A Comprehensive Solution to the Problem of Unmatched Medical Graduates

Today, Friday 3/18/2022, is Match Day.  This week tens of thousands of graduates of US MD and DO schools will find out what residency program they matched to.  And many tens of thousands will find themselves having not matched to any position at all.

These situations are always tragic- medical students have worked hard for 8 years (4 years undergrad, 4 years medical school), are often hundreds of thousands of dollars in debt, and find themselves with no clear path to making enough money to clear that terrible burden. Especially combined with projected future physician shortages AND the reality of current lack of access to care.  Many will declare that we need more residency spots to train more doctors. And we do. But the situation is more complex than it first appears, and reflexively increasing residency spots actually won't help anyone.

Instead, I propose these four steps, which will address the root causes of the problem, and create enduring change.

1. Funding for more residency spots, accompanied by a national healthcare workforce commission empowered by congress to conduct and analyze research into future workforce needs, and allocate (and re-allocate) GME funding for MD and DO residencies accordingly
2. A multi-step match. 1st step: US MD and DO grads. 2nd step (1 month later), unmatched US MD and DO grads along with Foreign Medical Graduates (FMGs) and US citizen graduates of international medical schools (IMGs).  This should be accompanied by a repeal of the NRMP provision requiring programs to offer all of their positions in the match.  Meaning if they want to specially recruit an FMG due to research or other relationships, they can still do so.
3. For all US MD and DO graduates, IMGs, and FMGs who pass Steps 1, 2, and 3: guaranteed entry into special Physician-Assistant Qualification pathways, which would be 6 months long and allow them to gain licenses as PAs.
4. Forgiveness of all educational debt for any US MD or DO graduate AND any US IMG who does not match 2 years in a row, and who does not qualify for a pathway to practice as a PA
5. If we retain preliminary residencies, then for all graduates who successfully complete a 1 year preliminary or transitional year residency (which do not lead to board certification and consequent allowance of independent practice), we fund and establish guaranteed reserve family medicine 2 year residencies.  These "completion" residencies would ONLY be open to such graduates, and would have to go unfilled otherwise.

I truly believe this multifactorial approach is the best path to solving the issue of unmatched medical school graduates. Simply expanding residency spots will never work.  But to explain why is complex.

Residency spots are not truly in shortage relative to US MD and DO applicants

To begin, simply expanding residency spots will not work.  We actually have ~37,000 residency positions for 27,000 US MD and DO seniors.  It's only when you take into account IMGs and FMGs that the total number of applicants is greater than the total number of positions.

(Results and Data: 2021 Main Residency Match)

While FMGs and IMGs make great physicians (I have several in my family), we obviously can not as a nation guarantee a residency spot to anyone in the world who wants one, and we should try to educate the physicians we need here, in the United States.  We should not send US citizens to the Caribbean or anywhere else, especially if US government backed educational loans are paying for their education.  Nor should we make a practice of taking physicians educated at the expense of the government in India and Mexico- to do so deprives these nations of physicians.

And indeed, all this leads to an obvious conclusion: the problem isn't that US MD and DO grads don't have enough residency spots.  It is that US MD and DO grads are applying for desirable positions that they have a lower chance of matching in.  Consequently, less desirable specialties and programs wind up interviewing FMGs and IMGs, and taking them- and thus are then unavailable to US MD and DO grads if they don't match

Furthermore, expanding residency slots will almost inevitably lead to more FMG and IMG applicants- just as building more highways to reduce traffic congestion leads to more people driving and consequently the same amount of congestion.  Thus, simply expanding residency spots (which to be clear, we should do) will not solve the problem.

And in fact, this brings up a critical question: which residencies should we expand?

Residency and fellowship spots in some specialties are in shortage relative to the healthcare needs of the nation

While many many specialties are in shortage, some are not. Radiation Oncology and Emergency Medicine are in fact widely projected to be in oversupply.  We need more family medicine physicians, and we need even more if we are to reduce patient panels and give FM docs enough time to actually think about each patient. Do we need more surgeons?  Probably.  But we're probably training too many pediatric surgeons and endocrine surgeons, among others.

The only way to sort this out is to truly decide on a national level what our needs are, and try to allocate accordingly.  This can be spectacularly off: notably, cardiothoracic surgery was in shortage, until the rise of endovascular techniques led to interventional cardiology taking over a lot of volume, leading to a surplus of CT surgeons in the 2000s.  Vascular surgery has exploded in demand as fewer and fewer general surgeons are comfortable dealing with vascular emergencies, and more and more of the field turns to endovascular techniques which are highly specialized and increasingly in demand.  Gastroenterology similarly exploded as routine endoscopic screening for colon cancer became established.  The entire field of HIV medicine may one day disappear as the drugs we have to treat it become better and better.  Nonetheless, unless we as a nation at least try to project and allocate future needs, we will be left with our current system- where the number of residency positions available is based primarily on how many residency spots were available last year, and how many residents the program feels like they could use and train.  No other developed nation leaves it up to individual programs and medical centers to determine how many doctors we will need- because training doctors can not work in a free-market system.

Fundamentally, medical centers that pay for and train doctors are training their competition.  That it happens at all is because of altruism and the professional ethos of medicine which goes back to the Hippocratic Oath: "I swear by Apollo... to impart precept, oral instruction, and all other instruction to my own sons, the sons of my teacher, and to indentured pupils who have taken the Healer’s oath, but to nobody else"

The last phrase of course shows that guild-like protection of physician incomes by restricting supply goes back millennia as well.

Still, there are reasons not to train too many doctors.  Every study ever done has shown that volume drives outcomes- surgeons who operate a lot are generally have better outcomes.  Older internists who keep seeing a normal number of patients instead of cutting back have lower mortality.  To be clear, as a nation we are nowhere near this situation.  We have far too few physicians in most specialties, and it's getting worse- especially in family medicine.

But, we need to keep this in mind because it drives a critical reality: we will never have and should never base the number of available residency spots on the number of medical students who want to go into that field.  Doing so will lead to oversupply and worse outcomes.  We have to base it on national workforce needs.

This leads to the first two proposals: a national, legislatively empowered commission to allocate and re-allocate residency funding, and a multi-step match which gives US MD and DO graduates first choice- and only then allows IMGs and FMGs together with unmatched US MD and DO graduates to look at the remaining spots.  Thus, US MD and DO grads who do apply for more competitive specialties or who initially don't look at certain locations can do so.  If they do not match, they have a chance to re-align their expectations, and try for spots more in line with their statistics.  This gets into an entirely separate issue of how to pick applicants for residencies, and that really is beyond the scope of this piece.  But what about those applicants who don't match despite this system?

Not every applicant can match or should match and complete residency.  We need to provide off-ramps and make use of their education as well.

There is a lot of legitimate criticism about the way we evaluate residency applicants in the United States.  Nonetheless, the reality is that there are medical students who graduate medical school, but who nonetheless should not or can not complete a residency.  To be clear, at present, these are vastly outnumbered by those who would make excellent physicians.  

Historically, the match was entirely optional to practice.  Not every aspiring physician had the resources and capability of upending their life to go wherever they had to to complete a residency.  To become a fully independent general practitioner did not require a residency- it only needed a medical school diploma, and later on only an intern year, which were widely available.  Consequently, the match was only for medical students who wanted to become specialists.  Any other medical school graduate could do an intern year and go into general practice.  But in the 1960s and 70s, the specialty of family medicine was born, and over the years the pathway of practicing right out of medical school has closed off, and rightfully so as medicine has become more and more complex.  There remains some options for graduates of medical school who do one year of residency to take up roles in wound care or in rural areas, but by and large a residency is required for most jobs.

Becoming a physician is hard, and it should be hard, and we can not expect that all graduates of medical school should get a residency and full independent practice.  It's hard to describe just how much power and damage a physician who isn't significantly dedicated to their patients can do.  The drive to work, to stay current, and to do the right thing for patients must be paramount.  It is difficult to predict which physicians will do well, and there is no correlation between how high someone's board scores are and how good a physician they are.  But there is a correlation between failing a licensing exam and failing to become board certified, and correlations between failing to become board certified and negative outcomes.  It is not unreasonable for program directors to view negatively applicants who have failed licensing exams, or who have other red flags such as criminal convictions.  These students may graduate medical school, but find themselves unable to match.

And as noted above, many Foreign Medical Graduates want to come to the US, or are already in the US- their partners may have gotten jobs and visas here.  Not taking advantage of their skills is foolish, even if we can not provide them with residency training positions.  For all these categories of students, we need to provide pathways for productive use of their knowledge, or at least measures that will keep their life from being ruined.  This gives rise to proposals 3 and 4.

For US MD and DO students, and FMGs and IMGs who can not secure a spot, we should make use of their training by giving them a guaranteed "off-ramp" to practice.  Thankfully, we have a profession, Physician Associate, which receives broadly similar training to MD students- their curriculum is indeed in large part a subset of the MD curriculum, as they train in the medical model.  They take roles in the healthcare system that require supervision.  Most MD and DO curriculums far exceed the requirements and curriculum of PA schools.  With some additional curriculum focused on practical skills (PA schools require orthopedics and familiarity with splinting as well as more hands-on readiness to assist in surgery for example) many medical graduates could easily be ready to become fully licensed PAs- with the ability to meet the needs of the population in an established role.

Finally, for those graduates who can not pass their exams, and who can not take the pathway above to becoming a PA, we must address the massive educational debt that will forever burden them and keep them from starting families, buying homes, and contributing to society in other roles.  If any US MD or DO school, or foreign medical school will not refund the tuition of students who can neither match nor become PAs, then the US should bar those schools from being able to receive US government backed education loans.  This will encourage schools to be more selective about admits, mitigating current predatory practices which saddle too many students with far too much debt and little realistic hope for securing a residency.  Guaranteeing this will actually help with significant disparities in healthcare workforce recruiting.  At present, many underrepresented minorities in medicine (URM) are scared off of applying due to the prospect of financial devastation if they fail to complete medical school and match.  By giving them a guarantee that should they fail despite their best efforts, they won't be burdened by debt forever, we can improve the diversity of the profession and give opportunity to all.

Conclusion

The US needs more doctors. The US needs more residencies.  How to finance residencies is a separate topic altogether.  The issue of preliminary surgical and medical residents raised in my fifth proposal is also a complex one, that I won't go into further in this post.  Nonetheless, simply adding more residencies will never address the root causes of the disappointment, pain, and fear on match day.  It will not address the reality that the US is underutilizing talented healthcare professionals.  Nor will it address the reality that the US is not training the doctors it truly needs for a healthy population to have good access to care.  Only by determining the healthcare workforce needs of the nation, prioritizing US MD and DO grads in a two step process, and giving off-ramps to applicants who cannot match can we fundamentally fix the system.