a16z Podcast - How Transparent Pricing Drives Healthcare Change

Episode Date: July 15, 2020

Dr. Marty Makary—surgical oncologist at Johns Hopkins University School of Medicine, and health policy and innovation expert—has long been a passionate advocate for transparent pricing in the healthcare system. We don’t talk enough (or really at all) about price in healthcare, says Makary (instead, we talk about cost). But shedding a light on prices in healthcare—from not just what those prices are but how prices are set and the value we all receive as consumers of the system overall—can help us measure quality in medicine, and be a driver for real behavioral change in the healthcare system, correcting many of the unintended consequences of a fee-for-service system like surprise billing or unnecessary medical procedures.In this conversation with a16z General Partner Julie Yoo, Makary and Yoo discuss what price transparency in the healthcare system could really do; how we can "steer" towards the good physicians who are not just highly skilled, but make the right judgment calls based on need and holistic health, not cost; how we might distinguish between high value and low value through medical appropriateness; and how we might gain clinical wisdom from other kinds of scientific discovery beyond randomized controls, especially during the wartime protocol of COVID-19.

Welcome to the A16Z podcast. This is Julie U, and I'm really excited to have Dr. Marty Macquarie on the podcast with us today, someone who has been a advisor, mentor, colleague, friend for many years, and also someone that many of us have seen on the public sphere in the last few months speaking about the impact of COVID and what we as a society need to do to respond to it and also be prepared for similar situations in the future. Marty is also an expert on health policy, has written a number of books, and specifically wanted to open up with a bit of context on his latest book, which is called The Price We Pay. So much of the book is really like a call to arms on transparency and, you know, incentive and structures, and particularly

price transparency in health care as a driver for behavior change. Really what it gets at is how incentives drive behavior in health care and oftentimes very unintuitive behaviors in health care. The fundamental problem, in my opinion, that's always bothered me, is that the that we don't talk about price, and we don't use the right lexicon. For example, we talk about health care costs. We should be talking about medical prices. We talk about preventable adverse events. We should be talking about medical care gone wrong.

But it's ironic with COVID-19 that health care costs are actually one of the major reasons we are massively underprepared. And that's because why are we shopping in? in China for medical supplies and PPE. Why are half of antibiotics made in India and China? It's because of healthcare costs. And right now, we've got this entire industry, like other industries right now, asking themselves, do we really need to be doing everything that we do?

Part of the challenge is that when you do transparently display what is quote unquote price, it just opens a whole can of worms about like, how does that link to value? And perhaps that's of the big tripping points that previous efforts have had is even when you are able to display a number that somebody interprets as a price, it really doesn't get at the core of the issue that you're describing. Well, you know, I'm a surgeon. I specialize in pancreas surgery and pancreas

cancer. And there's really two operations that me and my partners do in very high volumes on the pancreas. The number one operation is called the Whipple. Now, the Whipple is a well-known procedure that any physician is very familiar with. It's a long procedure. It's very involved as an expert in that Whipple procedure. I can tell you 5,000 facts about that operation, except for one fact, and that is how much does a cost, even in my own institution. So we have had this complete blindness about costs, and it's not because people are diabolical. It's because the market has never required hospitals to identify their true cost. to give a price. So as a result, like no other business, hospitals have simply eyeballed revenue,

eyeballed expenses, and hope they come out on top. And if they don't, they fight harder at the negotiating table with insurance companies. And we're left with this incredibly pathetic system where we have what I call the markup discount game in the book, or simply known as the game. we inflate prices for the sole purpose of giving ourselves negotiating leverage at the insurance negotiating table to use bigger discounts. So mark up discount, mark up discount. Hey, we're going to increase your discount this year from 45% to you, the employer or insurance company. From 40%, we're going to give you 5% more discount. And then we go around and we dial up the price is 7%. That crazy game, not only results in this cloud where we can't have honest competition, but it also requires

armies of people to play that game. And where's that money come from? It comes from that doctor-patient interaction of that money. When I was reading your book, I found myself getting angry and upset because it's just completely opaque in terms of what we think the value is of the services that we're getting. But there is a legitimate set of points around this concept of, well, you know, certain patients are sicker, even if it's the same procedure, the context around that particular patient might have a big influence on whether or not you actually spend more, whether it's on the equipment, whether it's on the number of people that you need to support that care, whether it's even the things beyond just the surgery, the procedure itself, right,

all the downstream implications of that patient situation. It seems like at the end of the day, it's sort of a trade-off between specificity of how you define the type of encounter and, you know, the nuance of just, you know, humans and the fact that we're all so different in all over different context. How do you sort of reconcile that gap? Well, I really do think we have good people in health care at every level working in a bad system. Imagine the airline industry does not post prices. And instead, when you go on the travel sites, there are no prices with each flight. And instead, the airlines argued, we have to bill you after the flight.

We simply don't know that the cost is going to be. We don't know the fuel price that day. We don't know if you're going to consume a beverage. We don't know if you're going to be diverted or delayed. and the pilot could have to work harder in Billmore RVUs in their coding of the billing of the flight after they land. I mean, we would say it's not fair. That's what's happening in healthcare right now.

One of the counter arguments is, oh, we can't give you a price. We don't know if you're going to be in the ICU. Well, of course, if you're shot in the chest, I as a surgeon am not going to give you a price. And I shouldn't. We've got to take care of you. But 60% of health care is shoppable. And so the tension between this exciting movement

towards price transparency, which is exciting, revolutionary, reducing costs dramatically in some parts of the country, sites that allow you to shop for diagnostic testing and basic procedures and what we call shoppable services, which is 60% of all medical care, is market domination or monopoly pricing through mass consolidation of health care providers, in part by private equity, in part from hospitals, just expanding with piles of cash that, they've never seen. Why are we bailing out hospitals with $25 billion endowments to their institution, sitting on piles of cash reserves, sue patients who can't afford to pay their bills? And meanwhile, rural hospitals are closing across America. So I think people want to see some honest competition.

So first of all, as we know, there are federal policies that have been put into place that will compel hospitals to publish certain aspects of quote-unquote price. And it's a combination of, you know, what the hospitals will charge, but also the dark art of payer contracts that also get at what if the actual reimburseable price for these procedures and services. And, you know, what that's doing is really atomizing the individual procedures, which seems to sort of fly in the face of the whole concept of value-based care, where we are trying to move away from pricing individual things that are done to us as patients. Let's say, you know, fast forward, whatever it is, 10, 20 years, who knows when we move to a situation where we're majority in a value-based care

type payment model across our health care system, will the notion of price transparency still be a benefit in a world like that? Well, first of all, I think price transparency ushers in quality transparency. If you buy a television, and that's the only TV in town, your brain thinks, do I buy it or do I not buy it? If there's one for $500 and $550, your brain thinks, what are the differences in the specs, in the consumer experience, the ratings. And so price transparency ushers in the badly needed quality transparency. And it needs that jumpstart from price transparency. Now, remember, billing quality is measurable. Billing quality is medical quality. And financial toxicity is a medical complication. And in a piece that my team and I put out

in JAMA a few months ago, we proposed five measures of billing quality. by which all U.S. hospitals can be measured. And it could be as simple as a five-star rating on those five measures. These are ways to reward the medical centers and the hospitals performing well in the marketplace, treating patients fairly. Price transparency is a disinfectant against price gouging, which is a massive problem. If you as a medical service can move your services off the master hospital bill, off of the insurance contract, you can get a lot more money.

That is the dirty secret in health care. That has been the game. And that's why labs have moved off the master hospital bill and off of insurance and some physician groups and air ambulances. So that is the game where there's a demand in the market for people to have a good user interface and a good experience. So why can't we measure that? Do you get a surprise bill? Johns Hopkins, my hospital, zero surprise.

bills. It's an institutional value. Even some community hospitals have made it a value that they require of their private groups that contract with the hospital. They say, look, we don't want our patient's surprise bills. You can't do that if you want to practice here. About one-third of U.S. hospitals will go after you if you cannot afford to pay their bill in court, garnish your wages, put a lien on your home, take money right out of your checking account through a court order. Now, only about 8% of hospitals do it that aggressively. But we've asked, all U.S. hospitals to stop all predatory billing. And we've argued that the suing of patients to garnish their wages violates the sacred trust between a doctor and a patient. I personally have

volunteered to be a pro bono expert for anyone who is sued by their hospital for a bill they cannot afford where the price was not disclosed ahead of time. Many of those hospitals I've described have changed their ways and stopped that practice after people came out. Does the hospital show prices for common shoppable services when patients ask, right? If you don't give somebody a price for a basic service that's elective, you don't have a right to shake them down and ruin their credit history. That is our belief. Does the hospital charge patients directly when there's a complication that stems entirely from the fault of the medical system or provider and no fault of their own? You know, like the retained sponge after surgery. No patient should be billed and have their

credit history ruined over the bill for that complication. Yeah. You mentioned something earlier that's sort of curious, and I want to tie it to this point, which is that people who have been maybe implementing these types of predatory, quote unquote, pricing practices, ultimately realize that just by getting rid of those, that they actually increase their revenue opportunity and just business health, let's call it. Has there ever been any work or studies done on the actual sort of the yield from these types of efforts, like the suing of the patients? It seems like it's a pretty highly capital intensive effort. It's a lot of call center agent. it's humans that you have to hire to do these outbound calls, a lot of back and forth. And, you know, my understanding is that the patient liability, like the component of a hospital's revenue that comes from these direct bills to patients is relatively small in the grand scheme of things. It's typically 10% or less. Like, how much is actually being spent on these efforts? And at the end of the day, if you were to net it out, can there just be an ROI argument as to why those things should go away? Oh, it's funny to say that because every now and then,

when I've gotten into a back and forth with a hospital system, we hear the argument, we have to do it for our financial sustainability. And then I point out, and we have the data on this, that all the money they get from garnishing the wages of people that they sue. Remember, these are number one in America, our research shows Walmart workers, number two, postal workers, number three, food service workers. You know, half of America lives paycheck to paycheck and has less than $400 of cash on hand. Those are the people we went into medicine to help along with the communities in general that we serve. So we do hear that we have to sue patients.

And I point out that it's less than half of the CEO pay in one year, all the money they collect from garnishing wages. And that usually stops the conversation right there. But your point is a great point because the larger question of how much money do we spend on these billing, coding, elections practices and what percent of the overall cost is that? That is the million dollar question. That is a beautiful question. And we don't have studies on that. But if you look at companies like MD Save that are posting prices online now for services, they're observing that hospitals are posting prices 30 to 40 percent below what they otherwise would bill for. And why are they doing that because they are telling those entrepreneurs that we don't have to pay for the coding,

billing, collections, selling debt, that whole world. That's an army. The customer service for all of that, that's a lot of money. So that's a proxy that all of those games, the billing, coding, collections, customer service games are 30 to 40 percent of the overall cost of a service. So tremendous opportunity there to reduce health care costs overnight. You know, it's a great segue into the notion of the fact that consumers are bearing more of the health care dollar out of our pockets. And that's what's driving a lot of this change is the fact that we're feeling the pain ourselves versus when you're in a fully insured scheme, I should say, you know, oftentimes you don't actually feel the brunt of it. You know, right now we're obviously in this unfortunate macroeconomic situation where employees are getting laid off. A lot of companies are going through massive cost containment efforts. And, you know, one of the sets of predictions that have come out of that is that health benefit subsidies in general from employers are going to fall dramatically. And many of those people will not find new jobs and will ultimately likely enroll in Medicaid

and other types of direct coverage programs. But is there a silver lining in this where obviously it's unfortunate that so many people are losing their coverage, but if they end up in a situation where they are bearing more of the cost, will we get to a critical mass where there will be much more coordinated demand

for transparent services? And what do you think that will look like in the next few years as we come out of this pandemic? I do think that all of those things you mentioned increase further the demand for honest pricing and honest medical care. The big question of why does health care cost so much and why is it burdening America in ways that it doesn't burden other countries. The conclusion I made from all of the research I did was, number one, it's pricing failures. Number two, it's unnecessary medical care, which is estimated to be 21% of all medical care out there in a national survey of

U.S. physicians that we conducted from Johns Hopkins. Pretty scathing result. I mean, when people in an industry say that 21% of all their services are unnecessary, and that comes from the frontline workers, namely physicians, that's quite an indictment on the state of the variations in quality of care. And the third was care coordination. And I think employer-based health care is, is the most exciting thing going on right now, because not only are you seeing employers use a direct primary care, but they're doing steering, which is when you have primary care physicians able to steer you to what we call high-value physicians, the doctor that is going to do the knee replacement only when you really need it, or the doctor that's going to tell

you, here's an N-set, I could do a knee arthroscopy, but it's of no value, so I can't ethically do that. Those are the doctors you want, the ones with great judgment. And that is the great opportunity in revolution right now with direct primary care. It's not just giving holistic care. That's important. It's not just food as medicine and talking about sleep and all those things. That is very important that we need to do that. It's also those things and steering. And steering is one of the great opportunities right now. Yeah, absolutely. And this is a good hat tip to your previous book, Unaccountable, and that so much of what you talked about there was the fact that the data needed to make that very assessment of what you described of who

are the physicians who perform well, who provide good service, are cognizant of best practices and quality measures and whatnot. Oftentimes, that information is solely known by peers, by other physicians, not necessarily things that you can discern from data, nor based on patient surveys, which often maybe actually focus on the wrong things like, do they have good parking options and at the office clean and things of that sort. Do you still believe that the premise of getting that information in a peer type motion is the right way to go about making those assessments? And have you seen any successful efforts to actually implement that in real life? I do see now informed patients that we really didn't see the magnitude of which when I

penned that book. And that is, back then, people used to come to me and I would talk about we could do the operation laparoscopically. We could give it six months and see if there are any problems with someone's hernia. We would give many options. And at that time, many patients would tell me they'd almost be tuned out when I give them the options. Doc, whatever you want to do, whatever you recommend, I will do. Now people are coming in saying, hey, I went to my physician. He recommended this. I looked it up. I found out there's other ways of doing it and there are other options. I'm here to discuss that. So we are living in the era of the empowered patient. We should be using a patient-centered lexicon. And that really drives the issue, right? All of a sudden,

now you can resonate with everyday folks, just like in the banking industry. Once we started using real terms, instead of collateralized debt obligations, we could actually say spending your money, you know, that they're using your money to take risks where the American people bear the risk of those risky financial moves. So I think it's an exciting time, and it's also thanks to the many entrepreneurs that are facilitating this. Maybe switching gears to kind of the current moment and time and the fact that, you know, so many of the guidelines that are in practice today haven't necessarily been updated to reflect, you know, modern practice, modern patient dynamics, things of that sort,

or don't have, you know, real credible RCT type evidence behind it and therefore people sort of dismiss it saying there's not a trial and therefore it must be wrong or it must not work. And you had proposed a different construct in your book around And it's like the wisdom of the crowds of people who had been practicing these things in real life, leveraging their expertise and experience, and, you know, coming up with sort of consensus-based approaches to those kinds of guidelines with COVID. This is an unfortunate situation where things are happening so real-time. And day by day, we're getting more information that might completely change our perspective,

our opinions in a pretty binary fashion. Do you see ways for that framework to be applied in situations like this pandemic, where things are moving so quickly, where we don't have a body of real-world evidence and experience to pull from? One of the barriers to scientific advancement, in my opinion, has been the dogma that randomized controlled trials are the only type of scientific method that can tell us whether or not something is true. And that old-fashioned approach, which is not wrong, it's just incomplete, has really been a

major barrier. For example, new therapies for COVID-19 have been going through the same protocol. that we have for lifestyle medications and obesity drugs and chemotherapy that may be 1% better than the last chemotherapy. You know, when somebody jumps out of an airplane and you ask the scientific question, does a parachute work to save their life? You don't do a randomized control trial, right? We are in a war right now with this COVID-19 infection. So we need to use wartime protocols, which means, yes, we do the randomized control trials, but can we learn from other types of observational studies? Can we learn from case series? Can we learn from those who received

a medication for compassionate use? Can we learn from before and after studies? Can we learn from historical controls? So there are other ways of inferring scientific discovery, and one of the ones I'm a big believer in is the marriage between solid scientifically sound data and clinical wisdom. And it's amazing the ideas that live within clinical wisdom. I'm talking about the observation that penicillin saved the lives of many soldiers in World War I because the idea was that this fungus was known to kill bacteria on the augur plates and labs. And when soldiers took it, they otherwise would die of these severe infections and many of them were living. they didn't do a randomized controlled trial, yet they used many scientific inferences, and they used

clinical wisdom. I personally think medicine can learn a lot from wisdom in ways that we're not tapping. You know, at Johns Hopkins, there's a well-known case that's reviewed in the neurosurgery department of a patient that survived a stage four glioblastomy multiformate, which is basically an incurable brain cancer. And this patient is alive 15 plus years in a situation where really nobody survives more than five years. The average survival is less than two. Why did that guy become a long-term survivor? It turns out that guy had a bad infection after the tumor was resected. And they went back and cleared the infection from the surgical bed in the brain.

maybe something happened. Maybe the bacteria triggered the immune system. Maybe the bacteria sort of ate into the tentacles of the cancer and the surgical resection bed. I don't know, but we can probably learn from that case as much as we can learn from using an old-fashioned chemo in a randomized control trial on patients with brain cancer. We can learn as much from outliers

as we can learn from our traditional models of comparing groups. So I do think there are big implications of that for medicine today. For example, the role of food, the role of environmental exposures, can we use clinical wisdom to develop appropriateness practice patterns? And that is the work of our team at Hopkins in the appropriateness in medicine project. We go to doctors, we ask them, tell us about an area of low value care or something that's not effective. And can we study it looking at patterns of physicians in big data? For example,

in a physician's practice, what percent of their spine surgery patients underwent physical therapy at least once in the year preceding that elective spine surgery? It should be 80 percent of the patients. If it's not, that's a signal in the data to do a deeper review. We now have over a hundred ranging from things like C-section rates, crossing a certain threshold, to how often procedures are parsed into two separate days when they should be done in the same day, to whether or not a baby has a belly button hernia closed by a doctor routinely or selectively. These are all developed by clinical wisdom. This is the exciting area of identifying high value versus low-value care in the market now, studying the appropriateness of a practice pattern.

On the point of appropriateness, there was that really stark example in the book around the vascular screening screenings that are done in churches, essentially lead generation for stenting procedures. And, you know, it's funny that concepts that are sort of good general business principles in the rest of the world are those that we consider inathema and health care because they have either, you know, ethical implications that make us uncomfortable or they literally are tied to inappropriate clinical practice as you're describing. And, you know, I think that's one of the interesting concepts here is that everything that you're describing about the future where we need to head, transparency, you know, appropriate pricing, things like that,

is actually good general business practice and good general business principles. And yet we're trying to drive down so much of the unintended behaviors that come from the fee for service medicine engine that do drive people trying to optimize revenue and optimize volumes and things of that sort. I guess my sort of high level question is can we have it both ways, you appropriately balance those things while still giving individual providers and provider groups reasonable economic incentives that have a capitalist nature to them, but without violating the trust and the ethics surrounded with some of the practices that you described? There's really no perfect incentive when it comes to medical care to pay physicians

and hospitals. On one hand, people should be paid for their hard work and compensated. On the other hand, if it is a completely flat model, as we've seen, in some experiments, the incentives can go the other direction. And ultimately, we rely on the ethics and the sense of personal calling to the profession that lives within side each human being that practices medicine. So I'm not convinced that there is a perfect payment model for physicians. I think any model has perverse incentives. We do have two problems, however, today in medicine. Overtreatment and under treatment. But by far, the problem of overtreatment is a much more massive problem. And it is fueled by the extreme in the fee for service payment model system that we have

inherited. And it's gamed and it's maximized. Something that really makes me sick to my stomach being a surgeon is seeing somebody who had something that I know they didn't need. And we know that most doctors do the right thing or always try to. But if you have one physician out there doing a lot of unnecessary stuff, they may be siphoning a lot of money that should be going to be in liars. So the problem of outliers and practice patterns is a real problem. I don't know of the perfect solution for it, but I do love the idea of patients being educated and empowered and physicians at the primary care level steering to what we call high-value physicians. At the same time, we develop metrics to study appropriateness in data. Now, some areas of medicine, we simply

cannot measure a physician's practice pattern, you know, a really good child psychiatrist versus a poor child psychiatrist. We have certain surrogate measures. We can look at how often they prescribe medication combos that shouldn't be prescribed or how often they're prescribing an antipsychotic for someone with a mood disorder. Those are things we can identify in the data, but a lot of it is great bedside care. It's judgment, skill, listening. Those are the characteristics of a great provider.

And we rely on all of these things together to figure out how to redesign health care and create accountability. Yes, amen to that. I think a lot of physicians unfortunately end up in that situation where they feel the need to do things that if they were to rewind back to the days of when they were training to be a doctor are probably things that they would have also cringed at or felt sick over. But because we burdened

physicians with so much administrative overhead and poorly designed workflows and tools and things of that sort, I think the big call to arms is really how do we take that off their plate such that they can focus on the humane elements of what they can uniquely do, which you just described. And I think that's ultimately the Holy Grail coming out of all of this. One of the exciting things is that we found when we show physicians where they stand on the bell curve on the appropriateness measures, they auto-correct their behavior. And I think it's a certain competitiveness. You know, medical training has a certain competitive bias. And I think we don't like to be outliers when we're being measured by something that is endorsed by our peers. And so we are seeing this tremendous improvement in ways that we've never seen before. We just have to get off this old way of measuring things on it. is do, never do clinical pathway. We cannot rely on a master set of clinical pathways for every decision in medicine. Sixty percent of medical decisions are discretionary. And for many things,

we can identify patterns, but we can't micromanage the care at that level. Yeah. We invested for over a decade in implementing these electronic health records and these infrastructure systems to collect massive amounts of data. And what we have not been able to do yet is use that data and actually you put it in front of people on the front lines to do something with it. And that, to me, is really where we're at in the healthcare space is how do we now focus on the higher order problems now that we have that infrastructure layer in place and provide feedback such that we can actually optimize the way that we deliver care on the front lines using that data.

So with that, thank you so much for joining us on the A16D podcast.