Seth Berkowitz: The Power Problem

Harlan Krumholz: Welcome to Health & Veritas. I’m Harlan Krumholz.‌

Howard Forman: And I’m Howie Forman. We’re physicians and professors at Yale University, and we’re trying to get closer to the truth about health and healthcare.‌

Our guest today is Dr. Seth Berkowitz, but first, we always check in on hot topics in health and healthcare. So what have you chosen for us today, Harlan?‌

Harlan Krumholz: Oh, I got a good story for you today, Howie.‌

Howard Forman: Go ahead.‌

Harlan Krumholz: Are you ready?‌

Howard Forman: Yeah.‌

Harlan Krumholz: All right. Look, this week the AI medical company OpenEvidence raised $210 million in Series B funding, lifting its valuation to, wait for it, do you know?‌

Howard Forman: I think you told me so I’m not going to try to guess, but billions.‌

Harlan Krumholz: $3.5 billion. $3.5 billion.‌

Howard Forman: Yeah.‌

Harlan Krumholz: And already, they claim that more than 40% of US physicians use it to make daily decisions. It’s the kind of traction in health tech that’s almost unheard of. And it’s not just because it’s a clever AI; it’s because it seems to be that people just really tack to it.‌

And here’s what I want to tell you that it strikes me that no matter what, at the end of the day, the product matters. If you’ve got a really good product, people find their way to it. Look, it comes back to Google. Google wasn’t the first search engine. There were a whole bunch of search engines that were out there.‌

Howard Forman: Right.‌

Harlan Krumholz: Then people started using Google and they’re going like, “Holy moly, this thing really gives me the answers I’m looking for.” There have been a variety of companies out there who have been designed to help doctors ask questions and get answers, and they’ve been okay. By the way, the foundational models, I think, in the end of the day I thought are just as good as anything. Use ChatGPT or Gemini or Claude. ‌

But then OpenEvidence came out directed towards physicians, and it is pretty amazing. So basically, the premise here is that it’s trying to solve the problem that doctors are overwhelmed with information. New medical paper published every 30 seconds. Clinical guidelines constantly evolving.‌

You’ve got a patient in front of you, you got a specific question you want answered, you don’t want to have to go searching anywhere. You want to be able to just express it like you do in the regular LLMs. I’ve got a question—I want to frame it the way that I want to frame it. I want to get the answer that’s for the person in front of me. So where did all this stuff start?‌

You have to understand its founder, this guy Dan Nadler. So back in 2013, this guy’s finishing a PhD at Harvard and he’s in political science really, I think, and he launches this thing called Kensho Technologies. He’d done an internship at the Fed. And he appreciated that a lot of people were making high-level policy decisions, without really the information that they needed.‌

And then they were using Google, but they really weren’t getting the answers that were most germane. So he starts thinking about like, “Well, wow, what if we put a platform together that allows traders and analysts to ask complex, data-driven questions using natural language?” And so they combine machine learning, entity recognition, structured reasoning to extract insights from a chaotic market group of data.‌

And he put this thing together that, again, productized in a way that people really liked it. And in 2018, look, the guy’s a PhD student in 2013, starts this thing. S&P Global acquires it for $550 million.‌

Howard Forman: Wow.‌

Harlan Krumholz: And incorporates it into their platform.‌

Howard Forman: Wow.‌

Harlan Krumholz: There’s another lesson here. This guy, after doing this, pivots now to medicine and wonders, “Can I do for medicine what Kensho did for economics, and for traders and for the market?” But the other thing is I sometimes go to these hackathons, and people who’ve had no experience think that they’re going to start something grand and big, and sometimes it works.‌

He did this with Kensho. But also, as he comes into medical field, which is much more complex, he’s coming with this experience and the track record. And so he comes together with another guy, Zachary Ziegler, who becomes his chief science officer. And they put together this thing that has trusted content, breadth, and depth, and again, the productization is really good. You get the answers you want.‌

Howard Forman: It’s very simple.‌

Harlan Krumholz: I’ve used it a bunch. I don’t see that it’s hallucinating.‌

Howard Forman: Not once.‌

Harlan Krumholz: And the model is basically that it’s free to docs. They’re making, they say, $50 million a month on ads.‌

Howard Forman: Wow.‌

Harlan Krumholz: And they’re taking off like crazy. I’ll just say quickly here, because then I’m interested what you think. First of all, kudos to him, kudos to the team, done a great job. I’ve often wondered about the moat, because in the end, a lot of the LLMs have the capability of being able to produce something like this.‌

But one of the things was, he has licensed exclusive content rights from JAMA and New England Journal, and he’s engendering a lot of clinician trust. And I think with everybody who’s coming on with queries, he’s also finding out that people like the answers and so they’re getting smarter faster.‌

So the questions will be, will they be able to maintain the advantage? Will they be able to fend off competitors? It’s going to be very interesting to see, but look, this guy has just done a remarkable job.‌

Howard Forman: It has no peer right now.‌

Harlan Krumholz: By the way, immigrant Canadian. His grandfather, by the way, was in Auschwitz. This is a guy whose personal history of family couldn’t get into U.S.‌

So they were sent to Canada. Comes to Canada, ends up in the U.S. as an immigrant, built companies, amazing.‌

Howard Forman: It has no peer right now, and it is remarkable. I personally, when I have a tough thing that I’m trying to figure out, if I put in just very, very simple information, it can give me very good, focused differentials to consider.‌

And then you can ask subsequent questions, as you can with other LLMs, and it continues the conversation with you and helps you get to an answer. I think it’s highly effective. Now, I agree with you that there may not be a moat around it, but as you said, sometimes brand name can be a moat.‌

Google has certainly proven that. For 20 years, the Google name has been a moat. It’s a brand.‌

Harlan Krumholz: Well, but they still, you use the search engine all the time, so if it degrades you would know it. So you still have to be a good product. I’ll tell you one last thing that’s interesting to me is, so who had the natural advantage here? Elsevier.‌

Elsevier owns thousands of journals, got all this content. They launched ClinicalKey AI to really do this thing, but if you’ve used this and OpenEvidence, there hasn’t been.... the last time I used Clinical AI was maybe six months or eight months ago, and I didn’t keep using it because I didn’t find it useful.‌

Howard Forman: This is very simple.‌

Harlan Krumholz: But this one, it sings, and we’ll see if it continues to sing, but kudos to the team.‌

Howard Forman: I agree.‌

Harlan Krumholz: Yep. Hey, let’s get onto our interview, Howie, some timely stuff here.‌

Howard Forman: Dr. Seth Berkowitz is a primary care physician and an associate professor of medicine at University of North Carolina at Chapel Hill, where he also serves as vice chief for research in the Division of General Medicine and Clinical Epidemiology. He’s an expert in health equity and social policy. His research examines how material conditions and health-related social needs, such as food insecurity, affect health outcomes.‌

And how these conditions are shaped by broader economic and policy institutions. He studies both individually focused clinical strategies like medically tailored meals and structural policy reforms aimed at addressing both the root causes and downstream consequences of health inequities.‌

Dr. Berkowitz received his bachelor’s and medical degrees from the University of North Carolina, Chapel Hill, and completed his residency in primary care, internal medicine at University of California, San Francisco. And earned an MPH from Harvard while completing a fellowship in general internal medicine and primary care there.‌

So first of all, I just want to welcome you to the podcast. It’s great to have you here. I will, by background, just point out we’ve been avidly following the Healthy Opportunities Pilot in North Carolina. We’ve reported on it a few times over the years, and one of the most recent times was when you published your JAMA paper on that earlier this year.‌

And I thought maybe just for continuity’s sake, if you could start off by just telling us what you found in that early publication in this longitudinal evaluation of the program? And maybe even just remind our listeners of the program.‌

Seth Berkowitz: The Healthy Opportunities Pilots are an initiative within North Carolina Medicaid, that are done under the authority of a Medicaid 1115 waiver, which allows for programs and services that are not typically part of Medicaid to be done under the condition that they’re likely to substantially advance the goals of the Medicaid program, i.e., improve health for Medicaid beneficiaries and be budget-neutral overall, so that’s the statutory requirement. North Carolina is in the midst of a Medicaid transformation in a number of different areas through its Medicaid 1115 waiver. But one part of it that I specifically focused on and was the lead evaluator for, is what has come to be called HOP or the Healthy Opportunities Pilots.‌

And HOP is a program that allows for the use of evidence-based, nonclinical services to address social risk factors like food insecurity or housing instability or transportation barriers. And so it essentially operates by contracting with a network of community-based organizations, what we call HSOs.‌

Human services organizations, who provide things like food boxes or medically tailored meals or housing remediation services or transportation services for Medicaid beneficiaries, who have both clinical criteria and who also have social risk factors like reporting food insecurity, so that’s the program overall.‌

The paper you referenced was the result of what was called the interim evaluation, so the halfway point evaluation of the program in North Carolina. In North Carolina, the program itself was started later than intended because of the COVID pandemic. And so that was supposed to be operating for four or five years in total, that was shortened a little bit because it started a little bit later.‌

And so the interim evaluation was for about the first 18 or 19 months, so about a year and a half of services. So that’s the background of it.‌

Harlan Krumholz: And what did you find? Because I think it’s important for our listeners to understand that it’s never purely binary.‌

Seth Berkowitz: In our interim evaluation at that time, about 13,000 Medicaid beneficiaries had enrolled in the HOP program, so a fairly large number. I should mention also that HOP as a pilot, was only available in certain parts of North Carolina. It wasn’t available throughout the entire state.‌

So three regions, largely rural regions, one in the western part of the state and two in Eastern North Carolina. And so the overall evaluation focuses on a number of different things. And I’m going to mention this even though the paper doesn’t focus on all of these, just because I want to be clear that these programs can affect many different areas of health.‌

So though we’re going to focus on one specific area, it’s important to understand the bigger picture. So we focus on a lot of different things, like patient-reported outcomes, the resolution of health-related social needs, healthcare utilization, care quality, care costs, and those kinds of things. In this interim evaluation, when we needed to go with the things that we had the most numbers for, and in the paper a subset of the evaluation, in particular, we focused on things like healthcare spending and healthcare utilization. Particularly, utilization of acute care like emergency department visits, and hospitalizations and those kinds of things. And what we found is that one, the cohort who enrolls in HOP, who have to meet both clinical and social risk criteria, are a relatively high-cost cohort overall.‌

So we leveraged the fact that HOP was only available in certain areas of the state to create a comparison group that consisted of people who reported health-related social needs just like the people in HOP but who lived in different parts of the state and so weren’t eligible for that reason. But were otherwise North Carolina Medicaid beneficiaries, were otherwise exposed to other changes in the North Carolina Medicaid program, were exposed to other factors going on in North Carolina and the country at the time, like the waxing and waning of the COVID public health emergency and unemployment and inflation, and all those things. So tried to keep all those the same, but they just lived in different parts of the state. Even with that, though, it’s a relatively high-cost group who enrolls in HOP.‌

And healthcare spending is high around the time of HOP enrollment. Some of that actually predates HOP enrollment, and you can see in the trends that spending starts increasing a little bit beforehand. And so it’s likely the case that some kind of health issue is tipping people over into the situation. And then so people enroll in HOP. At the time of enrollment, spending is relatively high.‌

Some of that is medical spending that I think essentially would have happened anyway, just based on the underlying conditions and the prior trend, but some of it is spending for HOP services that start. So the food boxes kick in or something like that. But then if you follow that over time, you start seeing that participation in HOP bends the cost curve, so to speak, to use a common phrase in the health policy world, where the decline in the trend of spending goes down more for people when they’re enrolled in HOP than we think it would have had they not been enrolled in HOP. So if we estimate a counterfactual based on people’s pre-intervention spending and based on what’s going on for other similar Medicaid beneficiaries in the state, we think the spending would have been higher.‌

The trend in spending would have been higher by about $85 per person, per month in the trend. And so when you take those together, the higher spending at the beginning and then the decreasing trend over time, you wind up with, around month eight of HOP enrollment, you get this break-even point.‌

And then beyond that, spending is lower on a monthly basis, including both medical spending and spending on HOP services than we think it would’ve been in the absence of HOPs. So that’s the main takeaway, I think.‌

Howard Forman: Getting results eight months in is pretty good. You can get people framing around thinking about surgery. There’s surgical procedure, there’s a lot of activity around that, that costs you money upfront. You’re doing it to improve health downstream. There’s no surgical intervention that doesn’t end up—let alone saving money. And you’re basically saying you’re getting savings.‌

I want to give a shout-out to your book, Equal Care. It raises some really interesting things and maybe I want to just delve into this a little bit. So you spoke about your experience during residency. I’ve seen the juxtaposition between cutting-edge medicine and unmet needs, and then that’s shaping your path. In Equal Care, you write about that you once saw health and equity as a knowledge problem but now see it as a power problem.‌

And I wonder if you could just help us understand what that means? It’s not really about ignorance, but it’s about lack of power to act on what we know. And to allow the continuation of these patterns of treatment and social context that continues to lead to these differences. Can you just articulate a little bit more about what you were thinking about, maybe even why you wrote the book?‌

Seth Berkowitz: Yeah, totally. Yeah. So as you mentioned, I had a book called Equal Care come out with Johns Hopkins University Press just last year, available wherever books are sold, Amazon, Johns Hopkins Press website, wherever. Yeah, and so as you mentioned, it was kind of this trajectory. So I have my medical training, and then I have some public health and masters-level epidemiology training.‌

But I’m not an economist, I’m not a public policy guru or anything like that. And really, I’m a primary care doctor in my clinical work and I do research. And so I got interested in essentially trying to address the consequences of broken social policy, because I was just seeing that in primary care. I was seeing a lot of food insecurity that was making it hard to control diabetes.‌

Or I was seeing a lot of housing insecurity and poor housing conditions that were exacerbating COPD and asthma and those kinds of things. And so then the researcher part of me, which was that knowledge piece, was like, “All right. Well, what do we need to know about how to better take care of these situations?” And I do think there is stuff to know.‌

In doing the HOP evaluation, we learned a lot about how to run what are still essentially clinical and not really public health or structural programs. And so there is more to know about the nitty-gritty, the food as medicine space and using food insecurity interventions to improve health. There are some interventions that seem to be working well, there are some that have not had their intended effects.‌

And there are nuances to intervention modality and how easy it is to adhere to, and the amount of benefits and stuff like that. So I don’t want to say that there’s nothing to know, or nothing more to learn. And as a researcher, you always expect, “Well, you’re going to call for more research and things like that.”‌

But at the same time, it just became more and more apparent as I kept chasing the thread of, “All right, if I want to address this issue....” “All right. If I need to address food insecurity, what do we need to do that?” And I was like, “Well, why are so many people experiencing food insecurity?” It’s like, “Well, how could you stop people from experiencing food insecurity in the first place?”‌

And just going on and on like that, it just became very clear that big picture, we know how to do a lot of this stuff. We know relatively straightforward, I don’t think they’re “easy” in a sense of getting everyone to agree on them or getting the votes for them and stuff like that. But in a technical sense, they’re very feasible to accomplish policies that are used in many places around the world with great success, to just make it so that people have the conditions in which they need to be healthy.‌

And so the power problem that I was talking about was although there can be nuances and it’s always possible to improve and tinker with policies and make them a little better on the margins, and make them more efficient and those kinds of things, big picture, we know enough to know what we should be doing to create these conditions in which people can be healthy.‌

The book specifically goes into like, “Well, what would those policies be and whatnot?” I’m happy to go into detail about that too here, but don’t want to totally derail things. But big picture, we know what those things are. But it’s this power issue of not, I think, having enough people collectively feel like that’s what we want to do and can actually implement that as policy.‌

Howard Forman: I want to just give you an opportunity to speak about the moment that we’re in right now and hopefully cover just two different issues at once. One is that one of the central issues around your book is health equity. And health equity has been blended together with all the politics around the phrase DEI right now.‌

I personally, in the things that I get to do at Yale right now, have pushed back and said that you just can’t compromise on the term health equity because it has absolutely nothing political involved in it. I think you do a great job laying out what health equity is and the underlying justice issues that do exist there.‌

But at the same time, we also now have this One Big, Beautiful Bill Act that Harlan loves to talk about. He loves calling it a “Big, Beautiful Bill” and it’s now passed, and there are so many things.‌

Harlan Krumholz: In part, Howie, because that’s its name—I’m just saying!‌

Howard Forman: I know, I know, I’m kidding with you. We know this. But there is so much in there that is right up your alley in terms of your area of expertise. There are cuts to SNAP programs, which is the Supplemental Nutritional Assistance Program. There are cuts to Medicaid programs, there are changes in the way that even 1115 waivers, I think, can be evaluated.‌

There are so many different things that are in there. But I do want to point out that one of the arguments that they would make—and by “them” I mean the people who voted it forward and signed it into law—is we’re going to give you more money upfront in terms of Earned Income Tax Credits and children credits.‌

In other words, money for families that might need it, but we’re going to take away money in other places. And one of the things that you have argued for at times, is that payments can be at times preferable to food or other types of specific programs.‌

I just wondered if you want to just give an overview of your thoughts of what the current moment means for the work that you are doing?‌

Seth Berkowitz: Sure. Yeah, I think it’s a complicated time. We’re talking about major changes in bedrock social and public policies across the U.S. that’s going to have just a lot of impacts in a lot of different areas. As you say, a lot of these are projections of what will happen.‌

Some of the Medicaid changes may not come into effect until 2026, others at different times and things like that. And there are going to be a lot of moving targets and those kinds of things, but I do think there are a couple, I feel like I keep using the phrase “big-picture.”‌

But there are a couple of big-picture things that I try to focus on, just given how much is going on, how much uncertainty there is and everything else. One is that, so I fundamentally do believe in the importance of income support programs for improving health. I’ve described myself as trying to preach the gospel of social insurance.‌

And I do think a major reason that population health in the U.S. is not what it could be is that we have a lot of gaps in how we support people when they run into the normal life-course circumstances that often prevent people from having a sufficient income. On one side, people say, “Well, the reason that you don’t have a sufficient income is you’re lazy. You just don’t work enough or something like that.”‌

But if you look at demographics and things like that, it’s actually not how it works. So about 40% of the population in the U.S. is either over the age of 65 or under the age of 18. So those are populations that you don’t generally expect go out and get a job as the right solution for income.‌

And then about 10% of the overall U.S. population falls within what you might call working age, age 18 to 65, but has what people understand as a good reason or a reason that they’re not going to support themselves primarily through paid labor. They either have a work-limiting disability, they’re a full-time student, and so are looking to work later but are just in school right now.‌

They’re providing unpaid but socially necessary caregiving, like taking care of children and things like that, or they’re unemployed. So temporarily they’re out of jobs, they’re actively looking for jobs, but they’re just not in a job right now. So that’s 50% of the population that doesn’t have income from what we call earnings or paid labor.‌

And if your public policy doesn’t support that 50% of people, then folks are at very high risk of not having the income they need to be healthy. And so you may have trouble buying food, you may have trouble affording housing, and those kinds of things. And this has been a recognized issue for over 100 years.‌

The birth of social insurance really goes back to the 1870s or so in Germany and the U.S. throughout from 1900s on, with major changes around the New Deal era in the 1930s. These risks have been understood risks of being in situations where you can’t earn a sufficient income, and public policy responses are there for them.‌

So we have things like Social Security, old age and survivor’s insurance to deal with the risk of old age. We have disability income programs, we have unemployment insurance programs. But the way we do them in the U.S., there are a lot of administrative burdens. They are hard to access; the benefit levels are insufficient.‌

And then there are other big gaps, are we don’t have child allowances like most other countries do. We experimented with that in 2021 with the expanded child tax credit that put out payments to families on a monthly basis, and it was a big success. Much lower food insecurity, much lower child poverty, but then that expired and hasn’t been put back.‌

And disability programs are hard to access, and so there are just a lot of problems with the way we do things. In lieu of doing them in the social insurance way, we try to do these what are sometimes called means-tested or targeted policies that instead of trying to prevent you from being in these circumstances, which I think is what the goal should be.‌

We wait until you’re in dire straits, but then maybe offer some help. So SNAP, I do a lot of work on SNAP. SNAP helps a lot of people. I’m not against SNAP in any way, but you have to be in a pretty low-income situation to qualify for SNAP. And then because it only goes to people in these situations, there’s a lot of room to make it harder to access and add more work requirements or cut people off for a certain period of time.‌

And really work requirements, I think, should be considered work reporting requirements. So the evidence is pretty clear that most people who should be “working well on these programs” actually are. And most of the people who lose coverage when work or work recording requirements are put in place, lose it.‌

Not because they wouldn’t actually meet them, but because they have some trouble just actually doing the reporting, is cumbersome and complicated and those kinds of things. And so the way that we do some public policy is this wait till you’re in a bad situation and then try to get some help. I think one, just in and of itself creates problems.‌

But two, when you have people who are looking to offset fiscal space for tax cuts and things like that, say, “All right. Well, if we just make it a little bit harder for people to get SNAP. If we make it a little bit harder for people to get Medicaid, then we won’t spend as much on these programs and our fiscal projections will look a little bit better.”‌

And so that’ll open up some room, which doesn’t make any of the problems those programs are trying to address go away. People are still going to not have sufficient money for food. People are still going to need healthcare, but it frees up some budget space and so you can get people to vote for it in that way.‌

But I think it’s just very likely to be harmful to folks in the long run, with a caveat that there’s a lot of uncertainty right now.‌

Howard Forman: Just wanted to thank you for being here, for shining a light. You shine a light on a very important topic. You mentioned Atheendar Venkataramani in your preface to your book, and he’s somebody that we’ve had on the podcast as well.‌

But this is one of those topics that we need to do more to illuminate and then to solve. But thank you for all the work that you have done and will continue to do.‌

Harlan Krumholz: Thanks for being a thought leader in this area. It really does make a difference. So we have people who are on the front lines of caring for patients.‌

Also, producing knowledge to help us tactically improve. But also at a higher level, like you said, and more in the normative area, how are we actually thinking about these things? So thank you.‌

Howard Forman: Just great work.‌

Seth Berkowitz: Yeah, thanks so much. Really enjoyed the conversation.‌

Harlan Krumholz: Hey, great interview.‌

Howard Forman: He’s awesome. He’s great.‌

Harlan Krumholz: He’s awesome. He is awesome. He does so many things well.‌

But let’s get to one of my favorite parts of the show, hearing what you think. So what’s on your mind this week?‌

Howard Forman: Yeah. So as the dust settles around the passage of the One Big Beautiful Bill Act of 2025, as we talked about earlier. And it’s now signed into law, there’s many subtle and not-so-subtle effects on healthcare. We’ve talked a lot about Medicaid, now today even about SNAP. But tucked inside this massive bill is something that absolutely will affect our medical education system.‌

And frankly, I only first became aware of this a couple of months ago, when a medical student told me that they were changing their specialty choice considerations because of these changes that they were anticipating. And so let me just break it down really quickly. First, let me just say, I was really fortunate to go to a seven-year public medical education program based out of City College in Stony Brook.‌

The grand total cost of my tuition for college and medical school was around $17,000, and by that I mean for all seven years, tuition and fees, $17,000. I lived at home, and by that I mean with my parents in Brooklyn. So room and board was covered by Alice and David Forman, inclusive of doting love and appreciation. I had a great deal, in my opinion, and I sincerely feel very fortunate.‌

Harlan Krumholz: So that’s probably another $2 or $3 million.‌

Howard Forman: Absolutely. Well, more than that, but nonetheless, most people don’t have that benefit. And even if you factor in inflation, you would find that actually tuition for the program I went to has gone up way more than inflation. City College up 500%, Stony Brook up 400%, in the same time that consumer prices have only gone up 160%.‌

Medical education, undergraduate education are getting more expensive. Bottom line, if you want to go to medical school and you’re not lucky enough to go to NYU or Einstein, or a few others that can offer full rides, you’re going to be taking on financial debt unless you come from a wealthy family.‌

Students have historically funded medical school from a host of sources, but the leading ones are the federal government, who have reduced interest rates and other generous terms. You may not think they’re generous, but they’re more generous than private lenders. And greater accessibility have made this appealing particularly, again, to those that don’t come from a wealthy family.‌

The median medical student graduates with about $200,000 in debt, and the median four-year cost of medical school is now around $400,000 from a private institution. The first line of support is what we call federal direct unsubsidized loans, which carry a market-based interest rate, but is generally readily available to medical students.‌

And while there is a limit to what they can borrow, it does afford a lot of medical students access to medical school. For the remainder, if that’s not enough to pay for medical school, many students rely on what’s called grad PLUS loans. Again, this is a loan available to professional graduate students, including medical students.‌

It’s only been available since 2006, and grad PLUS loans have a cap up to $200,000 lifetime and a slightly higher interest rate. But they’ve been a key source of financing resulting in over $1 billion a year in medical student funding. So what’s changed? The Big, Beautiful Bill Act has many changes, including now a cap on direct unsubsidized loans.‌

And the grad PLUS program is gone completely. So students who otherwise might have relied on that program will now have to rely on private loans, which are generally less generous and often more inaccessible. Then there’s something called the Public Service Loan Forgiveness program, which also was going to actually be axed during the negotiations but was spared by the Senate parliamentarians.‌

So that’s the good news. The Public Service Loan Forgiveness program, which allows people to write off their loans if they work within a not-for-profit system, that remains and still allows our students to pay off their loans to a degree, but there is yet more drama, and it’s very serious here. In March, the president issued an executive order asking that the Department of Education revise the criteria for Public Service Loan Forgiveness.‌

And just two weeks ago, the Committee of Jurisdiction at the Department of Education completed a negotiated rulemaking meeting that will now issue a proposed rule that will come up very soon, which will be open to public comment. And then a final rule on the way to implementation next July 1st.‌

In the simplest of terms, the outcome is likely to be that any institution that is engaging in, and I’m using my hands here to give scare quotes, “a substantial illegal purpose” will no longer allow the borrowers to have their loans forgiven. Now, what do they mean by this? We believe it’s going to mean anything from if they provide gender-affirming care, caring for undocumented immigrants, or providing substantial material support to marginalized communities.‌

And I could go on, but I think you get the idea. If they deem Yale New Haven Hospital ineligible for Public Service Loan Forgiveness, that could have a substantial impact on the residents who trained here and the fellows who trained here, should they otherwise be seeking that loan forgiveness. I do personally think, and I’m not a lawyer, that the courts would not readily accept the current definitions that they’ve been using and have been promulgating.‌

But I do think this is still a big concern. Ever the optimist, I’m hopeful that this will not result in the nakedly political and ideological objectives of the current administration. But if it does, it could dramatically shake up our residents and the institutions that we all work at here and across the country.‌

Harlan Krumholz: Wow, that’s such a great segment. I missed that part of the bill. I didn’t appreciate those changes, and those are important.‌

But I’ll just say one thing, Howie, shouldn’t we just break this whole system? It’s a problem. Look, no other country in the world does this.‌

Howard Forman: It’s so expensive to train people and then the care we give is expensive, no question.‌

Harlan Krumholz: If you’re in Europe and you go to medical school, they have much the experience that you had in terms of that.‌

Howard Forman: Right, low cost.‌

Harlan Krumholz: If this is a social good for our country, shouldn’t we be saying that we do have expectations of people who become doctors? And we are going to make sure that there are no financial barriers to become—not even just barriers. We should say, “It doesn’t cost anything to become a doctor.” But in exchange, we have expectations of everyone who becomes a doctor doing certain things that are going to be part and parcel contributing to society.‌

Howard Forman: Can’t agree with you more, Harlan, but I just think that reality, we’re not going to change very much in the very short run.‌

But we do have to push back at some of these things that are really trying to disrupt in a way that’s going to harm our healthcare delivery system.‌

Harlan Krumholz: Yeah, I know. So I’m 100% on the short term, what do we do to mitigate the things that are coming down the pike? But I’m trying to get us, look, if we wanted to project to what the world looks like in 25 years.‌

We’ve talked about how medicine’s going to be so different because of AI and all the advances that are going to be made, but also what do we think education ought to be like? How can it be made, not just affordable, but maybe we should be rethinking the entire approach?‌

Howard Forman: Totally. And we’ve talked about, look, NYU has the three-year program now. We need to probably think more about that.‌

If it’s expensive, figure out how to shorten it. How do we do it in other ways? I agree with you.‌

Harlan Krumholz: Yep, yep. Anyway, very good topic. You’ve been listening to Health & Veritas with Harlan Krumholz and Howie Forman.‌

Howard Forman: So how did we do? To give us your feedback or to keep the conversation going, email us at health.veritas@yale.edu.‌

Or follow us on LinkedIn, Threads, Twitter, Bluesky, or whatever social media you use.‌

Harlan Krumholz: And we love your feedback, questions, anything about the things that you’ve heard here today, feel free to rate us and often we’ll get back to you.‌

Howard Forman: Very true. If you have questions about the MBA for Executives program at the Yale School of Management, reach out via email for more information or check out our website at som.yale.edu/EMBA.‌

Harlan Krumholz: Health & Veritas is produced with the Yale School of Management, Yale School of Public Health. Thanks to our superstar undergraduates, Tobias Liu and Gloria Beck.‌

This week, we had an amazing high school student from Groton, Miles John, who helped us with—at least helped me with—my segment. And thanks to our wonderful producer, Miranda Shafer. Talk to you soon, Howie.‌

Howard Forman: It takes a village. Talk to you soon, Harlan. Thank you.‌

Harlan Krumholz: When they’re working with us, it takes a village.‌