Skip to content

The NIMH Director’s Innovation Speaker Series: Pursuing an Innovation Agenda: A New Healthcare Architecture

Watch on YouTube.

Transcript

JOSHUA GORDON As Zoom lets everyone in.  [silence]

JOSHUA GORDON Okay. Welcome, everyone. It's a pleasure to welcome you to this edition of the NIMH Director's Innovation Speaker Series. I'm Joshua Gordon, the director of the National Institute of Mental Health. And this series is oriented at bringing speakers with varying points of view and out-of-the-box thinking from the perspective of a mental health research organization to talk to us about what they do and how it might impact what we think about here at NIMH. So it's really a pleasure for me to welcome today Barak Richman who is an economist and a law professor. And he's currently at Duke University, although I think he's currently calling us from California where he recently finished a sabbatical at Stanford. Barak did his undergraduate work at Brown before going to Berkeley to get a PhD in business administration and then Harvard-- actually, sorry, reverse that order, where he got a law degree.

JOSHUA GORDON And his work has been on something that's a term that I wasn't really familiar with until our discussion just about an hour ago, relational exchange. And more recently, he's turned that aspect of economics and the law to attend to the issue of healthcare and how it's delivered in the United States. He's going to talk to us today about how to pursue an innovation agenda in creating a new architecture for healthcare. So Barak, thanks for coming to speak with us today, virtually. Really looking forward to hearing what you have to say and to engaging in a dialogue after. Just as a reminder to everyone, you can type in questions into the Q&A function in the chat, and we'll monitor that and address those questions at the end of the talk. But you can put them in at any time. Right. Thanks.

BARAK RICHMAN Yeah. Thanks so much. It's really terrific to be invited. And it's especially terrific to engage with this audience. So I just want to make two very preliminary comments before I start the heart of the talk. One is this: I guess I first conceived the kernel of this talk about a year ago, maybe a little more, before COVID. The talk, I think, is different in a post-COVID world, and that's one thing that will be very clear to all of you as we go through the talk. But also, it's something that I think deserves a lot more reflection, so I'm excited to give this talk now. Actually, it's funny to put it that way. I'm excited to give this talk now that we live with COVID. I mean, we view these issues differently now, and that's one thing that I'm very eager to get feedback on. The second preliminary remark that I want to make is that I'm very eager to speak with this audience where many of you are mental healthcare professionals. I was talking with Josh before. You all have unique vantage points into understanding the problems with the current delivery of care. And many of you have been at the forefront of pioneering improvements in the delivery of care. And I hope that much of what I talk about resonates with that. And I would very much welcome further conversations on those topics even after the talk, certainly during the Q&A but even after.

BARAK RICHMAN All right. So we are talking about pursuing an innovation agenda, and in particular, a new healthcare architecture. One of the fields that I've been active in is the world of organizational innovation. And we distinguish organizational innovation. How organizations or how an organizational framework can change can be both innovative and also can adapt to new technological innovations. And we distinguish that from what's called modular innovation. A modular innovation might be when we come up with a new medical device, we come up with a new drug, but we use that device or that drug within the same system of care. Architectural innovation is how the component parts change with each other. So I want to talk about that at the outset. But before I do that, I want to create or describe a dichotomy that I think is very useful. It's a dichotomy between health resources and health services. And this tends to be the dichotomy that we rely on too often in thinking about health reform. A health resources approach thinks about the resources we have, in particular, payers, hospitals, physicians. And we think about how they interact with each other. We think about what role they play in the delivery system. We think about how we might improve each of those given the resources that we have.

BARAK RICHMAN An alternative approach that I think is really important is thinking about services, how we might be able to organize services around, in particular, a patient, or whatever kind of objective we have. The main distinction here is not thinking about how we can improve care given the resources we have, which is a very much a constrained kind of question. But instead, thinking in an unconstrained way how can we organize care in a way that meets the service objectives that we have? And if we think about it that latter way, then we are no longer constrained by the organizational framework we have, by the architecture that we have. So this dichotomy, I think, is very useful. And when we think about a business architecture - and this is something that I'm really relying on that mostly comes from the business economics or business strategy literatures - a business architecture is defined by the interrelationships of components and individuals within an enterprise, or more colloquially, how a business is organized, which often reveals where its power center lies and how its certain actions are coordinated. A business architecture reveals a business's priorities and authority.

BARAK RICHMAN Now, very rarely has the logic of business architecture or architectural economics been applied to healthcare, which is really unfortunate because as you all know, as behavioral therapists know better than anybody, the relationships and the interrelationships are really critical in providing high-quality care. And I'm not just talking about the provider-patient relationship. I'm also talking about the relationships between providers and the relationships that patients have with wherever they access their care. Certainly, location of care contributes significantly to cost of quality, and therefore, we want to think about how we might locate care and locate the nexus of care and how the organization of care deeply affects patient behaviors, patient inconvenience, and the utilization of new technologies. So when we think about how we organize care, we really want to resist thinking about only how we deploy our resources. Instead, we might want to think about how we can deploy or create or provide high-value services, and that often involves an organizational component, an architectural question. How should we structure care? How do we encounter the-- how do we encounter the patient? And how do we organize care in an efficient way that really meets the patient's needs?

BARAK RICHMAN So as we think about architecture, let's think about the present architecture. And to a large degree, we have a hierarchical architecture. We have a hospital-centered delivery system. Hospitals are the most capitalized player in the delivery system. That's where most of our healthcare costs go. And for the most part, we have a hub-and-spoke model where hospitals are engaged and cooperate with physicians. We have a number of intermediaries, mostly financial intermediaries between patients and doctors. Those are insurers and PBMs. And in addition to this slide suggesting a hierarchical structure, a pyramidal structure, I think what it really captures, for me at least, is how the hospital tends to be the epicenter of where we deliver care. And when things are serious, we kick it up. When things become less serious, we're able to kick it down. That's the way we think about things. Now, this is the architecture. This is the delivery system that we've had for a long time. And I want to spend some time on this slide, which in many ways is the most remarkable slide in the presentation.

BARAK RICHMAN This is a very, very simple analysis that looks at how we delivered care in the year 1996 compared to the year 2012. And what we see is, for the most part, we have delivered care in exactly the same way. The volume of services and the organization of services, services we provide relative to other services has pretty much been exactly the same. Now think about what happened between 1996 and 2012. That's when something called the internet started to take shape. We had enormous growth in digital technologies. We had things like Google and, I guess, Facebook also. But, I mean, imagine Google, and we had smartphones. All of that developed between 1996 and 2012, and it really did not affect the productivity of healthcare and the organization of healthcare. That shows either amazing resilience or amazing ossification. The way we have delivered healthcare has not really responded to dramatically significant technologies that have really transformed and reorganized most other sectors of the economy. And then when you think about what kind of organizational structure we currently have with healthcare-- I don't mean to pick on Duke, but I'm picking on Duke.

BARAK RICHMAN This is a picture that is taken from Duke Health, Duke University Hospital, and we are so confident with our organizational structure that we have put it in brass. We have put it up on the wall. And really, the way we organize hospital medicine, let alone medicine in general, is really the way we organized it in the early 1930s. We have a medicine department. We have a surgery department. We have psychiatry with pathology. We have not adjusted the way we organize healthcare to not just technologies that developed after 1996 but even technologies that developed before that. So when I diagnose one problem-- I don't want to use the word diagnose in a metaphor. I don't want to pretend to be a physician. But when I diagnose what is problematic with our healthcare system, one thing that I observe is that it has remained very ossified. The delivery has not-- the organization of the delivery has not changed. Now, obviously, the way we provide medicines, the medicines we provide, the surgical interventions we provide, techniques have changed dramatically. But if the last 40 years of innovation have taught us anything it's that we also need to-- we also need to pursue organizational innovation. We need to develop new architectures that can really harness new capabilities and improve quality and reduce costs and improve productivity.

BARAK RICHMAN So just asking the very basic thought question, what could digital technologies mean for medicine? Well, for starters, digital technologies allows for the proliferation of information. In a pre-digital world, professionals, medical professionals mostly, were the sole repository for information, but now, we can directly link patients to information sources and other kinds of guidance. We also have an amazing capacity for data analytics. We can access digital records, including personal health records, that cannot only help us with diagnoses that no longer really need to be tethered to a human interaction, but it also can automate guidance. And if we link patients with data and data services and data analytics, there can also be enormous value in the health delivery system. And we also might be able to finally really assess in a much more rigorous way an ability to distinguish high-value from low-value care. Now, we can think about how certain digital technologies can encourage behavioral changes, can nudge people towards alternative delivery systems. And importantly, traditional professionals don't always have the skills to pursue these high-value interventions.

BARAK RICHMAN One theme of the conversation I had with Josh before, which is obviously a conversation that you have on a very regular basis, is how do you integrate different forms of care, say behavioral interventions from physical interventions or mental healthcare with physical healthcare? The traditional model relies on professionals. The way you do it is making sure that you have professionals available in a somewhat integrated setting. But digital technology doesn't make that necessary. And not only would it enable lots of consults, which means you'd have different kinds of hub-and-spoke models, but maybe even you could automate some forms of care. Which would allow either the mental healthcare provider or the physical healthcare provider - and I'm just speaking in terms of professional identities and training - to be able to offer more organic care if there's a full use of digital technologies. So let's think about how the incorporation of certain digital technologies might be able to influence the architecture that we have. Well, what has been happening recently is that hospitals have been purchasing physicians. This is what, in many ways, we have called the traditional ACO, the accountable care organization.

BARAK RICHMAN I'll say parenthetically, I think accountable care organization is a somewhat Orwellian term. It's used to describe lots of organizational arrangements that are not necessary accountable, don't necessarily provide care, might not even really be functionally organized. But unless we use the word ACO as a way of trying to describe different structures, different architectures of how healthcare is provided, then the traditional ACO that really has been very prominent over the last 15 years has been hospitals purchasing physician Groups. And this has not only been widespread in America, but it's really, I think, been underappreciated as how significant it is in organizing care. From just 2012 to 2016, the number of physicians that are employed by hospitals have gone from 26% to 42%. Another way to think about this is the number of physicians who were independent from the year 2000 to the year 2016, basically, it went from two-thirds to one-third. We no longer have physicians acting independently from hospitals to nearly the same degree we had before. Increasingly, physicians have been integrated with, or you might even say tethered to, hospitals in the way they operate in the delivery system.

BARAK RICHMAN Now, it's not the only kind of new arrangement or new accountable care organization that's emerged in the marketplace. Another has tethered or integrated doctors with insurers. And this is what I think has been a more promising arrangement. This is where our large insurers, especially UnitedHealth has really been very aggressive in this area. Insurers have purchased physicians, physician groups, and have tried to integrate the financing of healthcare with physicians. And to some degree, this has brought the financing arrangements and so-called value-based payments a little bit closer to patients. And doctors have been more actively involved in trying to decide particularly what kind of care should be provided and how care should be paid for. But this is an alternative arrangement that you might say has been tweaking the present architecture. But if we think about a new kind of architecture, one suggestion that I propose - it would be really central - is to no longer have the hospital at the epicenter. What we really want to do is try to even sever the hospital from the main architecture that patients are involved in.

BARAK RICHMAN And we can think about this in a couple of specific ways. We can imagine doctors and insurers collaborating with patients to really shop around for hospitals instead of going to the local hospital. And in fact, actually, certainly, instead of having the local hospital control the pyramid of care, we could imagine doctors and insurers working with patients to try to think of tertiary care more as destination care. And this is especially important. And one way I've really come to health policy was through my lens as an antitrust scholar. One very significant problem with American healthcare is that most hospital systems are local monopolies. I don't know how many of you saw 60 Minutes just this past Sunday, but this was highlighted as really the cause, the primary cause, for healthcare cost inflation over the last 20 years, probably even 25 years. It's been consolidation. It's been that hospitals have controlled the delivery system locally and have been able to charge monopoly prices. Well, if we think about severing the hospital from the local architecture, that opens up a new set of possibilities to contract with other hospitals, not the local one, not even the regional one, and it would stimulate a lot of hospital competition that we no longer have. But even more significantly, it really removes the hospital as the core organizer, as the epicenter of power for the delivery system.

BARAK RICHMAN So it not only would achieve lots of antitrust remedies or solve a lot of antitrust problems but it also-- it points towards a different kind of way of organizing care. But even more so, we can also think about ways that we can use digital technologies - and the slide here suggests it's through insurers but it doesn't have to be through insurers - where we can really link patients up directly with certain resources and, in fact, make healthcare delivery less provider centric. Certainly, providers would be the main organizers, the main visionaries of healthcare, but that doesn't mean they need to play a central role in the day-to-day delivery. In many ways, what this slide really suggests is that we currently have an architecture that largely flows from the center down or from the top down. But with digital medicine and different kinds of arrangements, different kinds of relationships, we can think more of a bottom-up approach. We can really think about how we can organize delivery around the needs of patients and then bring resources to them in a cost-effective way and only in a way where there are resources that they need.

BARAK RICHMAN So I want to remind us about this first slide and think about how our healthcare architecture, not just our healthcare policies but healthcare architecture largely relies on thinking about how to deploy resources that we have, whereas a bottom-up strategy, you would think about how we would organize services around the patients. It's a different orientation. And it suggests at trying to develop a different kind of architecture. So I want to map out three specific scenarios of what an alternative architecture would look like. The year is 2025, although, frankly, now we might say the year is 2023. We're really not looking as far into the future as I originally thought we would when I first conceived this idea, this presentation, a year ago. The year is 2025 and we've decided to use our smartphones. Let's think of three illustrations of what an alternative architecture would look like. Well, scenario number one. You're a parent. You have a child at home and it's sick, has a fever. So typically, what do you do? Certainly, what I did when I was a young parent, well, when the baby spikes a fever, you take her to the ER or you take her to some urgent care clinic. You might call ahead of time and the doctor tends to say, "Well, look, it sounds pretty serious. It might not be serious but let's take a look at her."

BARAK RICHMAN So you take the baby to the provider, to the clinic. The provider looks at her. The provider then offers a prescription. You take the baby back in the car. You strap her up. You go to the physicians-- you go to the pharmacy and you get a prescription, some kind of antibiotic. It's a story that we've all gone through. It's so mundane that we don't think there's anything unusual about it. But just imagine using your smartphone. Well, you FaceTime the pediatrician. Pediatrician makes some kind of preliminary diagnosis, and then the prescription is delivered by UBERDOC or by Amazon. There's no real need. I mean, obviously, there will be needs to take a baby to a healthcare center, to some kind of urgent care clinic. But this typical story really would never require taking the baby out of the home, taking the baby into a clinic where there are lots of other germs. In many ways, it's the worst thing for the baby. It certainly imposes a lot of discomfort to the baby, a lot of anxiety to the parent. All of this can be virtually organized. What is necessary is having the conduit digital connection to the patient as opposed to trying to get or needing to get patients to providers and to pharmacies to get the services that the baby needs.

BARAK RICHMAN Think about another scenario, one that's certainly been on my mind a lot over the last couple of years, caring for an aging parent. A colleague of mine is in the middle of writing a book called How to Care For Your Parents Without Killing Yourself. This is something that can be very stressful, very taxing. Imagine using our phones to do that. And we can imagine not only-- I'm really not talking about FaceTiming with your parent, although that certainly provides a lot of value. But imagine also having lots of monitors to the parent - monitoring blood sugar, cardiac irregularity, just any kind of motion detector that detects a fall - things that can actually trigger an emergency response without you but also keeping you posted. This is not really difficult technology. None of these scenarios involves very complicated technologies, technology we've had for at least 15 years. The key innovation is just having a different arrangement, a different architecture. Caring for an aging parent, if we use digital technologies around the parent, where the parent lives, we can really imagine offering independent living at a much safer and a much more cost-effective manner.

BARAK RICHMAN Third scenario is the delivery or this pursuit with the need of tertiary care. And I'm in California so I can make fun of California, and you'll see that I'll soon show my homerism about North Carolina. But hip replacements are provided in a fairly standardized way, but still, most people get hip replacements from the local provider, the local hospital. And the local hospital knows that. And most local hospitals, as I mentioned before, are monopolies for general inpatient services, and they charge monopoly prices. And you can imagine a high-quality hospital that does not have necessarily hip replacement expertise and still, nonetheless, charges monopoly prices and, frankly, might not even have terribly high quality. There's no reason we can't have some kind of destination medicine. Some people call this medical tourism, which I think is kind of a derogatory term. All it really is is expanding the sphere of competition of removing one from the local control of the local hospital and just being able to shop intelligently. This is a very natural consequence of not having a local hospital at the epicenter of the delivery system.

BARAK RICHMAN So here's the argument in a nutshell. Digital technologies - and not just digital technologies but especially digital technologies - enable a reorganization of the architecture of the US delivery system. It could lead to lower costs, better quality, greater access, yadda, yadda, yadda. But that health delivery system cannot improve on its own. We need a new healthcare architecture. If we're going to incorporate digital technologies, we need a new architecture that will really nurture those digital technologies. And the other thing is that the scholars of organizational innovation, of architectural innovation, will tell you it's really hard. It is fairly easy-- or at least it's easier for an organization to replace one drug with another drug, a better drug, one intervention, one medical device with a better medical device. It's much harder to break the relationships that structure an organization. Organizational innovation is really hard, and for that reason, what we tend to see in a lot of industries where there is organizational innovation is we have something that's called creative destruction. We have disruptive innovation. Again, these are terms that-- I don't want to go all Silicon Valley on you. I think these terms are both somewhat colloquial and it's also somewhat misleading.

BARAK RICHMAN The basic idea is that in order to pursue meaningful innovation, sometimes it requires a reorganization, and reorganizations are very hard for organizations to do on their own. I just want to offer one illustration of how architecture can really matter. Consider the story of electronic health records. Electronic health records were found to be the solution to paper. It was a bipartisan agreement that paper kills, paper records are problematical. What we need is to convert all of our medical records from paper into digital records. Now this happened, of course, when most of the world had already-- most industries have already gone to digital records, but medicine had not. So there was an effort to stimulate the uptake of electronic health records, but lots went wrong. There was lots of money to develop electronic health records. Money was given directly to providers. There was a notion that we're going to allow interoperability and allow seamless communication. But after at least eight years of working on this, there was even more money, but it didn't work.

BARAK RICHMAN We have an electronic health record system that is highly siloed. Interoperability is a very distant goal. And there's lots of evidence - in fact, I've been doing some research of my own - that showed that electronic health records actually increased administrative costs. It's not done a whole lot to reduce errors, and it has not been the kind of digital infrastructure that we really thought it would be and think it should be. Why? Well, basically because hospitals build EHRs the same way they built everything else. They built the electronic health records to duplicate and reinforced the kind of architecture that we currently have. Well, if we think about an alternative architecture that was pursued by another country, let's think about France where everybody had an insurance card, and on the insurance card were their individual health records. That meant that they could go to any provider they wanted, and they would be bringing their electronic health records with them. Instead of being an architecture that reflected the provider or the hospital infrastructure, it was built around the patient.

BARAK RICHMAN And what's really interesting about the possibilities that this kind of architecture offers is you can imagine individuals effectively uploading their anonymized health records into some kind of commons that would allow all sorts of digital analysis, all sorts of analytics, and analytics that would produce specific instructions to patients themselves. Having a digital infrastructure organized around patients means that we can have lots of digital services that tend to patients. It would be a patient sort of architecture instead of a provider-centered architecture. Some people think that what Apple is trying to do in the electronic health record system will do that. It has not yet really taken off, but you can imagine that instead of electronic health records being somewhere in the cloud that's organized by your hospital system, instead, it's something that walks around with you every day.

BARAK RICHMAN Now, Apple is not the only one who's thought that electronic health records might be a useful way they could enter into this space, and more generally, the promise of incorporating digital technologies into healthcare delivery has excited a lot of big players. All of the large digital companies, electronic health, software companies, are trying to get into healthcare delivery. They're making enormous investments. And it really kind of begs the question whether they have the capacity to reorganize or offer an alternative digitally-constructed healthcare architecture. I mean, I don't know if that's where the future will take us, but these folks who have been enormously successful in entering and creating new markets certainly see the healthcare space as a potential new market. And for sure, they are going to be entering with new architectural strategies, strategies that are very different from what current hospital leaders have been providing.

BARAK RICHMAN Now, here is my second to last slide. This is something that I think we really need to get our head around. As I said in the very beginning, this whole talk about the kind of relationships we have with each other and the kind of relationships we have with our healthcare providers looks very differently from this side of COVID-19. And this has been a really tragic national story, no question, but there also have been some valuable lessons. Some lessons have been very, very painful. One lesson is that telemedicine is here to stay. Many of you I'm sure have been pioneers in providing telemedicine. Behavioral health providers have been among the leaders in telemedicine. You know better than anybody that it can be very effective and been cost effective too. Well, now the rest of America knows. And telehealth visits surged starting in March and continued dramatically to surge, and it's not going back. We now have an infrastructure, not a complete infrastructure but an infrastructure that really does support a lot of telemedicine. A lot of providers now make it central to their strategies, and patients seem to like it. So all of what I'm describing now seems, to me a least, a lot more realistic than it did just a year ago where telemedicine still seemed to be a novelty.

BARAK RICHMAN Another lesson in this, very painful lesson, is that our parents and grandparents are not-- well, maybe it's an overstatement to say they're not safe, but they encounter or have encountered over the last six to nine months real health threats by living inside retirement homes, and it has made it all the more urgent to invest in making independent living easier and safer. It's been long known that the problem of elderly social isolation is a significant comorbidity. And to the degree that taking the elderly away from independent living and moving them out of their environment that they're familiar with disrupts social relations. There's always been a real health risk to bringing people inside retirement homes. And, well, now we know there are other health risks, and it makes it all the more urgent to try to sustain independent living. The COVID crisis has also accentuated the need to provide accurate and reliable health information. This is the dysfunction of the information ecosystem. It started with politics and it went to COVID. We cannot have a public healthcare policy where these are the folks that everyone's relying on to get healthcare information.

BARAK RICHMAN Now, I don't mean to put any blame on any one person, but ever since COVID broke out and there's been all these conflicting sources of information, I have not gotten a single email or phone call from my provider, my general practitioner, my PCP, saying, "This is what you're supposed to do." And what we really need to do is shift the source of information that people get from these talking heads and to trusted providers. You can imagine all sorts of pictures I could put here, but these are providers that we trust that we have relationships with. The entire information ecosystem really needs to change. And there's no reason that providers, our providers who know us and know our communities that we have relationships with, cannot be much more active in disseminating information to us. In fact, actually, it's a bit of a crime that they haven't been.

BARAK RICHMAN So let's think about what a new architecture would look like in a world post-COVID. Well, we certainly want to find any excuse possible not to go to the hospital for all sorts of reasons. We want to embrace all the possibilities to telemedicine, whether it's the local doctors or whether it's physicians that we can engage with directly. They certainly don't have to be local for engaging with physicians and providers through telemedicine. And for the most part, you can also imagine all different kinds of integrated telemedicine offerings where behavioral medicine, where nutritionists, where other kinds of behavioral interventions are really integrated into the core of primary care. We also could imagine actually removing our doctors or not having a hierarchical relationship with our doctors, maybe perhaps having a more direct relationship with our doctors through insurers or through some kind of information ecosystem and having them-- even though we don't see them physically regularly, we have them much more integrated into our daily lives, whether it's through telemedicine or through some kind of digital interface or some kind of information system. But having them much more actively engaged in providing us with information and direction and being much more patient focused.

BARAK RICHMAN So the conclusion. We live in interesting times. There are lots of possibilities that digital technologies offer, and there are lots of alternative delivery mechanisms that I think are on the verge of entering the marketplace. They might need some help. They might need some definitive policy directions. There might need to be some painful technological disruptions. But we also can think much more creatively or it's a time where you can think very creatively about how we organize the delivery system in ways that really deviate from how it's traditionally been offered. Like I said, I welcome questions now. I welcome questions after this. And I look forward to continuing the conversation.

ALEXANDER DENKER Thank you so much, Barak, for an excellent talk. So let's look at some of-- before we look at some of the questions-- and I believe you should be able to see some of the questions in the Q&A, and many of our attendees did upvote some questions so you can see some of the most popular ones. But I want to touch on, for a moment-- you've gone through a lot of these different possibilities in terms of how new technologies can be implemented, and some of the questions that-- this alludes to some of the questions that have come up a lot already. That all depends on adoption and what challenges are facing adoption. And efforts like Microsoft HealthVault and Google's Project Nightingale have faced incredible challenges with privacy concerns. And so I wanted to see if you could touch on issues of privacy and how that's playing into the US healthcare system and whether that is a misunderstanding of the system and that's why adoption is not happening or if we're just not engaging stakeholders properly, in general.

BARAK RICHMAN Yeah. I mean, a lot of this presupposes a different kind of economic relationship that not only we have with our providers but our providers have with our payers. None of this would happen if we continue to rely on fee for service. I just don't see any kind of adequate inducements to invest in a new array and standalone arrangements, whether it's through telemedicine or through digital apps or information mechanisms under the current fee-for-service system. So that is, you might say, the easy part. You certainly need to have arrangements, financial arrangements where investments in these new models would be profitable. And they largely would be profitable only if there's some kind of risk-bearing, only if there's some kind of return on reducing healthcare costs. So that's one thing. But in addition to-- but new financial models and new financial incentives are a necessary but not sufficient condition. I think we also need to orient ourselves more towards the world of population health.

BARAK RICHMAN The reason we have fee for service is because we still think of the unit of analysis of delivery as the patient-physician encounter. And that's not going to work in a population that has lots of health needs that continues to grow and where the access to providers is just really expensive. We're going to need to figure out ways to scale. We're going to need to think about how healthcare delivery moves away from the one-on-one interaction, to where the population, the community is the unit of analysis. And look, I mean, you can't do back surgery-- well, actually you can do back surgery. You can turn it into a-- you can have specialized centers. You can have what are called focused factories. And to a large degree, the large multispecialty hospital really is kind of on the verge of being extinct. It's very hard to do lots of things really well. So you can even scale up things like back surgery, hip replacements, and the like. But what we really need to do is to scale up more primary interventions and certainly behavioral interventions. We need to think about this in terms of scale, and that does require a different skill set.

ALEXANDER Could you talk a bit about how this concept could apply to care of marginalized patient populations such as the uninsured and the homeless? You briefly started to allude to that when you were talking about focusing on specific communities and populations.

BARAK RICHMAN Yeah. That's a terrific question. It's a really important question. And this audience knows better than I do how central integrated care is for the Medicaid population, for homelessness, and for homeless patients. There is never just one element that the costliest patients are suffering from. In fact, actually, I certainly know there are lots of comorbidities. But Josh was telling me before that, invariably, one of the comorbidities involves a mental illness. So in a world where low income and low SES populations are suffering from a number of comorbidities, a number of different health problems, both physical and mental health problems, it's all the more important to have a number of different touchpoints, to have an integrated strategy. It also is all the more important to make sure that there are efforts to be interventionist, to act upstream. And I think the individual provider cannot do this. So to the degree that the social determinants of health are increasingly important, to the degree that integrated care is increasingly important, that only accentuates the need to have a comprehensive approach to delivering care in a different kind of architecture that's patient centered. So how to do it? It's going to be hard, for sure, but all of the issues, all of the intersections between the social determinants of health and acts and healthcare delivery are going to have to be a part of developing this new architecture. By the way, I hope the leaf blower in the backyard isn't being problematic. This is another casualty of COVID.

ALEXANDER DENKER Can you talk a little bit more about the privacy concerns with both big data and healthcare and big tech companies getting involved? And also, what role do you think the federal government plays or local governments even play in regulating these data?
BARAK RICHMAN Yeah. So the story about privacy is really, really fascinating. In many ways, it's the primary issue of the 21st century, or at least of the next decade. When we think about-- can you still hear me? Because the leaf blower's right in my left ear right now. I'll talk louder.

ALEXANDER DENKER You're okay. We can see him in your left ear too. You're good. [laughter]

BARAK RICHMAN Well, hopefully, he's not too distracting. So it's a big issue behind healthcare. There are a couple things about privacy that make this situation even more difficult than you would think it is colloquially. First of all, we have a regulatory system that heavily regulates private health information, PHI, but does not heavily regulate non-PHI or at least pieces of information that are identified as PHI. So biometrics are PHI, healthcare delivery. Your claims data is PHI. What you eat is not PHI. Where you are is not PHI. These are things, of course, that Google knows all about you. Who your friends are are not PHI. Facebook knows all about that. But of course, where you go, who you interact with, what you eat, those are very significant determinants in health outcomes. So we have a regulatory system that heavily puts under lockbox certain information that is determinative of health outcomes and has no real security over lots of other pieces of information that healthcare providers want.

BARAK RICHMAN And, in fact, there is a really interesting lawsuit going on right now where the University of Chicago teamed up with Google trying to reduce hospital readmissions, and said, "Google, maybe you can help us predict who's going to be sick after we release them." And Google, of course, can do that. And that might be a real violation of privacy. So we have a regulatory system that really is out of whack, in large part because it was established before smartphones. So given the need, the general need to really update how we regulate private health information or our private information in general, there are going to be some real concerns about how not only that information is gathered - and by the way, it's already being gathered by Google and by Facebook - but how it's utilized. And I have a couple ideas of what needs to be done. What I will say right now, though, is that you might say in addition to reorganizing the architecture of healthcare delivery, we also need to reorganize the architecture of regulation. It makes no sense that the FDA and CMS and the Justice Department are all doing different things on health privacy all on their own. They really should be doing it in a comprehensive way.

ALEXANDER DENKER Thank you. We have a number of questions about the doctor-patient relationship and various issues surrounding the doctor-patient relationship both with how digital technologies can impact that relationship. And also, a longer question about how people might fall through the cracks during a transition from more in-person care to digital care. And how might we get ahead of that issue while digital healthcare is really young right now, when we can plan for that even.

BARAK RICHMAN Yeah. I mean, the questions reinforce each other. I mean, of course, yeah, I can imagine the gold standard being the one-on-one relationship between the individual patient and the individual provider having time with each other. But we have limited resources and we're going to have many, many people fall through the cracks, to use that language, if that's the only way we can get access to a provider. Now, telemedicine might expand the ability to have individual relationships in patients and providers, but people are still falling through the cracks for that also. There just is a lack of availability. World of limited resources, not just limited dollars but also just limited number of professionals. We are always going to have unmet demand. So it is a necessity. I think it's an economic necessity to figure out how to organize delivery of care, broadly speaking, in some kind of scalable way where patients can really access what they need in effective ways.

BARAK RICHMAN But we're not there yet in the sense that even if we implement the architecture that I envision, even if everybody walks around with a smartphone-- and of course, not everybody has a smartphone. And of course, even people who do have smartphones don't have great internet access. That's really important too. But assuming that everyone walks around with a smartphone with great internet access and knows how they can press a little app and access some kind of conduit to a provider, whether it is first some kind of AI-generated product or whether it's a behavioral therapist or a nurse practitioner or ultimately a physician or a psychiatrist, even if we had that infrastructure setup, we don't really know how to engage people yet. I mean, this is really the world where you operate in, where your research is so important. How can we create some kind of very meaningful or effective interaction of 20 minutes, whether it's computer generated or whether it's done in groups, perhaps one provider with multiple patients communicating virtually together? That also might provide some kind of social support that would be helpful.

BARAK RICHMAN Of course, there's a confidentiality problem that would have to be worked through too. Whether it's computer, whether it's group settings, whether it's individual conversations with nurse practitioners or professionals or ultimately with psychiatrists, that suite of offerings, it has not yet been determined what's going to be effective. I know there's a lot of research trying to answer that question, and hopefully, with the possibility of establishing a digital architecture where that's really possible, where some suite of offerings, partly digital and partly human, can tend to the needs of a large population. Once we set up the infrastructure and the payment models, then I'm fairly quite confident that the actual products will follow because that's where you come in. I mean, that's where there's a lot of active research. And I am confident that-- although there will never be any kind of perfection, there will always be people who continue to be marginalized or whom continue to have unmet needs, I'm confident that that kind of infrastructure can really invite a lot of creativity and a lot of effective care.

ALEXANDER DENKER There have been a couple questions to clarify what you were talking about, about what people do when they need to travel for healthcare and how that might be covered under a new system, those expenses and tackling that issue.

BARAK RICHMAN Yeah. That issue's already been addressed in a lot of ways. Charlotte is a place with a highly-concentrated healthcare delivery system. Carolinas healthcare, which is now Atrium, is a local monopolist. And Lowes, which is a competitor of Home Depot, is based in Charlotte. And Lowes realized that there was a lot of care that their employees were getting locally that was fine care but it was really, really expensive. So Lowe's made the very basic decision that, "We're going to fly people to the Cleveland Clinic to get certain kind of care. And not only will they get great care at the Cleveland Clinic because the Cleveland Clinic is world renowned, but the cost, even including not just airfare but the hotel stay and the like, was actually a lot less expensive. The kind of healthcare prices we're spending in this country are so extraordinary, so off the charts compared to other OECD nations, that it's very easy to pay for lots of care and frankly a recovery bonus and still spend less money than we would if we just procured the care locally.

BARAK RICHMAN Now, there are some additional logistical problems. What happens when your surgeon is 1,000 miles away and you have some kind of complication? There needs to be some kind of connectivity to being attended to locally. But the digital world makes that really easy also, whether it's speaking to your surgeon or some kind of care team that is 1,000 miles away, or speaking to somebody locally and having a local provider communicate with your surgery team that's 1,000 miles away. We don't really have those connections yet. But we should and we could and we certainly would if travel became much more common, more commonly associated with parity with getting care. I'll add to that, by the way, that most visiting rooms, most hospital rooms will soon have Zoom hookups where you can be visited by your family members who might be 1,000 miles away also.

ALEXANDER DENKER Sure. We also have a number of questions about technology literacy. How do you tackle-- how would you tackle those components of technology literacy and how approachable technology might be, especially when it comes to elder care, or as you were talking about earlier with taking care of an aging parent and how technology can assist with that?

BARAK RICHMAN Yeah. I mean, I certainly don't know how to do that. I don't know how to even train my own mom how to navigate through Zoom or FaceTime, but there are lots of people who do. I mean, just think about the world that Apple has developed and the kind of accessibility to digital products that they've been able to create because they're brilliant designers. Now, I will tell you that I do not think Epic and Cerner has a bunch of brilliant designers and it shows. But there are people who think about this very, very deliberately, very thoroughly, and think about how to make these sorts of digital products intuitive and helpful. And I'll also say that the elderly in 10 years will be very differently from the elderly we have now. And maybe, frankly, there might be some other care strategies that would involve some grandchildren teaching their grandparents how to use stuff. Maybe that would even be reimbursable.

ALEXANDER DENKER So we have time for a couple more questions. Josh, I don't know if you have any questions also? Where do you see the role of community health centers in the near future?

BARAK RICHMAN Yeah. I think community health centers are outstanding platforms where this could take place. Now, community health centers take all shapes and sizes. But the community health center that I think of, especially when I think about some of the research that I've done in rural North Carolina, I think about community health centers being staffed by community leaders or people who are familiar with the community, have some rudimentary skills about delivering healthcare or at least healthcare information, but having zero resources. And I think about a place like that could be really, really valuable in being some kind of focal point instead of a hospital, instead of a physician office, where one individual is encountering one provider in a fairly intimidating and discomforting setting. I think about these community centers as being terrific take-off points where you can imagine lots of telemedicine being provided, not just the physical infrastructure but also, perhaps, doing it in group settings where lots of local providers, nurses, community health workers, are getting supervision from more highly-trained providers that might be in a nearby metropolis.

BARAK RICHMAN I also think schools, public schools, could be really useful platforms, and this is exactly the conversation that gets me excited. If we were to imagine a different architecture, where would be the points of interface? It could be the smartphone, but it can also be lots of other places where community members frequent. And you can think about lots of really valuable community interventions, population and health interventions, that would be more effective in these community settings than anywhere else. That's where I get excited, thinking about what one might be able to do with these community institutions and how we might be able to integrate them-- I was going to say integrate them into the delivery system. What I really mean is building some kind of delivery system around them.

JOSHUA GORDON Well, I think with that note of excitement, we should draw it to a close. Other folks do have places to go. As wonderful as this hour has been, I want to thank you very, very much, Barak, for your really inspiring talk. It is compelling to think about what kind of a system we could design for care, for health in general, and for mental health, particularly if we went about it without necessarily trying to build it on top of the architecture we already have. And so thank you very, very much, and we'll see everyone else at the next all-hands meeting next month. Thanks, Barak.

BARAK RICHMAN Thank you, again. And for questions that I wasn't able to answer, I welcome emails directly.

ALEXANDER DENKER And I will send you the questions that came in. I can create that report and send that to you as well.

BARAK RICHMAN Perfect. All right. Thank you again for the invitation.

ALEXANDER DENKER Thank you, everyone.

ALEXANDER DENKER Bye-bye now.