he goal here is to ensure that the motivations which are obvious in private practice (you get paid for the work that you do) are mirrored in some way in a large academic practice. The method that your group uses is, as Amr E. Abouleish, M.D., points out on page 10 of this NEWSLETTER, the method that rewards people for the work that they do in your practice. I am happy to discuss what we did with anyone, but the plan your group uses must make sense in your own “backyard.”

Where Are You Going?
While this question may seem intuitive — and after you have listened to all of our other contributors, rather mundane — it is vitally important nonetheless. That is, one must think long and hard about what one wishes to incentivize. The plan that I will describe very briefly here has been in development for more than two years, and we still do not think it is complete or finished. It has accomplished much of what was intended, but it has had some unintended outcomes as well. I will attempt to describe both of these.

The purpose of the plan was several-fold. Primarily it was to ensure that stable and survivable growth in the practice plan was not impeded by a lack of anesthesiologists and that those same anesthesiologists were rewarded in an equitable manner. In order to achieve these two primary goals, two secondary goals were to examine prior working patterns (both academically and clinically) and from that pattern establish a productivity standard for both areas.

Origins
It is important to set the environment in which this plan was developed as it is uncommon in academic departments in the United States.

The genesis of this program was born from previous experience in England in the early 1980s. There the departments of anesthesiology were divided along two separate lines. One is what we would think of as a traditional academic department in the United States and the other what we would classically describe as a private hospital. (Of course all of this happens in a very different environment so the use of those terms is loose, but the analogy does hold.) Thus the development that preceded the incentive plan was the division of the department into two arms: academic and clinical. This division was not a part of the incentive plan but required the plan to accommodate it.

As a part of this division, physicians were allowed to decide which of the two arms they wished to pursue. The goal was to allow people to select how they wished to see their careers play out. Some may wish to participate in the training of residents only in the operating room and to have little or no responsibility for the didactic training program. They would choose to pursue the clinical arm. Those involved with teaching the medical students and residents, both intraoperatively and didactically, would choose to pursue the academic arm.

The basic salary structures also are different between the two groups, with the clinical arm earning more in base pay. Those in the academic arm have protected time based on their position; however, they are required to prepare a quarterly summary of achievements to receive their incentive. These achievements are then converted to a point system, and they are rewarded (paid) for the appropriate use of the time. The point conversions are loosely based on hourly efforts but are not exact. Table 1 contains several examples of how these two arms are rewarded within the system and how the clinical incentive plan is woven into this fabric. The “X” in the table allows the reader to input any dollar amount he/she wishes as a starting point. For example, if one used $30K for X in the table, the total salary range for a board-eligible clinical instructor would be $90K-150K. Due to space constraints, the Associate and Full Professor ranges are not published, and while they function similarly, they have different percentages at risk. There also is an opt-out for call once faculty members reach the age of 65 (based on years of service).

Development
The first step was to determine how much money was available for the clinical incentive. This was negotiated with the CEO of the health system. In those negotiations, we demonstrated that our group currently resided in the 75 percentile of work effort for private practice groups using data collected by Dr. Abouleish. Secondly, we began the process of deciding what would be rewarded. This step, as Dr. Abouleish describes, is critical to the success of the plan.

As this is a clinical incentive plan, clinical work is rewarded, but on what basis: days, hours, units, cases, sites? We ultimately decided on the billable hours measure derived from the University of California-San Francisco model. The rationale for this is that it is the most reliable way to assess all types of activity regardless of source (the department does not “cover” any critical care, although some members serve in both departments). The other reason is that it rewards the most efficient use of scarce resources (i.e., nurse anesthetists, residents, physician assistants, nurse practitioners). Pain productivity, however, was based on relative value units, as billable hours were not as relevant here. We also adjusted the average salary to account for this change, so from the faculty perspective, this was an “earn-back” rather than additional salary. We also ensured that the average amount “at risk” would not be more than 20 percent of one’s previous salary.

Once this was done, we started to collect data to use for our standard calculations. We decided to go back one year prior to the introduction of the division of the two arms to avoid confusion of this new method of allocating resources. FY 2003 average billable hours would form the standard from which current calculations would be based. Initially we were going to use a continuous scale to reward effort beyond the base year; however, our preliminary data suggested that one of two unpleasant outcomes would result. Either the starting point for additional compensation would be high enough that many would see no additional compensation or there would be a cap on earnings. Neither of these seemed palatable for an incentive plan, so we compromised and developed a tiered approach [Table 2].

The method of determining comparative efforts was to use the standard year average (“Percent of Standard” in Table 2) as the denominator of a percentage calculation to develop the tiers (i.e., if the average for the site was 1,400 billable hours in FY 2003 and the faculty member generated 1,600 billable hours in the current FY, this would give them a percentage effort of 1.14). In Table 2, one can see that there are differences between the academic and clinical tiers, as the academic group has both a clinical and academic incentive. We did choose to fix two points identically in both groups, which is the tier at 1.3 and the bottom at 0.9. The top end is theoretically unlimited. The current tiers are different, as we have continued to develop and refine the plan.

Implementation
Once the plan was communicated to all members of the department, there was a period for comments, additions and changes. We allowed about two months for these changes and then began modeling the outcomes for each site. The average value used as the basis for all calculations was different for each site. This reflected many realities, including differing missions at the sites, differing penetration and utilization of nurse anesthetists and residents, presence or absence of trauma and transplantation work and differing surgical populations.

We provided each chief and his/her respective faculty members with the effect that the plan would have on them for one full year prior to actual implementation. This allowed both the chiefs and the faculty time to evaluate, discuss and plan for the eventual effect the plan would have on their earning potential. I strongly urge that while other steps are optional, this is one that should not be omitted.

Finally, at the end of the modeling year and concomitant with implementation, we asked each chief to review the data and to give it a “smell test.” That is, were the people who theoretically received the highest compensation concordant with the chief’s beliefs? In almost all cases, this was true. We did need to make a number of small changes to accommodate certain vagaries at specific sites. At one site, for example, only a handful of practitioners practiced at their outpatient ambulatory surgical center. Thus a separate mean was needed to adjust for this. Also a special correction was needed to adjust for part-time faculty.

Outcome
In the final analysis, the outcome that one gets is what really counts. We are now at the end of the first year of actual implementation for the clinical incentive plan. While it resulted in a number of problems (including a recall petition circulated by an anonymous member of the faculty), greater than 50 percent of faculty earned at or above their salary prior to implementation. Total caseload was up, however, so there was more overall work done for the same pay, as many of my faculty would point out. Many feel, however, that they have done well by working smarter rather than harder.

As expected, the biggest complaint was and is that the tiering system does not allow for each hour of work to be recognized on a one-to-one basis. If the financial situation of the department allows, we will move in this direction, but it will result in lower incremental amounts than the tiers. Also the system does not allow those clinicians who choose to pursue academic endeavors to be fiscally rewarded for this effort; this also applies to the academic clinician who achieves more than the required number of points for their allotted nonclinical time. All of these situations require attention and appropriate reimbursement, but there are insufficient dollars for this currently.

Perhaps most gratifyingly, the academic productivity of the department is growing and accelerating. Each area of the department is now contributing to the academic “bottom line” in a way that could only be dreamed of five years ago. The academic point total per faculty is now approximately 10 percent higher than it was in FY 2003. The department also has moved from 12th to 4th place in National Institutes of Health funding among anesthesiology departments, there is a new Anesthesia Patient Safety Program grant in the department, five times the number of abstracts were accepted at the ASA Annual Meeting compared to fiscal 2003, and the department has been asked to sponsor a new medical student course on the basic science of health care and quality.

While nothing about this plan is perfect, it is an acceptable alternative to our previous compensation system. The goal of the plan was to allow people to “set their own compass,” as one of my hospital chiefs has stated. This is quite true, and further, it allows for people to decide when that compass needs to change and under what circumstances. It also has helped to align the incentives of all of players in our system along one axis. It has actually restored control of a significant portion of each faculty member’s life. That alone is a great step in the right direction.

John P. Williams, M.D., is Peter and Eva Safar Professor and Chair, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.