Dr. Graham Walker, an emergency physician, software developer, and the visionary behind MDCalc, shares how he bridges the gap between technology and medicine. Dr. Walker delves into the evolving dynamics of healthcare, highlighting generational shifts among doctors and the rise of employed physicians over traditional practice ownership.

Are healthcare providers’ passion and dedication being taken advantage of? How has COVID reshaped the mindset of a new generation of doctors? And is there a balance between personal fulfillment and professional devotion in modern medicine? Dr. Palmer, Dr. Jerkins, and Dr. Walker explore these notions and more in this thought-provoking episode.

Here are five takeaways from Episode 9 featuring Dr. Graham Walker:

1. The Exploitation of Medical Professionalism

Dr. Walker and the hosts discuss how the healthcare system often exploits doctors’ sense of duty and professionalism. The expectation that physicians will go above and beyond—working unpaid hours or attending unnecessary meetings—is framed as part of their calling, but this has led to burnout and disillusionment.

2. Generational Shifts in Medicine

There’s a noticeable shift in how younger doctors view their careers. Unlike previous generations who saw medicine as a lifelong calling, many newer physicians treat it more like a job due to systemic pressures, lack of autonomy, and overwhelming debt. COVID-19 also played a role in reshaping their clinical experiences and expectations.

3. MDCalc’s Origin and Impact

Dr. Walker created MDCalc to simplify clinical decision-making by digitizing medical scoring systems. What started as a personal tech project has become a widely used tool among physicians, with over two-thirds of U.S. doctors using it monthly. The platform now includes hundreds of validated scores and continues to grow.

4. AI in Healthcare: Promise and Pitfalls

The episode explores the role of AI in diagnostics, particularly large language models (LLMs). While AI shows promise in identifying rare conditions, its effectiveness depends heavily on input quality and physician understanding. Dr. Walker emphasizes the need for proper education and feedback loops to build trust in AI tools.

5. Empowering Physicians Through Transparency

Dr. Walker introduces Off Call, a platform aimed at helping physicians regain control over their careers. It provides transparency around compensation, work expectations, and contract negotiation. The goal is to support doctors in making informed decisions and navigating employment in a system increasingly dominated by corporate interests.

Transcript

Graham Walker:

That oath that we have taken to help our patients has absolutely been exploited by the healthcare system.

Michael Jerkins:

Welcome back to another episode of the podcast For Doctors, By Doctors. Dr. Michael Jerkins joined by the one and only Dr. Ned Palmer. Ned, are you there, Ned?

Ned Palmer:

I am here, Dr. Jerkins.

MJ:

I’m glad you’re here.

NP:

I am also glad I’m here. How are you today?

MJ:

I’m good. I have a total non sequitur to start this amazing show we have with our amazing guest, but I was actually— I don’t know what you call it in your neck of the woods. In my neck of the woods, we call it “rolled.” Some people call it “teepeed.”

NP:

Yeah, wow, okay. Real Devil’s Night classic here.

MJ:

Yeah, so Devil’s Night—I don’t know about that, but I know you guys in Michigan say that phrase.

NP:

Detroit specifically, but sure.

MJ:

You can have that, Detroit. It hasn’t happened since high school, so I didn’t know what was going on. I was kind of an angry old man and cleaned it up last night.

NP:

Who are the culprits here? Do you think this was a random act of homeowner violence?

MJ:

I know who the culprits are. What’s different from when we were in high school is that there are doorbell cameras now—and I have one.

NP:

A lot of them.

MJ:

So I do know who it was, and it’s someone I know and it’s friendly. But it was just a blast from the past, and I had to clean toilet paper from trees in my yard for a very long time.

NP:

I hope you didn’t get egged too. That used to be the one-two punch if someone was really upset with you.

MJ:

That’s next level. That’s too much.

NP:

It’s too much. And frankly, I’ll be an angry old man here for a sec—with the price of TP and eggs anymore, who’s even got the capabilities? The level of waste there… We went from hoarding toilet paper for a whole year in 2020 to just throwing it around the yard. I guess that’s progress, but it doesn’t feel like it.

MJ:

You’re right. It kind of prices people out of that particular type of vandalism—there are forms that are not as expensive.

NP:

That’s true. Are you comfortable naming and shaming the culprit on this podcast?

MJ:

I am in general, but not on this podcast. I promise you they’re listening though.

NP:

So was it directed at you or your son?

MJ:

It was still directed at me—25 years later.

GW:

It was me.

MJ:

Rolling—we call it rolling. Got toilet paper roll.

NP:

I could make that leap, yeah.

MJ:

That seems a little aggressive. Why would you call it that?

NP:

We used to burn down the city of Detroit. Through the 70s, 80s, and 90s, abandoned homes would get set on fire on the night before Halloween—October 30th. I thought everybody did this, but it turns out it’s hyper-local.

MJ:

That seems like a joke.

NP:

I don’t think anybody was kidding. A lot of people did it for insurance checks—absolutely committed fraud. The history of Devil’s Night is actually fascinating. I’m going to start a separate podcast: Devil’s Night for Doctors, By Doctors with Ned Palmer, where I’ll interview Michael Jerkins about his preconceived notions around Detroit.

MJ:

Well, I feel like it’s a good segue into our interview. Doctors’ lives are stressful in lots of ways, including having your house vandalized. I’m very excited about our guest, Dr. Graham Walker, who is someone I’ve used an app from for a very long time—MDCalc. He’s a pioneer in technology and medicine and an advocate for doctors. I’m excited to pick his brain and hear his story.

NP:

Same here. I don’t think we’ve had another guest whose end product I’ve used more than Dr. Walker’s MDCalc. Maybe the only other comparison would be whoever founded UpToDate, and I don’t even know if that’s a known person.

MJ:

It was an individual physician, but now it’s a conglomerate. MDCalc is similar in that people can upload their own tools. We had Mark Cuban on and talked about Cost Plus Drugs, but honestly, I use MDCalc more. Let’s get to the interview with Dr. Graham Walker.

Welcome, Dr. Walker, to For Doctors, By Doctors. We’re excited to have you. For those who don’t know, Dr. Walker is an emergency physician in San Francisco. He trained in emergency medicine at St. Luke’s Roosevelt Hospital Center in Manhattan and attended Stanford University School of Medicine. He’s also a software developer and created MDCalc, the NNT, and Ofqual—online resources helping physicians worldwide. In his free time, he writes about AI, technology, and medicine, and created the Physicians Charter for Responsible AI. Dr. Walker, welcome.

GW:

Thanks for having me. What an introduction.

MJ:

You’ve got quite a long list of accomplishments. Ned and I were talking before you got on, and we decided we’ve used your work more than any other guest—MDCalc at least once or twice a week.

NP:

And the NNT—that got me through grad school. That’s amazing.

GW:

That’s right.

MJ:

We’ve covered your history in medicine. Walk us through your journey from applying to medical school to practicing clinically today, and how you developed your other work.

GW:

My technology interest comes from my father, a very introverted psychiatrist. We always had computers from when I was four or five—floppy disks, DOS before Windows 3.1. I’m 44, so an elder millennial right on the Gen X cusp.

The medical side comes from both parents—my dad a psychiatrist, my mom a psych nurse and dental hygienist. They were both first-generation healthcare workers. I liked the idea of medicine but didn’t fully understand it until later. I wanted to combine helping individual patients with helping groups of people.

I was a social policy major in college, did pre-med requirements, and worked for a nonprofit on healthcare reform before medical school. At Stanford, my focus was community health and policy, and I built small tech projects in my free time. I thought all doctors were tech savvy, but quickly realized that wasn’t the case—I was fixing friends’ printers, making an online directory before social media, so classmates could share contact info and addresses for study groups and parties.

MJ:

It was like the Facebook for med students.

GW:

Yeah. I should have developed Facebook instead of a student directory!

MJ:

MDCalc is still pretty cool.

GW:

It is. We’re used by about two-thirds of U.S. doctors every month. Med students, interns, and fellows sometimes take selfies with me when they find out I created it. I don’t wear MDCalc swag at work or bring it up, but colleagues sometimes introduce me that way.

MJ:

Tell us the story of how MDCalc came about and evolved.

GW:

Yeah, shout out Monica Devacaruni. I think Monica’s an interventional cardiologist, still at Stanford now, I believe. She’s still in the Bay Area. But yeah, she was my chief resident. She was asking me questions about a score. And I think her expectation at the time was like you memorized these scores. The scores were very simplistic, all like plus one point, plus two points to keep the math simple for humans, which made sense.

You either memorized them or you had those little pocketbooks that fit in the deep pocket of your white coat—those tiny, cute little three-ring binders—and you would look at the eight-point font to find the score for Ranson’s criteria or some of the older chest pain and ACS (acute coronary syndrome) scores.

I taught myself web design. I’ve been doing websites since I was probably 17 or 18. I saw this as a niche I could solve by building a site that just had all these scores. There couldn’t be that many at the time, maybe 40 or 50 in the original version. I would go through those binders and find every single one and add them, even if I didn’t know what they did. Now there are thousands—between eight and nine hundred on MDCalc. The backlog is exponentially growing as research explodes, so keeping up is a challenge.

NP:

Yeah, absolutely. So within the MDCalc sphere, was the initial game just to grab every score that existed at the time? Or was there any quality control or vetting? Like, you know, we never really liked this score? Because especially with those three-ring binders, I assume you mean what was probably the red book or the green book, MGH’s Hospital Medicine?

MJ:

Mine was green, I think the first one.

GW:

I think mine was blue—maybe that was just the year it came out or something. Yeah, there was absolutely no quality control at the time. If it made it into one of those internal medicine handbooks, it was clearly being used by people, so I just added all of them. Then I asked interns if they knew other scores I could add.

NP:

It went through color eras.

GW:

Then I’d go on PubMed, look them up, download the PDF, and turn it into a calculator. No quality control then, but now we have a lot because the growth of scores changed a lot. My co-founder Joe Haboush, another ER doctor, and I understand we have a moral and ethical responsibility to do it right. MDCalc is trusted by tons of physicians, which is wonderful but comes with responsibility. If it’s on MDCalc, people assume it’s reliable, validated, and safe to use on patients. We have a thorough review process for every score: we build, test, and validate it because we’re the place people go to solve this problem.

NP:

Yeah, even before now it’s like, don’t double-check your math. For straightforward ones like CURB-65, which are plus one, plus two, or minus one, I delegate full authority to MDCalc to free up some neurons, hopefully.

GW:

We also have warnings and alerts for every input field because we know people can fat-finger stuff. Occasionally, people tag me on social media saying, “You need to fix MDCalc because my patient’s glucose was 1500 and your cutoff is 1000.” They think we didn’t account for that case, but we do want some wiggle room for typos. Like, you might have meant 150 instead of 1500. We want to warn people to double-check. We always provide the actual calculation, the algebra, or logic so people can verify or calculate themselves. If you have a crazy calcium score of 30, which probably isn’t compatible with life, you should double-check.

NP:

Yeah, and unit handling too. With MDCalc’s global exposure, different places use different units. In the English medical system, glucose units aren’t the same as in the U.S. So you need to manage both ends. From a data standpoint, is there a team continually doing QA/QC and refining the site? What does that process look like?

GW:

We frequently look for bugs and respond to user reports. Most errors now are technical — server issues with AWS or configurations — not calculation errors. The score logic is stable and tested across platforms before release. We have an internal team that reads papers, reviews scores for accuracy, validation, and clinical usefulness before adding them.

During COVID-19, especially in 2020, hundreds of scores appeared, like a COVID anxiety risk score—how much of the patient’s anxiety was due to COVID? I wondered how clinically useful that really was. MDCalc aims to solve real doctors’ problems.

NP:

Right.

GW:

If someone can show me how a score solves a clinical problem, we’ll add it. If it doesn’t affect management decisions, it’s probably not valuable for MDCalc.

MJ:

I see a lot of your writing online, especially about how technology in the last two years evolved quickly. There was a recent JAMA study you commented on about how much a large language model AI might help diagnostically. There was a headline implying the AI might be superior to doctors in diagnosis. Can you talk about that study and how you respond to that idea? Is AI better at diagnosing than doctors?

GW:

Shout out to my friend Jonathan Chen, an author on that paper. He was an intern when I was a med student at Stanford; we’ve known each other for decades. The study was interesting. They gave three groups: one had just the large language model (LLM) given a full case presentation including labs and imaging to come up with top three diagnoses. The other groups were human physicians—one allowed to use the LLM to assist, and one using whatever tools they normally use, like Google or UpToDate.

The New York Times headline was “AI defeats doctors in diagnosis,” which was inflammatory and clickbait but sparked discussion. A key point Jonathan made was that physicians didn’t really know how to use the LLM well. They didn’t just copy-paste the entire case but used it more like a search engine, typing in keywords or details.

Number one, this shows physicians need more education on using LLMs. Those of us familiar with terms like ChatGPT are already one or two standard deviations ahead of most of medicine.

Number two, the cases used were from a 1990s New England Journal paper testing diagnostic tools then, with intentionally difficult cases—not common stuff like pharyngitis or pneumonia. For example, cholesterol embolization syndrome—a case many interventional cardiologists have never seen. The authors selected cases you’d typically consult a colleague on because you didn’t know what was going on.

For those rare, complex cases, LLMs could be very helpful. But for common cases, an LLM might not help or could mislead by suggesting rare diagnoses instead of common ones.

MJ:

So it’s like it overweighs zebra diagnoses if used that way.

GW:

Maybe. We need rigorous studies on how these tools perform. Practicing medicine is not the same as reading a detailed case report. Clinicians gather and transform data from patient interviews. If the patient says symptoms have been going on a while, you need far more detail. This interaction is key. If an LLM gets garbage input, it’ll output garbage.

MJ:

That’s where I’ve seen people use AI mostly—looking at imaging or objective outputs rather than the subjective patient interview and exam. There’s a lot of nuance and colloquialisms in human conversations. It’s a catchy headline but in real clinical settings, it’s complicated. Are there uses now where AI analyzing imaging or objective data is good enough?

GW:

I think I haven’t looked nearly as closely into the radiology space because I know there are many radiologists and people smarter than me who know that level of nuance and detail. I think there is certainly value not in having an AI model read a study, but in having an AI model prioritize studies.

You could imagine if you’re a radiologist with 10 head CTs or 50 chest x-rays to read, you’d want to know if a model that’s good at detecting intracranial hemorrhage, mass effect, pneumothorax, or batwing pulmonary edema flags those first. That helps your colleagues get those reads sooner. Maybe the normal studies take longer because you’re not reading them in the order they come in, but in order of reprioritizing and triaging them.

I think that’s a potentially transformative use case because you still have the human in the loop — it’s still the radiologist interpreting the study. But the AI model can say, “We think there might be something here, can you take a look at this one first?”

NP:

In radiology, as three non-radiologists talking radiology, it’s an incredibly dangerous space. There was an interesting discussion — I’m sure I’ll mis-paraphrase — about the “gorilla in the X-ray,” like if there’s an overt positive finding, can you miss secondary or tertiary findings? That was probably 10 to 15 years ago, well before the language we have now.

There was conversation about what comes next to prevent diagnostic anchoring or bias, which we all have in our own specialties. So leveraging AI from a triage standpoint and to help stay wide open in cataloging possibilities seems like a fascinating opportunity as this develops.

GW:

It’s going to be really interesting — good or bad interesting — to see how this plays out. You’re right, there’s technology bias.

If we leave radiology and go back to emergency medicine, I could imagine an AI model telling me, “Hey, we think this person has sepsis,” but the patient has normal vital signs. So I ask, “What do you mean? Is this the right patient? Am I missing something?”

Then it gets back to the management question — what do I do about it? Do I listen to the AI and order lactate, blood cultures, start antibiotics immediately? Or do I go back and ask the patient about infectious symptoms I might have missed? Or do I ignore it as a false positive?

How much will I defer to the AI model and stop using my clinical gestalt? Will it happen suddenly, or will I slowly get lazy and just say, “Oh, AI says sepsis, okay,” and start clicking buttons without deeper thought?

These challenges aren’t just about AI and human-computer interaction — they’re about the human condition. Humans tend to trust tools once they make life easier. You trust MDCalc; you don’t double-check it. I don’t double-check a calculator. If it says 872 times 560, I trust it. That’s just how our brains work.

You can try to build systems to fight that or force humans to double-check — like TSA sometimes putting fake guns in scans to keep screeners alert — but it’s hard when things are right most of the time.

NP:

That rightness is an interesting shift. Early sepsis warning tools have been in EHRs for a decade or more and are often nuisances. I tend to see them pop up after I’ve made a decision, and I’m like, “Cool, an orange window because the patient is breathing 21 times a minute, with an arbitrary cutoff at 20.”

It’s that sense of rightness I think we have to reframe. I don’t check a calculator because it’s been right. I don’t assume it’s drifted off into wrongness. What level of shift will it take before we really change our comfort with these alerts?

Right now, if I see an alert in Epic, I’m like, “Okay, cool.” It’s almost tongue-in-cheek. What will it take for us to really start trusting or questioning these LLMs that have demonstrated strong accuracy when used correctly?

GW:

I think it takes a level of accuracy — whatever “highly accurate” means — at least 80%, probably higher. Then it takes feedback to the human.

If I ignored a sepsis warning 10 times but got feedback that it was actually right 9 of those times, that would build trust. Health systems don’t currently provide those feedback loops. They don’t remind humans when alerts were right or wrong. That’s an additional challenge that will probably be required during a transition period.

As more models get integrated into EHRs, we’ll need a phase where humans can adapt. We can’t jump directly to a black box system because it’ll be too untrustworthy.

MJ:

I’m interested in how generational differences affect views on technology in healthcare and being a doctor. There was a Wall Street Journal article recently describing how younger doctors view their work, and how later-career doctors disapprove of younger doctors. Did you see that article? What’s your take? Is there truth behind that?

GW:

First, props to mainstream media’s ability to get people talking despite social media. That Wall Street Journal article made rounds — multiple people texted it to me to make sure I saw it.

I do think there are generational differences. As an elder millennial, I feel both sides. I wonder if other generations feel both sides too or if that’s a millennial thing, being caught between opposing views.

Common themes I saw from physicians were:

A reductive, derogatory “boomer” take that boomers sold out younger physicians by selling practices to private equity and not giving younger docs autonomy or control like older docs had.

A sadder version where medicine is just another capitalist pursuit, like everything else in the U.S. And that worries me — you really need your doctor to care.

Doctors need to be curious and interested on a level beyond clocking in and out. If you see a weird rash or symptoms, you need to want to figure it out. Not just say “I don’t know” and pass the patient on.

If medicine is just a job — “I clock in, I clock out, not my problem” — that should worry patients and the profession.

But that’s also a response to healthcare incentives. If incentives were different, younger generations might not see medicine as just a job. It’s the system shaping those views.

NP:

You gave the clinical presentation, but I’ve thought about this. You’re an ER doc, who often makes decisions impacting patients’ civil liberties — judgments on safety, competence, capacity, sometimes incarceration. You want that person to have that extra drive or “umph,” whatever you call it.

We give ER docs authority few others have, often even police defer to your medical judgment.

GW:

That worries me and keeps me up at night. Sometimes you really need to be a doctor — look someone in the eye and say, “You’re making a bad decision,” and pass judgment.

It’s easier to just work your shift. For example, a patient with potassium six and renal failure who refuses admission — okay, next patient, click discharge.

I remember my attending Tommy Wong teaching me a lesson: a guy had a recent stroke, was hemiparetic, and wanted to go home. As an overwhelmed intern, I thought, “Great, one less patient.” But Tommy said, “Graham, he can’t go home. You have to tell him he has to stay.”

It’s that fine line — you have power and need to do the right thing. Sometimes you have to wield that power responsibly, tell people things they don’t want to hear, have hard conversations, even risk upsetting patients or bosses.

If it’s just a job, you might avoid those conversations because the patient’s made up their mind, or you don’t know them, so why bother? But I was taught you have to have those hard conversations. Make sure they’re informed, then if they still choose their path, it’s a free country. But you need to do the work of engaging.

MJ:

I agree. Well, I will gently stir the pot here. I mean, I think the job and calling is an oversimplification, right? For conversational uses, it’s a spectrum. It totally makes sense — job versus calling clinically. I think where people are pushing back, and maybe a thread in that article we mentioned about generational differences and reactions, like you said, is the non-clinical risks of it being a calling.

NP:

Yeah.

GW:

Yes, I’m —

MJ:

Hey, we’re having a committee meeting on some random thing that probably no one will be affected by. But because you care and you’re a doctor, and we expect the most out of you, you’re going to show up for an unpaid hourly meeting once a month. Then, five years later, you look back and ask: how many non-clinical duties have you accepted because someone made you feel guilty — because it’s a calling — that “we need you to do this, if you don’t, no one will”?

Those kinds of tasks outside direct patient-doctor interaction, I think, a lot of people are starting to push back on.

GW:

Michael, I could not agree more. And I’ll stir the pot with you as well — I’ll help you stir. We’ll stir the pot together. I couldn’t agree more. You know, we talked about incentives — it is that as medicine has become more transactional and more challenging, as you’re asked to see more patients with fewer resources, and not only that, but the patients you see are way more complex.

I fully agree. I pulled up this article because I literally link this article at least once a month when this conversation comes up. It’s from Dr. Danielle Offrey, a fantastic writer in New York, and just the headline from the New York Times —

NP:

She was our last guest on the show. My God, beating you out the door here. So I’m really excited to hear what you had to say. She’s wonderful. Shout out.

GW:

Amazing! This article is incredible. It’s an opinion piece, and it’s pre-COVID too — it’s from 2019. The headline is: The business of healthcare depends on exploiting doctors and nurses. The subtitle: One resource seems infinite and free — the professionalism of caregivers.

I could not agree more. There’s not a single word I would change in her opinion piece. That is absolutely true. And you’re totally right, Michael, that oath we have taken to help our patients has absolutely been exploited by the healthcare system.

On my podcast, we interviewed Eric Bricker, a fantastic guest, who said he has been in rooms with healthcare administrators who have used that against doctors. They’ll just tell doctors that the patients’ [bad outcomes] will happen if they don’t comply. It’s a great motivator because they know it’s our weakness — we do things because we took this oath that says, above all else, the patient’s problems matter more than our own. And it has unfortunately been abused.

That’s the other trend driving this calling versus job debate to the surface:

“We’re doing this with a societal contract that it won’t be abused — that we’ll work extra hours, stay late, not get paid for it, but the system won’t make it too overwhelming. We won’t stay late every day, but the times that we do, it’s because it’s the right thing to do.”

But yeah, it’s been abused.

MJ:

Do you notice a trend in trainees? Like, did the residents you trained with view their clinical work differently than residents now?

GW:

Good question. I work with our own residents at my hospital, plus visiting residents from UCSF who are ER residents. I think there’s the same level of dedication from residents for sure.

I think part of this was COVID — residents don’t feel like they’re allowed to be as hands-on or own things as much. And part of that is probably because the residents now were in med school during COVID. I remember in 2020, med students were told, “For your own safety, go home. Don’t see patients in person.”

NP:

Go home, yeah.

GW:

Right. That translated to residents too, at least early in COVID. There was an agreement that only attending physicians should see certain patients, like COVID rule-outs, so the fewest number of people were exposed.

I get that — it makes sense, especially with what we saw in New York. But there are consequences to every choice and every policy.

MJ:

Yeah, makes sense. We talked about “off call” at the top. This seems like a response to the issues doctors face. We’d love to hear more about what it is, how it benefits doctors, and maybe the story behind it.

GW:

Yeah. Off Call is a platform launching hopefully in two weeks — knock on wood. We’re getting very close. I co-founded it with my co-founder Jake Horowitz. I’m an ER doctor, Jake’s wife is an ER nurse. That’s how we met, through emergency medicine circles.

The idea is physicians have lost some leverage, power, and control over their jobs, careers, and how they practice medicine. That loss has made many physicians leave medicine or reduce their practice time. Part of that is lack of knowledge around their pay — how they’re paid, why, and if it’s fair.

We’ve gone from 20% employed physicians to 80% in just a decade. Not only did we not get training on running a private practice in med school or residency, but now we don’t get training on how to be an employee and what reasonable employer expectations are.

We want to give people transparency — not just numbers of dollars or RVUs — but how pay is structured, patients seen per day, call frequency, supervision of non-physician practitioners. That info really can make a job great or terrible.

We also want to bring in financial information for physicians. I think we’re all anxious that we’re behind on finances and worried about screwing it up, while also being targets for questionable insurance products and such.

NP:

Questionable instruments, yeah.

GW:

Yeah. The next phase of Off Call will help doctors explore side jobs, consider leaving medicine or trying something new, and help match them with side jobs. We’ve talked to physicians who have made that transition on our podcast.

We also want to build community. The physician lounge is gone.

NP:

Yeah.

GW:

Journal clubs too — we used to have a journal club in my group, not even academic. That sense of community and connection — sharing clinical questions, like, “Are you using this new drug?” or “How are you handling this patient?” — has fallen by the wayside.

I’d love a place for physicians to connect about clinical and non-clinical stuff. We tend to trust each other about stuff more than non-physicians, even if it’s maybe not the smartest choice. Like, “What kind of car do you like?” I see physicians asking each other that a lot. We tend to trust each other a lot.

NP:

I think the worst financial advice I’ve heard has been from other physicians — even people who prey on doctors don’t recommend things as bad as I’ve heard in some areas. Like this crypto thing I got into six months late and now I’m trying to drag you in.

There’s definitely an element of that. I’m really excited for Off Call. I signed up and added my last contract info, trying to populate your data set.

What excites me is tying this back to what we talked about before — how many doctors are W2 employed now. What does that mean about whether medicine is just a job? What does it mean to negotiate a contract with groups who know and prey on your idealism?

An example: Michael and I, both med-peds, negotiated a pediatrics contract — a very normal W2 pediatric contract. It was said explicitly that if you take more salary, “What will happen to the children?” Like it’s a zero-sum game. If you pay me more, something else is taken away.

But the group going through that negotiation said, “Oh my God, we need to rethink this.” There’s no stopping point to manipulative tactics like that.

I’d love to hear — since I added my info to Off Call — do you see it tangibly helping physicians through that process? Giving them the data to be better positioned, especially pediatricians, who are notoriously terrible negotiators?

GW:

Yeah, it is certainly — I’ve told Jake — a very “doctory” pitch: trying to help ourselves and healthcare by giving more data. But I think it will resonate because that’s how we view the world: we need data to make decisions.

If we can give people data, that’s necessary but not sufficient, Ned, like in your genetics class. It’s necessary to have the info.

The other piece we want to bring in is contract review and negotiation support — because you’re right, it’s not enough to just give info; you need to help make it actionable.

Here’s how you could negotiate if they don’t want to give more pay: maybe more education days, vacation days, or a few administrative hours a week to catch up on charts. There are ways to negotiate beyond dollars and cents.

You’re totally right. We do have this weakness of, “What about the children?”

MJ:

Another thing is sometimes you end up negotiating against your old program director — super awkward, and no one wants to push the envelope.

So funny you mention it — the data is important, but it’s nice to have an expert to translate, rewrite contract lines, and negotiate against someone you know, even if it’s awkward.

NP:

Yeah.

GW:

That’s a great point.

MJ:

Or frankly, you just don’t know how to negotiate. So Panacea Legal now exists to do that.

GW:

That’s great. Congratulations, guys.

MJ:

Panacea Legal helps doctors. Thank you. Yeah, Panacea Legal can look at your contract. There’s a group that has experience helping doctors and also works for employers, so they know the tips and tricks on the other side of the table. They also help negotiate. We have folks who say, “That exact scenario I described — I got info from Off Call, but I didn’t know how to ask for more because I’ve never done this, and I know the person hiring me and think they’ll hate me. Can you help?” And yes, we can help with that.

NP:

And it’s awkward.

GW:

It’s absolutely a skill set, right? It’s not something we know innately, and it’s something you can get better at over time. I think that’s how we have to frame it — like, you weren’t good at intubating or doing central lines or reducing shoulders when you first started. It’s a skill set. It’s a growth mindset skill set too.

NP:

A growth mindset skill set, and you can learn it or you can call a consultant. I’m probably not going to do a shoulder as an internist.

GW:

Yeah, if you can’t get the shoulder in, I’m going to her.

MJ:

Well, Graham, we’re probably running overtime. Sorry. I appreciate all you’ve done for medicine and the work you’ve done on behalf of doctors. Thanks for joining us. I do have one last question we ask all our guests: what is one thing you’ve recently changed your mind about? We get tons of information and data every day. Does it reinforce previous thoughts or actually lead to change?

So, Graham, what’s one thing you’ve maybe changed your mind on recently?

GW:

I’ll say this might be middle age or something, but I used to think there were direct cause-and-effect relationships — like, if I did this, that would definitely happen. Now I feel way less sure about the world.

Like, if you raise the price of bread, you’d think people would eat less bread. But now I’m like, maybe they eat more bread and less of other things. I feel way less sure where the world is going. That’s scary in some ways, but I’m more open to possibility and change than I was before.

MJ:

Yeah, makes sense. It’s been a crazy tumultuous last five years. The previous structures we thought would lead to predictable outcomes did not. So that makes total sense — good and bad.

Dr. Walker, thanks for your time and for joining us on the show.

GW:

Yeah. Thank you, Ned, great to meet you. Michael, great to see you again.

MJ:

Ned, that was awesome to talk with Graham. Lots of interesting takes. Curious if you have any salient points you’ll take away from that conversation.

NP:

Absolutely. Before I get to salient points, I really appreciate that we get to talk to medical celebrities. Graham Walker could probably walk around the world and not be accosted — but within our incredibly specific, narrow group, MD Counts has such a broad impact. People stop and take selfies with him in the ER while he’s actively at work.

I appreciated that a lot. It speaks to our community and how we value the output of these incredible things developed organically. The anecdote he gave was incredible.

MJ:

That’s hilarious. I think it’s a life goal for many doctors — to walk around and have other doctors want a selfie with you because you made their life better.

Also, the conversation about older versus younger generations of doctors and how we view work —

GW:

Other people.

NP:

Hey.

MJ:

I didn’t say this before, but I think some of these things are repetitive. Over time, older generations in any industry tend to think the younger generation doesn’t work as hard or has some deficiency. Then the younger generation says the same about the next one, and history repeats. There’s some truth, but it’s a cycle.

I like how Graham framed it — newer generation doctors’ views of medicine are a reaction to how healthcare has changed. Not just expectations, paperwork, or the shift from independent practice to mostly employed, but also debt. That changes how people see their job or calling when they have $300k–$400k in student loans to pay off.

NP:

Similarly, the incentivization structure — we joked about press Ganey scores suffering if you do certain things. Physician metrics often aren’t about quality outcomes. You can do the right or hard thing, but we’re usually incentivized on interventions, throughput, or shortening length of stay.

For example, he mentioned the emergency room coaching a family through something that’s worse financially across multiple metrics. It might extend his shift, lead to admission, and no procedure is billed. That’s probably worse financially under his group’s incentivization.

It’s hard to swim upstream or operate against that headwind given these incentives.

MJ:

Yeah. Definitely something many of us are talking about and hopefully fighting to change and improve. It’s not either/or — living totally in hospital or not caring. It’s somewhere on a spectrum. Hopefully, we push for the best balance for patients, sustainable healthcare, and doctors who aren’t exploited but can deliver good care in a system that values us — which can be questionable sometimes.

On that happy note — there are good things. I don’t want to end on a bummer. Graham and previous guests really care and are working for change. We see little progress here and there.

Really appreciate his time and talking to me as a medical celebrity. Thanks for joining us on another episode of the podcast for doctors by doctors. See you next time.

NP:

Thanks for joining this episode. You can catch the podcast Four Doctors by Doctors on Apple, Spotify, YouTube, and all major platforms. If you enjoyed this episode or learned anything, please take a moment to rate and subscribe so you don’t miss future episodes.

As always, thank you for listening. 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].