Rethinking Productivity

Friday, September 21, 2018 | Joel Sauer

Today and for the past decade or more the prevailing currency to measure physician productivity has been the work Relative Value Unit, or wRVU. This single metric is also the predominant compensation tender for employed cardiologists, used nearly exclusively for compensation funding in more than 80 percent of employed groups according to recent MedAxiom membership data. We’re now nearing the end of 2018 and that much farther down the volume-to-value transition path. Does relying so heavily on this single metric to measure physician work and value make sense? I would argue no.

Here are my issues with such singular devotion to the wRVU as the driver of cardiology compensation:

  • The wRVU is only a volume indicator. Period.
  • There is no connection to the outcome of the care rendered or whether it was delivered in a satisfactory manner. All that matters is whether a CPT event was performed.
  • The service is rewarded whether or not it was even warranted. Conversely NOT providing a service – even when that is the best course for the patient – has no reward.
  • wRVUs pretend that the value of care is at the individual provider level. There is no recognition of team or sharing of care delivery.
  • Due to the individual nature of the wRVU, it tends to divide our teams and create internal competition at a time when value is requiring much more collaboration. This is particularly true with regard to Advanced Practice Professionals (APPs) such as nurse practitioners and physician assistants.
  • Although a very laudable attempt has been made to accurately separate the effort between physician services (aka CPT codes) – and frankly it’s pretty good – the wRVU system places more value on certain activities and less on others. This can create priority problems.

There are certainly others and I’m sure many of you can come up with your own as well, but this list will suffice for my purposes today.

So here’s an alternative, or better yet, a companion production measure to the wRVU: Panel Size. What is this? It’s a group measure of the active cardiology patients in a practice, looking only at unique face-to-face encounters over an 18-month period. So if patient Joel Sauer was seen by a general cardiologist and an EP physician during the preceding 18 months, that would count as one (1) active patient. If during that same time period I only had an EKG read by a cardiologist with no face-to-face encounters, I wouldn’t count at all. Figure 1 shows the current Panel Size percentiles from the most recent MedAxiom MedAxcess survey.


Why do I like Panel Size as a cardiology production measure?

First, Panel Size is about recognizing the cardiology team, not just individual work. Thus it is somewhat of a population measure, which is part of the Triple Aim in terms of where healthcare is trying to go. As such, it is less about who sees the patient and what they do, but instead focuses on managing that patient from an overall sense. Additionally it does not reward individual work, just that of the team, since the metric is measured at the group level.

This leads to my second reason for supporting Panel Size as a production measure. On a regular basis I run into underutilized APP resources in integrated cardiology practices. When you dig into why this occurring, inevitably the compensation funding model stands in the way. For compliance reasons, many hospitals and health systems have created a huge economic hurdle through their compensation funding with physician activities being fully rewarded and those performed by an APP either not rewarded at all (aka $0 per wRVU) or under-rewarded (aka annual stipend).

Again, this is ostensibly for compliance reasons and the need to only pay physicians for work “personally performed.” It is not the subject of this blog post, but go to {} to find some legal pushback on this issue. These lopsided economics lead inexorably to the full power of APPs being significantly throttled. Panel Size is agnostic to who actually performed the service. Whether an APP or a physician, the work gets credited equally. The clinical judgement of which medical professional(s) should deliver the care is appropriately up to the physicians to decide and the reward system does not economically “judge” these decisions.

Third, Panel Size rewards new patient volumes. The more unique patients I can bring into the system, the more I build that active patient list and am rewarded. Certainly you could argue that wRVUs do the same thing, but there are flaws – two being noted above. In addition, wRVUs can reward over management; seeing the patient regularly even if the clinical value of these additional visits is suspect/unwarranted. With Panel Size, these additional encounters would all get rolled up into a single active patient.

Let me pause here to make sure some things are understood. I’m not accusing anyone of anything in particular. We create incentives for a reason: to elicit and reward certain behaviors and activities. Since this is the case, we need to be very careful that we’re rewarding – and understanding what we’re not rewarding – appropriately to get the outcomes we want. In my mind our reward system needs to be scrutinized beyond just what actually happens, to include what one could perceive might happen. Over-treating falls into this latter category. Regardless of the production measure, I advocate that a meaningful part of provider compensation should come from actual outcomes, service and value (aka the Triple Aim). This keeps all of our eyes on the right ball.

My fourth reason supporting Panel Size is that it rewards the cardiovascular team and efficient patient management. This may be somewhat redundant with the above, but I feel it has enough nuanced difference to be separately noted. The more a group can streamline the process of seeing and treating patients, the more active patients it can manage. When paired with outcomes and service, using Panel Size as a production measure appropriately rewards this efficiency.

Perhaps my biggest reason for advocating on behalf of Panel Size (number five if you’re counting) is that it rewards no treatment or procedure when no treatment or procedure is the right course for the patient. Our healthcare system in general must find ways to compensate providers for doing nothing when that’s the right decision and wRVUs simply fall short in this regard. As long as my provider (physician or APP) has laid eyes on me and determined that nothing more is needed, I count toward that active patient tally as equally as if I received a diagnostic test or intervention. To me that’s powerful.

To reiterate and to close, I want to be clear that I believe strongly that our providers – and for that matter our administrators – should be held accountable in terms of compensation to the Triple Aim: better outcomes (both individually and at a population level), lower cost and high levels of patient engagement and satisfaction. Regardless of the production measure, there should be compensation at risk for these value initiatives.

I get what a big change introducing Panel Size as a production measure would be for organizations and could have a significant impact on compensation, which is always scary. For that reason, perhaps a phased approach as shown in Figure 2 might be in order. Whatever the method, as the reimbursement system for the care we deliver changes and puts more emphasis on overall value, our compensation model needs to evolve too.

Given the unique team (aka subspecialization) aspect of cardiology, I feel that Panel Size as a production measure is worthy of a status equal to the wRVU. Is it perfect? Heck no, but has anyone really felt the wRVU has been ideal??


Addendum: How to Use Panel Size in a Comp Funding Model

Several members have asked me how specifically Panel Size could be used in a compensation funding model. To answer this I put together the below illustration. This utilizes general data from our most recent survey, but you could certainly tailor to your specific circumstances.

The box with the green header shows median values from the most recent survey, which I then used to calculate a “Total Comp per Active Patient” metric of $368. Then the boxes with peach colored headers use these data in a sample comp model. Both the wRVU and Panel Size (Active Patients) “contractual rates” are roughly 40% of the median survey data, in keeping with the weights expressed in my sample plan.

Personally I don’t like having value compensation tied to production measures, such as wRVUs. To me that makes it a defacto production measure, which is contrary to the point of value. Thus in my example below the “Value” compensation of $116,000 (roughly 20% of total comp) is a fixed per-FTE amount. As you can see, the total compensation earned in this example comes out very close to the median cardiology compensation of the 2018 survey.

It’s important to point out that in most cases not all of the value compensation available is actually earned. In 2018 the median percentage of incentive compensation actually earned was 84%; in other words, 14% was forfeited. Thus, when putting together compensation models this fact must be considered, otherwise the amount actually earned may turn out lower than expectations.


The MedAxiom blog post is published every Thursday at https://www/



Illustration: Lee Sauer


About the Author
Joel Sauer

Joel Sauer, MBA, is Executive Vice President of MedAxiom Consulting. Joel consults around the country in the area of value-oriented physician/hospital partnerships preparing health organizations for the value economy. His work includes vision and strategy setting, creating and implementing effective governance and leadership structures, co-management development, joint venture and other innovative partnerships, and provider compensation plan design. Beyond the above, Joel has a wealth of experience in service line development, clinical strategy development, provider workforce planning; including care team creation and physician slow-down policies, MACRA and bundled payment planning, and operational assessments.

To contact, email:

Leave a Comment

« Back