Physician Compensation; Is complexity the enemy of good?

Physician compensation, like healthcare payment structures, has become complex, cumbersome and contentious. What is the Venn diagram intersection of reality, equity and ego?

Physician compensation structures have mirrored healthcare insurance, reimbursement and payment designs during the past decade, and are complex, variably interpreted and difficult to unpack.  The results are highly variable compensation structures, often not firmly based on reliable data or well defined quality, safety and efficiency measures.  In addition there is a large difference between the lowest and highest earning Physicians.  In our health system, this difference is 4X for employed Physicians, and as much as 10X for independent Physicians.

In order to address some of these issues our health system (18 Hospitals in both urban and rural settings) is undertaking a comprehensive redesign of Physician compensation.  The current task force approach is to reduce complexity, increase reliability of metrics, drive equity and align incentives.  Initial focus will be on the employed Physician group, followed by PSAs (Professional Services Agreements; comprehensive contracts with independent Physician groups), and finally influence peddling with other Physician contracts.

We have looked at single payor systems in Europe for Primary Care and Specialty payment designs, as well as metrics for individual assessment of performance.  Large Physician owned multispecialty groups in the U.S. have excellent, but often overly complex KPI metrics.  Health insurance companies have lots of data, but are unwilling to share much that is meaningful.  Most elusive is how to reduce compensation inequity, as this necessarily involves increasing pay for low earners, and decreasing income for high earners.

Anyone else out there grappling with these issues? (I know from another post that some group is at least tackling salary versus incentive compensation and the resulting motivational differences)

  • Reduce complexity?
  • Use reliable measures/data –> a real source of truth?
  • Improve compensation equity?
  • Allow for some individual choice?



Electronic Health Record Implementation


Mismatch in resources and openness

Participant comments on Physician Compensation; Is complexity the enemy of good?

  1. I like your description of the Venn diagram where reality, equity and ego intersect 🙂
    This is spot on.

    I certainly do not have all the answers, but we do not have overly complex KPI metrics attached to physician compensation. Physicians are either paid on a fixed salary + overtime and on-call rates (public sector) or via fee-for-service prices (private sector). Doctors can charge whatever private fees they feel like the market warrants, but the private health insurance rebate is fixed per DRG and patients pay a variable ‘gap’.

    In an effort to reduce the variability of Physician related operating expenses (mostly salary) in the pubic sector, our organisation has very recently introduced a very tight clock on/clock off arrangement, where overtime must be pre-arranged and pre-approved. This went down like a lead balloon the other month when it was introduced, but on the flip side, overtime costs have significantly decreased in the medical groups.

    Let me know when you work it out 🙂

  2. I agree with your assessment of the difficulties.
    Here are a few things we have found helpful:
    -Medical Group Compensation Committees: this has been very helpful in aligning compensation to fair market value and the needs of the organization while giving the physicians a voice in their compensation. They can’t set their own compensation but are able to influence based on system guidelines. These are not paid positions and we haven’t had difficulty getting physicians to attend. This group also arbitrates physician compensation disputes.
    -Compensation based on fair market value: we use three surveys (MGMA, AMGA and ECG). We benchmark against a three year rolling average to lessen the swings in payment.
    -Our markets are mainly fee for service currently.
    -We have two basic methodologies for comp:
    * Coverage model (salary) with 10% at risk for Quality and Service. Hospitalists for example.
    * Production model with 10% at risk for Q&S. Primary care is 80% production, 10% panel size, 10% Q&S.
    The compensation committees choose the Q&S metrics with attention to simplicity. Preferably easy to pull from Epic.
    We have other markets that are highly fee for value and they pay 90% on panel size with 10% Q&S.
    We have worked to make the process transparent and connected FMV benchmarks.
    I think the biggest factor helping to improve the process has been the development of robust physician compensation committees.
    Two years ago one of our comp committees voted not to give compensation increases due to the financial status of the region. Because of the successful communication from the comp committee this ended up being a “non-event”.
    Hope that is helpful.

  3. I don’t think anyone has figured this out for sure.

    My two pieces of advice are that any metrics tied to data need to be VERY clear and understood by both sides. I also think the more frequently the metric is fed to the physicians the more effective they are. Reporting quarterly is probably not enough.

    May the quality/service components meaningful. I think many models don’t have set at a high enough percentage to be meaningful. I think at least 10% is needed.

  4. The issue is certainly recognizable. The difference of 10x versus 4x for independent or employed physicians seems rather large but most likely you would otherwise not be able to compete with the top ten high salaries independent physicians. From the reimbursement side one could use cost pricing method to reimburse but this would allow efficient doctors to earn more but probably not ten times as much. For the employed doctors one could introduce a hybrid system recognizing that high paid doctors cannot do without the less paid doctors at some point and financially recognize the low paid doctors for that.

  5. This is a very prevalent problem in US especially with mergers of hospitals into large health systems and employed model for physicians. We have a RVU based model with 10% for quality and 10% for good citizenship and this has been miserable. Ultimately the hospitals have to decide on what exactly they expect from their physicians. If offering high quality complex procedures are part of the goal and vision, then retaining highly skilled, quality physicians is the KEY. If having the ability to provide all services to all patients in the community is the goal, then standardizing salary will work. But an important thing to remember is not to hold physician responsible for metrices that they cant control.

Leave a comment