17 Conference Report #techonomy17

What Can Be Done About the Rising Cost of Care?


Dan Munro

Mario Schlosser
Cofounder and CEO, Oscar


David Kirkpatrick
Founder and CEO, Techonomy

Session Description: The U.S. spends twice as much on health care as other rich countries, and gets less care for the money. What will it take to make positive change?  How can insurers and doctors use predictive data to help patients and reduce costs? Is consumer-friendly, affordable health insurance possible?

The full transcript is also available here as a PDF.

Kirkpatrick: Now we’re going to look at why the American healthcare is so screwed up and what we can do about it. So please come out Dan Munro and Mario Schlosser. The statistics are shocking about how bad American healthcare is vis-à-vis other countries and in a minute Dan will recite those. But the fact is that it’s unconscionable, it’s a huge cost that we really shouldn’t have to bear as a nation and as American business and as consumers and people who are trying to stay healthy. All of the above are not working correctly in this country. And what we have up here are two experts on that, one is a practitioner and the other is a very polished observer.

Mario Schlosser is the CEO of Oscar which is a company, a health insurance company that has a lot of ideas about how to do things more efficiently using technology. Essentially it emerged as a response to Obamacare and so he has a lot of understanding about what’s really happened there, what it’s intended to do, what it might do, and we’re going to hear a little bit about that. Famously his partner is Josh Kushner so it gets a little extra coverage because of that and recently the two of them wrote an op-ed at Axios where they basically said, you know, that the president’s activities were having both positive and negative effects on Obamacare. But just the fact that you even implied that there was a little negativity there made a lot of news because Josh signed it along with you.

Dan Munro is a writer who has written about a lot of tech-related stuff and other topics but has become really an expert on healthcare and American healthcare policy reform in the last couple of years. He wrote a book in 2016 called Casino Healthcare, which is a powerful indictment of the American system and an argument for universal coverage. And I won’t use other buzz words because we’ll talk about the buzz words. But quickly, Dan, what are the statistics about how bad it is here?

Munro: Probably the first and the easiest one to kind of grab ahold of is the fact that the U.S. healthcare system as an economic unit if you measured it is basically the size of Germany’s GDP. And even abstracted one level lower than that, just the budget for Health and Human Services, which is now the federal government’s largest single department, just HHS is the GDP basically of Spain. So those two components coming together give you a sense of the size of the cost issue that I reference as being both a four-alarm fire, a category 6 hurricane, and on the Richter scale, a 9.0.

Kirkpatrick: And in the piece that you wrote for us in the last issue of our magazine, we had a chart that showed health expenditure per capita on one axis and life span on the other axis. And it is worth looking at that because it is so shameful. The U.S. spends so much more money and gets so much less for its money than any other major country. You know, Japan, U.K., all these countries spend less and people live like four years longer, five years longer. It’s just crazy. And we spend dramatically more than any of these other systems per capita. So what we’re here to do is to talk about what can we do about the rising cost of healthcare. So Mario, talk to us about what Oscar is trying to accomplish.

Schlosser: Think about in abstract healthcare costs, there will be three components that go into healthcare costs. There’s risk of population, there is utilization per risk, how often do you go to the doctor for a given condition, and there is unit costs in essence, so for every utilization what are you going to pay. When we compare the U.S. to other countries in the world at the abstract level then from a risk perspective the population isn’t that different.

In fact the U.S. population is on average, the median age is three years younger than most of these countries, rich countries in the world. From a utilization perspective we go to the doctor a little bit less, we get discharged from hospitals a little bit less than the rest of the OECD countries. And so most of the issue is in unit costs. That goes right back to the fundamentally incorrect sort of incentives in the healthcare system. Every provider isn’t in every insurance company’s network; there’s very little value and comparative pressure on the value chain. And the reason for that is that as an end user you don’t really pay for your own healthcare and don’t have much choice in the matter of how much you want to pay for that healthcare.

So the way we address this sort of like too confusing and too costly system is in three different ways starting with consumer engagements, we want Oscar to be the organizer of your healthcare. We care about 45 percent of all first-time physician visits—Oscar has a first-time physician visit, we can prove that we were the ones who recommended to you the physician that you ended up going to. We do alot through telemedicine. About 70 percent of all routine rashes and infections are now done through telemedicine for Oscar members and engagement is very different where we become the entry point of healthcare. We build different networks, for example; we just launched a healthcare plan with the Cleveland Clinic; we share risk 50/50 between us and the Cleveland Clinic, which is the perfect way in which you will—

Kirkpatrick: You mean financial risk?

Schlosser: Financial risk, exactly, which is the perfect way in which a provider system and an insurance company basically acts in a cooperative and coordinated way because you both share fully in the upside and the downside of the outcomes that you create. And we do all this on a very different technology stack. We talked about this a bit earlier but right now if I wanted in healthcare to do what Amazon, Uber, others in consumer industries are doing every single day, which is pricing based on demand and supply, I couldn’t do it as an example because the underlying infrastructure isn’t there. As an example, the typical claims formats by which a physician practices, so it’s a claim to the insurance company, doesn’t have a time of day in that format and so I couldn’t actually give you a discount if you went to a doctor in off-peak hours. Something as basic as that.

Kirkpatrick: And that’s sort of a rigidly predetermined form that can’t easily be changed.

Schlosser: We can change it because with the insurance the fees beyond Oscar from the beginning was the insurer is in a good position to become that organizer of your individualized, consumerized, personalized healthcare because the insurer controls the flow of data and the flow of money throughout the healthcare system and therefore can reroute those in smarter ways.

Kirkpatrick: So what is the fundamental incentive for a consumer to use Oscar versus another insurer?

Schlosser: Better experience and a lower price, as simple as that.

Kirkpatrick: It’s lower-priced but it’s also a more technologized, more user-friendly, and consumerized experience because I know consumerization is a key thing that you advocate that is basically absent in the American healthcare system.

Schlosser: Exactly. We essentially go to the member and we say we are going to build you a more customized, more narrow healthcare network. We work, for example, only with the Cleveland Clinic not with the other hospitals in Cleveland, Ohio. We work only with Mt. Sinai in Manhattan, not with the other hospitals in Manhattan. We therefore get lower unit costs from those providers but then because we have your attention as a member have a better, more scientific way as well to get you to the right provider for the right reason.

A simple example, again, if you go to the typical insurance company, you can try this today, go to the web, search one of the big incumbents and you look at the physician finder, very simple trick, you see they will distinguish about 100, 120 or so physician specialties, that isn’t enough to properly care routes, you know, the various conditions you may have as a member. You need about 400 different specialty distinctions and so we can build those in that tree, that hierarchy of medical sort of like concepts in conjunction with our healthcare systems because we built a narrower network, in exchange it got better unit costs.

Kirkpatrick: Just quickly, how does Obamacare fit into this?

Schlosser: In a very simple way.

Kirkpatrick: As quick as possible.


Schlosser: Yeah, in a very simple way, it was essentially the first really consumerized, individualized market where you didn’t have to go through your employer or through a bunch of other middlemen to select your healthcare. You can pick one plan, actually stay with it for a long period of time, and so in my opinion, whether the ACA survives in this current form or not is almost irrelevant, it points into a future where you as an individual can vote with your feet as to which insurance company and which healthcare subscription package—that’s what insurers really are—you want. For us it means about 40 percent of our members in New York, for example, have come through word of mouth now. You know, so we can have viral growth in a way you would never have if insurance stays in non-consumer products, if you get it through your employer.

Kirkpatrick: What percentage of your customers now, individuals, are coming through Obamacare?

Schlosser: I’d say it’s about 90 percent or so. We have an employer market or an employer plan now as well that’s starting to grow, but it’s about 90 percent.

Kirkpatrick: Effectively at this point, any individual who goes to Oscar is in effect using ACA.

Schlosser: Yeah, that’s right. I mean, so, if you ask me how they come to us I’d say maybe about 20-25 percent have come through a government website.

Kirkpatrick: Okay.

Schlosser: The rest come directly to hioscar.com, you can sign up there in five minutes or one of our people on the phone.

Kirkpatrick: Okay, Dan, what is your fundamental prescription for reform? Because I know you have a pretty good way of synthesizing this.

Munro: It’s hard to see this from the vantage point of where we are today. There’s a tendency to want to think of healthcare reform, systemic healthcare reform, as a single piece of legislation or a single administration. And the reality is that it’s actually going to take longer than a single piece of legislation or a single administration. The fact is that it’s probably a 40 to 50 year horizon or timeline for what I would call systemic healthcare reform. We’re in the middle of that timeline.

Kirkpatrick: Wait, but what’s the endpoint, obviously.

Munro: So the endpoint ultimately is if you look at other industrialized countries is to get to a model that revolves around universal health coverage and I make that distinction of using the word coverage because there’s a tendency to wrap universal healthcare and in effect blanket that as the solution when universal healthcare can be confused and often is confused, intentionally, with a single-payer healthcare as the only prescription that is on the road ahead. And the fact is we can get to universal health coverage without single-payer, we can do it as a multi-payer model.

Kirkpatrick: Which clearly seems to be what this country wants but the bottom line is you take it as a given, an inevitability, that we will get to universal coverage which seems so ludicrous that we’re not there already which every other industrialized country is already at. But where do you come up with the 40 to 50 year timeline?

Munro: Yeah, in a sense you look at the history of other countries that have duplicated this or that have gone on their own trajectory and their own timeline. And the best example is probably Canada and it took Canada literally about 40 years to get to a process of universal health coverage. And I wouldn’t recommend necessarily single-payer, I would absolutely recommend a multi-payer model for our system because it’s more of a cultural fit. We’re not a country that is likely to embrace fully the concept of single-payer, in part because there’s too much of an overhang politically with the idea of government run, government owned healthcare. We’re not culturally fit for that kind of a model and we don’t need to be in order to get to universal health coverage.

Kirkpatrick: So that all sounds right to you?

Schlosser: That’s exactly right. See, you need a system where the end consumer knows what he or she is paying and what value he or she is getting and where I have the power of putting pressure, a bit of pressure on the entire value train and that’s completely absent right now. If the employer with a big tax break pays for a broad network where there are no providers to compete for any value, you end up with a fee-for-service model where insurers pay for every single thing on the bill, you end up in the system we have right now where every single actor in the system has incentive, because you earn a percentage of the overall cost, for the cost to go up. And that’s how you end up with, I think the CAGR, the annual growth rates of just the employee contribution to employers for some healthcare between 2010-2016 has been at 10 percent.

Kirkpatrick: Ten percent annual growth in what individual’s pay for their health insurance?

Schlosser: Contributes to the health insurance to the employer and the employer’s cost has gone up as well.

Kirkpatrick: Okay, really quickly, how do you both feel and how positive has it been that Bernie Sanders has actually started talking and created almost a movement around this idea of Medicare for all which seems to me to be a very positive sign in that it has actually gotten some traction.

Munro: Absolutely. In fact, I’m thrilled in the sense of the kind of education that that has brought to American consumers and the American population at large. It’s conceptually understanding the idea of universal coverage. I’m not convinced and, in fact, Lawrence Lessig had a great quote referencing Bernie Sanders in this specific way when Lawrence said, “You know, everybody rolled their eyes when Bernie said Medicare for all and it wasn’t because it was a bad idea, it was because everybody understands that in this money-driven political system it’ll never happen.” And that’s a key element of the change that Bernie is bringing in the sense of the overall dialogue to the next—

Kirkpatrick: But it’s something that ordinary Americans understand and they don’t understand single-payer or single-pricing, they don’t even understand universal coverage, but Medicare, pretty much everybody understands. What do you think?

Schlosser: Well, I think you have simple issue in the U.S. economy right now and that is that healthcare is 20 percent of it. You see this very nicely when you drive into Pittsburgh and there’s the U.S. Steel tower right in the middle and that obviously was the headquarters of U.S. Steel when people were still manufacturing staff in the country and it’s now full of hospital administrators.

Kirkpatrick: Is that right? In Pittsburgh, yeah?

Schlosser: Yeah, it’s a UPMC system and it’s not even a hospital, it’s just administrators basically. And what better metaphor for like the switch, the shift of the economic forces you’ve had in this country. And if you want to get the cost of the system down, somebody’s cost is somebody else’s revenue, you’re going to have to shrink a whole bunch of connectivity back down to where it ought to be and that means capacity reduction. Right now you have an overbuild of MRI machines, you have an overbuild of emergency rooms in this country, and that’s going to have to shrink more towards primary care and more towards telemedicine, more toward even nurse practitioner–based practices and that’s a very, very painful process when—the Steven Brill quote—when the highest paid guy in the community is the hospital CEO. And like Germany, for example, we talked about this earlier, the way they do it is they’re a multi-payer system. I joined an insurance company in Germany because it had the name technician, technology in the name, literally, and I knew about it—there were not any more technology sides than anybody else but I liked the name basically. But they limit provider pricing, the government says this is what we’re going to charge for hip surgery and that’s it so you can make money at this level if you’re efficient, if not you go out of business. And so that’s how other countries choose to base the rate limits, the consumption of healthcare services.

Kirkpatrick: Right. Really quickly because I want to go to the audience but one of the great points you made, Dan, is that we always hear about the U.K. system as kind of the alternative, right? That’s the universal coverage, single-payer that we ought to be like, people seem to think. You reject that.

Munro: Yeah, I do. If fact, in part, because I don’t think culturally, again, we’re a good fit for a government-run, single-payer system and the good news is that we don’t need that in order to get to universal health coverage.

Kirkpatrick: And that’s the distinction between U.K. and Germany is [that] it’s government owned, hospital systems, everything in the U.K., in Germany is essentially a commercial system the government highly regulates but especially in terms of pricing.

Schlosser: Just in terms of pricing, everything else is basically free market. In Germany 10 percent of GDP healthcare, in U.K. 9 percent. So there is a bit of a difference but not that much.

Kirkpatrick: Okay. I want to talk tech. We haven’t gotten to that but who has questions or comments—any thoughts or questions?

Smythe: Roy Smythe from Philips.

Kirkpatrick: Oh, an expert.

Smythe: I’m not an expert. There are no experts as it turns out. This question is for Dan. So, Dan, wouldn’t you say that universal health coverage is necessary but grossly insufficient? I mean, just giving everybody coverage and layering that on top of a market-based healthcare delivery system is really what we have. You’re not really fixing the delivery of care, you’re just insuring everyone.

Munro: I think in a lot of ways, what you’re doing though is you’re removing a key incentive that’s at conflict with ultimately what healthcare is. Healthcare is not a consumer product.

Smythe: That’s how we treat it here.

Munro: That’s how we treat it, in part, because we’ve done it with other industries and we assume that we can do it simply in healthcare by tiering pricing and the whole basis of tiered pricing is assuming that you can extract and maximize the revenue and profits associated with tiered pricing. We do it for airlines, we do it for hotels, we do it for cars, we do it for appliances, we do it for mattresses. So we do it in all these other industries and the assumption has been that, gosh, we should be able to do that in healthcare. The trouble that it creates within a fee-for-service model is that when you have inelastic pricing, everybody drives to a quarterly beat and it’s the quarterly metronome that’s driving our healthcare system to make choices that aren’t in the long-term interest of both the country nationally and then in terms of our individual health, which is the real risk.

Kirkpatrick: Please feel free to jump in but the observation that I would make based on the fact that Techonomy has been really getting our hands dirty in this area for a couple years now that maybe I’m surprised was a revelation. But the real issue is that the U.S. is a fee-for-service ecosystem and the services are priced pretty much based on some weird factors that don’t have much—really don’t have anything to do with the health implications, the outcome of the pricing. So we are not paying at all for health, we are paying for services, which is why there are so many extra emergency rooms, so many extra EMRs, etcetera.

Schlosser: Yes, exactly. If I may add something to this as well. In a system where no single actor is responsible for organizing your healthcare or for the outcomes that the system produces, nobody becomes any good at managing all these various services and coordinating them. Here’s a good example from just recently, CMS, the kind of federal regulatory body, has a program in Medicare called ACOs, you know, these are physician practices that basically start taking some risk for the outcomes they produce and you can either go fully at risk, meaning you share upside and downside with CMS or you can sort of a light version where you would just share in some savings basically.

And 95 percent of all these ACOs across the country are merely in the shared savings thing and CMS gives them more money than they actually save. In other words, if somehow somebody out there figured out the administration and management of your care, which service at what point in time, end to end, basically to get you healthier at lower costs, they will be levering up on this like crazy. This is like a hedge fund manager who doesn’t want to put his own money in his own funds, you know? And that is what you get in a fee-for-service system where the consumer has essentially no role in the setting of cost and the demanding of value.

Kirkpatrick: Okay, quickly, Josh.

Reynolds: My question is do we have to agree first as to whether or not healthcare is a right or a privilege to achieve truly sustainable healthcare reform given that being healthy is a choice but we’re now trying to decide whether or not to regulate access to health if you’re not healthy. Do we have to agree to that first before we can reform and what does that do to your time horizon?

Schlosser: It’s almost like—I almost think that I could go into lots of philosophical thoughts here but the fact of the matter is we spend way more than any other country on earth on healthcare and we probably are treating it the least as a right. So clearly whatever we have been doing has not been working, somehow we don’t let people die when they get sick. They can go to the ER, get care there, it’s horrible, suboptimal care and it ends up being a lot more expensive than if we probably organized a system that invests more in your long-term future and health. The other thing by the way, one additional thing to that as well, is right now no insurance company is able to really invest in your long-term health because once you change jobs and employers, you change insurance companies and boom you’re somebody else’s problem or creative investments in a sense, that all goes back to the same thing. So my opinion, whatever we’ve been doing, treating it not as a right for the most part has not been working and we ought to try something different.

Reynolds: And to your point, Atul Gawande wrote a recent article for The New Yorker that talked about this exact point and I think, yes, ultimately we need to come to a polarity or a plurality that healthcare is a right.

Kirkpatrick: Two quick things I’d like to just hit before we wrap and we do have to pretty much run. And one thing, Eric Topol, who’s a great, great thinker on this and I hope he’s in the audience, but said at the section on data and ownership of data earlier was we don’t currently put your data all together. And if we do, which is a relatively simple technological challenge, we would be able to get radically improved health outcomes and the reason we can’t do it is because of the organizational inefficiencies of the system. And then I wanted to lead into the point you made to me which I found shocking because you’re really making some really disturbing discoveries by operating as an innovative insurer and the one you said to me about the names. Quickly, you know, the doctor’s names.

Schlosser: Oh, yeah, yeah. Right, well one of the big systems we’re working with has an Epic installation, Epic is the most kind of popular EMR system in the hospitals and everything. And basically they identify their doctors in that system by first, last permutations by text strings, you know, that’s how the matching of physicians still works. So in other words that’s as if Facebook knew no difference between various John Smiths.

Kirkpatrick: There’s 30 David Kirkpatricks.

Schlosser: In a sense, exactly. And so it just shows the very, very crude setup of data systems and therefore lack of interoperability that you still have in healthcare nowadays and it all goes back to the fact that none of the formats in healthcare, data formats, have ever been devised for any kind of clinical interpretability of clinical payloads, it’s all purely for payments. Hence, also, that’s why you don’t need a time of day on a claim, you know?

The other thing I think I mentioned earlier is which I found pretty crazy, the government basically tells insurers in Medicare Advantage, in individual, a small group, you’ve got to risk code your members. You’ve got to tell me how risky your population is, that can be done by pulling sort of charts and physician practices, which basically has to be done manually, you send the company into a physician practice that takes photocopies and the majority of these records gets put in—sort of like it scans but not actually optically character recognized and so there’s data about your physician notes and your care history sitting in a server somewhere but it’s all sort of like in image formats, it’s not really electronic, digitized so it could be used for proper care management and things like that.

Kirkpatrick: Okay. Is Eric in the room? Where are you? Okay, I just want to give you an opportunity if you have any final observations because you’re an acute observer of this whole situation.

Topol: Well, I think the unified points that I got from Mario and Dan is that it’s really about pricing and the cost of—it’s not even care, it’s health services. We don’t give care, that’s part of the problem. But what I think hasn’t been touched on enough, one is that there’s a third of all this $3.3 trillion that’s total waste, which there’s some things that are being done about that. And then there’s another big chunk that would be, if we used technology which the medical community doesn’t want us to use because it challenges the incumbents and all the cost structures like getting rid of hospital rooms by remote monitoring, getting rid of sleep labs which are complete, ridiculous drain of cost, the number one revenue centers in most hospitals and things like that. We don’t challenge the medical community with respect to these practices and so there hasn’t really been a movement to deal with the cost problem.

Kirkpatrick: Well, we would love to help get a movement started and I hope that you’ll stay with us and you guys will both continue to be part of our deliberations because it’s just inexcusable what the situation is today and thank you so much, Roy, and others, for participating in these discussions.


Tags: , , ,