|Download a PDF Version of the Article to Share With Your Team|
Physician Compensation Risk Assessment: New Final Rule Creates Challenges in Determining Fair Market Value and Commercial Reasonableness
Healthcare systems in the United States are still reeling from the financial impact of COVID-19 along with workforce burnout and shortages. On top of these environmental challenges, healthcare systems have been thrown the strongest curveball it has seen in more than a decade. The Final Rules released by the U.S. Department of Health and Human Services and the Centers for Medicare/Medicaid Services addresses the desire to prioritize value and care coordination. However, these new regulations mean that healthcare executives must take the time to re-evaluate physician compensation agreements to remain in compliance with shifting rules.
On November 20, 2020, the U.S. Department of Health and Human Services (HHS) released Final Rules for Physician Self-Referral Law (Stark Law), the federal Anti-Kickback Statute (AKS), and the Civil Monetary Penalties (CMP) Law (collectively referred to as Final Rules). On December 2, 2020, HHS/Office of Inspector General (OIG) published its AKS Final Rule, “Revisions to the Safe Harbors Under the Anti-Kickback Statute and Rules Regarding Beneficiary Inducements”, and Centers for Medicare/Medicaid Services published its Stark Law Final Rule, “Modernizing and Clarifying the Physician Self-Referral Regulations” in the Federal Register. These new rules, which significantly amend the existing laws, are a direct result of HHS’s “Regulatory Sprint to Coordinated Care” initiative. Needless to say, the Final Rules are considerable in terms of breadth and impact.
HHS has a stated goal of reducing regulatory barriers within our nation’s healthcare system and accelerating “the transformation of the health care system into one that better pays for value and promotes care coordination”. As HHS’s statement indicates, value-based arrangements and transactions are the focus of this episode of Stark Law and AKS revisions. That said, other areas and central ideas of the Stark Law and AKS are also significantly impacted. Many of these rule changes affect the daily work of those in the healthcare industry who are concerned with the fair market value (FMV) and commercial reasonableness (CR) of provider compensation.
Physician Compensation Salary Surveys…It Isn’t Always That Easy
Consulting “multiple, objective, independently published salary surveys remain a prudent practice for evaluating fair market value” as stated in Stark II, Phase III. Still, the Stark Final Rule indicates that salary surveys are not automatic—regardless of the percentile at which the compensation in question falls. According to CMS, we continue to believe that the fair market value of a transaction—and particularly, compensation for physician services—may not always align with published valuation data compilations, such as salary surveys. In other words, the rate of compensation set forth in a salary survey may not always be identical to the worth of a particular “physician’s services.”
In relying exclusively on salary surveys, many hospitals and health systems across the country have drawn a line in the sand and set a base compensation threshold at the 75th percentile. If base or guaranteed compensation does not exceed the 75th percentile for the physician’s specialty, as published by a survey source like the Medical Group Management Association’s Provider Compensation Survey, then hospitals and health systems do not seek a third-party fair market value opinion because they consider the compensation to be automatically fair market value. They believe they are fine, and their work is done. They can move on to something complex because this was way too easy. Other organizations have been slightly more conservative with total compensation and mandated in their physician contracts that they will not provide total compensation (base compensation plus all bonuses) above the 75th percentile (an actual “ceiling”).
According to CMS, some of the commenters on the Final Rule asserted that, “a safe ‘harbor’ based on a range of values in salary surveys would be consistent with what they stated was established CMS policy that compensation set at or below the 75th percentile in a salary schedule is appropriate and compensation set above the 75th percentile is suspect, if not presumed inappropriate.” To these comments, CMS responded, “For the reasons explained in Phase I, Phase II, and Phase III, we decline to establish the rebuttable presumptions and ‘safe harbors’ requested by the commenters. We are uncertain why the commenters believe that it is CMS policy that compensation set at or below the 75th percentile in a salary schedule is always appropriate, and that compensation set above the 75th percentile is suspect, if not presumed inappropriate. The commenters are incorrect that this is CMS policy.”
Clearly, from CMS’ perspective, both referenced policies are misguided. It is inaccurate for a hospital or health system to believe that just because base or guaranteed compensation is below the 75th percentile, there is no risk and the compensation they are providing is automatically fair market value. Likewise, a belief that paying a provider guaranteed or total compensation above the 75th percentile is not fair market value is also misplaced.
Bottom line, fair market value determination can be difficult and is often complex. The data and information reviewed and analyzed should be more than just dollars and Work Relative Value Units (wRVUs). Yes, base compensation, expected total compensation, and productivity targets are essential and carry a significant amount of weight in FMV determination, but other factors are weighty too. In addition to FMV, compensation arrangements with physicians and advanced practice providers (APPs) must also be commercially reasonable. According to CMS description in the Final Rule, “commercially reasonable means that the particular arrangement furthers a legitimate business purpose of the parties to the arrangement and is sensible, considering the characteristics of the parties, including their size, type, scope, and specialty”. In the Final Rule, CMS also reiterated that, “the determination of commercial reasonableness is not one of valuation.” An arrangement can be fair market value, but that does not mean that it is commercially reasonable. Conversely, an arrangement must be considered fair market value to be commercially reasonable. Neither the fair market value nor the commercial reasonableness question is straightforward in every situation.
Physician Compensation Risk Assessment
To begin mitigating compliance risk, we recommend that hospitals and health systems have a standard and systematic approach to FMV and CR determination. This process starts with an inventory of all compensation arrangements, followed by a comprehensive assessment of risk associated with each arrangement. Listed below are the ten characteristics, data points, and critical questions we believe are central to assessing FMV and CR risk on the front end of each arrangement and/or transaction. If not, a retrospective review or audit may be necessary.
1. Reason(s) for the arrangement/transaction.
Why is the health system employing this particular physician? What are the reasons for the transaction? Is the physician filling community need and increasing access to care? Is the physician necessary to secure emergency department call coverage? Is physician fulfillment of licensure or regulatory obligations? Is physician transitioning with a retiring physician? And lastly, is physician necessary for the provision of charity care, and/or improvement of quality and health outcomes? You’ll notice that driving patient volume, increasing revenue, and referral capture are not on this list.
2. Level and duration of salary guarantee.
What percentile of a weighted average of multiple applicable and reliable salary surveys is the provider’s base/guaranteed compensation for his/her specialty? How long is the duration of the guarantee? One, two, three, four years…or the term of the agreement?
3. Base salary adjustment mechanisms.
Contractually, is there a mechanism in place for base salary reductions if targets are not achieved? If so, how often is the base salary reset (quarterly, semi-annually, or annually)? How much does the base compensation right-size versus protect the providers? Does the organization have the wherewithal to implement reductions?
4. Demonstrated and/or expected practice losses.
Is the provider’s practice losing money? If so, how much per provider, and how does that compare to the rest of the practices within your network? How does the level of losses compare to benchmark data for employed practice losses in the provider’s specialty? If losses are higher, by how much and what are driving those losses? Are the drivers of losses payer mix and substandard reimbursement, poor revenue cycle, high overhead, or is provider compensation out of line? If the latter is true, you likely have an issue that will require changes to the model.
5. Compensation per unit of service (i.e., compensation per wRVU).
If the provider is paid, either totally or partially, through a bonus structure based on units of service, such as per wRVU, what is the rate? How does the rate compare to benchmark data? Does the rate result in a compensation percentile that is more, less, or equal to the percentile of production when compared to applicable and relevant benchmark data? If compensation levels exceed the level of production, by how much?
6. Production–and the alignment between compensation and production.
Continuing the line of questioning and discussion in number 5, what is the alignment between compensation and production? At what percentile of survey data is the provider’s production? And how does that compare to the level of resulting compensation? What happens to total compensation percentiles if production rises to higher percentiles—i.e., the 75th to 90th percentiles? Does total compensation remain well-aligned at these higher percentiles, or does misalignment (compensation levels exceeding the level of production) start to occur? If compensation levels start to exceed production levels, by how much? Is the level of production over the level of compensation more or less than ten percentiles? If more than ten, is it more the 25th percentiles? When the level is more than ten, our level of concern grows, as does the intensity of our questioning.
7. Other compensation sources.
What are other compensation sources influencing and driving the level of compensation? Are there medical directorships and/or committee participation payments in place? If the provider is a physician, is he/she paid for APP collaboration and oversight? If there is a medical directorship—how much? What is the medical director’s hourly rate? Are the hours tracked and reported? Is the payment based on or reconciled according to these documented hours? Is there a documented list of expectations, duties, and responsibilities? If there are APP collaboration/supervision payments, is there a documented list of expectations? Does the physician perform the duties and responsibilities of those expectations? What is the level of APP supervision payments? Are the APP supervision payments paid as a stipend or by some other mechanism? With all of these “other” compensation sources, are they legitimate and needed—is the organization receiving the service and value it expects? And do the expectations of these other duties make sense in terms of the provider’s primary responsibilities? Can the provider complete these other duties and still see a full load of patients under their expected clinic schedule and expectations? In other words, is what we call “compensation stacking” an issue? If the answers to these questions are not favorable, changes are likely in order.
8. Work attribution.
The question of work attribution (i.e., wRVU credit) applies to both physicians and APPs—though sometimes from opposite perspectives. For physicians, the question is often, is the physician receiving credit for services actually rendered by an APP and billed under the APP? This clearly cannot happen. Never. This is not an appropriate structure for APP supervision. If this is happening, it must be remedied immediately. Slightly grayer areas are the attribution of “shared” and “incident to” visits. We recommend no allocation of incident to visits (visits rendered by an APP but billed under a physician) to a physician and at least a 50/50 allocation of service credit (i.e., wRVUs) for shared visits. Lastly, for APPs in some specialties and practices, work attribution is limited or non-existent. In these instances, matching production levels (i.e., wRVU production levels) is challenging at best and impossible in many cases. In these situations, the APP’s contribution and efforts are not sufficiently recognized by the units of measurable production; they are directly attributed. Often their contribution falls under global billing and/or their efforts help to make the physician or physicians more efficient and productive.
9. Non-productivity (i.e., value and quality-based) incentive details.
How do non-productivity-based incentives impact total compensation? What proportion are nonproductivity-
based incentives of total compensation? How does that proportion compare to benchmark data? Are the goals and targets easily obtainable, or are they “stretch” goals? Metric goals that are too easily obtained could be considered guaranteed compensation—which directly impacts fair market value and commercial reasonableness.
10. Required hours of service.
What is the provider’s Full-Time Equivalent (FTE) status and/or how many hours or days per week or per year is the provider required to work? Paying a provider to be in the office seeing patients or in the hospital rounding on patients two days per week is different than five (5) days per week, and compensation, particularly base compensation, should reflect that. Likewise, an emergency medicine physician working 1,800 hours per year could be expected to make less than an emergency medicine physician working 2,300 hours per year.
Clearly, some specialties and services (i.e., hospital medicine, emergency medicine, critical care, and urgent care) are better measured by shifts and/or hours, as opposed to patients or wRVUs. These specialties typically cannot control or influence patient volume. Additionally, incenting volume in these settings can have unwanted and sometimes negative consequences on culture, teamwork, quality, and patient satisfaction.
Evaluating each of these ten factors collectively helps determine the level of risk an organization has with its compensation arrangements. None of the ten factors stand-alone—collectively, they tell a story. As we’ve discussed, nothing in fair market value and commercial reasonableness determination is automatic. All of a situation’s characteristics and truths must be taken into account and considered. As much as we’d like there to be a bright line, or a formula, or calculation that brings clarity and certainty—it doesn’t exist. Fair market value and commercial reasonableness determination is more art than it is science. It is difficult to describe, but you know it when you see it.