How to Save the Most Lives

And crucially, we in the US do not get better health outcomes, So we're spending lots of money without much to show for it. For a long time, people have looked at this problem and thought, what if instead of paying doctors and hospitals to treat us when we get sick, we could pay them to keep us healthy. Is there some way we could spend less on healthcare and get better results. I'm Jacob Goldstein, and this is what's your problem. My guest today is Farzad Mosta Shari. He's the co

founder and CEO of a healthcare company called Alde. Farzod's problem is this, how can you pay doctors to keep us healthy rather than treating us after we get sick. People have been struggling to solve this problem for decades, but for a bunch of reasons you'll hear about on the show, Farzad and his fellow travelers may be the

ones to finally solve them. Before he became an entrepreneur, Farzad went to medical school, but he told me that back in the mid nineties, when he was doing his clinical training, his residency and internal medicine, he started to question what he was and what he was not learning.

I kept looking at stuff that was happening and being like, this doesn't make sense to me, Like why, like you're telling you're teaching me how to you know, treat someone who's coming in here with difficulty breeding in the emergency room. I get that, Like it's really important that that person get their treatment, their nebulizer so they can breathe again, that's really I get that. But why is this person coming in today? Why is it all of a sudden

we're seeing people coming in with breeding problems. Why why is it that the people who tend to come in with breeding problems tend to be from these neighborhoods, not other neighborhoods. And it was those questions that were between medicine and public health and ibidemology that really grabbed me so far as Odd finished his residency and went into public health, and he had this this trait that at the time was kind of a superpower. He was really

into computers. So I came of age at a time when in a way, you know, asking the question can you use a computer for that was was like, uh, you know, unlocked a lot of a lot of value. And so I joined the New York City Health Department and I said, oh, public health surveillance, can we use a computer for that? Right? So one of my first startup, Jacob, was a within the New York City Department of Health

where I started. Was one of the group of people who started this national idea of why don't we why don't we track when when someone gets logged into the emergency room for just for their registering the patient in the emergency room, or when an ambulance gets dispatched, or when when when some item in the pharmacy goes beep across the scanner. Like before all those things were on paper.

Now they're all electronic, which means I can get a copy and if I can get a copy, then I can tell you in real time that something's happening to the health of the city. Right that was That was like consumed me for five to ten years, and it was really really cool. It was really really fun, and that system TILT today is the system that we use, for example, during COVID to see that what was happening

in the city. But then something fundamentally shifted in my life, which was Mike Bloomberg asked the Commissioner of Health at the time, who turned to me. I was like the data guy, and the question he asked was, don't worry about the politics, don't worry about the budget, don't worry about any of that. Your job is to answer the question. What's the question? And the question was what's the problem? What's your problem? As you might say, the question was, hey,

how do we save the most lives? And I was like can I swear? Oh yeah? Fuck? That was like what am I doing? Like? What am I doing? Why am I not starting with that question? That is such the right question, how do we save the most lives? And it turned out that no one in like, we're spending three four trillion dollars, We're doing all this activity right,

and no one was starting from that viewpoint. There were hundreds of thousands of research articles written about various things that we're doing, and no one was saying, what is this the scientific, evidence based answer to the question how do we save the most lives? And it turned out, in New York City in the year of Our Lord two thousand and two, the answer to that was smoking. The answer to that was we're going to attack smoking.

So that meant that we were going to dedicate more of our living waking hours to how do we reduce smoking in New York City than we do to anything else. And within the time span of four years, we drove I was a small part of a big effort to drive teen smoking from fourteen to seven percent in New York City, to drive adult smoking from twenty two point eight percent to sixteen point eight percent. That saved more lives than anything else we could have possibly done. And

then the next question was, Okay, that's public health. That's great, that's changing the environment within which people make decisions. But we know what we're spending all this money on healthcare? What can healthcare do to save the most lives? Just stay with that question. That's such a clarifying question that is, so that is my flaming sword, right, It's like, how can we save how can medicine? How can healthcare save the most lives? And again no one had the answer.

So we published this paper, Randy analysis. Do you want to take a guess, Jacob, as to what thing in healthcare? If we you know, we have thousands of evidence based guidelines and protocols and quality measures, and like, what is the one thing? And then phrase the question, well that if you took the performance of the American healthcare system from where it is today to where the best American healthcare institutions can perform, would reduce the number of premature

deaths by the greatest amount. Do not have any idea what that would be? Based on my preparation for this interview, I'm gonna guess lowering people's blood pressure. Yeah all right, see that's not fair because you've been listening. I did my homework. Homework, you did your homework. But the funny thing is that if you ask that question of you know, hospital CEOs, they'll say, like getting stroke patients to the

hospital ten minutes faster something. Yeah. Yeah, they'll say, like, let's not kill people in the hospital through infections, they'll say, right, well, because it's not really traditionally a hospital CEO's job to keep people out of the hospital, which is what you want to do if you want people not to die,

right exactly. But even if you if you talk to health planned people that they don't know right there, if you if you look at the number of quality measures that doctors are incentivized and measured against, right, like, there's not a there's not a ten x difference between the amount of importance placed on blood pressure control. Then there is the amount of importance placed on mamograms and colonoscopies and flu vaccines and all that other stuff. Right, it's

all good stuff. And you're saying the evidence as the efficacy of lowering blood pressure means there should be a ten x difference between blood pressure and right, it's just gigantic and it's sable, right like, And it's just science of blow Like we know there's cheap, cheap, safe drugs, there's life like, we know how to lower people's blood pressure.

And and then that was the that became my problem. Right, It's like, well, why aren't we why aren't we why aren't we controlling blood pressure, and I came to the wrong answer. And the reason I came to the wrong answer was because I tended to be the kind of person who said, can use a computer for that. You saw the computer as it was, It was your hammer. So every problem looked like a nail that could be

hit with the computer. And look, there was some truth to it, right, Like the truth to it was, you can't improve blood pressure control if you can't even know what blood pressure control rates are, right, Because you actually couldn't tell what blood pressure control rates were in any given practice, any given population, because the data was trapped in deadwood that there were you would write it down,

You would write down the person's bloodressure control. How would I even know, if I'm a doctor what percent of my patients had their blood pressure control? I literally couldn't, with a room full of paper charts, answer that question. I couldn't have quality improvement. I couldn't have decisions support that would prompt me when I'm seeing a patient. I couldn't have a registry that made a list of the patients to say, can you use a computer for that? Right?

As your friend that problem with the computer, yeah, so then so then we and we had a big concern around around what was now called health equity. So we went around to the doctors in New York City's poorest neighborhoods, in Harlem, in the South Bronx, in Central Brooklyn, bed Stye, and we said, hey, guess what you've been Medicine has been using paper and pen for thousands of years. We're going to give you a free electronic health record with

all of these functionalities built into it. And we spent the next year and a half building into the h ARE the functionality electronic health record, the functionality that we thought should exist in those electronic health records, and then we did that. We rolled it out to half of all those primary care docs in New York cities poorest neighborhoods. So mission accomplished. But you know, you want to guess what happened to blood pressure control rates? They did not improve.

They did not improve, And the number of strokes and the number of heart attacks and the number of kidney failures from that continued just as much. And so that was my continued frustrating He said, what was your frustration. That's my frustration is, um, I didn't feel like I felt like we'd done all this work, and at the end of the day, the reason we'd done all that

work hadn't hadn't been accomplished. And I reluctantly, you know, for twenty years, Jacob, I had like I was a public health guy, I was a government person, like like I was not a private sector person. And in fact, I kind of thought that the profit motive was the problem in many cases. And that's right, and it's wrong.

And what I reluctantly came to see is that the financial incentives are the water we swimming and and and that sometimes we and even see it likes it shapes the force of everything around us, the gravity that pulls on us or not right, like everything around us is there's this invisible thing which is the financial incentives of the system. And I was trying to ignore them. I

was trying to pretend like they didn't exist. And in fact, the financial incentives of the healthcare system are to treat strokes. They are not to No one makes billions of dollars preventing strokes. No one makes you make money waiting until literally waiting until someone's kidney fails, and then there's billions of dollars available to you. For treatment of like taking someone's blood out of their body, running it through a machine and returning it to them. There's a lot of

money in that, not needing that. There's not a lot of money doing that. You're talking about strokes and kidney fail because we know that lowering blood pressure, which is relatively cheap to do, lowers the risk of strokes and kidney failure. But you're saying we don't lower blood pressure, and that happens because that's literally the money in healthcare flows, because doctors and hospitals get money for treating strokes and

kidney failure and not for lowering blood pressure. Yeah. I mean there's one the one part of the healthcare system that is supposed to be in charge of the controlling the blood pressure, which is primary care. Right, it's the

primary care part. And guess what part of our healthcare system makes the least money, makes the you know, has the least prestige in in where I trained, right, it's it's Oh, those those people who you know they talk to patients, they don't they don't you know, they don't cut the patients, they don't do procedures on them, they don't do expensive scans on them, right, They sit in a little room and they talk to the patient about why they should take their medicine. Right. Oh, those those

those poor, those poor little primary care doctors. We spend five percent of healthcare costs on primary care, five percent. We spend as much on ICUs as we do on primary and even primary care doctors traditionally do not get more money when their patients stay healthy. Right. And I know primary care doctors are not motivated by money explicitly, but but still the incentives don't push primary care doctors to reduce their patient's blood pressure though they I'm sure

they want to, they don't get paid more if they do. Traditionally, until twenty twelve, big moment, big narrative moment happening, big moment, say healthcare economics, Cliffhanger. We'll get to that big moment and to how it led Farza to start his company in just a minute. Okay, now back to the show to find out what exactly happened in twenty twelve. In twenty twelves, the rules for accountable care are published by

the Center for Medicare and Medicaid Services. That basically says, if you get a group of primary care dots together, and you gain accountability for the total cost of care for the patients they care for. If those costs come down, we will give you half of that those savings. And it was like it was like the heavens had opened up, right, And I was like, oh my god. And I was like, primary care docs don't realize the power they have in

this new world. And the only problem, right was that if you're if you're an independent, small primary care practice, you don't really have the scale to to band together with one hundred other groups to pool your patients, to take that actuarial risk, to understand the regulations, to have the data systems to know what the playbook is to actually reduce costs, right. And I was like, yeah, but that's easy. That's the easy stop you for that. You

can use the computer for that. Someone's good. I was like, someone's gonna build a business doing this. So I just want to I just want to make sure that we're really clear about what rules are actually changing here, right Yeah, So okay, so um so the traditional way that almost everybody, including medicare, the federal program that pays for old people's healthcare. It's but probably the most important sort of payer and healthcare in this country. Right, they pay doctors fee for service.

You do your doctor, you do something for a patient, you get money. And so what's happening in two thousand twelve is the government is saying Medicare is offering doctors, primary care doctors in particular, a different way to get paid, right, and they call it accountable care, Like, what exactly does

that mean? So that means that if you submit an application to the government, your primary care doctor, you and your friends submit an application to the government, and the government says, oh, you care for these named human beings, right, you're the primary care doctor for these Medicare beneficiaries. So if you have ten thousand Medicare beneficiaries who are getting primary care from you, the Medicare actuaries, that's their job.

They look to see who are those people, what have their costs been in the past, how sick are they? And they're going to say, ah, next year, we think those ten thousand lives are going to cost us a hundred million dollars. Now, if the actual costs, once all the claims are submitted and all the costs are counted up, all of the hospitalizations, all the specialist visits, all the medications and scans and everything else that gets done to

the patient. If that actual cost comes in not at one hundred million, which was our budget, but at ninety million, there's ten million dollars of savings, and we will share those savings with these docs. We will give say half of those savings to the docks. So of that ten million dollars of savings, the government gets to keep five million they spent less, and the docks get five million. And you have to prove that you actually improved quality

while you reduced cost of care. It's a fundamental rewriting of the incentive structure of American healthcare. So you see this new thing in the world, and what do you think. I think, Who's someone's going to build this business that does this? And I went to the Brookings Institute where they had done a lot of the foundational work on this idea, and I created a learning network to teach other people how to do it, and we published a tool kit, and we had a learning collaborative. And I

kept waiting for nine months. I kept waiting for I was like literally every morning I'd wake up and I'd be like, Who's Who's starting a company to do this. Where where is that come? It's such an obvious idea, I thought to myself, it is like and the obvious thing for me was, well, I ain't going to be the hospitals, right, the most powerful people in healthcare today

are hospitals and health systems under one incentive structure. But getting those people, the incumbents, the ones who are powerful in the current system, the goliaths, getting them to change their economic model, well, that's never going to happen. Whereas you go to a you know, we got a group of primary care dogs, You're like, hey, you got nothing to lose but your chains. Yeah, they already don't like

the system. Right with primary care dots. If you're starting with people who are disgruntled as a as a general rule, are appropriately pissed off of the current system, doesn't help them take the care of their patients that they believe their patients should be getting. So okay, every day you're looking for somebody to start this obvious company doesn't happen. And so I went for a walk with the VC and that metaphor or literal did you actually go literal

without Yes? Yes I did, Yes I did. They got it right, away, and they were, you know, they saw the long term opportunity for combining information and incentives in a new way and creating a new business model, and so they like, literally we had one one pitch meeting after that, and they they they gave us the money to start the company. To to my co founder and

I to Guvees who had never started business before. Were there moments as your getting the company going, as it's growing, when there were things that were harder than you expected, things that didn't work that you thought would work. I mean, is this obvious idea to you? Yeah? Was it obvious to everybody else? Well, obviously not. And I think the biggest thing that didn't make sense to people was the going after these independent practices community docs as opposed to

the big, gleaming health centers. But the thing that really shook us was we we went out and we signed up our first groups of doctors in four states, and it felt just charmed. Everything was magic, And we built our software system and we started analyzing the data and we started doing the stuff that we knew was the right things to do for patients, and wholly commole hospitalization rates came down measurably within that first year seven percent.

It worked. It worked, and it was like and it was you know a lot of elbow grease and grit and whatever, but just just churning the focus of these primary care docs on reducing r visits, reducing hospitalizations was working. And I was like, oh my god, charmed life, right, Like, this thing is, this thing is gonna work, gonna totally

take off. And then we got the results from our first year's performance, and sure enough, total hospitalizations went down by seven percent, the biggest source of suffering in American healthcare. Quality went up and costs did not come down. Huh So it didn't work. It seemed like it was working, but the core bottom line things, they didn't come down,

and we only get paid. Like we had done all this work, burned all this cash, hired all these people, gotten all the hopes of these doctors up right, and and we and these investors and we told them, hey, it's working. We're gonna get paid. And you don't get paid for improving quality or reducing hospitalizations. You get paid if total cost of care comes down. And when we looked into it and we actually published a paper, you're like, this sucked. Let's tell the world, how much pain we're

in and what happened? And so what we learned was that the hospitals, when faced with a decrease in utilization, changed their coding so they got paid more for admission. And to be clear, changing their coding means changing the way they bill, so they're billing so that they get more. There are fewer patients coming to the hospital, they're billing more for each patient. For each patient. Yeah, so okay, so so far you're not making any money, right because

you make money by saving money. What do you how do you respond? Well, we had also in the in the intim we had, we had massively expanded, So not only were we waiting to see if we were going to get paid, but we went out there and on the success of the vision and the mission and the tools we built and the early results we got, we signed up a hundred practices in five new states and

we had to make it work. So we went to five states and fortunately in that first year with this new batch, one out of five states the docs got paid. If the initial problem is that, as the doctors you're working with, our lowering admission rates to hospitals, hospitals are effectively charging more for each admission. How do you solve that? How do you how do you fix that? There's only a limit to how much you can play those games,

how much the hospitals can play those games? Yeah? Yeah, yeah, so you so you just drive down admissions even more, even more. Yeah, And were you able to do that? Yeah? Yeah, So we're now down seventeen percent. Hospitalizations are down seventeen percent in for our patients compared to what they would be if they were just like everybody else in their communities.

So that's gigantic. So you're saying, same kind of patients you match the you know, big batch of patients, thousands of patients, if they're working with the doctors you work with, they're seventeen percent less likely to be admitted to the hospital. Correct? Is that at all a function of selection bias? The fact that docs who sign up with you already have a lower admission rate among their patients because that's what

they're into. I mean, that's such a big effect. It is, And this is compared to themselves, So it's it's it's comparing our practices compared to themselves before, compared to non practices compared to themselves before right. So it's the same docs, the same doctors, the patients get admitted to the hospital seventeen percent less often once they sign up with you. Then, yeah, how how do you do that? I mean, I understand that incentives are powerful. Is it just the money? No,

can use a computer for that. Um, it's it's actually creating data to target the right people for the right interventions. It's doing the basics better. So the fundamental thing that we're doing is we're increasing more access to primary care. So we get thirty four percent more visits to primary care doctors for the patients in particular who need it the most. And I just want it. There's there was a phrase you used in there that I just want to grab for a second. It was not just that

it's easier to see the doctor. It's easier to see the doctor for the patients that need it the most. That's right. It seems like that's a really important phase is figuring out who those patients that actually need to see their price marycare doctor or seeing them when they need to be seen like that seems hard, Yeah, and

that is where the technology comes in. My father in law was CEO of a health plan and before that he had started one of the health insurance company health insurance company, and before that he had started one of the first HMOs in the country. And he said to me, as far as how's this different than what we did fifty years ago? And I showed him our our technology and he was like, Okay, that's what's different. So how

does it actually work? So how does it work that a doctor knows what patients they actually need to see or need to be available for. Yeah, so nine of our practices use our platform, our software platform every day. Yesterday, our practices were in the tool and what they do. When you log into the tool, the first thing you see is you had three patients who went to the emergency room yesterday, and we think that within forty eight

hours you should call them. And here's here's the tool that helps you call them and you can now send them a text message. And just to be clear, like most doctors when their patient goes to the emergency room, how do they find out or that they don't They don't. They don't they don't because how would they be right?

It's just like how would the doctor is over here in some clinic and then the patient is just driving to the emergency room or whatever getting driven to the emergency room, and they're like separate islands exactly, even when even when the doctor's part of the same system as the emergency room, there's no system to inform them, there's no system to call the patient. There's no system for that. Why because there's no incentive to do that. Yeah, nobody gets paid more to if they do that, And now

there is, and now there is. Right, So that's that's one of the things you'll see. The other thing you'll see is you'll see it that says, hey, you know what, here's a list of patients who you haven't seen in a while, who you know they might need your help. They're pretty got pretty complex conditions, or their last blood pressure was pretty high or whatever. Right, here's a list. But you don't need to worry about that. Here's a list of patients for your scheduling person, the front desk

person between patients. When she's got a few minutes, she should call these patients and make an appointment for them to come back into primary care. Right, So that's the wellness work Listen, when they come in once a year, like let's have a visit. That's not based on you reaching out to me because you've got something wrong with you. But once a year, let's have a visit where we can just sit and we can talk, and we can

see how do we keep you healthy? What are the things that are going to be problems for you and prevent them? And you've found that it's profitable under this new system to do that. The practices make about fifty percent more per Medicare patient who's in these models fifty percent more, and they're they're like scrapping and begging for like a two or three or four percent rate increase, and they're getting a fifty percent increase through this model.

And is allidate's business model to take some chunk of the additional Medicare pay basically that the doctors get. Is that the model, that's the model. The model is we only get paid when the practices get rewarded and the government saves money. Are the practices on the hook if their patients wind up costing medicare more than expected? Great question.

So we do start with what's called one sided models, where it's it's it's like reward only, but then we move pretty quickly to two sided models and we backstop that risk for the practices. So so you take the you're on the hook and the practice is not on the hook. Correct. So I feel like I where you came from and where you are now, and I'm curious where you're going next, and in particular, you know, what are the things you're trying to figure out that you

haven't yet cracked. We're in this in order to reduce hospitalizations and complications and bad stuff happening to patients, and we need to continue and there's so much every day I see the opportunity that we're leaving behind. Tell me more about that, I mean, and specifically when you talk about seeing opportunities that you're leaving on the table, Like, what's a specific example of a thing there you look at you're like, oh my god, we can help people

and we're not doing it. What's an example of that? Kidney? Let me talk about kidney. So when someone's kidneys fail, it's a terrible thing, right where your blood is no longer being cleaned. They oftentimes it happens unexpectedly and the so called crash into needing dialysis, they get hospitalized. There's actually a big mortality rate for that when people actually die during that and never get a chance right to make it to the next stage. We can predict who's

going to crash into dioss with pretty good accuracy. So because we have all of this information on the patient before upstream of when that happens in the earlier stages of kidney impairment, we can do a really good job using machine learning AI models of predicting who's going to crash into diosis. We can steer those patients because we have the primary care relationship with them. We can do a handoff from primary care to kidney specialists and kidney

educators and work with the patients. And what we're seeing is at twenty four percent lower rate of hospitalizations twenty four percent for patients who've got that handoff into the kidney care management program to dialysis. And when that happens, presumably hospitalization goes down. Does mortality even go down? People die less if you do this. I haven't proven that yet, but I think it does well. So are you doing that already or is that a thing you think you

can do but that you're not quite doing yet. So we have a unit that does these tests, these experiments, and then when they work, we scale them. So this is literally hot off the presses. We got the results

from the pilot. It seems promising, and we're going to scale it and you can imagine like you can imagine that same sort of rinse and repeat right for a whole host of other of other things of just finding ways to be smarter about figuring out who is about to get really sick essentially, that's right exactly, and intervening either ideally to prevent them from getting really sick or at least to reduce the bad outcomes associated with getting really sick. You got it. We'll be back in a

minute with the lightning round. Now back to the show. Okay, let's close with the lightning round. Okay, Um, what you ran health? I t for the Obama administration. So what is the least glamorous thing about working in the White House? Um? We had to pay for our own water like literally, what's yeah? Why? Literally we had a we had because

the government is I mean, you really learn frugality. People have no idea Like in the government, we had a water club and you had to put in twenty bucks for to have the privilege of drinking from you know, the filtered water from from the water cooler. Uh, that is what you're That is the spirit of frugality that people don't understand. The government agencies have. What's why? Piece of advice you'd give to someone trying to solve a hard problem. Spend a lot of time asking yourself, what's

the question? What's the what's the heart of the problem? Um, and I think you should change the name of your podcast to what's part of the problem? What do you think I should change it to to to what's the heart of the problem? Huh, you don't like it? I like I wonder about the problem framing at all, frankly as a as a as a name for the podcast I've come. I love it. I love it. I love it. That's why. That's that's that's why I love your podcast.

I think it's the most important thing is to be working on the most important thing, and and we don't. We don't work on the most important thing. It goes back to that question you were talking about at the beginning, how can we save the most lives? That is that's my question, right, that's the that's that's that's the problem that I'm trying to solve. But for every every entrepreneur, what's the heart of the problem you're trying to solve and how can you how can you get to that?

And what I found is people don't really don't really ask themselves and iterate enough on do I have it right? Is that really the heart of the problem is that the question what was the last hike you went on? Ah? I love hiking, and I walk on the Capitol Crescent Trail and the Ciano Canal all day. Like some days I'll walk eight hours during the workday with my phone and my tablet and my backup battery. So the last hike I did, was it a long hike on the

Ciano Canal towpath? And was it, in fact a workday where you just walk in for eight hours and running your company? Yeah, that's a nice move. That's a very strong move. If everything goes well, What problem will you be trying to solve in five years hospitals? I think we've we've started where it's easier, where the incentive alignment is easier with independent primary care, and now we're adding

on people around that. But particularly in rural areas, we're going to need to solve the rural hospital problem and maybe maybe maybe we can help flip them to a different model. Farzad Mosta Shari is the co founder and CEO of Allidate. Today's show was produced by Edith Russlo, edited by Robert Smith, and engineered by Amanda k Wong. I'd love to get your suggestions for who else I should talk to for the show. You can email us at problem at Pushkin dot fm, or you can find

me at Twitter at Jacob Goldstein. I am Jacob Goldstein and we will be back next week with another episode of What's Your Problem.