Latest / Statecraft / How to Bring Down Healthcare Costs
Transcript
- 0:05SPEAKER_01: Hi, I'm Santi Ruiz, and you're listening to Statecraft.
- 0:08SPEAKER_01: Today, I'm joined by Anoop Malani.
- 0:11SPEAKER_01: He's a professor of law at the University of Chicago,
- 0:14SPEAKER_01: but he's currently on leave,
- 0:15serving as the first chief economist at the Centers for
- 0:18Medicare and Medicaid
- 0:19SPEAKER_01: Services.
- 0:21What that means is he oversees economic analysis for the agency
- 0:25that manages $2
- 0:25SPEAKER_01: trillion in annual health care spending.
- 0:29SPEAKER_01: That's 23%, more or less, of the entire federal budget.
- 0:34SPEAKER_01: The Centers for Medicare and Medicaid Services,
- 0:36SPEAKER_01: or CMS,
- 0:37SPEAKER_01: Medicare for 70 million elderly Americans,
- 0:40SPEAKER_01: runs Medicaid for low-income families,
- 0:43SPEAKER_01: and the health insurance exchanges where millions buy coverage.
- 0:46I wanted to talk to Anup about the transition from academia
- 0:49into the federal
- 0:50SPEAKER_01: government and to hear about what the day-to-day at CMS looks like.
- 0:54But I also wanted him to teach me a little bit about American
- 0:57health care policy.
- 0:58It's a field of policy I just know less about than some other
- 1:00topics that we cover here.
- 1:02SPEAKER_01: We talked about a couple maybe pretty basic questions.
- 1:04SPEAKER_01: For instance, the U.S.
- 1:06SPEAKER_01: spends like 20% of our GDP on health care.
- 1:08SPEAKER_01: Why is our life expectancy so bad compared to half of Europe?
- 1:12We also spent a lot of time talking about fraud,
- 1:15specifically Medicare fraud.
- 1:17How do you crack down on it without hurting patients who really
- 1:19need care?
- 1:21And how do you stop incentivizing private insurers to make
- 1:24patients look sicker
- 1:25SPEAKER_01: than they really are?
- 1:26SPEAKER_01: We also talked about some other stuff.
- 1:28SPEAKER_01: Skiing, committing fraud, or matching ACL injuries is a good time.
- 1:32SPEAKER_01: As always,
- 1:33you can find a carefully annotated transcript for this
- 1:36conversation and for many
- 1:37SPEAKER_01: others at www.statecraft.pub.
- 1:41If you like this episode,
- 1:42leave us a rating on your podcast platform of choice.
- 1:45SPEAKER_01: Okay, here we go.
- 1:47SPEAKER_01: Anup Malani, thank you for joining Statecraft.
- 1:49SPEAKER_01: Thank you for having me.
- 1:51SPEAKER_01: It's good to chat.
- 1:52SPEAKER_01: I realized in preparation for this conversation,
- 1:55SPEAKER_01: I think I owe you and my dad an apology because I'm a UChicago alum.
- 2:01SPEAKER_01: You taught.
- 2:03SPEAKER_01: there.
- 2:03SPEAKER_01: I mean, you teach there, you're on leave.
- 2:06SPEAKER_01: But I never took an econ class in undergrad at UChicago,
- 2:10SPEAKER_01: despite being at,
- 2:11SPEAKER_01: you know,
- 2:12SPEAKER_01: one of the famous economic institutions.
- 2:14So this is my public acknowledgement to my dad and to you,
- 2:17you know,
- 2:17actually taking your class.
- 2:19Well,
- 2:19can I share a fundamental criticism of an institution that I
- 2:22love that?
- 2:23SPEAKER_01: Yes.
- 2:23SPEAKER_01: Yes, that's what we do here.
- 2:25SPEAKER_00: Great.
- 2:25SPEAKER_00: And I now have two children going to University of Chicago.
- 2:28SPEAKER_00: Really?
- 2:29SPEAKER_00: Yeah.
- 2:30An 18 year old that's about to start in the fall and a 20 year
- 2:32old that's going to
- 2:33SPEAKER_00: start junior year.
- 2:35SPEAKER_00: And they love UChicago.
- 2:36SPEAKER_00: They love the culture.
- 2:37SPEAKER_00: They love the intensity.
- 2:38But I have one major criticism,
- 2:40which is you have to take a common set of classes.
- 2:43SPEAKER_00: sometimes called the common core or the core.
- 2:46And there are two things that are not included in that common
- 2:51core that are just shocking to me.
- 2:53SPEAKER_00: The first is you don't have to take an economic life.
- 2:55SPEAKER_00: You don't have to learn about markets.
- 2:58SPEAKER_00: The second thing is you don't have to learn about evolution.
- 3:01SPEAKER_00: You can take them optionally, but it's not required.
- 3:03SPEAKER_00: And to me, if you think about two mechanisms that govern life,
- 3:07that generally have a great deal of support,
- 3:09super important to understand.
- 3:10SPEAKER_00: It's those two.
- 3:11SPEAKER_00: And yet we don't teach that.
- 3:13SPEAKER_00: So.
- 3:14SPEAKER_01: Yeah, it is funny.
- 3:15SPEAKER_01: And I would have benefited from certainly the econ class.
- 3:19SPEAKER_01: I just always put it off.
- 3:20SPEAKER_01: I was like, yeah, no, I'll do that next year.
- 3:21SPEAKER_01: Fun courses this semester or this quarter, really.
- 3:25SPEAKER_01: I'll tackle it next year.
- 3:25SPEAKER_01: I just never got around to it.
- 3:26SPEAKER_01: So I think I've realized in practice,
- 3:28SPEAKER_01: a lot of what statecraft is,
- 3:30SPEAKER_01: is like me doing remedial econ education.
- 3:34SPEAKER_01: We had Dean Carlin on, who was chief economist at USAID.
- 3:38SPEAKER_01: You're the chief economist at CMS.
- 3:41SPEAKER_01: So in practice, I think it's working out.
- 3:43SPEAKER_01: I'm getting my education in markets, but it's taking a little while.
- 3:47SPEAKER_00: That's great.
- 3:48SPEAKER_00: Yeah.
- 3:48SPEAKER_00: And by the way,
- 3:49SPEAKER_00: there's an important intuition for economics,
- 3:51but I suspect that there is an overlapping but not completely
- 3:55overlapping intuition
- 3:56SPEAKER_00: for statecraft.
- 3:57SPEAKER_00: So for those of us that are consumers of your podcast,
- 4:00SPEAKER_00: it's not just you learning economics,
- 4:01SPEAKER_00: it's us learning about new intuitions from you.
- 4:03SPEAKER_00: So that's great.
- 4:05SPEAKER_01: I would hope so.
- 4:06SPEAKER_01: But it's good to hear from you.
- 4:07SPEAKER_01: But let me ask you.
- 4:08SPEAKER_01: So like I said,
- 4:08SPEAKER_01: we talked to Dean Carlin,
- 4:10SPEAKER_01: who was chief economist,
- 4:11SPEAKER_01: same title at a very different institution,
- 4:14SPEAKER_01: USAID,
- 4:14SPEAKER_01: doing global development.
- 4:16SPEAKER_01: You spent some time thinking about and researching global development.
- 4:19SPEAKER_01: But your role as chief economist at CMS is a very different role,
- 4:23SPEAKER_01: very different set of econ questions.
- 4:26SPEAKER_01: Let's start there.
- 4:26SPEAKER_00: What are you doing all day as chief economist?
- 4:29SPEAKER_00: Before I begin, let me explain.
- 4:30SPEAKER_00: This is the first time there's been a chief economist at CMS.
- 4:34So the very first thing I did is try to survey what other chief
- 4:38economists and
- 4:38SPEAKER_00: other institutions have done,
- 4:40SPEAKER_00: both in the government,
- 4:41SPEAKER_00: outside of government,
- 4:42to try to see what are the sorts of activities they engage in
- 4:44and how they can be
- 4:45SPEAKER_00: most useful to their organization.
- 4:48SPEAKER_00: Okay, having done that,
- 4:50SPEAKER_00: I break down my job into three parts.
- 4:52SPEAKER_00: The first part is to provide advice on prospective decisions.
- 4:56SPEAKER_00: Sometimes there's writing involved.
- 4:57SPEAKER_00: Sometimes there's just conversations involved.
- 4:59SPEAKER_00: Either case,
- 5:00there's some research that involves going to the literature or
- 5:03doing some back of
- 5:04SPEAKER_00: the envelope numbers to kind of advise on
- 5:07SPEAKER_00: immediate and pressing decisions.
- 5:09SPEAKER_00: Okay, so prospective decision making, that's item one.
- 5:12SPEAKER_00: Item two is projects.
- 5:13There are typically discrete projects where instead of just
- 5:17providing advice to the
- 5:18SPEAKER_00: decision maker,
- 5:19I'm making most of the decisions and maybe there's a kind of
- 5:22updating process that
- 5:24SPEAKER_00: goes on,
- 5:24SPEAKER_00: right?
- 5:24SPEAKER_00: I tell the administrator what it is that I'm doing,
- 5:27SPEAKER_00: just check in to see if he's got feedback,
- 5:30SPEAKER_00: but roughly I'm managing the project entirely by myself.
- 5:33SPEAKER_00: And often it involves
- 5:35improving economic decision making in different parts of the
- 5:38organization.
- 5:39SPEAKER_00: CMS, that's the second item.
- 5:41The third item is economic research on questions that are both
- 5:48retrospective and
- 5:49SPEAKER_00: prospective in the sense that,
- 5:51SPEAKER_00: for example,
- 5:52we may have done a regulation or there may be a statute and we
- 5:55need to figure out
- 5:56SPEAKER_00: what the impact of that is.
- 5:57SPEAKER_00: Or we want to just do some research to understand a topic that
- 6:03might have a decision forthcoming,
- 6:05but there's no concrete decision forthcoming.
- 6:06SPEAKER_00: So trying to understand how,
- 6:08SPEAKER_00: for example,
- 6:08SPEAKER_00: marketplaces work or how MA is functioning or things like that.
- 6:12SPEAKER_00: So that's the third.
- 6:13So the third item is general economic analysis,
- 6:16but not just for decisions.
- 6:18When you did your survey of chief economists and other kinds of
- 6:22institutions in the
- 6:23SPEAKER_01: government outside,
- 6:24SPEAKER_01: I'm curious,
- 6:25SPEAKER_01: what were your top line takeaways?
- 6:28SPEAKER_01: Before you went into the role, what did you think?
- 6:29Okay,
- 6:30it's going to be really valuable if I do X and I really need to
- 6:33avoid Y.
- 6:34So the main takeaway I got from talking to folks was that it's
- 6:38one of these jobs
- 6:39SPEAKER_00: where there's no clear definition.
- 6:41SPEAKER_00: So you have to make the job.
- 6:42And part of that is a conversation with the people that you're
- 6:45working with,
- 6:46SPEAKER_00: both the person that you report to,
- 6:47but also the people that you work in laterally with that are
- 6:50your colleagues.
- 6:51Try to understand what it is they do,
- 6:53how you can fit in,
- 6:54how you can be constructive.
- 6:55SPEAKER_00: You want to add value without being disruptive.
- 7:00SPEAKER_00: Okay, so that's the main takeaway that I got.
- 7:04The thing to try to do is to become a trusted advisor to the
- 7:07person that you report to.
- 7:10SPEAKER_00: They hire you because of your qualification,
- 7:12SPEAKER_00: but that doesn't necessarily mean that you have trust on day one.
- 7:15SPEAKER_00: Trust is something that you build up by doing a good job,
- 7:19SPEAKER_00: by getting the answers roughly right,
- 7:21SPEAKER_00: but you're building up that trust capital.
- 7:24And you really need to do that because if you want your work to
- 7:27have import,
- 7:28it's got to be that the decision makers who take it as an input
- 7:31actually value it,
- 7:32SPEAKER_00: right?
- 7:33SPEAKER_00: That it has leverage.
- 7:34SPEAKER_00: The thing to avoid is to impose your own preferences.
- 7:39SPEAKER_00: It's really important to understand that
- 7:42SPEAKER_00: You bring in a set of tools.
- 7:44SPEAKER_00: For me,
- 7:44it's economic analysis and understanding how healthcare markets
- 7:47work or how
- 7:47SPEAKER_00: healthcare financing programs work.
- 7:50SPEAKER_00: that doesn't give me the authority to displace people's preferences.
- 7:54They have different policy goals and I need to understand those
- 7:56goals and see how I
- 7:57SPEAKER_00: can kind of apply this analysis to those goals.
- 8:01SPEAKER_00: And part of that is also,
- 8:02SPEAKER_00: you know,
- 8:03SPEAKER_00: when you provide advice,
- 8:05if people don't follow your advice and they will sometimes not
- 8:08follow your advice,
- 8:09SPEAKER_00: you can't get angry about it because if you get angry about it,
- 8:12SPEAKER_00: they'll stop asking.
- 8:13SPEAKER_00: So the way to think about it is that you provide advice
- 8:17SPEAKER_00: input when they ask and you let them make the decision.
- 8:22And that's kind of the best you can do unless they give you the
- 8:25decision-making power.
- 8:27SPEAKER_01: Right.
- 8:28SPEAKER_01: I'm curious for you,
- 8:29how has it been to go from an environment in academia where
- 8:34there's decisions being
- 8:35SPEAKER_01: made all the time,
- 8:35SPEAKER_01: but they're not these kinds of as weighty political decisions,
- 8:40SPEAKER_01: decisions about reimbursement of healthcare insurance,
- 8:44SPEAKER_01: of
- 8:45SPEAKER_01: Who is covered under what?
- 8:47To go from an academic environment to an environment where
- 8:50major decisions are
- 8:51SPEAKER_01: being made and have to be made all the time under uncertainty.
- 8:55SPEAKER_01: How has that changed doing economic analysis?
- 8:59SPEAKER_00: Yeah, so you're exactly right.
- 9:01The main difference between academia and here is speed and the
- 9:05weight of decisions.
- 9:07SPEAKER_00: Almost every day, there's a decision that's important that's being made.
- 9:11SPEAKER_00: And in any given week or month,
- 9:13any decision that's being made is more important than almost
- 9:16any decision you'll
- 9:17SPEAKER_00: ever make in academia.
- 9:18SPEAKER_00: And of course,
- 9:19SPEAKER_00: I'm speaking from CMS's perspective,
- 9:20SPEAKER_00: where almost all items,
- 9:23SPEAKER_00: when you quantify them,
- 9:24SPEAKER_00: end with a B for billions and not hundreds or thousands or even millions.
- 9:29SPEAKER_00: Even when there are millions, there are usually three digits involved.
- 9:32SPEAKER_00: So I think that that's really what's going on.
- 9:35SPEAKER_00: Now, what does that mean for analysis?
- 9:38SPEAKER_00: When you're an academic, especially if you're an academic economist,
- 9:40SPEAKER_00: You do two things.
- 9:41You pick your questions and then you take your time to answer
- 9:44them to a certain
- 9:45SPEAKER_00: standard of quality.
- 9:47SPEAKER_00: Usually whatever the bar is for getting into a very good journal.
- 9:51SPEAKER_00: And in fact,
- 9:51you will discard questions when you don't think you can answer
- 9:55them to that level
- 9:55SPEAKER_00: of quality.
- 9:57That's not a privilege you have when you're working in a
- 9:59business or in the government.
- 10:01SPEAKER_00: There, you have to make the decision.
- 10:02SPEAKER_00: And given that,
- 10:03SPEAKER_00: your goal is to say,
- 10:04what is the best analysis I can do in the time period allotted
- 10:08to make that
- 10:09SPEAKER_00: decision?
- 10:09SPEAKER_00: But you have to do the analysis.
- 10:12So that means sometimes you're going to do back of the envelope
- 10:13calculations.
- 10:15SPEAKER_00: Sometimes you're just going to have to rely on theory.
- 10:17SPEAKER_00: And so the broader point that I would make is you spend a lot more time,
- 10:22SPEAKER_00: more than I anticipated,
- 10:23SPEAKER_00: thinking about theory and letting theory guide you.
- 10:26So basic econ 101 theory,
- 10:28to some extent,
- 10:28I would call it prize theory because I'm from Chicago.
- 10:31SPEAKER_00: And then you learn to do, you know, treat problems as family problems.
- 10:35SPEAKER_00: You just do back the envelope calculations as fast as you can.
- 10:38SPEAKER_00: You pray that the literature has already answered the question.
- 10:40SPEAKER_00: If not, you just got to,
- 10:42Get what numbers you can quickly and answer the question before
- 10:44the decision has to be made.
- 10:46SPEAKER_00: And real quick, define CRME problems.
- 10:49SPEAKER_00: I'm not familiar with that term.
- 10:50Fermi problems are problems where you don't have all the data
- 10:54that's available.
- 10:54SPEAKER_00: You just need to get an approximate solution.
- 10:57SPEAKER_00: And it's the art of coming up with solutions to questions.
- 11:00SPEAKER_01: Got it.
- 11:01SPEAKER_01: What do economists who study healthcare get wrong?
- 11:05Or what do they do that's unhelpful for decision makers in
- 11:09policymaking on healthcare?
- 11:11SPEAKER_01: Like now that you're on this side of the aisle,
- 11:14what do you see about the academic side of the profession that
- 11:17you think could be
- 11:18SPEAKER_01: improved?
- 11:20SPEAKER_00: Yeah,
- 11:20SPEAKER_00: I think you touched on a very important point,
- 11:22which is that there's a disconnect between health economics
- 11:26research and let's call
- 11:28SPEAKER_00: it American health economics research.
- 11:29SPEAKER_00: I don't want to comment on different countries,
- 11:31but at least the health economics research that I engaged in
- 11:34that my colleagues and
- 11:35SPEAKER_00: good friends engage in.
- 11:37SPEAKER_00: and the decisions that have to be made at CMS.
- 11:39And the main difference is not understanding the constraints
- 11:44that decision makers face.
- 11:46SPEAKER_00: And there's a myriad of them.
- 11:48SPEAKER_00: That's the first realization I had here.
- 11:50Beyond the complexity of the program,
- 11:52I realized that you can't just toggle switches on and off.
- 11:56SPEAKER_00: You're really constrained.
- 11:57SPEAKER_00: So let me give you a few examples.
- 12:00SPEAKER_00: When you want to change a regulation,
- 12:02there is obviously a whole legal process set up by the
- 12:06Administrative Procedures
- 12:07SPEAKER_00: Act,
- 12:07SPEAKER_00: notice and comment,
- 12:08SPEAKER_00: things like that.
- 12:10SPEAKER_00: But even outside that,
- 12:11SPEAKER_00: when you want to make a change,
- 12:13SPEAKER_00: you have internal clearances that you have to go through that add time.
- 12:17SPEAKER_00: On top of that,
- 12:17there's prudential concerns about making sure that you send the
- 12:20signal to the
- 12:21SPEAKER_00: private sector so that they can prepare.
- 12:24SPEAKER_00: You can't just pass a regulation and expect compliance right away.
- 12:27SPEAKER_00: There's a lot of work on the private side to be able to comply,
- 12:30SPEAKER_00: and you want to make sure that happens.
- 12:32SPEAKER_00: And so today,
- 12:33SPEAKER_00: if we want to make a decision,
- 12:35SPEAKER_00: and we made the decision today,
- 12:36SPEAKER_00: it might be two years,
- 12:3818 months to two years before you can actually see the
- 12:41effectuation of that
- 12:43SPEAKER_00: decision.
- 12:43SPEAKER_00: So you've got to make decisions way in advance.
- 12:46SPEAKER_00: And that slows down a lot of things.
- 12:48A second issue that really comes up is just there are political
- 12:52constraints.
- 12:53Now,
- 12:53this is not unique to this administration,
- 12:55but you can't just make great policy decisions.
- 12:58You have to think about the coalition that is going to support
- 13:03your decision making
- 13:04SPEAKER_00: and how your decision affects them because you exist in that coalition.
- 13:10SPEAKER_00: That's a second one.
- 13:11SPEAKER_00: A third one is just the resources for implementing things.
- 13:15SPEAKER_00: I think economists,
- 13:16SPEAKER_00: we just think,
- 13:17SPEAKER_00: oh,
- 13:17we'll change the premium schedule this way or provide this
- 13:20little mechanism to
- 13:21SPEAKER_00: provide incentive.
- 13:22SPEAKER_00: And we don't think about how do you implement that?
- 13:25SPEAKER_00: How many humans do you need?
- 13:26SPEAKER_00: What IT do you need?
- 13:27SPEAKER_00: What services do you need to have in place?
- 13:30What kind of education do you have to provide to the regulated
- 13:32party to understand
- 13:33SPEAKER_00: exactly how to comply?
- 13:35That process is complicated and itself either may not exist or
- 13:40takes time to construct.
- 13:41SPEAKER_00: And so I would say those are the three,
- 13:43SPEAKER_00: at least I can probably list more,
- 13:44but those are the top three categories of things that we have
- 13:47to think about.
- 13:48SPEAKER_00: So you don't have a blank slate on which to rewrite rules.
- 13:53SPEAKER_00: And I think that's the big constraint.
- 13:54And I imagine this affects regardless of which administration
- 13:58you work for.
- 13:59SPEAKER_00: When people come in and they say,
- 14:00SPEAKER_00: like,
- 14:00SPEAKER_00: okay,
- 14:00SPEAKER_00: oh,
- 14:01SPEAKER_00: you should just change things to meet this particular...
- 14:04SPEAKER_00: Oh,
- 14:04SPEAKER_00: my gosh,
- 14:05you have to understand those constraints and you have to
- 14:07understand when an
- 14:10agency's inability to take an action is due to constraints
- 14:13versus not being
- 14:15SPEAKER_00: proactive,
- 14:16SPEAKER_00: something like that.
- 14:17SPEAKER_01: Right.
- 14:18SPEAKER_01: That makes a lot of sense.
- 14:19And we spent a lot of time on this podcast talking about the
- 14:21challenges of implementation.
- 14:23SPEAKER_01: So I'm always glad to hear another guest ding the bell.
- 14:27SPEAKER_01: Things are hard to do in the real world.
- 14:29SPEAKER_00: So there are two things that flow from this.
- 14:32SPEAKER_00: OK, for academic economists, including health economists.
- 14:36The first thing that flows from this is thinking about the
- 14:39questions that CMS needs
- 14:41SPEAKER_00: answered or corporations need answered as opposed to what is economically
- 14:46SPEAKER_00: interesting in the abstract.
- 14:48OK,
- 14:48I think that would give research much more leverage,
- 14:52much more impact on the world.
- 14:53SPEAKER_00: It would increase the social impact of that research.
- 14:56SPEAKER_00: The second is,
- 14:57SPEAKER_00: and this is not confined to health economists,
- 14:59I think a lot more should focus on what you would call
- 15:02statecraft or I would call
- 15:04SPEAKER_00: implementation.
- 15:05SPEAKER_00: We need to think about the economics of implementation.
- 15:07SPEAKER_00: What is the value added by the APA?
- 15:10How do we think about decision-making in the context of
- 15:12political coalition?
- 15:14How do we think about what Eddie Lazier would call personnel
- 15:16economics in the
- 15:18SPEAKER_00: context of policy decision-making?
- 15:19SPEAKER_00: I think there's a lot of work to be done there.
- 15:22And one of the goals,
- 15:23I don't know if I'll achieve it,
- 15:24one of the goals is to try to bring
- 15:26SPEAKER_00: health economists through CMS so they can see that firsthand.
- 15:30And then my hope is that will then inform their research when
- 15:34they are back in academia.
- 15:36SPEAKER_01: that it seems like it would be very valuable.
- 15:39Let me zoom out and ask you a very basic question that I think
- 15:43will be familiar to
- 15:43SPEAKER_01: a lot of listeners.
- 15:44SPEAKER_01: The US spends a lot of money on healthcare,
- 15:47SPEAKER_01: quite a bit of money compared to many other developed countries.
- 15:51And we have worse life expectancy outcomes than that big bucket
- 15:55of other developed countries.
- 15:56SPEAKER_01: They spend less money.
- 15:57SPEAKER_01: Now, some of their systems are socialized, some of them are not.
- 16:01As far as I understand,
- 16:01it's not one obvious thing that they're all doing that we're
- 16:04not.
- 16:05SPEAKER_01: that would lead to those discrepancies in life expectancy.
- 16:09So at a very high level,
- 16:10what's going on here that we spend more and die sooner?
- 16:15I'm going to put my academic hat on for a second before I talk
- 16:17as a chief economist at CMS.
- 16:20SPEAKER_00: I think I want to be a little bit careful about making those assertions.
- 16:23Although I probably share your view that we could be much more
- 16:25efficient about the delivery.
- 16:28That is to see,
- 16:29do much better by both consumer and taxpayer in the delivery of
- 16:33healthcare.
- 16:33SPEAKER_00: So let's revisit some of those things.
- 16:34SPEAKER_00: So the United States roughly spends, you know, 20% of GDP on healthcare.
- 16:39SPEAKER_00: And I think for a long time, we used to worry about that 20%.
- 16:43SPEAKER_00: The exact number is going to be something like 18 or 19 percent.
- 16:45So I'm just going to round numbers,
- 16:47make it a little bit easier for listeners.
- 16:49SPEAKER_00: So they spend 20 percent.
- 16:50SPEAKER_00: But,
- 16:50SPEAKER_00: you know,
- 16:50an interesting puzzle for the last 10 years is that we've been
- 16:53stuck at 20 percent.
- 16:55And that's really remarkable because there are all these
- 16:57warnings back in the 1990s.
- 16:59SPEAKER_00: We're increasing as a percentage of GDP,
- 17:02SPEAKER_00: our spending on health care,
- 17:03SPEAKER_00: and it's going to become unsustainable.
- 17:05SPEAKER_00: But then for a decade, we plateaued.
- 17:08SPEAKER_00: And we're still trying to figure out why it is that we plateaued.
- 17:11SPEAKER_00: And so I want to be a little bit careful about just that assertion.
- 17:15SPEAKER_00: But on top of that,
- 17:16it's important to remember that there is no natural percentage
- 17:19of GDP that you'd
- 17:20SPEAKER_00: spend on health care from a purely economic perspective.
- 17:23SPEAKER_00: It just depends on the degree to which health care is a normal good.
- 17:27SPEAKER_00: And so if we think that it's a normal good,
- 17:29SPEAKER_00: and by normal good,
- 17:30SPEAKER_00: I mean that as you earn more,
- 17:32SPEAKER_00: you spend a higher percentage of your income on that.
- 17:35SPEAKER_00: And that's a normal good, or you spend more on that good.
- 17:38SPEAKER_00: Now,
- 17:38SPEAKER_00: healthcare just might be a good where you,
- 17:42SPEAKER_00: as you increase income,
- 17:43SPEAKER_00: you just spend a higher percentage.
- 17:44It could be more valuable to higher income individuals for a
- 17:47variety of reasons.
- 17:48So I don't think that there should be a statement that this is
- 17:52too much,
- 17:53okay?
- 17:53SPEAKER_00: Yeah.
- 17:54SPEAKER_01: Well, let me let me amend because that's a useful correction.
- 17:56SPEAKER_01: But acknowledging that we are a lot wealthier than,
- 18:00SPEAKER_01: you know,
- 18:00SPEAKER_01: we're wealthier than the UK,
- 18:02SPEAKER_01: wealthier than South Korea.
- 18:03SPEAKER_01: These countries spend less on health care.
- 18:06SPEAKER_01: Right.
- 18:06This makes sense to me as a layman that we spend more on health
- 18:10care because we can.
- 18:10And my marginal dollar maybe should go to additional health
- 18:14care spending.
- 18:15SPEAKER_01: But map that with the life expectancy point,
- 18:18SPEAKER_01: that we die a lot sooner,
- 18:19like several years sooner than the South Koreans or the
- 18:22Italians or the Swiss or
- 18:24SPEAKER_01: the Germans.
- 18:25SPEAKER_00: Yeah.
- 18:26SPEAKER_00: So first,
- 18:27SPEAKER_00: let me caveat what I said before,
- 18:28just because there's no natural answer to whether or not you
- 18:31should spend 20
- 18:32percent of your income on health care doesn't mean that that 20
- 18:37percent isn't too
- 18:38SPEAKER_00: much.
- 18:39It just means that just saying 20 percent is a lot doesn't
- 18:43prove anything.
- 18:43You have to give an additional reason why it might be too much
- 18:45or even too little.
- 18:47SPEAKER_00: Sure.
- 18:48SPEAKER_00: Your second question is, why do we get so little?
- 18:52SPEAKER_00: And I would argue that that too is too quick a statement.
- 18:55Not that it's wrong,
- 18:56but it's too quick a statement when put that succinctly.
- 18:59SPEAKER_00: And the reason why is twofold.
- 19:01The first is you have to understand the inputs into health in
- 19:06the United States
- 19:07SPEAKER_00: versus other places.
- 19:08SPEAKER_00: And if the inputs into health,
- 19:10meaning the underlying health human cap of the population is
- 19:14lower to start with,
- 19:16SPEAKER_00: then it might not be that we're getting a bad result.
- 19:19SPEAKER_00: It could just be that we're doing pretty good with a worse base.
- 19:23SPEAKER_00: And if you don't mind a quick detour,
- 19:24I'm going to talk about a really nice paper that I think kind
- 19:28of addresses this.
- 19:29SPEAKER_00: Sure.
- 19:29To me,
- 19:31this is my usual response when I talk to folks when they say,
- 19:34hey,
- 19:34why does the U.S.
- 19:34SPEAKER_00: do so poorly on longevity measures?
- 19:37SPEAKER_00: So this is a paper by Alice Chen, Heidi Williams, and Emily Oster.
- 19:41SPEAKER_00: My colleague, Heidi Williams.
- 19:43SPEAKER_00: Yeah, they're fantastic.
- 19:45SPEAKER_00: And this is one of my favorite papers.
- 19:46SPEAKER_00: It's actually kind of a descriptive paper, but super insightful.
- 19:50So what they look at,
- 19:51I'm going to try to be as precise as possible,
- 19:53but succinct.
- 19:54SPEAKER_00: They look at why one year child mortality is higher in the U.S.
- 19:59SPEAKER_00: than in Europe.
- 20:00SPEAKER_00: And it is higher.
- 20:00SPEAKER_00: It's higher in the U.S.
- 20:02SPEAKER_00: than it is in Europe.
- 20:03SPEAKER_00: And when they do their breakdown,
- 20:04a really important result that comes out is that the gaps that
- 20:08emerge in mortality
- 20:09SPEAKER_00: rates don't emerge in the first month.
- 20:12SPEAKER_00: They emerge in months 2 through 12.
- 20:15SPEAKER_00: Why is that important?
- 20:17SPEAKER_00: If you think about when the health care system,
- 20:19SPEAKER_00: where we're spending our money,
- 20:20SPEAKER_00: when you think about where it's going to have the biggest impact,
- 20:22SPEAKER_00: you think the first month.
- 20:24SPEAKER_00: And the first month, you don't see the differences.
- 20:26SPEAKER_00: The differences show up when the health care system is not operating.
- 20:29SPEAKER_00: It's in months 2 through 12.
- 20:30SPEAKER_00: It's what's going on in the community.
- 20:32SPEAKER_00: back at home.
- 20:33SPEAKER_00: And what that tells you is,
- 20:35SPEAKER_00: again,
- 20:35SPEAKER_00: we may be starting with different baselines,
- 20:38SPEAKER_00: that our population might not have the same longevity in the abstract,
- 20:43SPEAKER_00: or they may engage in other behaviors that lead to outcomes being worse,
- 20:47SPEAKER_00: longevity being worse.
- 20:48SPEAKER_00: And so I think we want to take that into account.
- 20:51SPEAKER_00: That's a great paper.
- 20:52SPEAKER_00: You should read the paper.
- 20:53SPEAKER_00: Also, all three are great researchers.
- 20:55SPEAKER_00: Always fun to read.
- 20:56But that's an important takeaway is that health is not the same
- 20:59thing as health care.
- 21:01SPEAKER_00: And when we want to figure out if our health care is doing a bad job,
- 21:03we need to actually look at places where health care is having
- 21:05an impact and not
- 21:06SPEAKER_00: necessarily where just the community or families or activity differs.
- 21:10SPEAKER_00: Sure.
- 21:10SPEAKER_00: The second point I would make that's really important here is
- 21:14SPEAKER_00: I value health.
- 21:15SPEAKER_00: I'm shocked.
- 21:16SPEAKER_00: There's a selection on what you do in life.
- 21:19SPEAKER_00: But I understand that people make health health tradeoffs.
- 21:21SPEAKER_00: So, you know, for a lot of my life, I skied.
- 21:25SPEAKER_00: And skiing is a it's not just a health health tradeoff.
- 21:28SPEAKER_00: It's a health other happiness tradeoff.
- 21:30SPEAKER_00: And we often trade our health for other things.
- 21:34SPEAKER_00: And we need to account for that.
- 21:36SPEAKER_00: We can't just tell people you should always prioritize health.
- 21:38SPEAKER_00: That kind of lecture doesn't always work.
- 21:41SPEAKER_00: We need to sometimes take people where they're standing.
- 21:44SPEAKER_01: Well,
- 21:44I don't want to pull you away from your core expertise and make
- 21:47you psychoanalyze
- 21:49SPEAKER_01: the American public.
- 21:50SPEAKER_01: But I guess I am curious, what's your reading then?
- 21:53SPEAKER_01: If healthcare costs are not the only determinant of why,
- 21:59SPEAKER_01: you know,
- 22:00just could we spend more on healthcare doesn't necessarily mean
- 22:02we should expect to
- 22:02SPEAKER_01: live longer.
- 22:04SPEAKER_01: Because we might start from a lower base,
- 22:06but what's your understanding of why Americans are starting
- 22:09from that lower base?
- 22:10Are we more risk tolerant,
- 22:12less risk averse than our friends in Europe and East Asia?
- 22:16SPEAKER_01: We're skiing too much?
- 22:18SPEAKER_00: No, you know, an earlier version of myself would say never skiing enough.
- 22:23SPEAKER_01: I don't think I've hit that limit yet.
- 22:24SPEAKER_00: That's a hard question.
- 22:26SPEAKER_00: And honestly, I don't know the answer.
- 22:27SPEAKER_00: But we do have different populations.
- 22:29SPEAKER_00: We have a different population mix.
- 22:31SPEAKER_00: We have populations that are substantially transplanted.
- 22:35SPEAKER_00: We are also geographically meaningfully different.
- 22:38SPEAKER_00: We're much more spread out.
- 22:39SPEAKER_00: We drive a lot more.
- 22:41SPEAKER_00: Our production technologies are a little bit different.
- 22:44SPEAKER_00: than they are in Europe.
- 22:46SPEAKER_00: And I think all those things are inputs into our morbidity and mortality.
- 22:51SPEAKER_00: Doesn't necessarily make us worse off or better off.
- 22:54SPEAKER_00: To get to those, you need to ask things like, what is your income?
- 22:58SPEAKER_00: What is your consumption?
- 22:59SPEAKER_00: I like consumption better than income.
- 23:00SPEAKER_00: You ask about happiness,
- 23:02SPEAKER_00: things like that,
- 23:02SPEAKER_00: whatever your flavor is,
- 23:03SPEAKER_00: whatever preferred methodology you have.
- 23:06SPEAKER_00: But I don't think I can understand that.
- 23:07SPEAKER_00: But I would say the following.
- 23:09SPEAKER_00: My input into health is not just medication.
- 23:12SPEAKER_00: It's a whole range of things.
- 23:13And I just want to point out that this is what the secretary of
- 23:16HHS is talking
- 23:17SPEAKER_00: about when he talks about make America healthy again.
- 23:19It's just that there are other inputs other than medicine and
- 23:22hospital care that
- 23:24SPEAKER_00: affect your health.
- 23:24SPEAKER_00: And we have to think about that as well.
- 23:27SPEAKER_00: And I think that's probably true.
- 23:28SPEAKER_00: Now, can I define it?
- 23:30I'm not an expert in understanding how lifestyle specifically
- 23:34maps to health.
- 23:36SPEAKER_00: That's not an area of research.
- 23:37SPEAKER_00: There are people that do a great deal of research there,
- 23:39but I don't think anybody denies that there is an important
- 23:42relationship there.
- 23:43SPEAKER_00: Of course.
- 23:44SPEAKER_00: Well,
- 23:44SPEAKER_01: let me get back to an area of your core expertise,
- 23:47SPEAKER_01: which is the econ problems of our healthcare delivery system.
- 23:53SPEAKER_01: You and I have chatted before this call.
- 23:55And one of the things you brought up in those conversations was
- 23:58the fiscal
- 23:59constraints on the American health care system that exists just
- 24:03because of basic
- 24:04SPEAKER_01: demographic change.
- 24:06SPEAKER_01: I found that really informative when we chatted.
- 24:08SPEAKER_01: I would love you to basically explain that again to me.
- 24:11What is the constraint broadly demographically that the
- 24:14American health care system faces?
- 24:16SPEAKER_00: Yeah,
- 24:16SPEAKER_00: so I think most simply stated is that because our fertility rate is below
- 24:20SPEAKER_00: replacement,
- 24:21SPEAKER_00: it's currently at 1.6.
- 24:22SPEAKER_00: Immigration helps sustain the population,
- 24:26but even immigrant fertility rates are declining and converging
- 24:30to the kind of long
- 24:32SPEAKER_00: residing Americans fertility rate.
- 24:34SPEAKER_00: You know, you get various estimates of when U.S.
- 24:37SPEAKER_00: population will peak.
- 24:39If you ask the UN,
- 24:40it's going to be around when the world population peaks in
- 24:422080.
- 24:42But if you ask the CPO,
- 24:43you might get numbers that are as early as the 2030s.
- 24:47I tend to,
- 24:47just as a sidebar,
- 24:48I tend to believe the more pessimistic numbers.
- 24:51SPEAKER_01: We might do a statecraft on this in the future.
- 24:53SPEAKER_01: You might know, what is his name?
- 24:54SPEAKER_01: Jesus Fernandez Gilverde,
- 24:56SPEAKER_01: the demographer who thinks that,
- 24:58SPEAKER_01: you know,
- 24:59SPEAKER_01: he's made a very cogent argument,
- 25:00SPEAKER_01: the UN numbers.
- 25:01SPEAKER_01: The UN projections are way too optimistic.
- 25:03SPEAKER_01: And I think I believe that.
- 25:05SPEAKER_00: First, I agree with you.
- 25:07SPEAKER_00: Second, I'm very excited to listen to that podcast.
- 25:09SPEAKER_00: I think Jesus is fantastic.
- 25:11You know,
- 25:12I don't have all the same views,
- 25:13but are substantially aligned.
- 25:15SPEAKER_00: I have some slightly heterodox views on where we're headed.
- 25:18Let me put a pin in that and we'll have that conversation
- 25:20offline,
- 25:20but I would love to hear it.
- 25:22SPEAKER_00: Yeah, that'd be great.
- 25:23SPEAKER_00: I agree with you that the pessimistic numbers are probably correct,
- 25:26SPEAKER_00: but let's think about how that cashes out.
- 25:28SPEAKER_00: So what that means that our population, if your fertility rate is low,
- 25:32SPEAKER_00: That means your population begins to age,
- 25:34SPEAKER_00: and our population has already begun to age,
- 25:35SPEAKER_00: but aging accelerates.
- 25:37SPEAKER_00: And that has two impacts.
- 25:38SPEAKER_00: The first impact is,
- 25:40SPEAKER_00: especially when you have a program like Medicare,
- 25:43you are going to have a greater demand for federal spending on
- 25:48health care.
- 25:49SPEAKER_00: Health care for the elderly, because there's more elderly.
- 25:51SPEAKER_00: There are more elderly.
- 25:53SPEAKER_00: But the other problem is that at the same time,
- 25:55SPEAKER_00: there's greater demand for federal health care spending.
- 25:58There's also less supply of that health care spending because
- 26:01as your population
- 26:02SPEAKER_00: ages,
- 26:02your working population declines and it's your working
- 26:05population that largely
- 26:06SPEAKER_00: supplies tax revenue across a range of sources.
- 26:10SPEAKER_00: And so that's our basic problem is an affordability problem.
- 26:13SPEAKER_00: And the reason why this is this places general attention on this issue,
- 26:17SPEAKER_00: even if we had stable finances,
- 26:20You would still face an issue between medical spending today
- 26:24and medical spending
- 26:25SPEAKER_00: in the future.
- 26:26SPEAKER_00: But now with demographic change, it's even more acute.
- 26:29SPEAKER_00: And the way that you respond to that,
- 26:32SPEAKER_00: you know,
- 26:33SPEAKER_00: one approach,
- 26:33which I think is not the approach that is something the
- 26:38American voter wants,
- 26:39SPEAKER_00: is just,
- 26:39SPEAKER_00: hey,
- 26:40SPEAKER_00: let's cut spending.
- 26:41SPEAKER_00: That's one approach, cut benefit.
- 26:42But a second approach is to say,
- 26:43hey,
- 26:44before we think about any of that,
- 26:46can we make
- 26:47SPEAKER_00: American health care more productive for each dollar spent?
- 26:51SPEAKER_00: Can we give patients more value for every dollar that we spend?
- 26:56SPEAKER_00: And if we can do that,
- 26:58we can make this program a little bit more sustainable despite
- 27:01that pressure.
- 27:02SPEAKER_01: Well, let's talk about that.
- 27:03SPEAKER_01: And I should just flag again for listeners.
- 27:07SPEAKER_01: CMS is,
- 27:08SPEAKER_01: I mean,
- 27:10SPEAKER_01: how to put this,
- 27:11it's like one of the huge cost centers in the federal
- 27:15government for good reason.
- 27:16It's the hub for most of,
- 27:19not all,
- 27:19but most of the government's spending on healthcare.
- 27:22SPEAKER_01: So almost 70 million people are enrolled in Medicare,
- 27:26SPEAKER_01: about a trillion dollars a year,
- 27:28SPEAKER_01: give or take,
- 27:29SPEAKER_01: are spent in that program.
- 27:31SPEAKER_01: So that's like 14, 15% of total federal spending from last year.
- 27:35My understanding is like a little bit smaller percentage is
- 27:38spent on Medicaid per
- 27:39SPEAKER_01: year,
- 27:39SPEAKER_01: but taken together,
- 27:40SPEAKER_01: you've got like a third,
- 27:41SPEAKER_01: roughly the federal spend runs in one way or another through CMS.
- 27:46SPEAKER_01: Is that, do I have that ballpark right before we move on?
- 27:49SPEAKER_00: Yeah,
- 27:50SPEAKER_00: I would say about 20,
- 27:51SPEAKER_00: 23 percent of federal spend goes through Medicare,
- 27:54SPEAKER_00: Medicaid or the exchanges.
- 27:56SPEAKER_00: It is probably the if you combine the three.
- 27:59SPEAKER_00: So Medicare for the elderly, Medicaid for the poor.
- 28:03SPEAKER_00: I'm simplifying.
- 28:03SPEAKER_00: There's obviously disabled populations and crossover with duals,
- 28:06SPEAKER_00: but Medicare for the elderly,
- 28:08SPEAKER_00: Medicaid for the poor and exchanges for the near poor.
- 28:11If you combine those three programs,
- 28:13you're talking about $2 trillion worth of spending.
- 28:16That's the biggest single ticket item that the federal
- 28:19government spends on.
- 28:20SPEAKER_00: And that doesn't even include the $300 to $500 that the states spend.
- 28:25SPEAKER_00: And it doesn't even include
- 28:26SPEAKER_00: employer-sponsored insurance where we don't directly fund it,
- 28:29but we provide tax expenditures by giving employers a tax break
- 28:33for providing that
- 28:34SPEAKER_00: benefit and workers a tax break for providing that service.
- 28:37And we don't control that,
- 28:38but health is super important and CMS is the largest
- 28:41expenditure.
- 28:42SPEAKER_00: Yeah.
- 28:42SPEAKER_00: So critical entity for understanding total federal spend.
- 28:47SPEAKER_00: Right.
- 28:47SPEAKER_01: And by that same token,
- 28:50SPEAKER_01: like you said,
- 28:51SPEAKER_01: if you could improve the productivity of those marginal dollars,
- 28:54you get huge gains because there's just a huge amount of
- 28:57marginal dollars there.
- 28:58One thing that we talked a lot about on this podcast over the
- 29:02last six or eight
- 29:03SPEAKER_01: months has been Doge and the Doge push to cut waste,
- 29:07SPEAKER_01: fraud,
- 29:07SPEAKER_01: and abuse.
- 29:07SPEAKER_01: And I think
- 29:08There was a lot of criticism of the public messaging of Doge in
- 29:12the early part of
- 29:13this year for spending a lot of time on things that were not
- 29:16finding ways to make
- 29:17SPEAKER_01: healthcare spending more productive and less wasteful.
- 29:21SPEAKER_01: Whatever listeners or whatever you think of the Doge project itself,
- 29:26SPEAKER_01: we don't have to get into that,
- 29:27SPEAKER_01: but I think it's obvious.
- 29:29The work that you guys are doing at CMS has a lot more to do
- 29:33with improving the
- 29:35productivity and efficiency of the government than almost
- 29:39anything else that one
- 29:40SPEAKER_01: could be working on to reform in the federal government.
- 29:42I'm not going to make you respond to that,
- 29:44but I just want to put a pin there.
- 29:46SPEAKER_01: What levers does CMS have to tackle this problem?
- 29:50SPEAKER_01: Just thinking institutionally,
- 29:52what tools could CMS have to try and improve the productivity
- 29:56of American health
- 29:57SPEAKER_01: care?
- 29:58SPEAKER_00: Okay, that is a very complicated question.
- 30:01SPEAKER_00: And I think it's helpful to break it down by type of program.
- 30:07And really,
- 30:07you need to think about three programs,
- 30:09Medicaid,
- 30:10Medicare and exchanges.
- 30:12SPEAKER_00: Sure.
- 30:12SPEAKER_00: And each of them have their own distinct challenges.
- 30:15SPEAKER_00: And then there's going to be some challenges that are common across all.
- 30:19SPEAKER_00: So if we think about Medicaid, one of the problems with Medicaid is
- 30:25SPEAKER_00: is that it's a program with peculiar incentives.
- 30:29SPEAKER_00: The federal government pays for a majority of Medicaid,
- 30:34SPEAKER_00: but states control how Medicaid is operated.
- 30:40And one implication of that is that states don't fully
- 30:44internalize the costs of their decisions.
- 30:47It's like going to a dinner and your friend tells you they're
- 30:50going to pay 60% of the check.
- 30:52SPEAKER_00: You might order a little bit more than you need to.
- 30:54SPEAKER_00: Right.
- 30:55Or you might not be careful in making sure that you're getting
- 30:58the best deal when
- 30:58SPEAKER_00: you're choosing a restaurant.
- 31:00So I think some of those issues arise just by the structural
- 31:05features of Medicaid.
- 31:06SPEAKER_00: And Medicaid's an important program,
- 31:08SPEAKER_00: don't get me wrong,
- 31:09but I think we could do a little bit better in making sure that
- 31:12there aren't
- 31:13efforts to either increase the federal share or to engage in
- 31:17inefficient payments
- 31:18SPEAKER_00: like state-directed payments.
- 31:20So that's one way,
- 31:21and I think that has to do with the structural program.
- 31:23SPEAKER_00: I'm going to turn to Medicare, and then I'll do exchanges.
- 31:25SPEAKER_00: Because Medicare and Medicaid are the big ticket items,
- 31:28SPEAKER_00: close to a trillion dollars for spending each year on each of those.
- 31:32So in Medicare,
- 31:33the real challenge is that Medicare is actually two programs
- 31:37that are combined.
- 31:38There's something called traditional Medicare,
- 31:39where the government is the insurer.
- 31:41SPEAKER_00: And they cover hospital care and they cover physician care.
- 31:44SPEAKER_00: And after 2006 drugs, you get through Medicare Part D plans.
- 31:48SPEAKER_00: But still, that's traditional Medicare.
- 31:50SPEAKER_00: And then there's something called Medicare Advantage,
- 31:52SPEAKER_00: which,
- 31:52SPEAKER_00: you know,
- 31:53SPEAKER_00: if you're old like I am,
- 31:54SPEAKER_00: you'd call that Part C.
- 31:55SPEAKER_00: But the young kids call it Medicare Advantage.
- 31:58SPEAKER_00: And this is where the government is not the insurer.
- 32:00SPEAKER_00: Instead,
- 32:00SPEAKER_00: the government plays premiums to insurers,
- 32:03SPEAKER_00: private insurers,
- 32:04SPEAKER_00: who then provide,
- 32:05SPEAKER_00: you know,
- 32:06SPEAKER_00: hospital and physician and drug benefits,
- 32:08SPEAKER_00: among other benefits.
- 32:09SPEAKER_00: Okay.
- 32:10SPEAKER_00: So it's really two programs.
- 32:11SPEAKER_00: It's a public insurance program and a,
- 32:13SPEAKER_00: you know,
- 32:14SPEAKER_00: the government provides premiums for private insurance.
- 32:17SPEAKER_00: And people can choose which of those to go into.
- 32:19SPEAKER_00: And they're run differently.
- 32:21SPEAKER_00: Meaning in traditional Medicare,
- 32:24SPEAKER_00: where the government is the insurer,
- 32:25SPEAKER_00: we just kind of set a whole range of prices and it's fee for service.
- 32:29SPEAKER_00: So providers just decide what to provide.
- 32:32And the history of traditional Medicare has been the government
- 32:34realizing that
- 32:35SPEAKER_00: leads to too much expenditure.
- 32:37SPEAKER_00: And so we try to pull back
- 32:39So we go from cost plus pricing to prospective pay pricing to
- 32:45more recently you
- 32:46SPEAKER_00: have under the Affordable Care Act,
- 32:48SPEAKER_00: you have something called accountable care organizations.
- 32:50SPEAKER_00: And the first one was we just pay whatever your costs are plus.
- 32:54SPEAKER_00: Then we say, oh, my gosh, that's too much.
- 32:56SPEAKER_00: Are you charging too much?
- 32:57So we're going to pay per sickness,
- 33:00roughly,
- 33:00and we'll pay you a certain amount for sickness.
- 33:01SPEAKER_00: And then you've got to manage.
- 33:03SPEAKER_00: And that was still not controlling costs as well as we'd like.
- 33:06SPEAKER_00: So then we switched to something called accountable care organizations.
- 33:09And there we said the federal government is going to shift the
- 33:12risk associated with
- 33:13SPEAKER_00: extra costs from the government to providers.
- 33:17SPEAKER_00: OK, kind of like capitation, but not exactly.
- 33:20SPEAKER_00: If you remember back in the 1990s,
- 33:22SPEAKER_00: for those listeners that follow health insurance,
- 33:25SPEAKER_00: that'll make sense.
- 33:26SPEAKER_00: That'll have resonance.
- 33:27SPEAKER_00: So that's that's the issue there.
- 33:28The government is insurers trying to get the providers to
- 33:31behave efficiently in there.
- 33:34SPEAKER_00: You know,
- 33:34you could imagine that the biggest gains that we could engage
- 33:37in are expanding
- 33:38accountable care organizations and trying to get providers to
- 33:42be a little bit more
- 33:43SPEAKER_00: careful with their spending.
- 33:44SPEAKER_00: And there, that's the margin for improving cost or efficiency.
- 33:49I would say providing better care for customers without
- 33:52overcharging the government
- 33:53or without charging government for things that is not necessary
- 33:56or not a high
- 33:58SPEAKER_00: return.
- 33:58SPEAKER_00: But doing it in a way that empowers the physicians, right?
- 34:01Like you also have to think about making sure that people
- 34:02participate in the program.
- 34:03SPEAKER_00: But that's the balance that we're trying to do.
- 34:05SPEAKER_00: Okay.
- 34:07SPEAKER_00: Now,
- 34:07SPEAKER_00: Medicare Advantage,
- 34:08SPEAKER_00: which is the private plan thing,
- 34:09SPEAKER_00: there,
- 34:10SPEAKER_00: we're not dealing with the providers directly.
- 34:11SPEAKER_00: We're dealing with the plans, private insurance companies.
- 34:14SPEAKER_00: And we have to incentivize them both to take on patients that are costly,
- 34:19SPEAKER_00: but then to treat them well.
- 34:22And there we want to make sure that we're getting a good
- 34:24assessment of how costly
- 34:26SPEAKER_00: different patients are to make sure beneficiaries,
- 34:28SPEAKER_00: I should say,
- 34:29SPEAKER_00: to make sure that the insurance companies take them on.
- 34:31SPEAKER_00: And there we're doing risk adjustment with the insurance.
- 34:33SPEAKER_00: We're setting insurance prices and determining risk,
- 34:36SPEAKER_00: which is different than what we're doing with the providers.
- 34:38So we can do better with risk adjustment in that regard and to
- 34:41keep down costs again.
- 34:44SPEAKER_00: without harming patients.
- 34:45SPEAKER_00: In fact,
- 34:45maybe even making the patients better off because we try to
- 34:47make those insurance
- 34:48SPEAKER_00: companies a little better aligned with what the patients actually want,
- 34:51SPEAKER_00: what the beneficiaries actually want.
- 34:54SPEAKER_00: Exchanges look a lot like Medicare Advantage.
- 34:57SPEAKER_00: We're paying a premium support.
- 34:59SPEAKER_00: It's slightly different than in Medicare Advantage,
- 35:01but we're paying for people to get insurance through a private
- 35:04insurance company.
- 35:05And again,
- 35:06we want the insurance company to take on even high risk
- 35:08patients.
- 35:09SPEAKER_00: And so we have to think about exactly how we do risk adjustment.
- 35:12SPEAKER_00: OK, right.
- 35:13SPEAKER_00: So those are the big issues.
- 35:14But then you can step back and say,
- 35:15OK,
- 35:16but is there anything that's common across all of these?
- 35:19SPEAKER_00: Now, one thing that's common across all these is fraud, waste and abuse.
- 35:23SPEAKER_00: Let's just look at the fraud component of that.
- 35:24SPEAKER_00: That's the easiest.
- 35:25SPEAKER_00: When you charge for something that you shouldn't be charged for legally,
- 35:29SPEAKER_00: that can happen in all these programs.
- 35:30SPEAKER_00: And we just need to be a little bit more proactive about it.
- 35:34SPEAKER_00: You know,
- 35:35SPEAKER_00: traditionally,
- 35:35SPEAKER_00: I apologize for the long answer,
- 35:36SPEAKER_00: but we've been thinking about this retrospectively.
- 35:39SPEAKER_00: Usually what happens is,
- 35:40SPEAKER_00: you know,
- 35:41SPEAKER_00: Dr.
- 35:41SPEAKER_00: Smith,
- 35:42SPEAKER_00: not to pick on Dr.
- 35:43SPEAKER_00: Smith in particular,
- 35:43SPEAKER_00: but Dr.
- 35:44SPEAKER_00: Smith.
- 35:44SPEAKER_00: Sorry, Dr. Smith.
- 35:45SPEAKER_00: Bills for something that he or she didn't do.
- 35:48SPEAKER_00: And that's fraud.
- 35:49SPEAKER_00: Right.
- 35:49And what we do is if those buildings became big enough that
- 35:52we'd notice,
- 35:53we'd then start a suit.
- 35:54SPEAKER_00: You know, False Claims Act suit, prosecuted by DOJ.
- 35:58SPEAKER_00: It would take a few years.
- 35:59SPEAKER_00: Maybe there'd be a settlement.
- 36:00SPEAKER_00: Maybe there'd be a payment.
- 36:01SPEAKER_00: And maybe Dr. Smith would be kicked out of the program.
- 36:03SPEAKER_00: But that's a slow process.
- 36:06SPEAKER_00: It doesn't capture everything.
- 36:07SPEAKER_00: And importantly, the dollars already went out.
- 36:09SPEAKER_00: Yeah.
- 36:09SPEAKER_00: Okay.
- 36:10And what we want to do is we want to switch to a program where
- 36:12we're doing it excellent.
- 36:13SPEAKER_00: Look at the data.
- 36:14SPEAKER_00: We see patterns that's just fraud.
- 36:16SPEAKER_00: And before the dollars go out the door,
- 36:18SPEAKER_00: we try to capture some of that and just ask,
- 36:20SPEAKER_00: hey,
- 36:20SPEAKER_00: Dr.
- 36:20SPEAKER_00: Smith,
- 36:21SPEAKER_00: did you actually do this?
- 36:23SPEAKER_00: Things like that.
- 36:23SPEAKER_01: Carrible moment for Dr. Smith.
- 36:25SPEAKER_01: Just really awful to get caught.
- 36:27SPEAKER_01: Exactly.
- 36:28SPEAKER_00: And so we want to try to engage.
- 36:31So our Center for Program Integrity is working on projects like
- 36:33that to kind of do
- 36:34SPEAKER_00: it a little bit earlier.
- 36:35SPEAKER_00: So the dollars aren't out the door.
- 36:37SPEAKER_00: We can cover more population in terms of protecting them against fraud.
- 36:41SPEAKER_00: And we have a better sense of exactly what we're doing.
- 36:44SPEAKER_00: So beyond fraud,
- 36:45SPEAKER_00: waste and abuse,
- 36:46SPEAKER_00: there is also a,
- 36:48SPEAKER_00: you know,
- 36:48SPEAKER_00: going back to this idea that health is more than just health care,
- 36:52SPEAKER_00: that your behavior and your lifestyle can affect what your health is.
- 36:56SPEAKER_00: You know,
- 36:56SPEAKER_00: at the secretary level,
- 36:58we're strongly thinking about how do we encourage people to
- 37:01engage in more healthy
- 37:03SPEAKER_00: practices,
- 37:03SPEAKER_00: whether it's exercise,
- 37:04SPEAKER_00: better eating,
- 37:06SPEAKER_00: those sorts of choices that are also inputs into health.
- 37:08SPEAKER_00: But again, don't directly involve federal spending right away.
- 37:12SPEAKER_00: I think those are cross-cutting across all the different programs.
- 37:16SPEAKER_01: Sure.
- 37:16SPEAKER_01: Let me ask you to teach me a little bit more about this fraud question.
- 37:21SPEAKER_01: I'm in the Medicare Advantage program.
- 37:22SPEAKER_01: I come in to see you,
- 37:24SPEAKER_01: Dr.
- 37:25SPEAKER_01: Smith and Dr.
- 37:26SPEAKER_01: Smith says,
- 37:28SPEAKER_01: yeah,
- 37:28SPEAKER_01: we think you have this set of conditions as maybe easy.
- 37:31SPEAKER_01: If I really do,
- 37:32SPEAKER_01: it's hard to really tell,
- 37:34SPEAKER_01: but I'm classified as a higher risk,
- 37:37SPEAKER_01: sicker patient than maybe the facts would really firmly suggest.
- 37:41SPEAKER_01: And Dr. Smith gets reimbursed at a higher rate.
- 37:46SPEAKER_01: This has been a problem going back decades, as I understand.
- 37:49SPEAKER_01: It's not necessarily a partisan coded issue.
- 37:53SPEAKER_01: Medicare Advantage is a massive program.
- 37:55SPEAKER_01: You could save a lot of money to crack down on upcoding.
- 37:58SPEAKER_01: And Dr.
- 37:59SPEAKER_01: Oz,
- 37:59SPEAKER_01: your boss,
- 38:00SPEAKER_01: in his confirmation hearings,
- 38:01SPEAKER_01: said there's a new sheriff in town,
- 38:03SPEAKER_01: quote,
- 38:04SPEAKER_01: on medical upcoding.
- 38:06SPEAKER_01: And he firmly understands this is an issue and wants to crack down on it.
- 38:10SPEAKER_01: How do you actually go about doing that without costing the patient?
- 38:14SPEAKER_00: Okay.
- 38:15SPEAKER_00: So let me address that in two parts.
- 38:17SPEAKER_00: First is let me give you a simple definition of fraud.
- 38:20And then the second thing I'm going to do is talk about
- 38:22concrete changes you can
- 38:23SPEAKER_00: make and address both fraud,
- 38:25SPEAKER_00: but at the same time,
- 38:26keep in mind patient outcomes because you don't want to cut
- 38:29necessary care.
- 38:30SPEAKER_00: So let's think about that definition.
- 38:32SPEAKER_00: You know,
- 38:32one of the challenges with fraud is it's a whole litany of both
- 38:36federal and state
- 38:37SPEAKER_00: rules,
- 38:37SPEAKER_00: statutes that govern fraud.
- 38:39SPEAKER_00: It's very hard to teach because the rules do different things.
- 38:43SPEAKER_00: And so to understand fraud,
- 38:45I think it'd be helpful to have a simpler definition that
- 38:47captures what all the
- 38:48SPEAKER_00: different statutes are trying to do.
- 38:50And Jetson Leader Louise and I have this paper that we wrote
- 38:53just before I started
- 38:53SPEAKER_00: this job where we try to provide that.
- 38:55SPEAKER_00: And here's our attempt at providing a simple definition.
- 38:58So fraud can be simply defined as when the following three
- 39:03items do not match.
- 39:05That is,
- 39:06what the provider should have done for the patient given the
- 39:11contract with the payer.
- 39:12SPEAKER_00: Second, what the provider did for the patient.
- 39:17SPEAKER_00: And third, what the provider told the payer the provider did.
- 39:21SPEAKER_00: When any of those three things differ from the other two, that's fraud.
- 39:26SPEAKER_00: So let me give you three simple or very common categories of fraud.
- 39:30SPEAKER_00: Take upcoding.
- 39:31SPEAKER_00: Upcoding is simply,
- 39:33you may have done the right thing for the patient according to
- 39:36the contract,
- 39:36SPEAKER_00: but you told the insurance company you did a little bit more.
- 39:39SPEAKER_00: Okay, that's upcoding.
- 39:41Let's take the second most common class of fraud,
- 39:44which is medically unnecessary care.
- 39:46That's a situation where you told the payer that you were gonna
- 39:51do what was
- 39:52SPEAKER_00: appropriate,
- 39:53but you actually provided more care than is required and billed
- 39:58for that extra
- 39:58SPEAKER_00: care.
- 39:59SPEAKER_00: So you did too much.
- 40:01SPEAKER_00: The third is the opposite of that, which is substandard care.
- 40:05SPEAKER_00: You're supposed to provide a certain high quality of care.
- 40:08SPEAKER_00: And you did less than that,
- 40:09SPEAKER_00: even though the contract required to provide the higher quality care.
- 40:12SPEAKER_00: And you still billed for higher care, even though you provided less care.
- 40:15SPEAKER_00: And very often that happens when,
- 40:16SPEAKER_00: you know,
- 40:16a doctor has somebody that's not qualified or a nurse or just a
- 40:20tech do the
- 40:21SPEAKER_00: procedure that a doctor.
- 40:23SPEAKER_00: That would be a classic example.
- 40:25SPEAKER_00: And that's how you define fraud.
- 40:26SPEAKER_00: OK,
- 40:27SPEAKER_00: and I think that that really helps when you use those three categories,
- 40:29SPEAKER_00: because that captures almost everything.
- 40:31The only exception is something called kickbacks,
- 40:33which we can talk about if you'd like.
- 40:35Now,
- 40:36the second part of your question was,
- 40:38you know,
- 40:38what's going to be different?
- 40:40SPEAKER_00: And I already kind of answered that question to some extent.
- 40:43SPEAKER_00: Again,
- 40:43SPEAKER_00: in the past,
- 40:44SPEAKER_00: in general,
- 40:45SPEAKER_00: we've been doing fraud enforcement after that,
- 40:48SPEAKER_00: after it's occurred,
- 40:48SPEAKER_00: after the money's out the door,
- 40:49SPEAKER_00: and we usually use litigation tools to go after it.
- 40:52I think that an important thing that we're doing now is we're
- 40:55shifting towards a
- 40:56SPEAKER_00: little bit more ex-ante.
- 40:58Let's use the data that we get in,
- 41:00the claims data,
- 41:01to identify patterns that suggest fraud.
- 41:04SPEAKER_00: For example,
- 41:04if there's a procedure that is supposed to take 30 minutes and
- 41:09a doctor or a
- 41:10hospital or a specific provider is billing for that and the
- 41:13billing practice
- 41:14SPEAKER_00: suggests that the person worked 5,000 hours in a year,
- 41:17SPEAKER_00: probably fraud.
- 41:19SPEAKER_00: We should at least investigate that.
- 41:20I'm having to do the math real quick in my head to think how
- 41:23many hours do I work in a year?
- 41:24SPEAKER_01: 5,000 hours.
- 41:25SPEAKER_01: More than I do.
- 41:27SPEAKER_00: 100%.
- 41:27SPEAKER_00: 100%.
- 41:28SPEAKER_00: So typical full-time is 2,000.
- 41:29SPEAKER_00: Thank you.
- 41:30SPEAKER_00: Good to know.
- 41:30SPEAKER_00: Okay.
- 41:31SPEAKER_00: And also,
- 41:32I'm not sure I'd want a physician that's working two and a half
- 41:34full-time jobs to
- 41:35SPEAKER_00: do my procedure.
- 41:36SPEAKER_00: Surely not.
- 41:37SPEAKER_00: And the thing is,
- 41:38SPEAKER_00: that's a very simple example I gave you,
- 41:40SPEAKER_00: but there are many other examples of situations where we'd agree...
- 41:44SPEAKER_00: probably this is not occurring,
- 41:46SPEAKER_00: services are not being provided,
- 41:48SPEAKER_00: or at least not provided to the level of quality we want.
- 41:51SPEAKER_00: And so that gets back to the last part of the question.
- 41:52How do we make sure that when we cut spending,
- 41:55we're not hurting the patient?
- 41:57SPEAKER_00: Well,
- 41:57SPEAKER_00: the first thing you can do is just try to figure out,
- 41:59SPEAKER_00: is this provider actually plausibly providing these services?
- 42:02SPEAKER_00: Because if we really are inferring that he or she is not,
- 42:06SPEAKER_00: Then, in fact, the patient is not better off by that spending.
- 42:08And in some cases might be worse off if the person is like
- 42:11taking a 30 minute
- 42:12SPEAKER_00: procedure and trying to do it in 10.
- 42:14SPEAKER_00: The second thing that we can do is to set a high threshold.
- 42:17SPEAKER_00: I mean, there's plenty of work to be done on the fraud side.
- 42:20SPEAKER_00: And so if we say,
- 42:21SPEAKER_00: look,
- 42:22SPEAKER_00: if you bill for 5000 hours in a year,
- 42:24SPEAKER_00: you know,
- 42:25SPEAKER_00: we're going to we're going to go after you.
- 42:26SPEAKER_00: But if you're billing for like 4000 or 3500,
- 42:29at least we'll look a little bit more before we decide that
- 42:31this is fraud.
- 42:33And so you can have a high threshold and graduate both the high
- 42:36threshold and the
- 42:36amount of scrutiny you get to make sure that the thing that
- 42:39doesn't happen is that
- 42:41SPEAKER_00: the patient doesn't get care that he or she needs.
- 42:45SPEAKER_01: Sure.
- 42:45SPEAKER_01: Let me ask you a more specific version of this.
- 42:48SPEAKER_01: By revenue, UnitedHealth is the fourth largest company in the country.
- 42:52SPEAKER_01: It's after Apple.
- 42:53SPEAKER_01: It's huge.
- 42:54About one in 10 doctors work for it or are affiliated with it
- 42:57in some capacity.
- 42:58SPEAKER_01: It's like a huge organization.
- 43:00SPEAKER_01: There's been some good reporting,
- 43:01SPEAKER_01: I believe from Stat News,
- 43:03about the practices that UnitedHealth engages in through
- 43:07bonuses and performance
- 43:08reviews that basically encourage clinicians to identify more
- 43:11health problems in
- 43:12SPEAKER_01: patients.
- 43:13Even if those conditions are marginal or seem dubious,
- 43:16it's not totally clear.
- 43:18SPEAKER_01: When in doubt,
- 43:19UnitedHealth has a set of incentives for clinicians to say this
- 43:23patient is a higher
- 43:24SPEAKER_01: risk patient.
- 43:26SPEAKER_01: What tools does CMS have at its disposal to crack down on it?
- 43:29Should it just ban that kind of incentive structure internally
- 43:33so that you don't
- 43:34get an extra bonus as a doctor for identifying more health
- 43:37problems in your
- 43:37SPEAKER_01: patients?
- 43:38SPEAKER_01: How should I think about this from the outside?
- 43:42SPEAKER_00: Okay, two things real quick before I answer.
- 43:44SPEAKER_00: The first one is I don't want to pick on United in particular.
- 43:47SPEAKER_00: I'm not saying United is good or bad.
- 43:49SPEAKER_00: That's not my job to do.
- 43:51And so I want to be a little bit careful that the answer I'm
- 43:53going to give you is
- 43:54SPEAKER_00: unrelated to United,
- 43:56SPEAKER_00: that any particular instance,
- 43:57SPEAKER_00: you'd actually want to see,
- 43:58talk to someone who might be involved in a case or is actually
- 44:00an investigation,
- 44:01SPEAKER_00: but that's not me.
- 44:02SPEAKER_01: Of course.
- 44:03SPEAKER_01: I'm just mentioning it as one.
- 44:04SPEAKER_01: It's been publicly reported.
- 44:05SPEAKER_01: There's like a body of work,
- 44:07but I'm imagining this is a problem beyond any one individual
- 44:10insurance company.
- 44:11SPEAKER_00: Yeah,
- 44:11so I'm going to answer the spear of the question rather than
- 44:13that specific company
- 44:14SPEAKER_00: is what I'm saying.
- 44:15SPEAKER_00: The second thing I want to say is I want to be a little bit careful.
- 44:18SPEAKER_00: And I think you,
- 44:19SPEAKER_00: by the way,
- 44:19SPEAKER_00: hit with this question on an important topic,
- 44:22SPEAKER_00: which is United is an insurance company.
- 44:25SPEAKER_00: And it's not necessarily,
- 44:26SPEAKER_00: although it has some vertically integrated segments,
- 44:29SPEAKER_00: that includes providers.
- 44:31SPEAKER_00: So sometimes it's contracting with providers,
- 44:33SPEAKER_00: arm's length,
- 44:33SPEAKER_00: and sometimes it's actually an employer of the providers.
- 44:37SPEAKER_00: And obviously when it's an employer,
- 44:38SPEAKER_00: it can actually tell,
- 44:39maybe has a little bit more power to tell the provider exactly
- 44:42what to do.
- 44:42An arm's length contract,
- 44:44it has to do indirect incentives,
- 44:45things like that.
- 44:46But what you want to ask all the time is why United or a
- 44:50company like United,
- 44:51SPEAKER_00: okay,
- 44:51SPEAKER_00: so let's just call it an insurance company,
- 44:53SPEAKER_00: might have an incentive to get a doctor to report more or less,
- 44:56SPEAKER_00: right?
- 44:56SPEAKER_00: If the insurance company was acting on its own,
- 44:58SPEAKER_00: at least in hindsight,
- 45:00SPEAKER_00: within a plan year,
- 45:02SPEAKER_00: it doesn't necessarily want
- 45:03the provider to exaggerate exactly what the claims are because
- 45:06it affects the bottom line.
- 45:07SPEAKER_00: But if the insurer is being paid by the government,
- 45:12SPEAKER_00: say the federal government,
- 45:14SPEAKER_00: to provide insurance either through premium supports or an MA,
- 45:17SPEAKER_00: the Medicare Advantage,
- 45:19SPEAKER_00: the premium itself,
- 45:20SPEAKER_00: it might.
- 45:21But the reason is because that premium might itself depend on
- 45:25how much care
- 45:27SPEAKER_00: the individual is getting.
- 45:28So now we have to go back and discuss risk adjustment and then
- 45:31think about how we
- 45:32SPEAKER_00: answer that,
- 45:33SPEAKER_00: how we make risk adjustment.
- 45:34SPEAKER_00: What is risk adjustment?
- 45:35SPEAKER_00: Risk adjustment says if you decided in Medicare,
- 45:39the federal government was going to pay the exact same premium
- 45:42for every individual
- 45:43SPEAKER_00: that chose to participate in Medicare Advantage,
- 45:46insurance companies would have an incentive to try to find
- 45:49individuals who are low
- 45:51SPEAKER_00: cost and healthy.
- 45:52SPEAKER_00: And cover those guys,
- 45:54SPEAKER_00: but not cover those individuals who are high cost and not healthy,
- 45:58especially if they couldn't charge any more than what the
- 46:00government was paying in
- 46:00SPEAKER_00: premiums.
- 46:01SPEAKER_01: That's called cream skimming.
- 46:03SPEAKER_01: Sorry, cream skimming?
- 46:05SPEAKER_01: Cream skimming.
- 46:06SPEAKER_01: As in the really healthy guys and girls, we're going to ensure them.
- 46:11SPEAKER_00: Exactly.
- 46:11SPEAKER_00: Right.
- 46:12And the idea is back in the day when we used to think that
- 46:14cream was good,
- 46:15SPEAKER_00: we would skim the cream off and that's the best part.
- 46:17SPEAKER_00: Some of us still do.
- 46:19SPEAKER_00: Well,
- 46:19SPEAKER_00: you know,
- 46:20SPEAKER_00: for a lot of the rest of us,
- 46:21SPEAKER_00: we're like,
- 46:21SPEAKER_00: okay,
- 46:22SPEAKER_00: we're going to go with the skim milk,
- 46:23SPEAKER_00: which has the cream taken out of it.
- 46:25SPEAKER_00: But yes, so that's called cream skimming.
- 46:28You take the very best patients and you kind of keep them in
- 46:30your pool and you try
- 46:31SPEAKER_00: to avoid high risk patients.
- 46:33SPEAKER_00: That way you can be more profitable.
- 46:34SPEAKER_00: Now, we anticipate that.
- 46:36SPEAKER_00: The government does.
- 46:37SPEAKER_00: And so what we do is we engage in risk adjustment.
- 46:40We try to estimate how much each individual will cost the
- 46:43insurance company and
- 46:44then pay the insurance company a premium that's proportional to
- 46:48that cost.
- 46:50SPEAKER_00: And this should encourage insurance companies,
- 46:53SPEAKER_00: in theory,
- 46:54SPEAKER_00: to take on high-cost patients and not just low-cost patients,
- 46:57SPEAKER_00: to eliminate creams.
- 46:59SPEAKER_00: Now, the devil's in the details.
- 47:02SPEAKER_00: It depends on how well you do risk adjustment.
- 47:04SPEAKER_00: If you do risk adjustment poorly, you might not solve the cream skimmer.
- 47:07SPEAKER_00: However, sometimes you can make another error.
- 47:10And that is that you use a formula for determining risk that is
- 47:15manipulable by the
- 47:17SPEAKER_00: insurance company.
- 47:18SPEAKER_00: Help me understand that.
- 47:19SPEAKER_00: What would that look like?
- 47:21SPEAKER_00: Okay.
- 47:21SPEAKER_00: So let me give you an example from Medicare Advantage.
- 47:23So say I calculate your premiums or your risk based upon your
- 47:28past claims.
- 47:29SPEAKER_00: So what claims the Santi file,
- 47:32that is to say the insurance company file on behalf of Santi in
- 47:35previous years?
- 47:37SPEAKER_01: Let's make a specific.
- 47:38SPEAKER_01: I had a big ACL and meniscus surgery in the past year.
- 47:42SPEAKER_01: and otherwise have been very healthy.
- 47:44SPEAKER_01: So let me give you that.
- 47:45SPEAKER_00: Great.
- 47:45SPEAKER_00: You and I are actually, we overlap in ailments or afflictions.
- 47:49SPEAKER_00: I had an ACL reconstruction earlier this year too.
- 47:52SPEAKER_00: Too much skiing?
- 47:53SPEAKER_00: No, this was squash.
- 47:55SPEAKER_00: My new current alternative to skiing.
- 47:57SPEAKER_00: Okay.
- 47:58But anyway,
- 47:58we take a look at your past knee surgery experience and all
- 48:02surgery experience.
- 48:03SPEAKER_00: And we'd say,
- 48:04SPEAKER_00: okay,
- 48:04SPEAKER_00: well,
- 48:04SPEAKER_00: that tells us like,
- 48:05SPEAKER_00: you know,
- 48:05SPEAKER_00: if you've had this surgery,
- 48:06SPEAKER_00: you're a more active person,
- 48:07but you're more likely to have this surgery again or have other
- 48:10afflictions with
- 48:10SPEAKER_00: your knee.
- 48:11SPEAKER_00: So you're a higher risk.
- 48:12SPEAKER_00: OK, sure.
- 48:13SPEAKER_00: Now, that's fine if that was the truth.
- 48:16SPEAKER_00: But if an insurance company is likely to have you for multiple years,
- 48:20which is often the case for older patients who don't switch as
- 48:23much and your older
- 48:24SPEAKER_00: patients,
- 48:24you have a lot of afflictions that they can kind of attribute
- 48:27to you where we would
- 48:28SPEAKER_00: say,
- 48:28SPEAKER_00: oh,
- 48:28SPEAKER_00: yeah,
- 48:29SPEAKER_00: that's probably likely to be the case.
- 48:31What they could do is that this year they could mark up your
- 48:34claims and say you
- 48:35SPEAKER_00: claim for more than you actually did.
- 48:37Or they might have you go through some diagnostic tests that
- 48:39even if they result in no
- 48:41SPEAKER_00: positive proof of affliction at least shows that there's a concern.
- 48:46SPEAKER_00: So we can increase our claims this year, the insurance company can.
- 48:49SPEAKER_00: And knowing that while they're on the hook this year,
- 48:52SPEAKER_00: in future years,
- 48:54SPEAKER_00: they'll get a higher premium.
- 48:55from the federal government in MA premium because they now view
- 49:00Santi as a more risky person.
- 49:03SPEAKER_01: That's interesting.
- 49:04SPEAKER_01: Right.
- 49:04SPEAKER_01: And just to get a little bit more color here,
- 49:07SPEAKER_01: they might just push more diagnostics on me,
- 49:10SPEAKER_01: even if I'm broadly healthy,
- 49:12SPEAKER_01: because tell me if I'm getting this right.
- 49:14I'm generally healthy for an older guy,
- 49:16but there's these little risk signals.
- 49:18SPEAKER_01: And if you do enough diagnostic tests,
- 49:20you'll find something that looks like a risk signal and you can
- 49:22throw that in the
- 49:23SPEAKER_01: mix.
- 49:24So the way to think about it is that your risk score is a
- 49:26function of your past claims.
- 49:28SPEAKER_00: OK, or at least a bunch of different factors, including your past claims.
- 49:32SPEAKER_00: And the insurance company that gets the benefit of a higher risk score,
- 49:36SPEAKER_00: which is a higher premium,
- 49:38they might be able to manipulate the inputs into the risk score
- 49:40calculation.
- 49:42SPEAKER_00: OK, sure.
- 49:42SPEAKER_00: So the challenge with risk adjustment is to do two things.
- 49:46One is to eliminate factors in calculation of risk that the
- 49:51insurance company can
- 49:52manipulate or to increase the cost of manipulation for each of
- 49:57those factors.
- 49:59SPEAKER_00: So the two kind of approaches that you can take.
- 50:02SPEAKER_00: Now, the challenge is this is a game.
- 50:04SPEAKER_00: This is a back and forth, right?
- 50:05SPEAKER_00: Like,
- 50:05SPEAKER_00: you know,
- 50:06SPEAKER_00: once you set the risk score function up,
- 50:10SPEAKER_00: the insurance company manipulates,
- 50:11SPEAKER_00: you discover what they're doing to manipulate,
- 50:13then you close that loophole and then they find another
- 50:15approach to manipulate.
- 50:17SPEAKER_00: And this goes back and forth.
- 50:19And what we need to do is we need to constantly play that game
- 50:23or we need to reduce
- 50:25the incentives of the insurance company to engage in that sort
- 50:30of behavior and that
- 50:31SPEAKER_00: manipulation of risk scores.
- 50:32SPEAKER_00: Sure.
- 50:32SPEAKER_00: Okay.
- 50:33SPEAKER_00: So those are the two approaches that you can take.
- 50:35And those are kind of the changes that we'd like to see have
- 50:39happen going forward.
- 50:40SPEAKER_00: But again,
- 50:41SPEAKER_00: this gets back to,
- 50:42SPEAKER_00: again,
- 50:42what we talked about at the very beginning of our conversation
- 50:45about capacity.
- 50:46SPEAKER_00: You need to be able to do that quickly,
- 50:49even though you've got this large entity that's managing many
- 50:51different programs
- 50:52SPEAKER_00: and has some bureaucratic components to it.
- 50:55SPEAKER_00: And so that's the challenge with risk adjustment.
- 50:58SPEAKER_00: And you don't want to go too far,
- 50:59SPEAKER_00: by the way,
- 50:59SPEAKER_00: because if you undercompensate insurance companies,
- 51:02SPEAKER_00: like if you're too aggressive,
- 51:04SPEAKER_00: you're going to get cream skin.
- 51:05SPEAKER_00: So that's the balance that you need to play.
- 51:07SPEAKER_01: Well, let me ask one more question on this.
- 51:10This admin has finalized some rule changes,
- 51:13and there's work in the Biden admin on this.
- 51:15There's been work in the Trump admin on this to try and tighten
- 51:19those rules so that
- 51:20risk adjustment is a little bit more locked down and that you
- 51:24don't get the same
- 51:24SPEAKER_01: kinds of incentives for fraud on the margin that you might have now.
- 51:28SPEAKER_01: At the same time,
- 51:29this admin is going to pay much higher reimbursement rates for
- 51:33Medicare Advantage
- 51:35SPEAKER_01: to these private insurance companies than what the Biden admin proposed.
- 51:39SPEAKER_01: So by one account,
- 51:40SPEAKER_01: and again,
- 51:41SPEAKER_01: I'll link all these in the show notes,
- 51:42that should result in more than $25 billion in additional
- 51:46payments going to
- 51:47SPEAKER_01: Medicare Advantage plans in 2026.
- 51:50SPEAKER_01: Help me square that circle.
- 51:51We want to reimburse these private insurers more at the same
- 51:55time as we're cracking down on
- 51:58SPEAKER_01: potential fraud in the system?
- 51:59SPEAKER_01: How does that whole balance net out?
- 52:02SPEAKER_00: Yeah, great question.
- 52:04SPEAKER_00: Also gets at the complexity of the program.
- 52:06SPEAKER_00: So let's answer this in parts.
- 52:08How do you address risk adjustment gaming even as you're
- 52:11increasing spending?
- 52:13And the way to kind of square that circle is to understand that
- 52:16it is possible that
- 52:18SPEAKER_00: the composition of risk in the program in MA is increasing.
- 52:25SPEAKER_00: At the same time, there's also inaccuracy in estimation of that risk.
- 52:30SPEAKER_00: So imagine if,
- 52:31SPEAKER_00: and I'm not saying this is true,
- 52:32SPEAKER_00: I'm just going to give you a kind of proof by counter example,
- 52:35SPEAKER_00: but that illustrates how this could work.
- 52:37Okay,
- 52:37let's suppose that you're spending a dollar this year on
- 52:41average per person.
- 52:43And you know that the composition of the population is going to
- 52:47change such that
- 52:48SPEAKER_00: the actual cost of that population is going to be $1.10.
- 52:51SPEAKER_00: Okay.
- 52:53New people coming into the program from traditional Medicare
- 52:55could be new people
- 52:56SPEAKER_00: coming into the program because they're aging into the program.
- 52:58It could be just underlying changes in the population,
- 53:01whatever it happens to be.
- 53:02SPEAKER_00: Let's suppose that's the truth.
- 53:04SPEAKER_00: Okay.
- 53:05SPEAKER_00: And now that's on average.
- 53:07SPEAKER_00: Some people are going to cost a dollar.
- 53:09SPEAKER_00: Some people are going to cost $1.20.
- 53:10SPEAKER_00: Some people are going to be in between.
- 53:12SPEAKER_00: Okay.
- 53:13SPEAKER_00: And the problem we face is not merely that the cost is rising.
- 53:17SPEAKER_00: That's something that we just have to acknowledge and we have to pay for.
- 53:19SPEAKER_00: But at the same time,
- 53:20SPEAKER_00: it could be the case that some individuals who might cost $1.05,
- 53:26SPEAKER_00: their costs are exaggerated to $1.10 by the insurance company.
- 53:30through manipulation of claims so as to exaggerate their risk
- 53:34adjustment score.
- 53:36SPEAKER_00: So what we want to do is we want to say,
- 53:37SPEAKER_00: look,
- 53:38SPEAKER_00: on the one hand,
- 53:39we definitely want to account for changes in actual costs for
- 53:43the program.
- 53:44But at the same time,
- 53:45we want to cut back on fake costs,
- 53:48exaggeration of risk scores.
- 53:51SPEAKER_00: And so you can have both things happen at the same time.
- 53:54SPEAKER_00: You just have to be careful about it.
- 53:57SPEAKER_00: And I think that's what we're trying to do.
- 53:59Now,
- 54:00can I use this as an opportunity to provide another lesson
- 54:02that's important?
- 54:03SPEAKER_00: Why not?
- 54:04So we already talked about one margin just a second ago,
- 54:07which is the program is not static.
- 54:09SPEAKER_00: The population in the program is not static.
- 54:10SPEAKER_00: That's changes.
- 54:12In one place where it changes is people going from traditional
- 54:15Medicare to Medicare Advantage.
- 54:17SPEAKER_00: Medicare Advantage has a number of interesting properties,
- 54:21SPEAKER_00: useful properties that make it an attractive program.
- 54:23So one thing that makes a very attractive program is that it
- 54:27allows for allocation
- 54:29SPEAKER_00: of financial risk to both insurance companies and to providers.
- 54:36And to do that in a way that makes sure that both insurance
- 54:38companies and providers
- 54:39SPEAKER_00: are giving patients the best care at appropriate prices.
- 54:43So in traditional Medicare,
- 54:45you do that through what's called accountable care
- 54:47organizations.
- 54:48SPEAKER_00: I mentioned that earlier.
- 54:49SPEAKER_00: You shift the risk just to providers.
- 54:51But you can imagine that there's going to be some times where
- 54:53you want to shift the
- 54:54risk to the insurer because the insurer knows a little bit more
- 54:56information and
- 54:57SPEAKER_00: sometimes you want to shift it to the provider.
- 54:59You can't do that in traditional Medicare,
- 55:01but you can do that in Medicare Advantage.
- 55:03You can shift it to the private insurance company who can then
- 55:05take a portion of
- 55:06that risk and shift it to provider and optimally allocate it in
- 55:09such a way to get
- 55:10SPEAKER_00: the best deal for the patient and for the federal government.
- 55:13SPEAKER_00: OK, so that's one reason why you might prefer Medicare Advantage.
- 55:18Another reason that patients really like is that Medicare
- 55:21Advantage actually
- 55:22SPEAKER_00: provides more coverage than traditional Medicare.
- 55:24SPEAKER_00: So, for example, traditional Medicare is hospitals, physicians, drugs.
- 55:28SPEAKER_00: But maybe you're going to get some additional care through an MA plan,
- 55:32SPEAKER_00: additional care benefits like vision and dental and things like that,
- 55:35SPEAKER_00: that patients value.
- 55:36SPEAKER_00: Yeah.
- 55:37SPEAKER_00: OK, so that might be another reason why you like Medicare Advantage.
- 55:39SPEAKER_00: Anyway, there are a range of regions.
- 55:41One of the things that you might want to do is pull people into
- 55:43the M&A plan if you
- 55:44SPEAKER_00: think it's better for patients,
- 55:47SPEAKER_00: better quality,
- 55:48SPEAKER_00: lower cost,
- 55:49SPEAKER_00: both.
- 55:50SPEAKER_00: So that's one thing that's important.
- 55:52And to keep in mind when we think about that cost estimate,
- 55:54that's an economic point.
- 55:55But here's an accounting point that's also very important when
- 55:58you hear this number.
- 55:59SPEAKER_00: So when you hear a number,
- 56:01SPEAKER_00: something like we spend,
- 56:03SPEAKER_00: we meaning the federal government spends $1.10.
- 56:06I'm going to make up that number because there's that dispute
- 56:08about what the right number are.
- 56:09SPEAKER_00: But let's suppose that somebody asserts the government spends,
- 56:12SPEAKER_00: for every dollar spent on a traditional Medicare patient,
- 56:16SPEAKER_00: the government spends $1.10 on a Medicare Advantage patient.
- 56:21SPEAKER_00: I'm not going to deny that that's actually true,
- 56:23but I want to put that in context so you understand that that
- 56:26doesn't necessarily
- 56:27SPEAKER_00: imply that MA is inefficient or that it's not providing quality.
- 56:30SPEAKER_01: I was going to ask you about this, so I'm glad you're bringing it up.
- 56:33SPEAKER_00: Great.
- 56:34SPEAKER_00: Yeah.
- 56:34SPEAKER_00: Because,
- 56:34SPEAKER_00: you know,
- 56:34it connects to this point that it looks like the government is
- 56:37spending more on MA.
- 56:39SPEAKER_00: So when we think about how much is spent on the patient,
- 56:42SPEAKER_00: we should think about all possible spending,
- 56:45meaning you should think about total costs of treating the
- 56:48patient in traditional
- 56:50Medicare and the total cost of treating the patient in Medicare
- 56:54Advantage.
- 56:55SPEAKER_00: For simple comparisons,
- 56:56the problem is that the coverage of those costs varies in
- 57:02traditional Medicare
- 57:02SPEAKER_00: versus Medicare Advantage.
- 57:04SPEAKER_00: In traditional Medicare,
- 57:05SPEAKER_00: the government pays for part of the insurance,
- 57:08but there's a substantial amount of cost sharing that's
- 57:11involved in traditional
- 57:12SPEAKER_00: Medicare.
- 57:14SPEAKER_00: In Medicare Advantage, there's very little cost sharing.
- 57:17SPEAKER_00: And it kind of maps onto,
- 57:18SPEAKER_00: if you study health insurance,
- 57:19SPEAKER_00: it kind of maps on this idea that,
- 57:21SPEAKER_00: like,
- 57:22when you have a fee-for-service program where you're going to
- 57:24pay for anything you
- 57:25SPEAKER_00: want to,
- 57:25SPEAKER_00: you could go to any physician,
- 57:26SPEAKER_00: et cetera,
- 57:27SPEAKER_00: usually to avoid excessive spending,
- 57:29SPEAKER_00: you have a high coinsurance,
- 57:31SPEAKER_00: high deductible,
- 57:32SPEAKER_00: high copays,
- 57:33SPEAKER_00: et cetera,
- 57:34SPEAKER_00: to make sure that the patient at least sees part of that cost.
- 57:38SPEAKER_00: And when you do some sort of managed care,
- 57:40SPEAKER_00: whether it's a PPO or an HMO,
- 57:42SPEAKER_00: like what you might see in Medicare Advantage,
- 57:44SPEAKER_00: you don't have very much cost sharing.
- 57:46SPEAKER_00: Instead,
- 57:47you have the provider that determines whether you can go or the
- 57:50plan determines can
- 57:50SPEAKER_00: you go out of network or you have the provider that has to get approval,
- 57:54SPEAKER_00: give approval to go to a specialized physician,
- 57:56SPEAKER_00: things like that.
- 57:57SPEAKER_00: So those are the two different ways to control costs.
- 57:59But what that means is that in traditional Medicare,
- 58:02if you take total cost of a dollar,
- 58:04a higher percentage of it is paid by the patient and a smaller
- 58:07percentage is paid
- 58:07SPEAKER_00: by the government.
- 58:09SPEAKER_00: When you go to Medicare Advantage,
- 58:11SPEAKER_00: higher percentage is paid by the federal government,
- 58:12SPEAKER_00: a smaller percentage is paid by the patient.
- 58:14SPEAKER_00: And so if you just compare the total dollar amounts,
- 58:18it could look bigger in Medicare Advantage simply because the
- 58:21government's
- 58:22SPEAKER_00: accounting for a higher share.
- 58:23SPEAKER_00: And that's why it's a bit of an apples to oranges comparison.
- 58:27You really need to think about which of these two programs are
- 58:30appropriate for each
- 58:31SPEAKER_00: patient and then determine first whether or not we're spending that cost
- 58:35SPEAKER_00: appropriately overall,
- 58:36SPEAKER_00: regardless of who's actually paying the bill.
- 58:39SPEAKER_00: And then second,
- 58:39filter that down to the actual dollars being sent out of the
- 58:42treasury to providers.
- 58:45SPEAKER_01: That makes sense.
- 58:46SPEAKER_01: So we spent some time talking about tackling waste, fraud and abuse.
- 58:50SPEAKER_01: I think like you, I agree there's a lot to be done there.
- 58:54SPEAKER_01: But as you point out,
- 58:55a lot of the real meat or the real savings in health care comes
- 59:00from just bending
- 59:01SPEAKER_01: the cost curve and just being able to deliver good health care for less.
- 59:06Can you talk a little bit about how tweaking what we reimburse
- 59:10and how we design
- 59:11SPEAKER_01: those incentives can bend the cost curve?
- 59:13SPEAKER_01: I know that's been a focus of yours.
- 59:16SPEAKER_00: Yeah.
- 59:16SPEAKER_00: So let's talk about the term bending the cost curve.
- 59:19SPEAKER_00: That is roughly,
- 59:20SPEAKER_00: I think,
- 59:21broadly people should think of that as if we hold the quantity
- 59:26and quality of care
- 59:27SPEAKER_00: to patients constant,
- 59:28SPEAKER_00: how can we make that care less costly?
- 59:32SPEAKER_00: That's bending the cost curve.
- 59:34SPEAKER_00: And bending the cost curve usually falls into two buckets.
- 59:38SPEAKER_00: One is what economists might call a static bucket.
- 59:41SPEAKER_00: Let's take the existing technologies that we have,
- 59:44SPEAKER_00: treatments and therapies that we have,
- 59:46SPEAKER_00: and just make sure we use them to bend the cost curve,
- 59:50SPEAKER_00: okay?
- 59:51SPEAKER_00: Not invent anything new,
- 59:52not make any changes in how healthcare is provided from an
- 59:55innovation perspective.
- 59:57SPEAKER_00: And then there's the dynamic bending of the cost curve,
- 59:59which is how do we make sure that in a few years we have
- 1:00:02technologies that allow us
- 1:00:04SPEAKER_00: to provide the same quality and quantity,
- 1:00:05SPEAKER_00: but with lower cost?
- 1:00:07So when we talk about reimbursement policy,
- 1:00:10we're talking about existing technologies.
- 1:00:12SPEAKER_00: So we're really talking about the static question.
- 1:00:14SPEAKER_00: How do we reduce cost holding quantity and quality constant?
- 1:00:19SPEAKER_00: And there,
- 1:00:21I think what we're looking for is opportunities where a
- 1:00:25provider to take care of a
- 1:00:27patient is providing a particular therapy or a diagnostic or
- 1:00:32other broadly
- 1:00:33categorized treatment that is more costly than another one that
- 1:00:37would be equally as
- 1:00:38SPEAKER_00: good.
- 1:00:39SPEAKER_00: Sometimes this goes under value-based care,
- 1:00:42SPEAKER_00: which is,
- 1:00:43are we paying for care that is paying an amount that's more
- 1:00:47than the value from
- 1:00:48SPEAKER_00: that care?
- 1:00:49An example of this might be a rule that we recently had on site
- 1:00:52neutrality,
- 1:00:53SPEAKER_00: which is a rule that says,
- 1:00:55SPEAKER_00: currently,
- 1:00:55SPEAKER_00: it is the case that for certain procedures,
- 1:00:58the amount that we pay for it varies depending on where that
- 1:01:00procedure was done.
- 1:01:02We might pay more in a hospital setting than in a more
- 1:01:04outpatient setting,
- 1:01:05things like that.
- 1:01:06SPEAKER_01: And real quick, why is that the status quo?
- 1:01:08SPEAKER_01: Like what in layman's terms,
- 1:01:10SPEAKER_01: why is it that Medicare says,
- 1:01:12SPEAKER_01: yeah,
- 1:01:12we'll pay you more for the exact same service because you
- 1:01:15perform the service in a
- 1:01:17SPEAKER_01: inpatient or a hospital or an outpatient facility.
- 1:01:21So there are sometimes legitimate arguments for why you want to
- 1:01:23pay more in a certain setting.
- 1:01:25SPEAKER_00: So, for example, take an emergency department.
- 1:01:27An emergency department often costs more because it is a very
- 1:01:31capital-intensive department.
- 1:01:32SPEAKER_00: It allows for a lot of contingencies.
- 1:01:34SPEAKER_00: If something in that procedure goes wrong,
- 1:01:36the emergency department has the equipment and personnel right
- 1:01:40there to immediately
- 1:01:41SPEAKER_00: take care of it.
- 1:01:43SPEAKER_00: So,
- 1:01:43SPEAKER_00: for example,
- 1:01:43SPEAKER_00: if you go in and you have a gunshot wound,
- 1:01:46it makes sense to do an emergency department because who knows
- 1:01:49what's going on
- 1:01:50SPEAKER_00: inside your body that could have been affected by the gunshot wound.
- 1:01:53SPEAKER_00: But if you've got a toothache and you go to the emergency department,
- 1:01:56SPEAKER_00: That doesn't make a lot of sense.
- 1:01:57SPEAKER_00: We've got to extract a wisdom tooth or something.
- 1:02:00SPEAKER_00: You don't need all that extra equipment.
- 1:02:01SPEAKER_00: It's a very straightforward procedure.
- 1:02:03SPEAKER_00: Not that it's painless, but it's a straightforward procedure.
- 1:02:05SPEAKER_00: We know what we need.
- 1:02:06SPEAKER_00: There's not a lot of contingencies.
- 1:02:08SPEAKER_00: So it wouldn't be appropriate to pay more,
- 1:02:10SPEAKER_00: to do it at the emergency room and then pay more for it.
- 1:02:13And so what ends up happening is that there are going to be
- 1:02:16some situations where
- 1:02:17SPEAKER_00: we do want to pay more because we'd rather have the procedure done in an
- 1:02:20SPEAKER_00: environment where it is more costly to do.
- 1:02:22SPEAKER_00: But there's going to be a lot of things for which we don't need that.
- 1:02:25SPEAKER_00: And we need to move for those procedures.
- 1:02:27SPEAKER_00: We need to move reimbursements to encourage,
- 1:02:30SPEAKER_00: for example,
- 1:02:30SPEAKER_00: the wisdom teeth extraction to be done on an outpatient basis.
- 1:02:34And obviously,
- 1:02:34I'm giving very peculiar examples,
- 1:02:36but that's for purposes of illustration.
- 1:02:38SPEAKER_00: Sure.
- 1:02:38SPEAKER_00: And so site neutrality rules basically says for a given procedure,
- 1:02:41SPEAKER_00: you're going to pay the same amount regardless of where it is conducted.
- 1:02:45SPEAKER_00: And you do that for procedures where it should be done,
- 1:02:48especially procedures where it should be done in a more low
- 1:02:51intensity setting.
- 1:02:52And that then encourages the provider to do it at the low
- 1:02:57intensity setting where
- 1:02:58SPEAKER_00: it's appropriate.
- 1:02:58SPEAKER_00: Something like that.
- 1:03:00SPEAKER_00: And that's the purpose of those rules.
- 1:03:02So there's a good reason for why you would do it in some
- 1:03:04procedures,
- 1:03:04pay more for certain sites.
- 1:03:06SPEAKER_00: But for others, you don't.
- 1:03:07SPEAKER_00: And for those others,
- 1:03:09you want to have a kind of a site neutrality rule to encourage
- 1:03:11the provision of
- 1:03:12SPEAKER_00: that care in a more efficient setting.
- 1:03:15SPEAKER_00: And by the way, just as good for the patient.
- 1:03:18SPEAKER_00: And so that's the basic idea.
- 1:03:19SPEAKER_00: But again,
- 1:03:20SPEAKER_00: more broadly,
- 1:03:21SPEAKER_00: if we step back,
- 1:03:21SPEAKER_00: we're saying,
- 1:03:22SPEAKER_00: OK,
- 1:03:23SPEAKER_00: we want to bend the cost curve in a static sense.
- 1:03:27Given existing technologies,
- 1:03:28are we using those technologies in the best possible way?
- 1:03:33SPEAKER_00: What I mean by that is for a given level of quality and quantity,
- 1:03:36SPEAKER_00: are you able to provide that care at a lower cost?
- 1:03:41SPEAKER_01: Sure.
- 1:03:42Let me ask you about another example of trying to bend the cost
- 1:03:45curve and don't
- 1:03:47SPEAKER_01: want to make you spend too much time on it,
- 1:03:48but I'm thinking about all of these prescription drug price
- 1:03:53pushes in American
- 1:03:55SPEAKER_01: politics.
- 1:03:55And you've seen them from this administration,
- 1:03:56we've seen them all across the board.
- 1:03:58SPEAKER_01: Everybody wants prescription drugs to cost less.
- 1:04:01SPEAKER_01: There's like a bipartisan interest in bringing down drug prices.
- 1:04:05SPEAKER_01: The counterpoint,
- 1:04:05SPEAKER_01: of course,
- 1:04:06SPEAKER_01: is that if you reduce drug prices,
- 1:04:08SPEAKER_01: mechanically,
- 1:04:10SPEAKER_01: you reduce the revenues of the companies that develop those drugs.
- 1:04:13SPEAKER_01: And in the future,
- 1:04:14maybe you get fewer of those drugs because companies are less
- 1:04:17incentivized to
- 1:04:17SPEAKER_01: develop them.
- 1:04:18So you get cheaper prices today for drugs and maybe fewer
- 1:04:22really valuable medical
- 1:04:23SPEAKER_01: innovations tomorrow.
- 1:04:25SPEAKER_01: This admin has talked about trying to reduce prescription drug prices.
- 1:04:30Are there ways,
- 1:04:31other ways to incentivize biotech companies to invest in R&D?
- 1:04:35SPEAKER_01: If you do bring down those prices,
- 1:04:37SPEAKER_01: if you do say like,
- 1:04:38SPEAKER_01: hey,
- 1:04:39SPEAKER_01: we're going to pay a lot less for Ozempic,
- 1:04:42SPEAKER_01: say.
- 1:04:43SPEAKER_00: Yeah,
- 1:04:43SPEAKER_00: I think you've highlighted the kind of trade-offs that are involved,
- 1:04:46but I want to add two complications that make it so that it's
- 1:04:50not always the case
- 1:04:51SPEAKER_00: that when you lower drug prices,
- 1:04:54SPEAKER_00: that is to say the amount the payer pays drug companies for prices,
- 1:04:58SPEAKER_00: that you get less innovation.
- 1:05:00SPEAKER_01: As a side note,
- 1:05:01this is the problem with you economists is you're always
- 1:05:03introducing complications
- 1:05:04SPEAKER_01: to my toy models in that question.
- 1:05:06SPEAKER_01: It's very frustrating.
- 1:05:07SPEAKER_00: Well, my retort is that it's reality that does the complications.
- 1:05:10SPEAKER_00: All we're doing is providing a map to that reality.
- 1:05:13SPEAKER_01: Just explicating.
- 1:05:14SPEAKER_01: Okay.
- 1:05:15SPEAKER_00: Just explicating.
- 1:05:16SPEAKER_00: That's it.
- 1:05:17SPEAKER_00: We're just neutral bystanders.
- 1:05:19SPEAKER_00: So let me offer two complications.
- 1:05:21SPEAKER_00: The first complication is a really simple one,
- 1:05:23which is that the modern drug market is not like your
- 1:05:26traditional drug market,
- 1:05:27SPEAKER_00: which fits that model that you have in your head.
- 1:05:30SPEAKER_00: The modern drug market has intermediaries involved.
- 1:05:34SPEAKER_00: And intermediaries can be PBMs, they can be wholesalers, et cetera.
- 1:05:39SPEAKER_00: PBMs.
- 1:05:40SPEAKER_00: Pharmacy benefit managers.
- 1:05:42SPEAKER_00: Right.
- 1:05:42SPEAKER_01: OK, I understand just middlemen in this process.
- 1:05:45SPEAKER_00: Yeah.
- 1:05:45SPEAKER_00: But let me take a step back for a second.
- 1:05:47SPEAKER_00: OK,
- 1:05:47SPEAKER_00: because I think that,
- 1:05:49SPEAKER_00: you know,
- 1:05:49SPEAKER_00: a lot of time if I were a listener,
- 1:05:51SPEAKER_00: I would look at this and say,
- 1:05:52SPEAKER_00: why is the system so complicated?
- 1:05:54SPEAKER_00: Like, why can't we do things simple?
- 1:05:56SPEAKER_00: And the usual answer is,
- 1:05:58SPEAKER_00: as it turns out,
- 1:05:59each of these changes addressed an old problem that we've now
- 1:06:02forgotten about.
- 1:06:04SPEAKER_00: And so if we just come along without knowing that old problem,
- 1:06:06SPEAKER_00: it looks like the new system is just like this,
- 1:06:09SPEAKER_00: you know,
- 1:06:09SPEAKER_00: ball of yarn that is so knotted up.
- 1:06:13SPEAKER_00: You can't possibly understand why it ever got here.
- 1:06:15So let's just pause for a second to talk about PBMs and then
- 1:06:18talk about the problems with PBMs.
- 1:06:19SPEAKER_00: And then I'll connect it back to exactly what you said.
- 1:06:22SPEAKER_00: Okay.
- 1:06:22So imagine you're a payer and you have to,
- 1:06:26you're going to cover patients at local hospitals.
- 1:06:28SPEAKER_00: That's perfectly fine.
- 1:06:30You negotiate with those local hospitals,
- 1:06:31your deals,
- 1:06:31but you also have to cover drugs.
- 1:06:33Now,
- 1:06:33the problem is that the drug companies are located around the
- 1:06:35world.
- 1:06:36There's a lot of drug companies and you need to negotiate with
- 1:06:38them to make sure
- 1:06:38SPEAKER_00: that your patients,
- 1:06:39SPEAKER_00: your beneficiaries get access to a portfolio of drugs that are useful,
- 1:06:44SPEAKER_00: right?
- 1:06:44SPEAKER_00: That a health insurance plan would cover.
- 1:06:47SPEAKER_00: You can negotiate with them.
- 1:06:49SPEAKER_00: But, you know, as it turns out, there's a lot of them.
- 1:06:51SPEAKER_00: In addition,
- 1:06:52it's the case that other insurance companies also have to
- 1:06:54negotiate with the same
- 1:06:55SPEAKER_00: people,
- 1:06:56SPEAKER_00: same drug companies,
- 1:06:57SPEAKER_00: to kind of get contracts on coverage.
- 1:06:59SPEAKER_00: So these entities called pharmacy benefit managers emerge,
- 1:07:02SPEAKER_00: and they both negotiate on behalf of insurers with the many companies
- 1:07:07SPEAKER_00: drug companies,
- 1:07:08but they also then help the drug companies a little bit further
- 1:07:11and help them
- 1:07:12SPEAKER_00: figure out,
- 1:07:12SPEAKER_00: here's the portfolio or formulary of drugs that you ought to cover.
- 1:07:16SPEAKER_00: And that helps the insurance companies.
- 1:07:18SPEAKER_00: Okay.
- 1:07:18By the way,
- 1:07:19this is not the only type of intermediary that exists like
- 1:07:21this.
- 1:07:21SPEAKER_00: Hospitals have these, they're called GPOs.
- 1:07:24These are organizations,
- 1:07:25the purchasing organizations that help the hospital find
- 1:07:29SPEAKER_00: gurneys and sutures and band-aids and things like that, right?
- 1:07:33SPEAKER_00: They're just intermediaries that buy a bundle of things that you need.
- 1:07:36SPEAKER_00: So this is a phenomenon that exists in a lot of places.
- 1:07:38SPEAKER_00: The problem is that over time,
- 1:07:40SPEAKER_00: PBMs,
- 1:07:42you don't need a lot of them to do the negotiation because the
- 1:07:44duplicate of the
- 1:07:44SPEAKER_00: negotiations,
- 1:07:45SPEAKER_00: they ended up being a few small,
- 1:07:47SPEAKER_00: very large PBMs.
- 1:07:49And as often happens in that sort of situation,
- 1:07:51there's a concern that those PBMs
- 1:07:53SPEAKER_00: generate profits.
- 1:07:55SPEAKER_00: Now,
- 1:07:55I'm not going to make a contention that there's excessive
- 1:07:57profits or things like
- 1:07:58SPEAKER_00: that,
- 1:07:58SPEAKER_00: but there's this concern that exists.
- 1:08:00SPEAKER_00: And the important thing for our purposes,
- 1:08:02SPEAKER_00: that means that those PBMs,
- 1:08:04SPEAKER_00: which had a purpose,
- 1:08:06SPEAKER_00: may now be adding to the overall cost of drugs.
- 1:08:09SPEAKER_00: And so that if you reduce the spending on drugs,
- 1:08:12SPEAKER_00: you take a dollar out of that total spending,
- 1:08:15SPEAKER_00: some of that dollar is going to come from the PBM,
- 1:08:17SPEAKER_00: which doesn't engage in R&D,
- 1:08:19SPEAKER_00: instead of coming out of the pocket of the drug company,
- 1:08:21SPEAKER_00: which might engage in R&D.
- 1:08:23SPEAKER_00: And that's the argument made for why it is that you don't,
- 1:08:27SPEAKER_00: whenever you cut prices,
- 1:08:27SPEAKER_00: that doesn't translate immediately into a cut in R&D.
- 1:08:30SPEAKER_00: That's the first complication that I want to highlight,
- 1:08:33SPEAKER_00: which is why the simple Econ 101 model isn't exactly right.
- 1:08:37SPEAKER_00: Good intuition, but a little bit different than reality.
- 1:08:39SPEAKER_00: And let me provide another one, even more important, I think.
- 1:08:43So what's really interesting with drugs and with many products
- 1:08:47is that once you do
- 1:08:48SPEAKER_00: the innovation,
- 1:08:49SPEAKER_00: you figure out,
- 1:08:49SPEAKER_00: oh,
- 1:08:49this small molecule or this large molecule has some medical
- 1:08:53benefits and a
- 1:08:54SPEAKER_00: tolerable safety profile.
- 1:08:55SPEAKER_00: We should use it.
- 1:08:57SPEAKER_00: You don't need to do that in a bunch of different countries.
- 1:08:59Once you've done the discovery,
- 1:09:00you can go sell in a bunch of different countries.
- 1:09:02SPEAKER_00: sell in the US, sell in Europe, sell in Japan, you pick your area.
- 1:09:06And the prices that those countries pay,
- 1:09:10in some sense,
- 1:09:11it's compensating you for your R&D.
- 1:09:13So R&D is done in one place,
- 1:09:15but compensated by all these different countries.
- 1:09:18SPEAKER_00: Now,
- 1:09:19you could have a situation where the United States pays a very
- 1:09:23high rate of
- 1:09:23SPEAKER_00: compensation,
- 1:09:25SPEAKER_00: say 280% of what's paid elsewhere.
- 1:09:28SPEAKER_00: And what that means is that the U.S.
- 1:09:29is shouldering a much higher share of the total R&D than other
- 1:09:34developed countries.
- 1:09:38SPEAKER_00: And so you might say, OK, wait, hold on a second.
- 1:09:40SPEAKER_00: What if I decrease payment in the U.S.
- 1:09:44SPEAKER_00: by a dollar and increase it by a dollar in other countries?
- 1:09:48SPEAKER_00: That is not decreasing R&D.
- 1:09:49SPEAKER_00: It's shifting the R&D burden to the foreign market.
- 1:09:54And so when we have these discussions about sometimes called
- 1:09:57most favored nation
- 1:09:58SPEAKER_00: pricing,
- 1:09:58SPEAKER_00: sometimes called international reference pricing,
- 1:10:00it's partly about shifting the cost of that R&D and not so much
- 1:10:05reducing the R&D.
- 1:10:06SPEAKER_00: And so now we have a second reason to think,
- 1:10:09SPEAKER_00: okay,
- 1:10:09there's not an immediate translation between the changing of
- 1:10:11the price in the
- 1:10:12SPEAKER_00: United States and the R&D.
- 1:10:13SPEAKER_00: It either could be because it's coming out of the middleman
- 1:10:16Or it could be because what we're doing is shifting some of
- 1:10:18that R&D burden abroad.
- 1:10:19SPEAKER_00: And when we think about the arguments for and against these policies,
- 1:10:23SPEAKER_00: we have to consider those alternative impacts.
- 1:10:27SPEAKER_01: Sure.
- 1:10:28SPEAKER_01: But are there other levers that we can pull to incentivize biotech R&D?
- 1:10:32SPEAKER_01: Right.
- 1:10:32I mean,
- 1:10:33one of them is just like,
- 1:10:34here's how much we're going to pay you for the drug.
- 1:10:36And so if you make more of those,
- 1:10:37we're probably going to pay you a lot of money.
- 1:10:39SPEAKER_01: You know,
- 1:10:40SPEAKER_01: this is a loaded question I'm asking because at IFP,
- 1:10:43SPEAKER_01: at my parent institution,
- 1:10:45SPEAKER_01: we think a lot about encouraging biotech companies to innovate more,
- 1:10:48SPEAKER_01: to spend more on R&D.
- 1:10:49SPEAKER_01: But I'm hoping you'll answer this question for me as well.
- 1:10:52SPEAKER_00: Yeah, I mean, here's the thing.
- 1:10:54SPEAKER_00: We do have other tools.
- 1:10:54And I think one of the greatest examples is what the first
- 1:10:57Trump administration did
- 1:10:58SPEAKER_00: with Operation Warp Speed.
- 1:11:00SPEAKER_00: We needed a COVID vaccine.
- 1:11:02We went out and we supported companies and we made promises to
- 1:11:06buy vaccines.
- 1:11:07SPEAKER_00: We even went further.
- 1:11:08We like help them with their supply chains to make sure that
- 1:11:10they got access to low cost things.
- 1:11:12SPEAKER_00: We worked with for regulatory relief.
- 1:11:14SPEAKER_00: We did a whole range of things.
- 1:11:16SPEAKER_00: that we could do to spur innovation.
- 1:11:18SPEAKER_00: And an important thing to remember is each of these tools,
- 1:11:20SPEAKER_00: patents,
- 1:11:21SPEAKER_00: which is the traditional mechanism,
- 1:11:23which is will give you a monopoly for a certain number of years
- 1:11:25if you come up with
- 1:11:26SPEAKER_00: something new and that's patentable,
- 1:11:28versus giving you promised commitments to purchase or direct
- 1:11:32subsidies to do R&D.
- 1:11:34SPEAKER_00: Both have a role.
- 1:11:35SPEAKER_00: And roughly speaking,
- 1:11:37SPEAKER_00: and here I'm not providing any deep insight,
- 1:11:39SPEAKER_00: I think this is well known,
- 1:11:40SPEAKER_00: is that if you know what product you want,
- 1:11:43You can directly ask for it and say,
- 1:11:45give me this and I'll pay you money.
- 1:11:48SPEAKER_00: OK, and that's a way to incentivize innovation, that exact same thing.
- 1:11:50SPEAKER_00: So, for example, in the military, we do this.
- 1:11:52SPEAKER_00: We want a tank that can shoot this far and this rapidly.
- 1:11:56SPEAKER_00: We know that.
- 1:11:57SPEAKER_00: We want that for our military strategy.
- 1:11:58SPEAKER_00: We can just pay for that.
- 1:11:59SPEAKER_00: Defense companies will do the R&D.
- 1:12:01SPEAKER_00: They'll come and bid and will incentivize innovation.
- 1:12:04SPEAKER_00: Same thing with the COVID vaccine.
- 1:12:05SPEAKER_00: But there are going to be other times where we don't know.
- 1:12:07SPEAKER_00: what we want.
- 1:12:08SPEAKER_00: And when we don't know what we want,
- 1:12:10SPEAKER_00: we offer a patent and we say,
- 1:12:13SPEAKER_00: hey,
- 1:12:13SPEAKER_00: we'll give you a monopoly if you invent something that's useful.
- 1:12:16SPEAKER_00: We don't know what it is that you want to invent,
- 1:12:18SPEAKER_00: but you have a strong incentive,
- 1:12:20SPEAKER_00: that monopoly,
- 1:12:21SPEAKER_00: to go not only do the R&D,
- 1:12:22SPEAKER_00: but to find out what you should do R&D on.
- 1:12:25And that's kind of the magic of patents is that when you don't
- 1:12:27know,
- 1:12:27you ought to use that.
- 1:12:28SPEAKER_00: And each of these have their pros and cons.
- 1:12:31SPEAKER_00: So,
- 1:12:31SPEAKER_00: for example,
- 1:12:32when you think of the government deciding that the wrong thing
- 1:12:35needs to be built,
- 1:12:36SPEAKER_00: then we might think of that as waste,
- 1:12:38SPEAKER_00: right?
- 1:12:38Like you built something that society doesn't actually need and
- 1:12:40you encouraged R&D
- 1:12:41SPEAKER_00: that was a waste.
- 1:12:42So we should focus that first set that you're talking about or
- 1:12:45the set that IFP is
- 1:12:46SPEAKER_00: also talking about on the things that we know we want.
- 1:12:49And for things that we don't know,
- 1:12:50we want this more decentralized approach.
- 1:12:53SPEAKER_00: And art is in picking those.
- 1:12:56SPEAKER_01: All right, Anup, last question for you.
- 1:12:58SPEAKER_01: On the 30th of July this year,
- 1:13:01SPEAKER_01: there was a big White House event called Make Health Tech Great Again.
- 1:13:06SPEAKER_01: CMS was there and unveiled this big initiative to modernize health data
- 1:13:10SPEAKER_01: interoperability,
- 1:13:12SPEAKER_01: which is a colossal mouthful.
- 1:13:14But my understanding is this is a topic that you've been
- 1:13:17interested in for a little
- 1:13:18SPEAKER_01: while and you may have had something to do with.
- 1:13:20SPEAKER_01: So will you tell me a little bit like, why should I care about
- 1:13:24SPEAKER_01: Again, modernizing health beta interoperability.
- 1:13:26SPEAKER_01: I'm really working on my elocution in this episode, by the way.
- 1:13:28It's really useful to try and say these big,
- 1:13:31chunky phrases five times fast.
- 1:13:34Yeah,
- 1:13:35certainly healthcare offers a number of opportunities to work
- 1:13:38on tongue twisters.
- 1:13:39SPEAKER_00: Look,
- 1:13:39SPEAKER_00: I wasn't responsible what happened at the White House,
- 1:13:41but it is one of the reasons why I'm excited to work with this
- 1:13:44team at CMS.
- 1:13:46SPEAKER_00: So let's step back.
- 1:13:47SPEAKER_00: We want to bend the cost curve.
- 1:13:49SPEAKER_00: We want to do this dynamically.
- 1:13:50We want to affect future technologies that might provide care,
- 1:13:54same quality,
- 1:13:55but at lower cost.
- 1:13:56One way to do that is to ride on the kind of digital revolution
- 1:14:01that we've seen in
- 1:14:02SPEAKER_00: the last 25 years and then accelerate in the last five years with AI.
- 1:14:06SPEAKER_00: You know,
- 1:14:06SPEAKER_00: if I was an economist and speaking to other economists,
- 1:14:09SPEAKER_00: I would actually say what we want to do is increase capital labor ratios.
- 1:14:13That would be the term that we would use with the specific form
- 1:14:16of capital we think
- 1:14:17SPEAKER_00: about as software and data.
- 1:14:19SPEAKER_00: But the key thing is,
- 1:14:20SPEAKER_00: in lay terms,
- 1:14:21SPEAKER_00: it's we want to ride this digital revolution to improve care.
- 1:14:26And that ranges from everything like finding patients the right
- 1:14:30answers to medical
- 1:14:31SPEAKER_00: questions without even seeing a doctor,
- 1:14:33SPEAKER_00: to finding them the right doctor,
- 1:14:35SPEAKER_00: to helping doctors do diagnoses better.
- 1:14:38SPEAKER_00: et cetera.
- 1:14:39SPEAKER_00: None of this displaces the doctor.
- 1:14:41The goal is to actually empower the doctor to be more
- 1:14:43productive and to empower the
- 1:14:44SPEAKER_00: patient to find the treatment that's best for them.
- 1:14:47SPEAKER_00: And to do that by either raising quality,
- 1:14:49SPEAKER_00: holding cost constant,
- 1:14:50SPEAKER_00: or holding quality constant,
- 1:14:51SPEAKER_00: lowering cost.
- 1:14:52SPEAKER_00: Okay.
- 1:14:53SPEAKER_00: But data is the new oil.
- 1:14:56In order to do this digital revolution,
- 1:14:58you need to actually have access to data.
- 1:15:00SPEAKER_00: OK, we need access to both claims data and EHR data that patients have.
- 1:15:07SPEAKER_00: But at the same time, we need to respect two things in getting that data.
- 1:15:10SPEAKER_00: We need to make sure the patient's privacy protected.
- 1:15:12SPEAKER_00: That's very important.
- 1:15:14SPEAKER_00: And the second thing to do is we need to protect IP.
- 1:15:16Sometimes that data actually contains intellectual property
- 1:15:19beyond just the
- 1:15:20SPEAKER_00: information on the patient.
- 1:15:21SPEAKER_00: We need to protect that.
- 1:15:22So our big challenge is how do we make data both protected but
- 1:15:26available so that we
- 1:15:27can add these innovations that are going to create new gen AI
- 1:15:31healthcare tools?
- 1:15:33SPEAKER_00: Well, there's a lot that you can do there.
- 1:15:35And what we want to think about is broadly thinking about a
- 1:15:38healthcare data
- 1:15:39SPEAKER_00: marketplace that protects that underlying data,
- 1:15:41SPEAKER_00: but also allows it to be utilized to improve technology.
- 1:15:44SPEAKER_00: And that usually involves four things.
- 1:15:47SPEAKER_00: And this is, you can study other markets, agricultural markets,
- 1:15:51SPEAKER_00: equity derivative markets, water markets, energy markets.
- 1:15:54SPEAKER_00: There's all these markets.
- 1:15:55SPEAKER_00: Basically, we do four things.
- 1:15:57SPEAKER_00: Sure.
- 1:15:57SPEAKER_00: We reduce transaction costs.
- 1:15:59SPEAKER_00: We use network effects to kind of grow certain markets.
- 1:16:03SPEAKER_00: We provide some supply subsidies and we provide some demand subsidies.
- 1:16:07SPEAKER_00: And here, the thing that's being traded in some sense is data.
- 1:16:10SPEAKER_00: So let's take that first thing.
- 1:16:12That White House announcement on July 30th was an attempt to
- 1:16:14reduce the transaction
- 1:16:16SPEAKER_00: costs associated with
- 1:16:18SPEAKER_00: data markets for healthcare.
- 1:16:19SPEAKER_00: And that is to make sure that people have identifiers,
- 1:16:23SPEAKER_00: common identifiers for providers,
- 1:16:25SPEAKER_00: for patients,
- 1:16:25and that there's a security system that ensures that the data
- 1:16:28is protected and it
- 1:16:29SPEAKER_00: cannot be leaked.
- 1:16:31And that basically is like the equivalent of building rails for
- 1:16:34railroads,
- 1:16:35something like that.
- 1:16:35SPEAKER_00: But here, what you're transporting is data.
- 1:16:37SPEAKER_00: There's a whole bunch of other examples we can do on that.
- 1:16:40SPEAKER_01: But just to be precise here,
- 1:16:42what is the administration doing and announcing here this July
- 1:16:4530th event to build
- 1:16:47SPEAKER_01: those rails for the railroads?
- 1:16:49SPEAKER_00: Building a set of protocols for patient IDs,
- 1:16:52SPEAKER_00: for provider IDs,
- 1:16:53SPEAKER_00: and for encryption that ensure that patient data can flow.
- 1:16:57SPEAKER_00: So let me give you a very concrete example.
- 1:16:59My experience with going to doctor's office is that you spend
- 1:17:01the first 20 minutes
- 1:17:02SPEAKER_00: filling out a paper on a clipboard,
- 1:17:04SPEAKER_00: filling out some paper information that,
- 1:17:06SPEAKER_00: like,
- 1:17:07SPEAKER_00: I don't understand why they don't already have.
- 1:17:09And part of that is because there's a lot of friction to the
- 1:17:12flow of information,
- 1:17:13SPEAKER_00: both within and across organizations.
- 1:17:16SPEAKER_00: And what we want to do is we want to make it so that happens,
- 1:17:18SPEAKER_00: but at the same time,
- 1:17:19SPEAKER_00: information doesn't get leaked so that your privacy is protected.
- 1:17:22SPEAKER_00: So if we have a protocol for that,
- 1:17:23SPEAKER_00: that means that whenever you go to the doctor,
- 1:17:25SPEAKER_00: you know,
- 1:17:26SPEAKER_00: ideally,
- 1:17:26SPEAKER_00: they would just know you.
- 1:17:27SPEAKER_00: Oh, Santi showed up.
- 1:17:28SPEAKER_00: Okay.
- 1:17:29SPEAKER_00: And then you go to the dentist and Santi showed up.
- 1:17:31SPEAKER_00: No paper forms to fill out.
- 1:17:32SPEAKER_00: Maybe they'll ask you questions if there's something new.
- 1:17:35SPEAKER_00: but not your old stuff.
- 1:17:36SPEAKER_00: And so your experience with the physician should be much better.
- 1:17:40SPEAKER_00: But it'll transcend not just this,
- 1:17:42but a whole bunch of different interactions and encounters you
- 1:17:44have with providers
- 1:17:45SPEAKER_00: and patients.
- 1:17:47SPEAKER_00: So these are the sorts of things we want to reduce those frictions.
- 1:17:50SPEAKER_00: And this is just the tip of the iceberg.
- 1:17:52There's a whole bunch of other things that you can do to reduce
- 1:17:54those frictions,
- 1:17:54SPEAKER_00: those transaction costs.
- 1:17:56SPEAKER_00: But the next thing we want to do is try to figure out,
- 1:17:57SPEAKER_00: well,
- 1:17:58how can we encourage this market to work more efficiently and
- 1:18:01grow that market?
- 1:18:02SPEAKER_00: How can we encourage...
- 1:18:04SPEAKER_00: physicians and providers to gather data with greater fidelity,
- 1:18:08meaning like make sure you take proper patient records without
- 1:18:11increasing the time
- 1:18:12SPEAKER_00: that doctors have to do spend on it.
- 1:18:14SPEAKER_00: So for example,
- 1:18:15SPEAKER_00: AI scribes,
- 1:18:16can that make the doctor's life better and also give us better
- 1:18:18records at the same
- 1:18:19SPEAKER_00: time?
- 1:18:20Can we encourage hospitals to cooperate with each other so that
- 1:18:23when patients go
- 1:18:24SPEAKER_00: from one hospital to another,
- 1:18:25SPEAKER_00: their data is shared efficiently,
- 1:18:27SPEAKER_00: and so there's no medical errors?
- 1:18:29SPEAKER_00: So those are supply incentives.
- 1:18:30SPEAKER_00: And on the demand side,
- 1:18:31can we encourage providers to really start leveraging data to
- 1:18:36improve the care that
- 1:18:37SPEAKER_00: they provide to patients?
- 1:18:39So this is kind of a multi-pronged approach to build this
- 1:18:43infrastructure that we
- 1:18:44SPEAKER_00: think is important to make sure that we really can apply AI
- 1:18:48to improve healthcare,
- 1:18:50lower cost,
- 1:18:51maintain quality,
- 1:18:51or improve quality,
- 1:18:52maintain cost.
- 1:18:54SPEAKER_01: Sure.
- 1:18:54SPEAKER_01: And just to get one level deeper before we close,
- 1:18:56SPEAKER_01: what are the pieces of data that you're going to have to unlock here?
- 1:18:59SPEAKER_01: I mean,
- 1:18:59SPEAKER_01: one piece that I understand is like just I've got my medical records are
- 1:19:04distributed across a bunch of different hospitals and maybe
- 1:19:06that's integrated by
- 1:19:07SPEAKER_01: Epic sometimes in the back end.
- 1:19:09SPEAKER_01: But just like the facts about me medically are all over the place.
- 1:19:14SPEAKER_01: And maybe you can do some integration there.
- 1:19:16SPEAKER_01: But then there's also like the claims data, the insurance stuff.
- 1:19:19SPEAKER_01: That's a whole nother question, right?
- 1:19:21SPEAKER_01: What does my insurer have about me?
- 1:19:23SPEAKER_01: It's a separate.
- 1:19:24body of information so this is about my pay grade the question
- 1:19:28i'm asking you know
- 1:19:28the answer is gonna be about my pay grade but how do you go
- 1:19:31about building a system
- 1:19:32where all that data can be usefully integrated because i can
- 1:19:36imagine very nice for
- 1:19:38SPEAKER_01: me personally as a patient to be able to access all this stuff and i like
- 1:19:43SPEAKER_01: Doing my ACL, I've talked a lot to LLMs about my own health situation.
- 1:19:48SPEAKER_01: And in some cases,
- 1:19:49it's been meaningfully more useful than specific providers'
- 1:19:52perspectives have been
- 1:19:53SPEAKER_01: without indicting the medical profession.
- 1:19:56But what's the world in which you can actually pull all this
- 1:19:58stuff together and I
- 1:20:00SPEAKER_01: can dump it all in one place?
- 1:20:03SPEAKER_00: So,
- 1:20:03SPEAKER_00: I mean,
- 1:20:03SPEAKER_00: if you think a little bit more like a data scientist,
- 1:20:06SPEAKER_00: less like an economist,
- 1:20:07SPEAKER_00: you kind of say,
- 1:20:07SPEAKER_00: you know,
- 1:20:08SPEAKER_00: what would make my life easier to basically manage a database?
- 1:20:12SPEAKER_00: So you want your fields to be very similar.
- 1:20:15SPEAKER_00: You want the data to be in a machine readable format.
- 1:20:18SPEAKER_00: And that kind of uses a common format.
- 1:20:20You want to have field attributes that tell you how to deal
- 1:20:23with it and protect
- 1:20:25those data and what data you can transmit and what data you
- 1:20:27have to very much
- 1:20:28SPEAKER_00: protect.
- 1:20:29That's the way that I would view this problem when it comes to
- 1:20:33thinking about how
- 1:20:33SPEAKER_00: to build this.
- 1:20:34SPEAKER_00: Now,
- 1:20:34if you want to think about how you go from that ideal database
- 1:20:39where you're
- 1:20:39SPEAKER_00: protecting the data but using it to the best of its ability,
- 1:20:42SPEAKER_00: you also need to think about how am I going to get from here to there?
- 1:20:45SPEAKER_00: Now, in an ideal world, payers and hospitals would all get together.
- 1:20:50They'd agree on a protocol and shake hands and maybe have a
- 1:20:53beer after work to celebrate.
- 1:20:55SPEAKER_00: But that's kind of hard.
- 1:20:56SPEAKER_00: It's hard because they're at different levels.
- 1:20:58SPEAKER_00: Some people have already gathered data.
- 1:21:00SPEAKER_00: They've got their own databases.
- 1:21:01SPEAKER_00: Other people are not gathering data,
- 1:21:03SPEAKER_00: but they've got this great patient population that's really useful.
- 1:21:06So there are different stages of development,
- 1:21:08and they might not all agree on protocols.
- 1:21:11SPEAKER_00: even though they would all benefit from agreeing on protocols.
- 1:21:14SPEAKER_00: And part of the role that the government can play is just to say,
- 1:21:16SPEAKER_00: hey,
- 1:21:16SPEAKER_00: guys,
- 1:21:17let's all sit down in a room and figure this out and figure out
- 1:21:20some rules.
- 1:21:21SPEAKER_00: Obviously,
- 1:21:21SPEAKER_00: the benefits might not be evenly distributed,
- 1:21:24SPEAKER_00: and we have to address that in any kind of discussion like this.
- 1:21:27But the government can play a role in just making sure that the
- 1:21:29private sector does
- 1:21:30come together to have this kind of agreement to build this
- 1:21:34marketplace in.
- 1:21:34SPEAKER_00: And I think that's the role that we would want to play,
- 1:21:39SPEAKER_00: knowing that it's not about the government building a database.
- 1:21:41SPEAKER_00: That's not something we want.
- 1:21:43SPEAKER_00: What we want is the private sector to go build this database,
- 1:21:46SPEAKER_00: to build this data infrastructure,
- 1:21:49SPEAKER_00: and then use that to improve the care that they provide.
- 1:21:51SPEAKER_01: Sure.
- 1:21:52SPEAKER_01: And this has been a blast and we've gone way over time.
- 1:21:55SPEAKER_01: So I really appreciate your patience with my fumbling here.
- 1:21:58SPEAKER_00: No worries.
- 1:21:59SPEAKER_00: Thank you very much for the time.
- 1:22:00SPEAKER_00: I appreciate it.