Astrana Health, Inc. (ASTH) Earnings Call Transcript & Summary

I remember when we were writing the initiation for Cano Health, we wrote about DMG [indiscernible] medical group that started in one market and was really successful and then they had trouble expanding, and we wrote that as like a risk, what could happen to a primary care group in Cano ended up being in that exact scenario where they were successful in Florida, and they expanded into new markets and ran into trouble in terms of like how things performed versus expectations and ultimately, didn't have a good outcome. And so like how do you prevent that scenario from happening. Like what did -- I guess you don't know what they did wrong, but like, from an investor perspective, it's like you're expanding into a bunch of new markets. Clearly, the model works in one, but like where are you drawing a line like, okay, Nevada and Texas are ramping as we thought. And therefore, this can export. Like what is the data point that gives you that confidence?

Yes. So the really interesting thing is, there probably aren't, if any, certainly not many platforms that have made it work in multiple markets as successfully as they have made their first market work. However, there do exist models that work in each given market. In Florida, there is something that works in New York and Texas and wherever. So probably Optum is the only one that has managed to kind of make it work everywhere. And so the interesting thing is if there are models that actually work, why can they not work when under the same umbrella of one company. And I actually think the reason is because companies often become too dogmatic in terms of believing their care model is the only care model that works instead of understanding that there could be different care models, different modalities of engaging physicians and patients that need to be taken when you go to a different region. For example, hypothetically, if someone were to create a holdings company that just acquired assets in different states and just let them run as they were already running. In theory, that combined holdings company "would work" across the different markets. And one might argue that, that kind of looks like Optum for example. So why couldn't a company actually make it work is because they aren't flexible enough with their model or their model doesn't have the configurability to support different types of modality of how a community wants to experience their healthcare. I think that's where our model really shines. As you know, we have the Care Partners business. We can go out and affiliate with independent doctors, literally on an office-by-office basis or we can affiliate with large groups as we've done with CFC or BASS or others. We can go in an employee providers the de novo clinic setting or we can acquire clinics and integrate them into our clinic business in the Care Delivery segment. Here in Nevada, for example, tomorrow, we're doing a grand opening for our newest Blue Diamond clinic, a quick plug here, which was a de novo opening. And we can support providers via the Care Enablement technology platform as well flexibly, if that's what they want to do, which we did with CFC prior to the acquisition. So I think the key is, as we go into new markets, we're figuring out which modalities work the best in that market, at least to enter, rather than imposing the view that, for example, all markets must be led by a 20,000 square foot clinic and 10 of them in these areas. That flexibility, I think, essentially allows us to look slightly different in each market, but allows us to find economic success and sustainability in those markets.

One other plausible angle potentially that I've been thinking through like how that could be the case where there are groups in every county, throughout the country, that are making it work, but they can't all like expand and make it work themselves is like, when you're taking risk, you can lower cost, but like your benchmark matters a lot. And so like there is a scenario where in California, value-based care became very prolific, and they are a consolidation of groups and they have purchasing power or pricing power, where they can go to managed care and get a strong benchmark. And that's true in Florida and that's true of Optum, where they have 50% market share in MA in most of their counties. And so like is there a risk that like you do have good pricing power in L.A., and that's why the model works because you're getting a good benchmark, whereas you go to Texas, you go to Nevada, you go to these markets where you're much smaller and you go to the group Humana and you're like, you're paying me 90% in California, but here, I'm giving you 85% because you're really small, like is that possible that, that's the explanation or are you kind of getting reasonable benchmarks everywhere?

I think we're getting reasonable percentages of premium everywhere. And Nevada contract, Texas contracts don't look that -- frankly, that different from the California teams or the California contracts, even though they are different teams, obviously, from the national payers side. And I think it's possible that, that might be the case with groups that feel like they must grow kind of just for the sake of growing because then they become price takers and not makers. And so they might have to take on extraordinary risk on subs or they might have to take on a very low percentage of premium just to get a contract or whatever the case might be. I think because there's no -- we don't feel necessarily any kind of burning pressure that we must grow just because. We know we're free cash flow positive and we'll be around. I think we can wait and choose the markets that we think will actually have reasonable -- have a sustainable path to profitability.

So you're negotiating prior to entering like what is my cap rate going to look like?

Yes. Totally. We're modeling everything out prior to the entry. So if we enter, we think -- I mean it's a fair deal, but we think there is a path for us to get to profitability over time.

Okay. And then in terms of the California business, you recently got the Knox-Keene's license for both Medicare and Medicaid. When you've talked about on the earnings calls, which has been very helpful, like the percentage of premium that you currently are in full risk and where it's going. I've been trying to come up with like math in terms of what you're implying, and it comes out to around, I think, $800 million in incremental like run rate annual premiums off of 2Q? And so first, is that like a reasonable kind of like math to do in terms of the revenue opportunity, but then if you assume a 5% to 10% margin on that, it comes out to like $40 million or $80 million of EBITDA on top of whatever you're doing today. And so are those numbers reasonably directionally accurate? And what is the time line of achieving them? And what are like the big potential stumbling blocks of not being able to?

Yes. Yes. I think the potential opportunity is right, in that ballpark. We probably will not achieve the full potential because we probably wouldn't have every last member of the book in a full risk up-down arrangement. In particular, probably the government programs we will, over the next 12 months. The time line is shrinking as we make progress, but probably commercial, we will not be in full up-down arrangements anytime soon. So I think instead of $800 million, it might be $500 million, $600 million, but I think the potential sizing was correct. And then in terms of the EBITDA opportunity, there is coordination we need to do with the hospitals in order to implement our care model. So as an example, when we work with the hospitals, we'd like to have influence over the discharge process. Typically, if you don't have a hospitalist in the -- if you don't have your own hospitalist in the hospital or if you have no influence over the hospital, they can kind of discharge to any random place they want. Oftentimes, they're trying to rush people through the beds. They don't have enough capacity, so they'll just call like 20 different SNFs and see who is the first to pick up and then discharge immediately to that SNF. You have no control over where the patient is going. You have no idea if they're going to stay in that SNF for 2 months or 2 years and who knows and the cost kind of can balloon out of control. And you also can't implement any care programs that you've developed in-house. For example, we have a in-home care program. If they get discharged to a random SNF that we don't know, it's hard to get them now -- repatriating them back into our in-home care program that we think has benefits to the patient. So to financially incentivize hospital systems, which we've been doing as we move into full risk, to work with us in this model. We often share part of the risk [indiscernible] savings with them. So that probably means that we won't get to the full margin expansion that we might expect because, say, 20%, 30% of that will be shared back to the hospital. So you might get 70% or 80% of what you just said, but maybe not 100%. So I'd probably guide, with those caveats in place, to those improvements in both revenue and EBITDA happening probably over the next 2 to 3 years, probably kind of in a fairly even cadence across those years.

So to your point about like having the hospitalist -- like if you go to them and you say, "Hey, we're going to take full risk now and we're still going to give you some sort of shared savings, but it's less than what you've gotten before". Like are they pushing back and saying, "okay, now you can't have your hospitalist here?" Or like is there -- like what is changing in terms of dynamic? Or is it really that you're just giving them a rate cut and nothing else is really changing?

I think it depends. It's kind of -- this is not a great answer, I know, but I think it's very situational. Hospitalists, especially some of them are doing -- it's like a tale of 2 cities, and I'm sure you know you just cover the big players, but some of them are doing very, very well, and some are struggling, right? And so I think it doesn't really help us to have fewer options for patients in a given region because then you have 0 pricing power with everything owned by one person or one company. So I think we negotiate our arrangements in a way where we can ensure that community hospitals, that patients love, can sustain and stay open. But in a way that we think is fair for the amount of work relatively that we're doing.

But are you at all reducing like the number of people that are in hospitals that are affiliated with Astrana? Like the hospitalists, are they directly employed by you and that's not changing or...

We employ hospitalists even in hospitals that are not currently in arrangements with us. But again, kind of a chain-by-chain negotiation in terms of what they'll let us do, how much of the shared savings they would demand, et cetera.

So that's part of the arrangement. It's like even allowing me to have a hospitalist is part of like the whole structure? Right? Okay.

Or having the right data feed so that when someone gets admitted, we hear about it even before the ADT stuff, and we can deploy someone to help [indiscernible].

Given like what percentage of cost hospitals are. It would seem important that you need like [indiscernible] a person there and not having that would change something?

Very much so. Yes.

You're saying in general, like for most of your hospitals, that's not getting completely eliminated or?

Right. We would absolutely want to continue to do that and then supplement it with other methods, but obviously, nothing beats literally being there.

Yes. Right. Okay. And then in terms of the growth you were talking about before into new markets or even within California, like you've been doing a lot of deals in the last, I guess, 12, 18 months. And so what is the thought process around organic versus inorganically? Are you building any clinics ground up? Or is everything kind of like through a deal? And then I guess, how are you thinking about like the cadence and like sizing of M&A generally?

Yes. That's a good question. We're doing all of the above. Yes. So as I mentioned earlier, kind of the clinic tomorrow is a full de novo build. There are a couple of others in Nevada that have been full -- and California for that matter that have been full de novo builds. There have also been acquisitions. CFC was an acquisition of a network, for example, and we've acquired a couple of clinics here and there as well. I think we just see it as a -- we kind of choose the option that has the highest ROI, like either it's a buy versus build decision. Even an acquisition still requires some build, right? You have to integrate it, you have to do all the TI, you have to kind of hire all the providers in. So there's still some kind of dip in the J curve, just not maybe as deep. So what we do when we evaluate whether we're buying or building is just to figure out does it make more sense. First, is there anything to buy at a reasonable cost. And if there is at some cost, what do the curves look like and what do we think we're going to get more bang for a buck versus going out and building. I mean, I saw that -- I think one of the questions people often have is why even bother building and buying a network of providers? What are you even buying? Like you're not buying...

They can walk away or something...

You're not buying a clinic. You're not buying a piece of physical asset, I suppose. And I think the real question is really how much would it cost and how much time would it take you? And how much profitability opportunity costs are you losing by not having that network? And is that worth kind of what you're paying for it? When I pay 20x for a network? Absolutely not. But the CFC multiple was also 6x upfront and not like 8x. So I think there were -- we think about the ROI and kind of the hurdle is pretty high. So we think about all those factors when we go do that. The other thing is that network is often -- depending on the quality of the network, not all networks are made the same, but some are actually high quality and have a lot of stickiness with the providers and some don't. So I think we would try to verify kind of what that stickiness looks like.

All right. And I think we're running up on time. So I would like one last question here. Kind of going back to our earlier conversation about like what makes certain models work and what doesn't? We had like UnitedHealth Group here the other day, and we kind of asked them this question, and they were saying, we don't think it's our vertical integration that makes Optum Health work, meaning like the dynamic between UnitedHealthcare and Optum, but they said it was their horizontal integration, which could mean either scale or integration with like specialists or something like you're referring to, I guess, I don't know, pharmacy benefit management or surgery centers or urgent care. And I guess you do a little bit of that. And so how would you answer that question in terms of, is it important to be vertically integrated with the payer? And if not, like what about horizontal integration actually works? Because it intuitively doesn't fully make sense to me that being aligned with a specialist is actually a good thing because you don't have a misaligned incentive where maybe you want to [indiscernible]. I don't fully get how the horizontal integration helps, but would be curious to hear your thoughts.

Yes. That's a great answer from United.

They kind of have to say it, though.

Yes. Yes. I think I think vertical integration can mean different things, right? So in our risk-bearing entities, the Restricted Knox-Keene licenses, for example, they are actually licensed health plans in the State of California. They are not plans that can go to market and construct a benefit design and a plan design and kind of be in the marketplace or anything like that, but they are fully licensed health plans. And so in some sense, we've chosen kind of the parts of the vertical integration that we actually want. And what we want is the ability to take on full risk in a multi-payer setting -- in a multiline of business setting. By sitting outside of the kind of single vertical payer, and you may know that we used to have an MA plan -- our affiliates used to have an MA plan, but that retrained us, especially at our scale. We're not obviously United scale -- restrained us to that one vertical of our MA plan, our groups and our technology platform. By having a plan that can [indiscernible] 5 approximately percentage of premium with all of the other plans, we can kind of create a single payer vertical in which we are vertically integrated. We have the plan "that is taking 85% of the premium dollars". We have the medical group, both owned and affiliate, and we have kind of the infrastructure to support all of that. So there are elements of vertical integration that help. I think that's part of the story of having the Restricted Knox-Keene license and why that's beneficial, not having it i.e., not being vertically integrated and being only an outpatient group, definitely is worse than kind of having the plan and being involved. And that's something we're continuing. In Nevada, there's no restricted -- there's no Knox-Keene act. So there's no RKK, but we have the appropriate licensures in Nevada -- I mean in Texas, we have the appropriate licensures to essentially operate as a pseudo health plan. The other comment on horizontal integration is something we've also talked about before. The company is founded by specialists. We have a lot of specialists in the network, 2/3 of the 10,000 doctor specialists that we talk about. And even in our employee group, which is around 150 employee providers. We employ specialists too. We employ cardiologists, derms, endocrinologists, internists and obviously, PCPs. So I think there's merit to that as well. You maybe don't get the -- to your point about quality, we'd like to think that we obviously have high quality and we employ an affiliate with high-quality specialists. And by the way, we actively unaffiliate with low-quality specialists. And we have a medical committee that's obviously always looking at that. It may not be literally the highest quality like cardiologist in all of Los Angeles or all of the state. But I also think that there's something to be said about the balance between access and kind of literally like having the world's #1 cardiothoracic surgeon or something, because if you can get someone into access at a good enough quality, that's quite good for the patient. Not every patient literally needs the #1 surgeon, #1 orthopedic surgeon, whatever it is in the country. But what they do need is to not wait 3 to 6 months to see their cardiologist or their radiologist or whatever the case might be. So I think that's the balance that we strike in a value-based system. At the end of the day, we are responsible. We take on the great responsibility of -- hate to use this word, but kind of resource allocating our healthcare system to the public in a way that optimizes for their health. And so we think that our hybrid model, and this is probably what Optum and United are saying as well in terms of them also owning clinics, affiliating with different high-quality specialists and then kind of really integrating all of that into coordinated system, can help get to that more efficient healthcare system.

All right. Great. [indiscernible] the answer. And thanks so much for joining us.

Thanks, Adam. I really appreciate it.

Appreciate it.