Roy Bejarano: Welcome to the latest edition of the Scale Community Analyzing Healthcare podcast. I'm joined by Professor Kelli Frias, Professor of Marketing at American University's Kogod School of Business in DC, a very esteemed school of business. The reason I'm extremely excited to have Kelli here today is that she focuses on the most interesting topics in healthcare: the actual dynamics of the marketplace, its dysfunctions, pricing opaqueness versus transparency, and how that leads to issues with consumer behavior, decision-making, and affordability. It all starts with pricing. A lot of us are dealing with the consequences of mispricing and opaque pricing, almost like a river with trickle-down effects. So it's exciting to start at the very top. Kelli, welcome.

Kelli Frias: Thank you so much for having me.

Roy Bejarano: My first question: imagine we're having dinner with friends, four of us, each describing what we do for a living. What would you say?

Kelli Frias: Interesting. I'd say one element of what I do is teaching, and I spend a significant amount of time on that. The other big element is trying to understand how consumers shop for healthcare, particularly in a space where Americans, and insured Americans in particular, have never experienced shopping for healthcare. This is a completely new phenomenon, and as marketers we know little about it as well.

Roy Bejarano: So with the historical work you've done: where have we been, where are we today, and where are we heading?

Kelli Frias: Historically, consumers have followed the directives of their insurance providers, so in-network and out-of-network providers, and also their physicians, and rightfully so. Where my doctor tells me to go, how often they tell me to come in, and when I need to see a specialist is where we've all been for many, many years, if we've been lucky enough to be insured. Now the market is starting to tell us that we as consumers need to take on much of that role of shopping for and navigating healthcare. It's telling us that in a couple of different ways. One is regulation, so lots of—

Roy Bejarano: I can guess one. Deductibles. I think the increase in patient financial responsibility has a perfect correlation with time spent. The more money I owe, the more focus you're going to get from me. Fair?

Kelli Frias: Yes, that's one.

Roy Bejarano: But I interrupted you. Keep going.

Kelli Frias: No, that's okay. I was going to say, just in general, everything is more expensive. If I thought I had great healthcare 15 or 20 years ago and could afford a deductible, now I might be in a place where I just can't afford standard care, even with insurance.

Roy Bejarano: So the growth rate in average per capita deductibles is probably something you follow closely. No surprise, it's been higher than the annual rate of inflation, I'd estimate. Good estimate?

Kelli Frias: Yes.

Roy Bejarano: So the system is grabbing more and more patient attention through patient responsibility, but it's not evenly spread. The rate of growth in outpatient care deductibles is faster than in inpatient hospital care deductibles, which means the blind patient getting hundreds of thousands, if not millions, of dollars of hospital care continues. Fair?

Kelli Frias: Fair, and I think we're in a—

Roy Bejarano: Is that intentional or accidental?

Kelli Frias: I think it's part of a bigger problem within the system. The system has erred throughout, whether we're talking about the hospital system or insurance. I don't think this stems from one place. It's coming from a multitude of factors, all of which, at the end of the day, take away from the consumer's ability to just live, to cover their everyday expenses.

Roy Bejarano: Okay. Let's play as many sides of this debate as possible. Start here: patients have no business deciding on their own care. They don't know medicine, they haven't gone to medical school, they haven't completed residencies. What do they know about one orthopedic surgeon versus another? The counterargument is other forms of care where patients are entirely responsible. A good example is aesthetics. It's not covered by insurance, and you pick your own plastic surgeon. Wellness, preventive care. Even dental, if it's not Medicaid, is usually heavy on cash and light on insurance. Patients have been primarily, if not entirely, responsible for those kinds of care for decades, and they seem to be doing fine picking those plastic surgeons. Why can't they apply the same research and due diligence, especially in this day and age, to picking the right orthopedic surgeon for the right treatment? What say you?

Kelli Frias: Part of this has to do with context. For many years, marketing researchers would say quality indicators and price indicators are pretty stable. What we're learning from our research is that those indicators are not stable. They're very context-dependent. If I'm getting care here in the US versus, as in much of my research, in Mexico, the way we look at quality actually shifts with the market. Another example is the GLP-1 space, where consumers are seeking their own care. They're also shifting their quality indicators and what they presume to be a good or bad price in those markets.

Roy Bejarano: Explain that to me in English. Let's move from professor-speak to everyday speak. And I love that you're comparing us to Mexico, so let's drill down on that. Do Mexican patients have it better than us? Are they making easier, more efficient decisions? Compare and contrast.

Kelli Frias: Okay, getting away from my professorship. Let's use the example of a crowded waiting room. In the US, where the market is very appointment-driven, if I walk into my doctor's office and see 10 people sitting there, I'm thinking, gosh, I'm going to be here forever. This is somebody who doesn't know how to schedule their patients and doesn't value my time. If I cross the border into Mexico, which is heavily walk-in-based and depends on a fluidity of providers, patients actually say, wow, this waiting room is completely full. That means people are waiting for care here, which means this doctor must be really good. Same cue. I'm looking at quality in waiting room A and waiting room B, and I completely change my perception of what that crowded waiting room means.

Roy Bejarano: What's behind that? Let's unpack it.

Kelli Frias: Think about how that quality indicator operates at the market level, the macro level, which consumers aren't seeing. They're just seeing that one indicator. The reason is that this is a very complex, usually emotionally driven context. There's already heightened anxiety anytime we're talking about healthcare, so oftentimes I just pick the easiest cue.

Roy Bejarano: Got it, I'm catching up to you. I'd argue that both examples, Mexico and America, are horribly flawed, dysfunctional, irrational. You shouldn't be deciding which doctor to see based on the waiting room at all. It's an input, but one of the least important. Think about the hundred other inputs that should drive that decision: quality of care, the history of that group, the applicability of their experience, the actual treatment plan you have in mind, all the research you've done, and maybe cost as well. To say I'm in the right waiting room or the wrong one based on how many other people are there tells me the system is broken in both countries.

Kelli Frias: Absolutely.

Roy Bejarano: Let's bring that to today. We live in an age with so many ways to get instantaneous information. You can research anyone and everyone using AI. We've moved beyond Google: it's organized, edited, filtered. And we do see the influence of ChatGPT, Claude Anthropic, and Google on clinical care. Patient adoption of these tools is rampant. It's not well studied yet, and it's certainly not well regulated yet, so it's a bit of a frenzy. But we know it's happening. We're all doing it. Have we solved for optimal care selection, consumer-based healthcare, and price transparency? What's working and what's still not working?

Kelli Frias: For the last few years, we've talked about price transparency as if merely having more information is better. What we find in our research is that's not the case. Consumers having more information can actually be worse. Why? Going back to how much we can actually process, it feels really hard. So we feel like we've done a lot of work, and that makes us feel like we've made a good decision, when in reality we may not have made a good decision at all.

Roy Bejarano: What's interesting is you're blending price transparency into the broader question I asked. I don't think anyone is going on ChatGPT to compare prices on insurance-based medicine. I don't think that's happening at the patient level, because no one is doing the math. If I spend $15,000 on my shoulder surgery, what does that do to my deductible and copay next year, versus going with this other physician and spending $12,500? Plus, I'm not even sure which physician will charge me more, because neither can predict up front what the full treatment will cost. We're still stuck in the old age of price opaqueness. How do we solve for that?

Kelli Frias: This is one of the reasons we specifically looked at populations that practice cross-border care, because they do this regularly. What we found is those consumers know their medication prices down to the penny. They're some of the only consumers we were able to find who actually shop. Most of us have no idea, if I go to get an MRI, what the differences are between facility A and facility B. We don't even think about it. But these consumers actively do, which makes them a great model to study. How do we fix some of this? If we start thinking about these cue indicators, and how we identify quality indicators that are consistent, is there a way to take what we're seeing in the US–Mexico market and adapt it to the US? That's part of what we're trying to do with this research.

Roy Bejarano: Give me an example of that five years from now. We've seen top-down attempts at superimposing efficient, rational, price-based, resource-based decisions on patients. Value-based care and managed care are perfect examples. Insurers have played this role for a very long time, and so have employers: Don't worry about picking that orthopedic surgeon, Roy, we'll do it for you. The treatment will be complicated. There will be multiple images, MRIs, x-rays, plenty of PT, a whole slew of medications. We don't expect you to stop at every step and say yay or nay to the latest scan and then pick which location you'll get it from. We'll make those decisions for you: yay or nay to the latest PT plan, drug A versus drug B. We know you have a life to live and can't spend 12 hours a day researching the hundred different price decisions involved in all these claims. So say five years from now we've correctly superimposed the Mexico price-selection model here in the US. I'm still struggling with how involved an average patient will be in a complex treatment.

Kelli Frias: Your example of a complex treatment is probably one where we'd still expect variance in pricing, because, as you said, it is complex, and we probably don't have an answer right up front. But in other cases, say a patient who needs an antibiotic or a PT visit, is there a way to simplify that? I think there is. If we think of this as a continuum and take some of the care that works really well in Mexico and translate it to the US, I think that will be a very important space. I don't think this will just be a decision for regulators. We see it now in the GLP-1 space. Consumers are already shopping and are being expected to navigate different kinds of care. You also gave the example of aesthetics. I don't think we'll be able to pull all of that back. It's more about how we translate the information so consumers can actually digest it and make better decisions.

Roy Bejarano: That's a great example, and it answers my overly complicated question: anticipate that, as has been the case, patients will become increasingly involved in these decisions to different degrees. Take a fairly simple decision: Do I want a GLP-1? Do I keep using it? Is it working? Where's my source? Enough people are using it that, anecdotally plus ChatGPT, you can get a pretty good sense of what you should be paying today. All of that is ring-fenced around what I'd call a simple decision. Who's replacing my shoulder, and what kind of treatment will I go through over the next two years of my life? That's multilayered and multifaceted, and it probably requires a lot more in-person expertise that I just don't have. It puts me in a position of real insecurity as a so-called consumer.

That brings me to the top of the stream, where you ask the most fundamental question in healthcare: what should healthcare be built around? Who should occupy the central position? Most people would say the patient. Healthcare should be for the patient, by the patient, because we're the end consumer, goddammit. I'm one of those consumers myself, so I get it. Then physicians say it should be them. They went to medical school, they have the expertise, they should make the decisions. You get regulators supporting group A or group B. And there's a third group that's actually been in charge for the last hundred or so years, playing different roles: healthcare administration, the quasi-oligopolistic insurance industry, MSOs with private equity backing. I'd call that the managerial class. They argue neither of the other two groups should be in charge, because it's all too complicated for everyone. So you end up with a wrestling match over who should occupy the seat. What are we striving for? I worry we'll create patient-based healthcare and not be that happy with the results. What happened to public health?

Kelli Frias: All of these questions are critical. If we think of the patient as the center of this hub, making these decisions, is that the best model? I don't think so, and I don't think it's either-or. What we're starting to see, and I just saw an example of this yesterday, is that we're co-producing diagnostics and treatments, at least in areas of care where the patient can make a valid choice in more simplified settings. Dentistry and cavities might be a good example, where the patient says, yes, I want to move forward with this. I do think we'll move into a space where co-production is happening, and whether we like it or not, it will be demanded of us as marketers, business people, and providers. As you said, with the influx of AI, consumers are already coming into appointments with their own treatment plan, or at least their presumptions about what it should look like.

I also think we'll start thinking differently about what insurance providers do and whether we need them. Do we need these large insurance companies to administer these policies and contractual relationships, when smaller groups or collectives may be able to do much of it more efficiently? We've talked a lot about price, but another big factor, sometimes more important, is accessibility. Even if I have insurance and can afford to see my specialist, I may not be able to see them for three or four months. That's not accessibility.

Roy Bejarano: Yes, there are many variables to consider and no one answer to any of them. I do think we have a unique healthcare system here, which doesn't make it uniquely bad. I also don't think it's uniquely complex. The argument that the US system is broken because it's so complicated is upside-down logic. Healthcare is complicated everywhere, and the US system is complicated because healthcare is complicated. It's misleading to suggest otherwise.

Our system relies on a lot of for-profit and nonprofit hospitals, and they're all effectively for-profit. That creates its own dependency on an underwriting program: the insurance industry. If you replaced those 4,000 for-profit and nonprofit hospitals with government-owned ones, you'd de facto move that underwriting cost to the government. It's government-funded, government-paid; you just need to show up, Mr. or Mrs. Patient. That takes away a lot of the burden and pressure of needing enormous insurance with tremendous deductibles, because most of our insurance is catastrophic. It's for what happens if I need to go to the hospital and spend half a million dollars on my or my wife's treatment. I think that's a direct result of having a system that's 80% essentially for-profit hospitals, which is quite unique compared with the world around us. Most other countries have the inverse: most of their hospitals are government-owned, government-run, and government-paid.

Let me ask you about the quantity of care consumed in this country in aggregate. I know you've done a lot of research on this. Are patients consuming less healthcare than they would if they knew a lot more about the cost? Or are they consuming more because they don't know the cost, so they're irresponsibly and negligently buying more than they should? Or is it a net wash? If you added a lot of transparency, maybe you'd get a lot more care, because patients would feel safer going through the system and less vulnerable to surprises. Look at the pressure: aged patient balances, credit score issues tied to outstanding balances, and worst case, personal bankruptcies driven by healthcare costs. We still have a major problem with patients being charged fairly and accurately up front and having the tools they need to avoid all that. If we solved for all of that, net-net, do you think we'd see more healthcare consumption, the same, or less?

Kelli Frias: Interesting question. It's tough. If we compare this to, say, the auto industry, we make trade-offs between costs and deductibles, and we don't necessarily ask for the most coverage. But in healthcare, if my child in particular needs care, I want them to have all the tests. It's natural to want all of that. Do I think patients demand more care than necessary? No, I think that's an extraordinary case. What we see in all of our data is that people delay care in ways that are probably more costly in the long run. This is where Mexico becomes quite beneficial, because they can access care quicker and oftentimes cheaper than even their deductible price.

Roy Bejarano: What data do you have to support that? Give me examples of Mexican patients accessing more care. Is the frequency of primary care visits greater there? Is dermatology representation higher? What do you point to that shows it actually manifesting in the marketplace?

Kelli Frias: I wouldn't say it's a question of more or less, but of necessary care. Going to Mexico for care is not easy. There's still a significant amount of work in crossing the border, and we haven't gotten into safety, but there are real, legitimate safety concerns: violence, property crime, et cetera. So they're not making this decision lightly. In many cases, they're waiting hours just to cross. But once they're there, they know they can reliably get care. Once I cross into Juárez from Texas, for example, I know I can see a physician that day and get some sort of treatment plan that day. So I'm not sure it's more care, or just that they can access care. We saw lots of examples of people who couldn't afford or couldn't access care, so they'd hold off on getting care in the US until they could get to Mexico or to a different provider. As you know, that often becomes the emergency room visits that otherwise would have been unnecessary.

Roy Bejarano: So are we talking about a mostly US-based Latino population traveling to Mexico for healthcare because they're disenfranchised from the US system? It's too complicated, they're not necessarily covered by Medicaid. What's the scale of this migration to Mexico for care? Millions of people? What are the demographics? You've studied this. Tell me about it.

Kelli Frias: Accessing this information is incredibly difficult. We don't collect any data in the US about people going into Mexico. We only care about people coming here. Estimates are somewhere around a million visits a year. There's a little data indicating that once people get their tax returns, they start traveling into Mexico. We're starting projects that use foot-traffic data to discern whether people going to Mexico are actually getting care, but it's very hard to identify. So what do we know? We can see growth of US-based hospital systems on both the US and Mexican sides of the border. For example, a Dallas-based hospital system is now building facilities on the Mexican side because it knows it's catering to US-based consumers. You can see that in many border towns. There's what we'd call a cluster of healthcare, whether dental, medical, and not just aesthetics: a cluster of clinics as soon as you cross the border.

Roy Bejarano: So I'm a US resident living in Texas. I want to see a dentist, so I drive over the border and see a Mexican dentist. What's motivating me? Access: my US dentist will only see me two or three months from now, and I need immediate care. Okay, what else?

Kelli Frias: It could be a variety of things. Access is one. Another is price. You've checked the prices. You might even have received a treatment plan in the US for, say, several thousand dollars. You call a dentist in Mexico and ask the cost, and they'll tell you it's a fraction of that, usually a tenth. So you go to Mexico.

Roy Bejarano: You had me sold at a tenth.

Kelli Frias: A tenth, yes. And the third piece is: I'm just sick and tired of being treated so poorly in the US. The physician doesn't listen to me or address my concerns. I get two minutes with them. When I go to Mexico, they treat me like a human being. Those are the three groups.

Roy Bejarano: Of those million visits, what share are Medicaid, uninsured, commercial insurance, Medicare? What would you estimate?

Kelli Frias: That would be a really rough estimate. All the data we've collected is from people with employer-sponsored insurance. None of our data is from people with Medicaid or marketplace plans. We specifically looked at employer-sponsored plans because the costs are becoming so much higher. If I had to guess, probably 50%.

Roy Bejarano: Fifty percent are what? Insured, and 50% are uninsured?

Kelli Frias: I would guess, yes. Insured through their employer.

Roy Bejarano: And the rest paying cash?

Kelli Frias: Uninsured, or not insured through their employer.

Roy Bejarano: The insured ones are interesting to me, because if I'm insured, I assume anywhere from 50% to 90% of the bill is paid by someone else, my insurer. I'm left with the leftovers: the copay and the deductible. So if care in Mexico isn't covered by insurance and costs 10% of the total, I have to compare that 10% to my copay and deductible, if I still have a deductible. My copay will be around 10% too, so I haven't really saved that much. It makes perfect sense if I'm not insured. If I'm not covered, I'm not going to a US doctor. We see this all the time in state-by-state comparisons of primary care coverage. In states with very large uncovered populations, like Texas compared with California, the gap is clear. California spends an enormous amount on healthcare, and the headline is always fraud, and there is fraud and all kinds of shenanigans. But there's also a commitment from California, New York, and Massachusetts. Massachusetts is at 98% coverage, first in the country to pay for everyone's care at an ideological level, and you can applaud some of that.

Kelli Frias: That's powerful.

Roy Bejarano: If you're in one of those states where you're not so lucky and you need care, you avoid it. You see that in primary care representation: the share of people in Texas and Florida who don't have a primary care physician and haven't seen one in years is something like 20%, 30%, 40% of the population. Staggering. The share without a dentist who haven't seen one in years: 30%, 40%, 50%. Staggering. I'd think they'd make great candidates to go to Mexico and face very little underwriting risk, because 90% of the cost has been taken out of the equation. That makes sense to me. For patients who are covered, unless their copays are enormous, it becomes more suspect to me. Why go through the headache if you end up spending the same amount of money?

Kelli Frias: Let's add the amplification of care costs in our system. Much of that comes from the gatekeeper and referral infrastructure we have in the US. Say I go to my US doctor with shoulder pain. That might mean 10 visits before I get an MRI: two or three visits with the physician, then maybe five to seven physical therapy visits at $40 each. What started as a $25 copay could be $1,000 before I get my MRI, and I still don't have a treatment plan.

Roy Bejarano: I see. So we might have runaway-train usage here in the US. I can see that, because we still have plenty of fee-for-service here. I'd say even value-based care is probably more quantity-oriented than Mexican care, which I imagine is less American in its approach. I haven't studied it, but I can imagine it. I'd also throw in cultural preferences. If you're Spanish-speaking and you love going to Mexico and love Mexican doctors, US care is a coin flip: you might get a Mexican doctor, you might not. You might just love Mexico's approach to healthcare. Anyone who's spent a lot of time in Mexico, as I have at different stages of my life, knows it's an incredibly sophisticated place with lots of great value, not just in healthcare. So maybe that's a million people who love Mexican doctors.

Kelli Frias: Absolutely.

Roy Bejarano: There are immigrants all over the US, and I'm one of them, with fond memories of experiences in different countries. I still think about the dentists on Harley Street, because I grew up going to them. We all have cultural preferences, so I think that might be playing into it too.

Kelli Frias: That's definitely a piece of it. On top of that, our infrastructure allows physicians to spend just a few moments, literally five or six minutes, with a patient. That doesn't feel like a quality encounter. I want to be treated nicely, have a conversation that feels warm, and feel you're interested in me and my health. So there are elements of that as well.

Roy Bejarano: I love what you're doing here, Kelli. You're looking at this anomaly, this strange, quirky, idiosyncratic reality that a million visits are happening outside the system, and you're using it to ask a lot of really piercing questions. There are a lot of great answers to those questions, and a lot that needs fixing, that continues not to work for consumers in US healthcare. What else should I have asked you? You've done so much research at the top of the stream, where everything else starts. What else isn't working? What are you excited about? What do you think will be working in a matter of months? What about regulation? We've gone through a price transparency revolution from this administration, which started the last time it was in charge, and now it's enforcing it. Just yesterday I saw very large penalties directed at systems that continue to flagrantly ignore, abuse, or misrepresent their version of price transparency. So the screws keep tightening. This administration insists that if we just get clean data, a lot of waste and bad decision-making will come out of the system. It sounds right. As someone who's studied economics, better information usually leads to better decisions. What are we missing from a regulatory standpoint?

Kelli Frias: Better information, of course, is critical. What we're leaning on is more information, and more information isn't better for consumers at the point of service. This will sound complicated, but I'm actually excited about what AI is doing in this space. I think there's a lot of value in having agency over your healthcare. I'm a proponent of patients being very involved in their own bodies and in their family's care. We've been talking at the individual level, but there's a big place for family. We often call this the Hispanic health paradox: families that take care of one another actually have better health outcomes in many cases. So there's very much a place for that.

Roy Bejarano: It's going to be very interesting, because from a regulatory standpoint, AI is easy to hate and easy to want to stamp out. It's empowering a lot of patients to make great decisions and get a lot more care, which will lead to more employment. More leads to more employment. It'll also lead to more automated solutions, but the jobs will come. At the same time, it's destroying existing jobs that aren't needed anymore. That's the nature of markets, efficiency, and resource allocation. When those jobs start disappearing, politicians go into overdrive. Then add a few anecdotes, maybe more than anecdotes, of AI gone wonky: dreaming up answers and camouflaging them as real, fact-based answers. Misleading information is also dangerous. So there's a real risk that your dream, Kelli, of very empowered, AI-enabled patients won't materialize.

Kelli Frias: I don't know that it won't materialize. I think we're going to need to think about consumers as various segments. I really think we'll get better at targeting consumers and identifying their preferences. Unfortunately, healthcare has become so politicized. Some of my research suggests that the word "affordability" means different things to consumers who consider themselves conservative versus liberal, even when we're talking about the same dollar figure.

Roy Bejarano: Access definitely means something very different. Access for whom?

Kelli Frias: Absolutely. Words that carry all that meaning change our perceptions of whether a policy is good or bad.

Roy Bejarano: So we're going to keep arguing. That's the one prediction we can bet everything on.

Kelli Frias: Yes. We're very divisive in our rhetoric, and it just seems to be getting worse.

Roy Bejarano: I don't know. If you read the history books, there have been plenty of arguments in the past. I'm not sure that one is true.

Kelli Frias: True. I love a good art museum visit, where you can go back and see it through art.

Roy Bejarano: My word, plenty of arguments in those art museums. Kelli, thank you for your time today. It was a really interesting conversation, diving deep into the micro- and macroeconomics that support the entire healthcare industry. It's a fascinating place to swim, and I really appreciate it.

Kelli Frias: Thank you so much for having me. I enjoyed it