially in times of great uncertainty.

A 2018 analysis found that about 55% of physicians received at least some compensation based on their personal productivity.  Moreover, the author’s experience corroborates this finding, having presented ample evidence that many hospitals, health systems and medical groups now pay physicians for their clinical services using some variation on the following formula:

Clinical compensation = 

(National median compensation per wRVU benchmark for physician’s specialty from most recent MGMA/AMGA/other physician compensation survey report)a ×

(Physician’s personally performed wRVUs)a +

(Compensation for meeting specified clinical quality targets)

There are three compelling reasons for the widespread adoption of the productivity-based portion of this model:

  • It creates an incentive for physicians to see more patients and do more procedures, which inevitably leads to more revenue in a system dominated by fee-for-service reimbursement.
  • It offers an expedient approach to setting compensation levels.
  • By tying the value of each unit of service to a publicly available survey report, it helps create an impression of impartiality and correctness.

The problem with relying on benchmarks

These impressions overlook a fundamental flaw in the model: General benchmark data cannot, by definition, reflect the reality of each organization’s unique situation. In relying solely on such data, an organization effectively relinquishes control of one of its largest expenses.

Consider the hypothetical examples presented in the exhibit below.b

Percentage changes in compensation for two physician specialties over 5 years based on hypothetical benchmark data







Pulmonology – General and Critical Care

Reported median compensation to wRVU ratio






Percentage change from prior year







Reported median compensation to wRVU ratio






Percentage change from prior year






Based on the data in the exhibit, when the 2017 edition of this hypothetical survey report was released, a productive pulmonologist with 8,500 wRVUs and a productive neurosurgeon with 13,000 wRVUs paid using the median compensation per wRVU ratio for their specialty would suddenly have found themselves making about $37,000 and $70,000 more, respectively, for the same amount of work. It is unlikely that the market for physician services in these specialties or the organization’s financial situation would have changed dramatically enough in such a short period to justify these substantial changes. More important, while the physicians would have welcomed these increases, their organizations could not have budgeted for them — nor could they have prepared the physicians for the decreases coming in the next year.

Some might consider averaging data points from multiple years of a survey report or blending multiple survey reports to be an effective solution for smoothing out such aberrations. But again, such an approach cannot account for unique factors affecting any specific organization, such as its payer mix, payer contracts, utilization of advanced practice providers or overhead expenses.

Why is this particularly important now?

The COVID-19 pandemic’s effect on volume, revenue and physician compensation in 2020 will doubtless have a massive impact on the data shown in the survey reports, as will the recent changes in the 2021 Medicare physician fee schedule (including the wRVUs assigned to common CPT codes).c In the wake of the disruptions caused by COVID-19, the weakness inherent in these compensation models tied to survey data have become more evident. Organizations now face difficulties not only in foreseeing the changes in benchmark data but also in predicting physician compensation expenses. Such uncertainties also create an unnecessary stressor for physicians, who need a sense of stability like anybody else.

Taken together with the challenges outlined in the sidebar below, these forces raise questions about whether the survey reports for 2021, 2022, and even 2023 will provide a reliable look at the market for physician services. Though the surveying organizations are trying to “smooth out” the benchmarks, continuing to rely upon these data in physician compensation calculations could expose organizations to significant unpredictability. Simply put, healthcare organizations need a more sustainable model.

The shortfalls of publicly available survey reports: What they are, and what they are not

Publicly available survey reports are generally the best resource we have for understanding the market for physician services, but they are not perfect. Each report typically contains a lengthy introduction regarding how and from whom the information was gathered, and the limitations of the data. It is all too easy to skip over these pages and move directly to the data tables.

At a minimum, it is important to understand that:

  • Survey reports are based on a voluntary response and are not necessarily representative of the relevant market
  • Data are self-reported, and different respondents may interpret the same survey question differently
  • The reports are backward-looking (i.e., reports released in mid-2021 will discuss compensation paid in 2020)
  • The organizations doing the surveying can perform only so much quality assurance on the data

Accordingly, it is best not to use survey reports alone in setting compensation, but as one input alongside others.


A better solution

Healthcare organizations can take three broad actions, detailed below, to strengthen their productivity-based physician compensation model and prepare for an uncertain future. The approach described herein can be used by organizations large or small for a group of similar physicians within a specialty or for individual physicians.d

1 Replace compensation formula variables tied to benchmarks

The goal is to replace the general formula from the start of this article with the following formula:

Clinical compensation =

 (A fixed $XX per wRVU for a given specialty, specific to a given organization) ×

(Physician’s personally performed wRVUs) +

(Compensation for meeting specified clinical quality targets)

Although this model is straightforward, identifying and justifying the “right” number takes some work. Organizations should invest time and resources in a careful four-step process, led by executives responsible for physician services.

Step 1. Identify and analyze every aspect of physicians’ work and pay. The organization must first collect all relevant data, documents and other information from throughout the organization. This may involve the following areas:

  • Legal (e.g., written agreements between the organization and physicians)
  • Human resources (e.g., payroll data for employed physicians)
  • Accounts payable (e.g., payments to independent contractor physicians)
  • Revenue cycle (e.g., volume and revenue for physicians’ billable clinical services)
  • Compliance (e.g., timesheets for physicians’ administrative services, if applicable)
  • Operations (e.g., schedules and burden for physicians’ call coverage services, if applicable)

Gathering insights from physician and non-physician administrators helps the organization understand the whole story and resolve questions and “blind spots” regarding what physicians do and/or how they are paid. For example, these discussions could reveal important information at a level of detail rarely described in contractual agreements, such as the specific types of work a surgeon does on an inpatient and outpatient basis and while on call.

Taking time to gather and deeply understand this information allows the organization to present a revised compensation model that accomplishes three purposes:

  • Assures physicians that the organization has considered the full value of their services
  • Responds to any objections they might have
  • Anticipates how changes to the compensation model will impact physicians’ pay in various scenarios (as discussed further below)

Step 2. Determine what can and what should be paid. Some organizations have internal policies or guidelines on what can be paid for physician services. Others consult a healthcare valuation expert to determine the fair market value (FMV) of services provided, to ensure compliance with applicable laws and regulations.

In either case, the outcome will be a ceiling for what can be paid, which may then be adjusted downward to account for factors specific to the organization, such as its financial situation within the physicians’ specialty, historical compensation to the subject physicians, and internal equity. For example, while the FMV of the clinical services provided by internists may be $55 per wRVU, relatively low reimbursement and/or relatively high overhead may keep the organization from paying that full rate.

At this point, the organization has a clear understanding of:

  • What the physicians’ overall compensation should be
  • What the physicians are doing to earn that compensation
  • How that compensation is broken down into various forms (e.g., biweekly base salary, annual incentive payments, hourly administrative rates, daily call stipends, etc.)

Step 3. Back into an appropriate compensation per wRVU factor. While it is tempting for an organization to simply set a payment rate at or below a national median compensation per wRVU benchmark derived from multiple market survey data reports, the organization’s circumstances may pose an obstacle to this approach. For example:

  • Non-cash benefits (e.g., paid time off, continuing medical education, retirement plans, etc.) may be relatively “rich.”
  • Recruitment incentives (e.g., signing bonuses, moving allowances, loan repayment, etc.) or clinical quality incentives may be substantial.
  • The physicians may be unusually productive
  • Payments for call coverage, physician administrative, or other non-clinical services may be above the FMV of those services.
  • The organization may be located in a market where physicians are paid well below national rates for their services.

In such circumstances, the payment rate per wRVU would need to be adjusted downward to ensure total remuneration to the physicians is reasonable for the totality of their services. Likewise, the payment rate per wRVU could be adjusted upward if, say, the benefits were relatively “lean” or payments for non-clinical services were relatively low. Ultimately, the payment rate needs to be set correctly to ensure that the overall compensation model stays in balance.

Step 4. Simulate and adjust the model. The importance of simulating and adjusting the compensation model cannot be overstated. After completing the previous steps, the organization should have all the necessary information to perform this step. For example, if the organization was updating the compensation model at the beginning of 2021, it might use the following data from 2020:

  • Actual full-time equivalency (FTE) status, wRVUs, call coverage shifts, medical director hours, quality scores and any other factors that drive compensation in either the current or proposed compensation models
  • Actual total compensation under the current model

Comparing actual total compensation from 2020 under the current model with the expected total compensation under the proposed model for each physician, changing nothing but the model, helps leaders understand and explain the impact of the change on each physician.

At this time, the organization should also perform a scenario analysis to understand how elements such as higher or lower productivity, or more or less paid call, impact the proposed model. These simulations allow organizations to anticipate potential issues and adjust as needed.

2 Implement a thoughtful, transparent physician compensation oversight process

Creating a compensation oversight process that physicians trust is a best practice and can be a critical success factor when an organization moves away from relying solely on publicly available survey reports.

The form of the process and the extent to which physicians are involved will vary. However, an effective process encompasses two key aspects.

Evaluation of the physician compensation model on a regular basis. The model should be evaluated every two or three years to determine whether changes are needed. This evaluation should account for various inevitable changes in factors, including:

  • Trends in wRVU production, quality scores or any other factors that drive compensation
  • How the financial situation within a specialty (e.g., payer mix, payer contracts, overhead expenses) may have changed
  • Equitable pay among similar physicians
  • Market data gathered during recruitment and exit interviews
  • Trends in applicable metrics from multiple relevant survey reports

Clear and timely communication to affected physicians about the results of the review. Physicians should be fully informed about what the review process entailed, what changes are being made and why.e Physician satisfaction tends to be higher when physicians feel that someone is looking out for them, even when that someone must deliver bad news.

3 Remove payment formulas from physician contracts

Including the finer points of the physician compensation model within a physician’s contractual agreement is potentially problematic for two reasons:

  • Organizations must amend the agreement every time they need to make even minor changes to the compensation model, which can be an administrative headache.
  • The physician may think they can negotiate the compensation model, which is often not the case (as doing so would be inequitable).

The best practice in this area is to reference a standard compensation policy for a given physician’s specialty and to attach the current policy, which is subject to change. Compensation policies should be simple enough that any provider can immediately say what is expected of them, how they are paid, and what they could to do earn more. Of course, each organization should consult its legal and compliance departments on the right approach.

Bringing it all together

Physician compensation models tied directly to survey reports have become pervasive, due to their perceived benefits. This situation may create hesitance in organizations around moving away from reliance on surveys out of a fear they are moving away from the market. In truth, taking a more tailored approach can help them create a more grounded, sustainable compensation model that reflects their specific needs, with less risk to their financial standing and the quality of their relationships with physicians. This process helps position organizations for success in uncertain times—an outcome that is well worth the investment.


a.  wRVU stands for work relative-value unit, a measure of the time, skill, training, and intensity necessary to perform the various clinical services.

b. Note that these examples use hypothetical data based on widely used survey reports. The intent is simply to illustrate the concepts presented.

c. While the exact impact is not yet clear, it is reasonable to assume that wRVUs in the 2021 surveys will likely be lower than in 2020 while physician compensation levels will remain the same or at similar levels for significantly reduced production because many organizations provided some level of compensation guarantees to their physicians in 2020.

d. Although the approach may seem daunting for small organizations, the author’s experience has shown that the complexity of the work tends to correlate to the size of the organization involved, so smaller organizations should find it manageable using their relatively limited resources.

e. This regular review also provides a good opportunity to study whether the model itself should be changed to rely less on productivity and more on other factors, such as clinical quality.


googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text1' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text2' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text3' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text4' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text5' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text6' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text7' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-leaderboard' ); } );