Pediatric nephrologist Dr. Bryan Carmody joins us as we discuss what the future of medical education and residency selection looks like in addition to the rise of doctor unions and potential workforce changes. Is a lottery system the solution to a fair Match? What is causing the increase in unionization among residents and attendings? How will new pathways for IMGs change the doctor workforce? The Sheriff of Sodium walks us through the various possibilities of what may be to come.

Here are five takeaways from the conversation with Dr. Bryan Carmody:

1. Residency Match System Challenges

Dr. Bryan Carmody discusses the inherent issues with the residency match system, highlighting the “arms race” of CV stuffing and the dissatisfaction many feel when left unmatched. He suggests that while the algorithm itself works, the surrounding pressures and expectations create significant stress for applicants.

2. The Role of Physician Unions

The conversation touches on the rise of physician unions, with Dr. Michael Jerkins noting the increasing trend of unionization among residents. This shift is seen as a response to the changing dynamics of physician employment and the need for collective bargaining to improve working conditions and compensation.

3. Duty Hour Restrictions and Training

Dr. Bryan Carmody predicts a potential reduction in duty hours from 80 to 60 per week, similar to changes in Japan. This shift could impact the structure of residency programs, requiring adjustments in training duration or methods to ensure competency despite reduced hours.

4. Economic Impact of Residents

The discussion explores whether residents are a financial burden or benefit to academic medical centers. Dr. Carmody argues that residents enable revenue streams far exceeding their compensation, challenging the notion that they are a net expense.

5. International Medical Graduates and Workforce Changes

The episode delves into the potential impact of new pathways for international medical graduates to practice in the U.S. without traditional residency, as seen in states like Tennessee and Florida. This could significantly alter the physician workforce landscape, offering a solution to perceived doctor shortages.

Transcript

Dr. Bryan Carmody:
I don’t think the arms race ever ends. The best we can do is try to use metrics that—by competing—leave people better off. At the end of the day, those who don’t get a dermatology residency or become neurosurgeons don’t disappear. They go on to care for patients and contribute in other ways. The question is whether they’re left better or worse by the competition they’ve been part of.

Dr. Michael Jerkins:
Welcome to The Podcast for Doctors (By Doctors) featuring myself, Dr. Michael Jerkins, and Dr. Ned Palmer. Here, we have honest talks and interesting guests about the ups and downs of training, practice, and what matters to our community. Dr. Palmer, how are you doing?

Dr. Ned Palmer:
I’m doing well. Are you going to say it like “the Ohio State” every single time? Is this going to be a pattern?

MJ:
I hope so. I’ll try to remember every time—it is The Podcast for Doctors (By Doctors) after all. Today we’re excited to talk with Dr. Bryan Carmody, a physician who’s become known for his analysis of medical education, the physician workforce, and what’s working—and not working—in residency selection.

NP:
It was fantastic. I’ve been a fan of Dr. Carmody’s work for a long time. Few people dive into the nuances of things like the matching algorithm, rank choices, and the ripple effects of policy changes as deeply as he does. It was a real privilege to talk with him.

MJ:
What I appreciate most is that he’s willing to talk about topics doctors often avoid—like money, compensation, or even duty hours. For example, I remember during residency when we had a panel on finding a first job. I asked about starting salaries, and none of the four panelists would share. That was… less than helpful.

NP:
I remember that. It was probably the closest you came to being thrown out by our program director! But it proves the point—there are unspoken rules about what we “can’t” talk about. That silence hurts the community. Instead of supporting each other, we compete. And that’s exactly why shining a light on these practices is so important.

NP:
So, let me properly introduce today’s guest. Dr. Bryan Carmody is a pediatric nephrologist at the Children’s Hospital of the King’s Daughters in Norfolk, Virginia. He also teaches at Eastern Virginia Medical School and serves as a pediatric associate program director. But he’s perhaps best known as “The Sheriff of Sodium,” analyzing medical education and residency selection on X, YouTube, and his website thesheriffofsodium.com.

Dr. Carmody, welcome to the podcast.

BC:
Thank you for having me.

MJ:
We like to start with the burning question: how did you become so salty about medical education?

BC:
Mostly through experience. I’ve been teaching medical students and working with residents for years, and that brings you face-to-face with the problems and perverse incentives in our system. About five years ago, I started writing and talking more openly about these issues, and I guess that’s what led me here.

MJ:
You’ve written recently about residency match and selection. Could you give us a state of the union—what’s working and what isn’t?

BC:
That’s a broad question. At the most basic level—the actual running of the NRMP algorithm—the match works fine. The mechanics of processing rank order lists are solid. The problems people have with “the match” are less about the algorithm itself and more about everything surrounding it. There’s the constant CV-stuffing students feel they must do. There’s the reality that many applicants leave the match unsatisfied—or unmatched. Those factors create frustration with the system as a whole.

MJ:
Right. Ned and I were talking about this “arms race” in applications—the endless padding of CVs with abstracts and papers. Honestly, when I look at what students have now, I’m shocked I ever got into medical school. What could possibly stop this? Or should it even be stopped?

BC:
The only thing that would truly stop it would be a lottery system, which some countries have tried. For example, in the Netherlands, once applicants met a threshold, they were entered into a lottery for medical school admission. It’s fair in the sense that everyone has the same chance, but unpopular because it limits incentives to excel beyond the threshold and creates new inequities depending on where that threshold is set.

For medicine especially, where many stakeholders want the “best and brightest,” it’s hard to see a lottery system taking hold here. That’s why I don’t think the arms race ever truly ends. The best we can hope for is to compete using metrics that actually leave people better off.

Imagine if residency spots were decided by how many digits of pi you could memorize. Everyone would memorize strings of numbers, but at the end of the day, the same number of applicants would match, only now they’d all be carrying around useless digits of pi in their heads. It wouldn’t benefit their patients or their careers. That’s the danger of meaningless competition.

MJ:
That’s the best analogy I’ve heard so far. It captures how performative much of this feels—done for appearances rather than genuine growth. But then the big question is: can we design a system people can’t gamify? If not, what metrics actually work?

I really appreciated your write-up on the winners and losers of Step 1 moving to pass/fail. From your perspective as an educator, what metrics truly stratify performance in a meaningful way?

BC:
One of the challenges we face is deciding what we actually care about when selecting residents. Of all the possible attributes and dimensions of performance in a physician, which ones matter most? We don’t really have a clear gold standard. Because it’s hard to define, we often default to whatever’s convenient.

For example, I recently spoke with a group of general surgeons about this. Whatever criteria you use will inevitably be gamed. The question is: which metrics, when gamed, still leave people better off? Ideally, the process should prepare applicants not just for residency, but also for whatever they do if they don’t match.

Take situational judgment tests. Many medical schools now require them, and they’re creeping into residency selection too. Companies like CASPer offer them, and the AAMC had one for a while. These tests ask applicants to respond to scenarios and are meant to assess qualities that the MCAT or grades don’t capture. Data shows they add value. Students generally dislike them—maybe because they’re harder to game—but of course, they can be gamed like anything else.

BC:
The question then becomes: if applicants spend their time gaming a situational judgment test, does it help them? Probably not. In fact, it might have negative value. Compare that with something like laparoscopic training. Imagine a surgery program required applicants to log hours on a laparoscopic trainer—suture practice, timed tasks, measurable outcomes. At the end of the day, the same number of applicants would still match, but now both the “winners” and the “losers” would walk away with useful skills.

This matters because residency educators are being asked to do more with less time. Duty hour restrictions aren’t going away—if anything, they’ll tighten further. So it helps if residents start with a higher baseline of proficiency.

MJ:
On that point about duty hours—right now it’s 80 hours averaged over four weeks. You think that will actually be shortened? What kind of number are we talking about?

BC:
I think 60 is coming. Japan has already moved toward a 60-hour week for medical trainees. I expect we’ll see similar changes here in time.

MJ:
I’ll admit, I wasn’t great at submitting my duty hours in residency. Ned and I trained together, and I may have set the record for “most weeks in a row not submitting.” I’d forget, then retroactively guess. Usually, I just put “8 to 5, Monday through Friday” as a joke.

But seriously—if duty hours go from 80 to 60, is there evidence that patient care quality or training outcomes improve?

BC:
It depends on what you’re asking people to learn. Some things in medicine can be made more efficient—like studying for USMLE with better resources. But others, like procedural skills, can’t really be sped up. You learn them at the rate you learn them. Even cognitive skills are different in real life than on an exam.

On the USMLE, the critical information is condensed into a three-sentence vignette. In reality, that information is locked up in the patient. You have to extract it, separate useful from useless, order the right tests, and ignore red herrings. Those skills can’t be easily accelerated.

And even after residency, physicians continue to grow. A surgeon with five or ten years of experience will likely be far more proficient than someone just finishing training. The question is: where do we draw the line and say, “this person is good enough to practice independently”?

NP:
That sounds a lot like the admissions conversation—what metrics are we even using? Time seems like the wrong one. Toward the end of Michael’s and my training, competency-based education became the big buzzword. It sounded good, but in practice, measuring competency wasn’t easy. You ended up with checklists, bullets, or artificial scenarios—almost like the USMLE again.

So let me ask: if you had a crystal ball, how do duty hours actually change? Will it take another sentinel event, like the one decades ago that drove the first reforms? Or will the pressure from within the system be enough?

BC:
I don’t think another sentinel event will drive change. Instead, it’ll be incremental pressure from multiple directions. House staff unions will push duty hours down. Competency-based education will also play a role.

For example, in pediatrics, the ACGME is rolling out new guidelines this year that cut back inpatient and ICU time during training. When that happens, hospitals can’t rely on residents to provide nonstop, wall-to-wall coverage anymore. They’ll need to find other ways to cover services.

Once that dependency breaks—once hospitals can no longer structure residency purely around filling service needs—you suddenly have more flexibility in how to train residents. That’s been the biggest barrier to reform.

MJ:
That’s fascinating, because it highlights how much residents actually carry in terms of workload. If ACGME reduces inpatient and ICU time, someone else has to take care of those patients. We’ve seen programs close and then backfill with nurse practitioners and PAs. Do you think that’s how this shift will play out—replacing resident work with advanced practice providers?

BC:
Yes. I think that’s happening now, and over the next few years it’ll be fascinating to see. We’ll probably get some data on how different hospitals are managing this shift.

Take free-standing children’s hospitals, for example. Historically, residents fully staffed the ICUs, the ER, and the wards. That won’t be possible anymore. Depending on the size of your program, maybe you can cover the PICU or the NICU—but probably not both, at least not with full resident coverage. That means some units will have uncovered time.

There are lots of ways to address that. You could run a resident service, a nurse practitioner service, or use a hybrid where NPs or PAs cover at certain times. Once you take that first step away from “residents provide wall-to-wall coverage,” it becomes much easier to expand. You might start with part of a unit being run by advanced practice providers, with attendings overseeing care. The hardest part is breaking the tradition that residents will always be there, backed by a jeopardy system to guarantee constant coverage.

And that ties into another interesting question—one you’ve written about before: are residents a net expense or a net profit for academic medical centers? Do they actually cost money, or do they generate it?

BC:
Let me make a broader point here. With selective accounting, you can make almost anything look like a money loser. If you only tally costs and ignore the revenue that flows from those services, of course it looks bad on paper.

All over the country, you’ll find academic physicians convinced they’re being paid more than they’re worth. They’ll go to their chair and ask for a raise, and the chair will frown at a spreadsheet and say, “Well, the ER is a money loser. Dialysis is a money loser. The wards are a money loser. How could we possibly pay you more?”

In a narrow sense, maybe you can make that math work. But it ignores the ripple effects. If you don’t have the ER staffed, who’s feeding patients into the wards? Who’s sending cases to the OR? Services don’t exist in isolation. It’s like Costco’s $1.50 hot dog—you don’t measure its value by the hot dog itself, you measure it by the customers it brings in to buy everything else.

Residents are a classic example. They enable revenue streams far beyond what they’re paid.

BC:
I got an email recently from a resident negotiating for more money. They kept using the word “stipend.” That language bothers me. “Stipend” is the old term used to justify paying residents less, framing it as an educational allowance instead of a salary. But residency is a job. Residents are employees. The Supreme Court even settled that question, partly because of the tax implications for institutions.

And if you want proof that residents bring in more value than they cost, just look at hospital behavior. Even without new CMS funding for additional residency slots, hospitals continue to expand their resident complements. Hospitals don’t dig deeper financial holes just for the fun of it. They’re not adding residents to lose money—it’s clear they’re a net positive.

Yet physicians are often made to feel guilty for asking for more. We’re told medicine is a “calling,” not a vocation. So asking about compensation—or even whether that 6 a.m. Wednesday committee meeting is paid—feels taboo. That mindset is toxic.

MJ:
I think pediatrics can be especially tough. I’ve heard of negotiating tactics where doctors ask for a raise and the response is, “But what about the children?” You’re up against professional negotiators, while most doctors only practice negotiation once every few years. How do you prepare for that?

And to tie this together: we’ve seen a sharp rise in residency unions. Ten years ago, they were barely whispered about. Now it feels like every month a new program unionizes. Have you seen that play out, and what do you think of the gains?

BC:
Where I live, Virginia, residents generally can’t unionize—it’s a state law issue. But in most states, they can, and we’ve seen a lot of big institutions move in that direction. Some of those battles have been very public.

The archetype was Massachusetts General a couple of years ago. As soon as unionization started brewing, residents suddenly got a $10,000 raise, plus other concessions. Institutions may say what they like, but the math is obvious: they’d rather pay more than deal with a unified workforce.

More recently, physicians at ChristianaCare—a large health system across Delaware, New Jersey, and Pennsylvania—began moving to unionize. That’s a new frontier. But it tracks with the larger trend: fewer physicians are self-employed or working in small groups, and more are employed by large corporate systems.

As more doctors come into practice already experienced with unions from residency, I think this movement will only grow.

MJ:
That’s fascinating. I’m in Detroit right now, surrounded by union history—I can see a United Auto Workers logo from my window. So I’ve been following this closely. It feels like residency unionization could reshape things in a big way.

Do you see the AMA stepping in on this? I haven’t seen much enthusiasm from them about unionization for trainees.

BC:
I honestly don’t know. I haven’t followed their take on this. I guess I eventually don’t look to the AMA for insight on issues that I’m trying to understand. But yeah, I’ve got to think that the dynamics are shifting. Because, like I said, 30 or 40 years ago, I think it would have been very easy to see the AMA just rolling their eyes at unionization because most doctors were in business for themselves essentially. But now, as you all know, the single biggest employer of physicians is not even a health system—it’s Optum, far and away employing more physicians than anybody else, more than the VA, more than Kaiser Permanente. And so as trends like that increase, I think it’s very easy to see. Once physicians become a cog and easily replaceable employee, they have the same incentives to unionize as coal miners or auto workers, or anyone else who does work that’s valuable but is individually replaceable.

And do you, just going back a little bit on the bottom-line impact that residents have on these medical centers, at one point I think you had published a number—which, of course, is pretty impossible to really define for every single trainee across the country—but what is the rough ballpark that you think about regarding the impact on the bottom line that residents have to medical centers?

BC:
Well, I think the number that I use was from a RAND study, which was the best I found, where they comprehensively looked at the inputs and outputs. There’s a lot of variability. It’s obviously possible you could add a resident who doesn’t enhance revenue. If that resident simply divides up call among more people but zero new patients had expensive procedures or got admitted, that resident wouldn’t have a net financial benefit. But that’s not usually the circumstance in which most hospitals choose to add a resident. So yeah, the RAND study, I want to say, had a ballpark of about $200,000 per resident. Of course, it depends on federal subsidies and the specialty of the residents. I think that’s a credible estimate.

MJ:
Yeah, it’s interesting. I think about this personal story—I’ve told it a hundred times, so apologies, Ned—but as a second-year resident, my wife and I had a child at the hospital I worked at. Over the course of a month, you can imagine how much charges I generated for that hospital. And we got the bill, which, without insurance, would have been the same amount. This was for a healthy, full-term baby, one day hospitalization, no complications. The bill I got for my child was about the same as my take-home pay for an entire month. It’s very interesting to think about what I generated for the hospital versus what I was paid. And I feel like a lot of residents are starting to catch on to this. There have been studies showing similar numbers for neurosurgery residents, for instance. I think more and more are waking up to this reality.

BC:
Yeah, I think many people have had a moment like that.

And one thing I wanted to ask you about—because I saw you tweeted about this recently—is on overtime pay. As you know, we already talked about duty hours and working holidays and nights. Many residents ask about this. What do you think about residents advocating for overtime or holiday pay? What’s happened on the macro scale, and where do residents fall into that?

BC:
Well, residents have always been in a netherworld regarding some of those labor laws—for reasons both probably just and unjust. If you’re a resident, minimum wage laws don’t apply; you’re not an hourly worker. Residents can total up their hours and divide by their compensation and see that they make less than a wage in some localities. But the laws don’t apply to them; there’s a specific carve-out. I do think that will change over time, for the same reasons it’s changing elsewhere. Those political forces and desires to satisfy constituencies will eventually apply to medical trainees as well. And when that happens, it’ll be a powerful incentive to lower duty hours. If you have to pay overtime, it’s easier to staff differently.

MJ:
Absolutely. Then the bottom-line impact is very different. I know Ned and I were talking about this before and racking up a little on the match. I want to ask this real quick: when it comes to match rates and specialties that have had years of successful match rates, maybe the next year sees a dip—where have you seen trends in certain specialties on fill rates? Especially as you’ve outlined on your blog, “winners and losers.”

BC:
Many specialties are remarkably stable over time. If you look back at years of match rates for something like maternal medicine, it may fluctuate slightly with the number of applicants, but match rates are generally high and stay high. Medicine is a good example: even when the match rate is 96%, it doesn’t mean everyone goes home happy. There’s a spectrum of programs, and at the upper end, internal medicine programs can be as competitive as any specialty. Rates for MD graduates tend to be pretty good overall, but some specialties have grown more or less competitive over time.

The epitome of that is anesthesiology. In the 2000s, anesthesiology reached a point where, I want to say, 25–30% of positions went unfilled—something unheard of in any specialty these days. That was when people worried anesthesiologists would be replaced by CRNAs. Nowadays, anesthesiology has grown more competitive and sought after each cycle. Psychiatry has also become more competitive over the past 10–15 years. PM&R is moving in that direction, still small but gaining interest. On the other hand, radiation oncology has become less competitive; 15 years ago, it was highly sought-after with top applicants, PhDs, AOA, etc., but now some spots go unfilled. Orthopedics remains extremely popular, and family medicine and pediatrics still have leftover positions each year.

MJ:
Can we talk about emergency medicine specifically? They underwent a significant upheaval a year or two ago with reports on approaching saturation. Have you seen that trickle down to applicant enthusiasm?

BC:
Yes, there was a big shift. Two years ago, the story of the match was a number of unfilled emergency medicine positions—over 500, if memory serves. You didn’t have to go back far to find years with nearly zero unfilled spots. Emergency medicine is a large specialty, so even a handful of unfilled positions is notable. The decline occurred over a couple of cycles, with decreasing interest, especially among USMD applicants. This past year, the number of unfilled positions was significantly fewer, but the composition of applicants shows that USMD participation hasn’t fully rebounded to pre-2020 levels.

BC:
I don’t think that’s going to change. I think instead what will happen is that spots will be filled by a greater proportion of osteopathic medical graduates and international medical graduates. Emergency medicine is transitioning into a specialty that used to be almost exclusively filled by US MD graduates, and now it’s going to be more balanced. Many specialties, like internal medicine, have already found that balance. Internal medicine remains a very sought-after specialty for many medical students, and the applicants who match at the top programs are every bit as polished as those matching at top dermatology or neurosurgery programs. I think you’ll see the same thing in emergency medicine, with programs recruiting in a similar manner.

And you’ve commented on this before on your website or Twitter (X, sorry), related to recent legislation in Tennessee and Florida that allows physicians who did not complete a US residency to practice—though there are stipulations. How do you see that affecting the workforce? Do you see more states adopting similar legislation?

BC:
I think this is going to be the single biggest change in medical training in recent history. Many states have already adopted similar programs, and I expect more will. The incentives are clear: medical practice in the US offers economic opportunities that aren’t available elsewhere, so the supply of qualified applicants is effectively limitless. International medical graduates can increase the supply as long as they meet US standards, and the economic incentive to come is very strong.

At the same time, hospital systems face challenges providing resident coverage at past levels because residents are unionizing and working fewer hours. One way to staff efficiently is to bring in international medical graduates on these new pathways. In Tennessee and Florida, for example, these physicians will work in hospitals that train residents—not as residents, but in some supervised capacity. This could be nocturnists or hospital service roles, providing a somewhat cheaper alternative to wall-to-wall resident coverage. More hospitals will adopt this approach.

It’s unclear whether these physicians will be able to practice outside the state where they trained, because other states may still require a US residency for licensure. But hopefully, states will find ways to direct these graduates to areas of need. This could completely shake up the physician employment market, which, as we’ve discussed, is increasingly like employment in other sectors—physicians have skills, but they’re more easily replaceable.

MJ:
And if insurance companies don’t reimburse services for these physicians in the same way as board-certified physicians, that changes the economics of whether this pathway makes sense.

BC:
It’s hard to imagine board organizations blocking this, because they benefit from certifying more physicians—the more exams they sell, the more physicians are on the hook for MOC. Even if they didn’t have that incentive, lobbying from hospitals that supported these laws would likely influence the boards. These efforts have passed legislatures in every state where introduced, often framed as solutions to doctor shortages.

NP:
The board piece is interesting because boards have historically been self-regulating, with minimal oversight. There’s even precedent for creating alternative boards, like a congressman from Kentucky who started his own board after being dissatisfied with the American Board of Ophthalmology. So there could be some incentive for increased oversight over boards and their quality metrics, which would be an interesting push.

BC:
I agree.

NP:
We could probably talk about this for hours, but you probably don’t want to spend that much time with us, Dr. Carmody. Ned and I are big fans of your work, especially doctors advocating for our community. These are the people we train with and practice with, and they’re being affected by these changes. It’s valuable to have a physician’s voice educating on these topics.

I want to close with one last question: Being a physician, dentist, or veterinarian is always changing—our opinions, treatments, and approaches evolve. For you, has there been anything in medical education or these topics that you’ve recently changed your mind about or evolved your viewpoint on?

BC:
Well, actually, that’s where I mentioned it before—the physician shortage. It’s something I used to believe in, but I don’t really believe in anymore. I’ll probably be talking about the USMLA standard setting a little bit this summer. My opinion has shifted over time—it started in one place, shifted, and then shifted back, I guess that’s the best way to put it.

So yeah, I think my positions do change over time, but sometimes it’s hard to remember the way you used to think because your current perspective feels so familiar.

MJ:
Yes, that’s totally fair. We really appreciate your time shedding light on these topics that matter a lot to our community. Thank you, Dr. Bryan Carmody—the “Sheriff of Sodium” himself—for joining the podcast today.

BC:
Thanks for having me.

MJ:
Thank you, Dr. Carmody, for coming on the podcast. It was great to talk about all these topics we love to nerd out about, and even better to hear from someone who’s an expert.

NP:
Absolutely. I was really excited to hear Dr. Carmody’s take on the changes in physician employment, the shifting physician labor structure, and what that means for the future. It was also valuable to have an honest conversation about the rise of physician unions—three areas that don’t get much attention.

It’s interesting because the health system has so many stakeholders—residency programs, hospital systems, boards, insurance companies—all with competing interests. As doctors, we’re at the center of it. Hearing him talk about international medical graduates, how that could affect trainees, boards, board certification, and the economics behind it was fascinating.

MJ:
Very much so. I thought I was fairly in touch with healthcare trends, and it’s humbling to realize I’m not. Dr. Carmody is incredibly engaged and deeply knowledgeable. His comments on the future of residency work hours were a prime example—he’s way better connected in this space than I am. For everything I’ve read, I don’t think I’ve seen an analysis like his.

What do you think—60 hours down from 80 hours?

NP:
Sixty hours is surprising. If you reduce residency hours by 25%, do you extend training by 25% to maintain a fully qualified physician? That would turn a four-year program into five years. It depends on how you structure it.

As duty hours are cut back, people will realize how many physician extenders are needed to replace trainees. The amount of service and hours trainees provide will have to be absorbed by others, and it’ll be fascinating to see how hospital systems handle that.

MJ:
Well, that wraps up another edition of the Four Doctors by Doctors podcast with Dr. Ned Palmer and myself, Dr. Michael Jerkins.

NP:

Thanks, Michael. You can catch The Podcast for Doctors (By Doctors) on Apple, Spotify, YouTube, and all major platforms. If you enjoyed this episode, please rate and subscribe. Next time you see a doctor, maybe prescribe this podcast. See you next time.

Check it out on Spotify, Apple, Amazon Music, and iHeart.

Have guest or topic suggestions?

Send us an email at [email protected].