Demerit Scholarships: A Method for Weeding Out the Marginally Bad Resident

I propose a program that identifies weak residents, those ill-suited to the practice of orthopaedic surgery, and bestows upon them large sums of money.

Of course, the money comes with strings attached. In return for accepting these funds, the recipients must agree to withdraw from residency and find their happiness elsewhere. Indeed, the whole purpose of these “demerit scholarships” is to ease out those who do not belong.

Although the resident selection process in orthopaedic surgery is among the most rigorous in medicine, every now and then the process breaks down (or a resident himself suffers a breakdown) and we discover years later a once-promising candidate who really is unfit to continue. At times, the flawed resident’s performance is so egregiously bad that the program would be willing to pay any price, bear any burden and meet any hardship to assure the expulsion of the resident. Yet at other times—and more commonly—the resident’s performance is only marginally bad: bad enough that we would like him out, though not bad enough to motivate expulsion, and all of the hassles it brings. In the idiom of chemistry, the departure of these troubled residents is a thermodynamically favorable reaction, but there is a barrier of activation energy. The entire purpose of the demerit scholarship program is to catalyze the departure of residents who should be gone.

It is helpful to consider the problem from the perspective of the resident. Troubled residents are rarely willing to go without a fight. They have made substantial investments in their careers, and do not want to leave residency empty-handed. Because of these high sunk costs, they are likely to resist vehemently any invitation to leave, especially if given no good alternatives. Ideally, when residents are told it is time to go, we want them packing up, not hunkering down. Demerit scholarships can help cultivate that proper attitude.

Residency programs exist primarily for the benefit of the public. Accordingly, only residents who will serve the public well should be allowed to remain. Recall Michelangelo’s insight: to make the statue of David you must chisel away all of the stone that is not David. To form the ideal cadre of orthopaedic residents, we must chisel away all of the residents who are not ideal.

Demerit scholarships may sound like a Swiftian “modest proposal,” not suggested in seriousness, but only to make a point. To the contrary, demerit scholarships are practical and necessary. Think about them as a kind of severance pay. In that sense, demerit scholarships would be akin to the terminal year of salary given to assistant professors who don’t earn tenure. Demerit scholarships provide recipients the resources to facilitate a transition to another field.

Admittedly, raising the funds to sustain this program will be a problem. The federal government, which provides the bulk of the funding for residency training, would like to spend less on graduate medical education, not more; and philanthropic donors are apt to be resistant as well. (Donors are said to prefer constructing new buildings, whereas demerit scholarships are more like razing the old).

The problem of moral hazard must also be considered. Specifically, if residency selection committees were to know that they could, in economic terms, “externalize the cost” of bad selections—by asking the Demerit Scholarship Program to rid them of their poor choices—they might take riskier candidates. (Come to think if it, this appetite for risk, and with it, a willingness to consider atypical applicants, may be a net plus.)

Demerit scholarships deserve our attention and support. At present, it is very difficult to fire a resident, and on balance, that is probably a good thing. Program directors already wield tremendous power over the fate of a resident’s career. Still, the restrictions on program directors who want to weed out the marginally bad resident may be just a little too constraining. By creating a system that allows program directors and troubled residents to reach a mutually beneficial, forward-looking decision, a happy middle ground is met. Demerit scholarships, competitively awarded, may be just the trick.

May the worst man win.

William N. Levine MD

Professor and Vice Chairman, Department of Orthopaedic Surgery

NYP/Columbia University Medical Center

Joseph Bernstein is a bright, thoughtful orthopaedic surgeon who has fashioned his career on musculoskeletal education and has always been considered a resident mentor to students, residents, and fellows. I also consider Joe a friend, so it is with a great deal of reticence that I completely reject his “non-Swiftian” proposal to introduce demerit scholarships as a pathway to weed out the “marginally bad” resident.

While I can understand some of Dr. Bernstein’s motivation to introduce this concept, the fatal flaw of his proposal is that it is simply not necessary. The implementation of the Accreditation Council for Graduate Medical Education (ACGME) milestones on July 1, 2013 will further aid program directors, chairmen, and programs in doing exactly what Dr. Bernstein has recommended—identify the “outlier” early in the process with the goal to either remediate or ultimately remove the resident from the training program. Until now, the evaluative process has been extremely subjective and challenging, making it difficult to identify accurately those residents who are at-risk. Sometimes we do not identify these residents until it is too late. Anecdotally, I know of many residents who have made it all the way to fellowship interviews (fall or spring of the PGY-4) when they are then informed that they cannot pursue a fellowship due to their lagging performance. Clearly, that is a pathway that none of us would support. I believe the milestones (albeit with all of the angst, limitations, and uncertainty inherent to this new system) will finally give programs the “blueprint” necessary to institute corrective plans of action early in a resident’s career.

Additionally, Dr. Bernstein describes the challenges in firing the marginally bad resident, but in fact there are many programs in the country that have demonstrated that with appropriate and not typically overly onerous documentation, the fears of firing underperforming residents are rarely realized (fear of litigation, decreased resident complement, and the like).

Finally, and perhaps most critically, Dr. Bernstein fails to detail where the programs will get the resources to “bestow large sums of money” to the resident whose contract is not renewed. As Dr. Bernstein correctly points out, we face a critical time in graduate medical education, with fiscal responsibility becoming more and more difficult to achieve. We certainly will not see an increase in federal funding for such a program. The analogy to the Assistant Professor who does not achieve tenure is weak at best. The poorly performing resident is being told that orthopaedic surgery is not the appropriate career path while the Assistant Professor just did not meet the necessary milestones for that academic center.

In summary, Dr. Bernstein’s proposal is interesting (and certainly provocative), but ultimately flawed. It will not keep the “marginally bad” resident from being selected by a program, nor will it make it easier to eliminate that resident from a program once that resident’s training has started. It is incumbent on all of us to ensure that we do everything we can to identify the at-risk resident early, and act on this information appropriately, leaving dismissal (nonrenewal) as a step to be used only for those residents who fail despite these efforts.

James E. Carpenter MD

Professor and Chair, Department of Orthopaedic Surgery

University of Michigan

Dr. Bernstein has correctly pointed out that not every medical student who enters an orthopaedic surgery residency is well-suited to the field. Despite medical school performance evaluations, standardized testing, and brief in-person interviews, we remain unable consistently to determine who will develop into a competent orthopaedic surgeon following 5 years of residency training. While most residents successfully complete training and become proficient orthopaedic surgeons, there is (and has always been) a small percentage of entering residents that will not. As Dr. Bernstein suggests, it would be best for our health system, our patients, and likely these residents, if they were recognized and dismissed from orthopaedic training, or perhaps moved to another field.

One option to achieve this end, as Dr. Bernstein suggests, would be to create a buyout system, so popular with high-profile athletic coaches these days. While innovative, this idea is neither practical nor affordable. Another option to meet this need would be to make progression in training something to be earned by most, but not all trainees, rather a nearly automatic step, such as it is today. Just as not all students are accepted into medical school, and not all medical students are able to match into orthopaedic residency, not all residents should be expected to complete training. Instead of routinely moving to the next stage of training, progression could be based upon a series of deliberate decisions supported by objective knowledge and skill. These performance metrics would bring with them the expectation that not everyone will move on in sync. High stakes evaluations would be moved to each phase of training, rather than after training as part of the board certification process. Failure to meet these standards might automatically result in a 1-year delay. A second failure might result in termination of orthopaedic training. Firm cut points for scores and task performances would be required to avoid the hassles and potential litigation associated with “flunking out” of a training program. This could also be helped if it became a standard expectation for training programs that some portion of entering residents would need more time or would need to leave the program for another field.

Rather than view such a dropout rate as a sign of a poorly performing program, the ACGME might flag programs with no resident losses or delays for further scrutiny. Program directors could be aided in their effort to manage the variability by allowing flexibility in the adding or dropping of PGY-1 positions and by allowing graduate medical education funding for an individual to extend beyond the 5-year limit without a penalty. Additionally, certifying organizations, such as the American Board of Orthopaedic Surgery, might move to embrace a competency based criteria for certification, rather than the traditional time-based requirements.

Some may contend that this is a goal for the ACGME Next Accreditation System. However, the development of objective and validated performance evaluations is lacking at this point. Such that everyone knows all legal associates do not become law firm partners, and each new stockbroker trainee does not become a fund manager, our training systems should recognize and expect that not every PGY-1 in orthopaedics will become an orthopaedic surgeon.