Julie Barnes on the Policy Behind the Patient

Melanie Marcus: Thanks so much for joining us today, Julie. It’s great to have you on the podcast.

Julie Barnes: I’m so glad to be here.

Marcus: To ground us in why the policy behind the patient matters, let’s start with a story. Can you take us back to a moment—a story you’ve heard from a patient, a provider or a plan member—where care went off track specifically because a policy or data gap existed? What did that reveal about the system?

Barnes: Absolutely. I remember one conference I attended: Health Datapalooza, a Washington, D.C., conference focused on health tech. It was 2016, and it was a really big deal that Joe Biden, who was vice president at the time, was there speaking.

He was describing how his son—everybody remembers Beau Biden and his terrible experience with brain cancer—couldn’t have his medical records shared electronically between Walter Reed National Military Medical Center and MD Anderson Cancer Center while he was getting cancer treatment. Those two hospitals’ electronic health record (EHR) systems were not talking to each other. So the family of the vice president of the United States had to physically bring a disc with the medical records down to Texas so MD Anderson could treat Beau. I mean, can you imagine?

Marcus: No, I can’t imagine.

Barnes: This story had a lasting impact on me. The sitting vice president of the United States, with every available resource to him, couldn’t make two hospitals share his son’s records. That tells you the system is really not working for the patient.

That is why Joe Biden had such a personal, vested interest in interoperability and in prohibiting electronic record systems from blocking health information. And that’s literally why the 21st Century Cures Act of 2016—passed when Barack Obama was president and Joe Biden was vice president—required healthcare providers and software developers to allow patients and authorized users to access electronic health information.

Marcus: I came to Surescripts right after that, so I heard all the stories about it. It was quite a compelling story for everyone here as well. Thanks for sharing that, and we’re going to come back to it because there’s a lot there.

But let’s turn to your career. You’ve spent it at the intersection of law, policy and health IT. When you started, what was the most significant friction point patients and providers faced? And has it changed?

Barnes: My interest in health policy began when I was working on healthcare issues for a member of Congress in 1993. That’s when the Clintons proposed their Health Security Act—the first national healthcare reform proposal of my generation. There have been many since, but that was the first one for me.

It was a time when medical costs were totally out of control. There were millions of uninsured Americans, and people felt like they couldn’t leave their job because they would lose their health insurance. It was my first real job on Capitol Hill. I was a very young staffer, and I remember being pretty flabbergasted when the national health reform effort to fix those problems just totally fell apart.

I learned a lot about how politics plays a real part in policy. So I went to law school, and I realized that when big reforms can’t happen, smaller laws get passed. While I was in law school, the Health Insurance Portability and Accountability Act—HIPAA—was passed in 1996. It addressed that job lock problem by introducing the concept of creditable coverage to limit preexisting condition exclusions.

We all take that for granted now, but once upon a time that was a real problem. The P in HIPAA stands for portability. Now we all know it as the HIPAA Privacy Rule, and that is part of what I established my law practice on—and really what gave rise to what I’m doing now.

Marcus: I remember the Clinton health reform and when it fell apart, but from afar, not up close and personal. I’m sure it had a major impact on your thinking and the way you work today. So how has it changed now?

Barnes: Unfortunately, a lot of these same problems still exist. Medical costs continue to be out of control. There’s still a real issue with people not wanting to leave their employer-based health insurance—although you can, with COBRA, and because you can get individual health insurance on the exchanges, which works very well. Employer-based coverage just remains the gold standard in this country. So it hasn’t changed enough.

I do think the HIPAA law and the Privacy Rule, and then in 2016, on top of the Affordable Care Act laying the groundwork, the 21st Century Cures Act is really what changed the game for interoperability and for prohibiting the blocking of information. That has absolutely changed the landscape since I began in health policy.

Marcus: You started Maverick Health Policy. Why did you start it, and is there any meaning behind the name?

Barnes: There is a pretty good story about that. In 2018, I was watching Senator John McCain’s funeral on TV. It was held at the Washington National Cathedral, and if you don’t remember, it was epic. A moving service, amazing music, fantastic speeches. The most famous politicians in the country were delivering these incredible eulogies for John McCain.

I watched the whole thing and got pretty emotional. So to change my mood, I went out to run errands and ended up at a PetSmart adoption fair. There was a beautiful white furry puppy that looked a little like John McCain, and his name was Maverick—which is what McCain was famously called in the Senate, the Maverick of the Senate. So I adopted a dog named Maverick and named my health policy firm after him.

Marcus: On the day of John McCain’s funeral. That’s a great story. I’m glad I asked.

Barnes: I just like the name Maverick in general. It evokes something fiercely independent-minded, and that’s what we do. We offer independent, nonpartisan strategic advice about health policy to our clients.

Marcus: That’s great. Well, let’s talk about where the rules stand today. The CMS Interoperability and Prior Authorization final rule has major provisions now in effect as of January 2026, involving faster decision timelines and new denial transparency requirements. The biggest shift lands in January 2027, when the Provider Access and Payer-to-Payer APIs become mandatory.

I just said a lot. From the patient’s standpoint, what should change because of that rule?

Barnes: There are two really concrete shifts already in effect as of January 2026. There’s a faster decision timeline for prior authorization requests, and there’s denial transparency—health plans have to give patients and providers a very specific reason when coverage is denied.

Then there’s another big structural shift coming in January 2027, when the Provider Access and Payer-to-Payer APIs become mandatory. That’s when the computer systems between the providers and the health plans have to talk to each other. A patient’s data has to follow them even when they switch health plans, even when they switch providers.

So for patients, that really means their medical histories will finally follow them wherever they go and whomever they see for healthcare. They don’t have to re-explain their medical history every time they see someone new.

Marcus: So you don’t have to fill out the forms every time.

Barnes: Yes. It’s losing that clipboard and that memory test every single time. The idea is that won’t have to happen anymore.

Marcus: Is the world ready?

Barnes: No. I don’t think everyone is ready. There is definitely going to need to be a cultural shift from what we’ve all gotten used to, this manual communication process between the medical staff working on behalf of physicians and the health plans they’re interacting with.

It’s been a phone call, fax level of communication, just explaining what documentation is necessary to prove that a patient really needs coverage for a given service or procedure. And that needs to stop. It takes way too long, and there’s too much room for human error. So yes, we’re trying to streamline that process with these rules, and hopefully that will happen.

It’s sort of like when the banks moved from an in-person process to an electronic one. Remember when online banking didn’t exist? You had to physically bring a paper check to your bank to cash it. Now you can just take a picture with your phone and look online to see your balance. That’s what’s going on in the healthcare system.

Marcus: Right. In healthcare, that’s happened organization by organization but not across organizations, and I say organization because it could be a health system, a doctor’s office or a health plan. Each one has its own set of data silos that don’t interact.

I’ll use an example I’ve used a few times on the podcast. Even though health systems can exchange data, it’s still a change management exercise for the actual staff. I had a blood test in one health system that had to go to another EHR in another health system, and they wanted me to print it out and fax it. I said, “I know your systems. I know they work. Here’s what I think is happening. Can you just try one more time?” And they did, and they got it done 20 minutes later.

But that’s because I knew to ask. If we’re still making that change between health systems, do you think it will go any faster between providers and payers?

Barnes: Change management, cultural shift, whatever you call it … I think we’re all going to experience some speed bumps on this road to the modern healthcare system. It’s just the reality that people are used to what they’re used to, and it’s very hard for people to make the shift.

It happened to me the other day. A specialist asked me to put an image on a CD and bring it with me. I knew better, just like you did, and I said, “You don’t have to do that. It’s accessible by this website. Here’s the code.” I literally handed a piece of paper to the staff to say, “This is what you do.” And they said, “Oh, great. We can load it up. It’s all set.” Two different healthcare systems, both extremely sophisticated, both in the Washington, D.C., area. And I was still asked to do that.

It’s just this constant effort to educate people about the new way of doing things. And it’s like anything else. Change is hard. We’ll get there, but it’s going to take all of us driving the change and really insisting on it from the patient perspective. So the education is going to take a minute.

Marcus: I think so, for sure. You’ve also written about what happens when rules aren’t followed. In a recent blog post you used—and I love this analogy—Ted Lasso’s line about being a goldfish, someone with a 10-second memory who can shake off yesterday’s mistakes, to talk about the renewed push on information blocking enforcement.

You paired it with the story you just told about Joe Biden being unable to get information between healthcare providers. So if a vice president couldn’t make that happen for his own son, what makes you optimistic that enforcement today will get data moving when it didn’t before?

Barnes: It is different now. I really believe that information blocking is no longer just guidance and a good idea. There are real penalties attached. The Centers for Medicare & Medicaid Services (CMS) and the Office of the National Coordinator for Health Information Technology (ONC) have shown a real willingness—a real zeal—to act on violations. They’re not just describing the problem anymore.

A couple of people matter here: Chris Klomp and Tom Keane. Klomp is now chief counselor at the U.S. Department of Health and Human Services (HHS), in addition to running Medicare, so he’s basically overseeing all of HHS from an operational perspective. And Keane is the new national coordinator for health IT. Between the two of them, they have really made clear that information blocking enforcement is a huge priority.

The first thing Klomp did when he came in—and he said so publicly—was move money from CMS over to the HHS Office of Inspector General for the purpose of information blocking enforcement. So he put money where their mouth was.

There are proposed changes in HTI-5, one of the regulations ONC oversees for the certification of electronic health records. ONC updated its FAQs about information blocking guidance, affirmed that automated technology should have access to electronic health records and clarified which fees are allowed. ONC is issuing notices of non-conformity all the time—sending notes to hospitals saying, “You really need to make sure that you are not blocking information.”

Dr. Keane has said out loud that they are, quote, “singularly focused on how technology can make healthcare more affordable and can enable a patient-centered healthcare system.” They also use the term data liquidity. It’s another word for interoperability, but it’s supposedly the agency’s top priority, and I think we’re seeing that over and over again as they try to promote streamlined data flow across the healthcare system.

Between that and some of the bills happening in Congress—Senate Health Committee Chair Bill Cassidy is definitely raising concerns about information blocking—there are a number of things actually happening, as opposed to everyone just talking about it as a good idea.

Marcus: Can you back up for just a second and give a simple explanation of what you mean by information blocking?

Barnes: Electronic health record systems are products sold to hospitals that hold everyone’s medical records. They are proprietary systems that live within hospital networks. And just like the Joe Biden example, with Walter Reed not talking to MD Anderson, they’re different systems that do not connect and do not talk to one another.

So when one system is asked to send information to another, disparate system, sometimes it says, “No, we’re not going to transfer that information,” for lots of reasons. Sometimes it’s a privacy reason. Sometimes it’s a security reason. Sometimes it’s, “No, those are our owned records, and we’re not going to transfer them over to another record system.” That’s blocking information, and it’s not conducive to patient care.

Marcus: Thank you for that, just for those listening who aren’t necessarily familiar with it. So we’ve talked about interoperability and prior authorization. There’s also price transparency, and we do a lot of work on that here at Surescripts around the prescription.

Price transparency is part of the same picture, right? Helping patients and providers see the cost as part of the care decision itself. Where does price transparency fit in that whole reform puzzle?

Barnes: Price transparency is the third leg of the same stool. Interoperability gets the clinical data moving. Prior authorization reform speeds up the coverage decision. And price transparency data lets patients see how much things cost before they’re locked into a decision about what to do next. Without it, patients don’t know what they’re going to have to pay.

There are price transparency rules on the books now. There’s hospital price transparency, and there’s another rule for health plans called Transparency in Coverage, intended to let people understand how much things cost and how much they’re going to pay out of pocket. But that data is just not easy to come by right now, mostly because people have health insurance, the insurance pays for part, people pay for part, and it’s very difficult to know what parts people are paying for. So there are all kinds of rules and procedures embedded in those transparency rules that are still coming out and undergoing (again) the cultural shift involved in making those things happen.

Plus, the data is just extremely complicated. There are spreadsheets on spreadsheets, where people who write code and look at those kinds of data sets for a living are really struggling to pull down what it actually translates into for a regular person trying to understand what that service, at that moment, at that period of time, will cost them.

There’s a mechanism called advanced explanation of benefits that’s intended to help patients know their costs up front, but progress on that in particular has been very, very difficult.

Marcus: I’ve done some work in price transparency for prescriptions, and at least there’s a product associated with it, so there’s a price associated with that product. But care is harder. You go into the hospital for a hip replacement, and you don’t know what complications may come. It’s hard to really predict.

There have been lots of things happening around value-based care and bundled payment, trying to decide what that costs. But it really is very complicated.

Barnes: One hundred percent. Just like you say, it’s one thing when you have a product and you know how much it costs. So, part of what’s easier—still hard, but easier—is real-time prescription benefit. ONC finalized those requirements in another rule called HTI-4, and that’s great.

Real-time prescription benefit really helps move the needle, because it allows patients to access their drug cost information in the health record at the point of care, so you can make informed decisions about what that costs. And another big-ticket item is TrumpRx and this idea of self-pay pricing, so patients can understand the options available to them if they’re just paying out of pocket.

There are lots of things like that happening that are much more concrete. But as you say, care is hard and complicated, and there are a lot of gray areas. Something can happen in a clinic visit that changes the treatment plan completely. So how the heck do you predict what somebody is going to need as that progresses over time? Healthcare is not a black-and-white box. It’s going to be a process over time before we understand exactly how to price things out for people in a way that’s more comfortable than it is today.

Marcus: For sure. As we talk about this incredibly complex landscape of pricing, it strikes me that AI has some opportunities. So, let’s talk about AI. Healthcare has been faced with a deluge of state legislation on health AI over the past few months and years, and a lot of debate in general.

Where do you see the path forward on AI regulation in healthcare?

Barnes: Wow. That’s a big question. It’s certainly something our firm is dealing with every day. We serve healthcare organizations and technology companies, as well as venture capital firms, that need to know just this: How is artificial intelligence for healthcare uses going to be regulated? AI developers need to know how to move forward with their products. Hospitals and health plans need to know how to deploy artificial intelligence safely. And patients are expected to use it and to trust it.

It is the biggest thing happening in our system today, and we are constantly asked to produce reports, particularly about state AI legislation, because the federal government has not acted to oversee the use of artificial intelligence models in healthcare yet. Congress seems unlikely to preempt any of these state laws in the near term. HHS is trying to leverage its existing but limited authority to clarify its oversight of health AI.

Because there’s no uniform federal approach yet, state governments are reacting to doctors being nervous about artificial intelligence usurping their role in the healthcare system, and to constituents using ChatGPT and other chatbots to ask questions pretty constantly about their healthcare.

There’s a quote I’ve heard from Laura Adams of the National Academy of Medicine. She said, “Things are moving at the speed of patient desperation to get information.” It’s pretty powerful, and it definitely sticks in my mind. We have lots of problems in our healthcare system, and one of them is access to providers at the time we need the information. So people are availing themselves of what’s available, and it’s moving at a faster speed than it can be regulated. It’s very uncomfortable not knowing whether we can trust the information everybody’s getting.

So, we’re doing a lot of tracking, not just of what state legislatures are passing, but also what state medical boards, professional societies and other groups are doing at the state level to influence AI regulation.

Marcus: I think I’ve heard you speak about this before, and it makes me think: AI doesn’t have a state boundary, so how does this work? Can you give some examples of the regulations that are out there?

Barnes: Sure. The easiest one off the top of my head is mental health. It’s the major legislative topic right now. Illinois, for instance, passed something called the Wellness and Oversight for Psychological Resources Act, and that is something other states are emulating. Colorado, Maine, Rhode Island and others have something like it.

That model establishes restrictions on AI use in psychotherapy services. It’s basically permitting or prohibiting uses of AI by licensed practitioners as a consumer protection element: saying don’t advertise or misrepresent that AI is a doctor in a psychotherapy context.

Marcus: My mind is going to lots of places. Is it because they don’t want anybody’s private health information out there? Is it about privacy, or is it because AI is substituting for a professional opinion?

Barnes: It’s both. Definitely both. Part of what’s going on is a provider shortage in our mental health system. There are chatbots serving as therapists, and states are trying to make sure that when people avail themselves of these chatbot therapy opportunities, they understand exactly what it is and what it is not.

Some states are considering regulating AI as a licensed mental health provider by defining and emphasizing, when there are human-like features in those chatbots, what those interactive user conversations should look like, can look like, and are allowed to look like.

Marcus: Super interesting. What other examples are there beyond mental health?

Barnes: One of the trends is what they’re calling regulatory sandboxes. It’s an extension of the fact that states are the experimentation place for new things. State bills are authorizing these regulatory pilot programs, basically where you can study something, test AI solutions and understand how this would work in the real world, providing more or less a safe harbor for those things to be tried out before they run large throughout the system.

So, sandboxes are a big-ticket item for the states.

Marcus: Those both sound like positive things. I’m sure there are some that are inhibiting the use of AI, but how do we thread the needle between AI regulation that helps and AI regulation that hinders when this is so new?

Barnes: It’s a great question. Frankly, this is one of the reasons I started Maverick Health Policy. What really needs to happen is that healthcare organizations get comfortable with technology, and technology vendors need to understand so much better how the healthcare system works.

They are two different languages, two different sets of players, very much in different camps. The interoperability rules are really when things started to come together between these two disparate players. But it’s still true that they’re uncomfortable.

Think about people who work for big tech companies—the proverbial coders who are in their laptop all the time. They are not thinking about the healthcare system as healthcare. They’re thinking of it as algorithms to solve certain problems. And healthcare doesn’t think like the coders at all. They’re all about healthcare, and technology might help, but it’s sort of an add-on. What’s happening right now is that those two worlds are colliding. So, the technology folks need to get embedded with true medical experts, so those languages can become one.

The biggest example is the Utah sandbox. There’s a Utah Office of Artificial Intelligence Policy, and it needs to get together with Utah’s medical licensing board, and they need to talk to each other about what those pilot programs in that sandbox can look like. It’s one of the ones out front in the state universe of this regulatory landscape.

Marcus: It is. Well, as you can tell by the name of the podcast, we’re looking for better ways. So describe what happens when policy actually works—data flows, decisions come faster, and patients and providers have fewer obstacles.

Barnes: Honestly, I spend so much time in the problems that I’m not sure what a perfect solution looks like. But I think we’re getting closer on the prior authorization side. We’re in the middle of a number of conversations with the technology vendors on the ground who are working with both health plans and healthcare providers to make that much more of a streamlined process.

Again, we’ve got cultural issues there, but what’s really happening on the ground is workflow issues. When a patient is sitting with a provider and the provider decides this is what we need to do next, what happens is you end the visit, and the doctor goes out to his staff and says, “We need you to ask the health plan for that patient to approve this coverage.” And then the staff goes about doing that. So it’s once removed from the conversation, and that workflow is not necessarily understood by the computer processes right now. The electronic health record doesn’t assimilate to that process.

So they’re in the weeds, really trying to figure out how to make that streamlined process work. And honestly, they’re making some real headway, because there’s pressure to get it all done by January 1, 2027. I’m really impressed with the doggedness of the people doing that work. They’re having some real success. It’s just that our American healthcare system has so many different owners of all these different things. You get done with one system and one health plan, and you have to move on to the next one. It’s a lot.

Marcus: For sure. We are absolutely in that prior auth space on the medication side. Right now, you’re talking mostly about the care side, where all the complexity we talked about on pricing is also involved in prior authorization. But it’s coming straight to the medication side as well, and there, all the workflow challenges exist and progress is being made. We are doing that work now.

All right, let’s push on it a bit. What’s an assumption about federal health policy—that it’s too slow, too political, too disconnected from patients—that you think most needs to change?

Barnes: It is true that probably the biggest problem is accessing the data. So, it’s really everything we’ve just discussed. One of the things we have not talked about yet, and that is one of the solutions to the problem of getting information where it needs to go, is TEFCA—the Trusted Exchange Framework and Common Agreement. It’s the on ramp, the national way of exchanging records.

Marcus: And Qualified Health Information Networks, or QHINs.

Barnes: Yes. Surescripts is one of the QHINs. For many years TEFCA was this great idea that people really wondered if it was going to take off. And Tom Keane, the new national coordinator for health IT at HHS, announced that TEFCA has surpassed one billion healthcare records.

So the efforts over the years to create this trusted national interoperability framework are finally working. I think it’s a lot to ask of one system, so it’s going to be interesting to see how well it does as it keeps growing. But that is one way we’re seeing a solution to one of the serious barriers in our healthcare system.

Marcus: And that has taken a while to get going. It’s complicated, so it had to be put together. All the parties needed to align and develop and work together. And now that it’s together, it didn’t take long for it to reach a billion.

Barnes: Exactly.

Marcus: Let’s look ahead a little. Over the next 12 to 18 months, what should healthcare leaders be watching most closely on the federal policy front?

Barnes: I’m very cognizant of the political landscape and the major health reform processes I’ve lived through over the years. It’s been 33 years and about four or five different bites at major reform efforts. And I am very impressed with what this HHS has been doing to put speed to this data liquidity element and really get all the players invested. That’s different.

The reason this administration is doing things most differently, in my view, is the forced coordination of the sub-agencies of HHS. CMS is talking to ONC, ONC is talking to the Food and Drug Administration (FDA), FDA is talking to the Office of General Counsel. There is a very serious effort to make sure the jurisdictional parameters and guardrails around each of those sub-agencies blend with the others, so that where one agency’s jurisdiction leaves off, the other one picks up.

Artificial intelligence is one of these. FDA oversees medical devices, drugs and food. They have the Software as a Medical Device (SaMD) definition and process, and they’ve blessed over 1,000 medical devices that have some artificial intelligence element to them. But that’s where FDA lives—in medical device land. The second you get out of medical device land, it starts to morph into somebody else’s jurisdiction.

So one of the most important things I’m watching, and it’s just different than I’ve seen in all my years in this business, is that this administration is truly making those leaders sit down together. The FDA commissioner, the national coordinator, CMS Administrator Dr. Mehmet Oz, and all these other lawyers involved in interpreting the parameters of their jurisdiction are actually working together to figure out how to align, so that you have a more seamless set of policies.

I hope that narrative carries over into future administrations. People ask me all the time, “Is this just going to be a Trump administration policy that gets dumped by the next guy?” And my big answer is that if you look back over the years at health tech, it has absolutely carried forward from one administration to another. HIPAA privacy, electronic medical records, interoperability, information blocking. From 2004, when George W. Bush created the concept of the national coordinator and the idea that everyone should have access to their healthcare records, from 2004 to 2026, throughout whatever administration was Republican or Democrat, people have absolutely found that to be the true north.

So I’m hopeful that this administration, which has done such an incredible job coordinating those efforts like I’ve never seen before, is what carries into the next administration so this can continue. Because that is one of the problems: the jurisdictional parameters around legal authority.

Marcus: That is very good news, and we are seeing it too. It’s hard not to see it if you’re in this space. I completely agree that over time, if you look back, we are not seeing healthcare policy whipping from side to side. We see the build of it, which is very good news.

All right, let’s wrap up our conversation by going back to where we started: the patient. This is part of where our podcast really gets its name There’s a Better Way. So in that spirit, what’s your better way in healthcare? What is your ideal future for patients?

Barnes: I think the ideal future for patients is that they’re able to get the care they need when they need it, through modern delivery systems and technology-supported care.

We are always pining after the days of the Norman Rockwell picture—the primary care provider who knows someone from childhood through adulthood. It’s the Forrest Gump model, where the doctor who put braces on his legs sees him again when he’s an adult. That’s just not reality. That is not how we live. That’s not how we operate.

We need to move into a modern reality, even when it’s uncomfortable to trust some of these computer systems and electronic devices. And I think we’re going to be forced there by the generations that grew up with the iPad in their crib. My dad still goes to the bank to cash checks. I do not, and my kids have no idea why they would ever need to darken the door of a bank. So we’re in that transition generationally, of understanding and believing in the technology.

I think a great future for patients will be when they only have to go inside an institution when they need to actually be operated on, and everything else can be done from the comfort of their own homes. That is what I think would be a beautiful future for all of us.

Marcus: Wouldn’t that be nice? That would be great. Well, thanks, Julie, for your time and this conversation. It was a complete pleasure speaking with you.

Barnes: Thank you so much for having me.