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

Okay Great. We're hitting the home stretch here in the middle of the afternoon session here on day one of the Barclays Healthcare Conference. I'm Steve Valiquette, the health care services analyst. For our next session here, we have Apollo Medical. And with us to my right, we have Brandon Sim, one of the company's co-CEOs. This will be a fireside chat. So I think with that, I guess, we'll dive right in.

Awesome. Thanks so much, Steve, and obviously, thanks to Barclays for having us here.

Yes. So I think in this case, we'll go a little more high level. I think and all the investors know a ton about the company. But I think the biggest thing is, let's just start with having you walk through the growth outlook for 2023. And one of the key growth drivers you see especially breaking it down, maybe organic versus inorganic, and then we can kind of go up there.

Absolutely, yes. I mean the company has gone through -- I think, for the new investors in here, the company has gone through a lot of change over the last 4 years. I mean the company has been around for over 25 years, for most of its life as a private company. We went public via a reverse merger in 2017, a unique choice to say the least. And kind of since then, from 2017 to 2019, we didn't even do earnings calls, for example. So really, the amount of change that has happened from 2019 until now in terms of professionalizing the company, building the right management team, building a technology platform purpose-built for value-based enablement, especially in the mature -- the very mature value-based environment of California, where we were started and founded and continue to be headquartered, has been kind of a long journey in order to unlock a lot of the lay in value that has been in our populations in California and take that scalable model outside of California well -- as well. And so a lot of the growth that we're seeing from 2019, I think, from '19 to the midpoint of '23 estimates, 27% revenue CAGR, 28% EBITDA -- adjusted EBITDA CAGR. It's very strong growth while not -- while maintaining kind of the profitability that we've always maintained with over -- with around $90 million of free cash flow being generated last year out of the $140 million of adjusted EBITDA. And so it's a real model that we have worked really hard over the last 4 years to codify, to build the platform around and to take that into new regions in California, Central California, San Francisco Bay Area as well as into new states like Nevada, Texas and New York. Breaking down some of the growth outlook for next year. I mean, we've guided towards a fairly large range on the revenue side primarily due to a couple of factors. Organically, everything in the guidance is organic. So organically, we're looking at low to mid-teens pure kind of membership head count growth with the balance of the kind of growth made up by improved payer mix by some very incremental growth in our new regions in Texas and Nevada. I say that it's incremental because all of those are primarily still fee-for-service markets that we are working to move towards outpatient or 1-day global risk, and we expect that to happen in the next 12 to 18 months. And so to this date, they are going to be very minimally contributing to revenue growth. But over time, that is something that's going to be a meaningful driver. That being said, even in our core California regions in terms of membership growth and improvement in payer mix and kind of blended PMPM, that's going to already drive the -- I think, I believe, like 17.5% revenue growth at the midpoint. It's actually going to be a little bit higher than that. We've also baked in some of the Medicaid redeterminations into that growth number. We expect around 5% to 15% of Medicaid book of business. Medicaid represents around 25% of our revenue, so at the midpoint, 10% of that business to -- on a gross perspective to kind of be redetermined and maybe not qualify for Medicaid anymore, 10% of 25%, so 2.5%. We expect that impact to be spread across probably 2 years, '23 and '24. So not a huge impact -- not a huge headwind to revenue growth but still something. And we expect actually around 40% to 50% of those members to be recovered or recaptured into a different line of business. As you all may know, we are line of business-agnostic. We serve Medicaid members, commercial, exchange, Medicare Advantage, Medicare fee-for-service and our ACO. And so part of the benefit of that is that we can maintain care coordination as well as maintain them in a risk-bearing ecosystem even across different life circumstance. During COVID, for example, a lot of folks, unfortunately, lost their jobs. We saw them move from commercial coverage in the [ ApolloMed single player ] ecosystem to Medicaid or to a subsidized exchange plan, for example, it'll be similar here with Medicaid determination. The folks -- we do expect 40%, 50% of those folks to come back in another line of business for us. And so that kind of bridges us from the 1.1 million or so of revenue that we report in '22 towards kind of our baseline assumption for where we're going to land in '23. Another big factor that we haven't taken into account in terms of that guidance is the acquisition of the restricted Knox-Keene license that we announced late last year. For those who aren't unfamiliar in California, without a restricted Knox-Keene license provider group is actually not allowed to take on risk in both professional, which is outpatient as well as inpatient or hospital-related risks. We're only allowed to do outpatient risk or part Part B risk in Medicare lingo. And so what that means is we're actually only taking risk on 35% to 40% of the premium dollar instead of 85%, even though we're doing the work in terms of having the hospitals, inpatient care managers, the transition of care coordinators so on and so forth. And so pending regulatory approval, that's something that will also serve as the catalyst. That isn't even included in this kind of guidance forecast for next year in terms of revenue growth moving from 35% of the premium dollar to 85% on our 1.3 million value-based care members in California. Hopefully, that helps.

Yes. No, definitely. I've touched on a lot of subjects that we'll dive into a little bit deeper. One thing that's going to come up a little bit is within a lot of these publicly traded companies that are focused on value-based care, there could definitely be a wide array of arrangements with the physicians as far as contractual relationships, how many are directly employed versus just contractually affiliated? I think you guys -- we thought a pretty detailed Barclays value-based care market model, and I think you guys kind of stick out as a company with a fairly large number of primary care physicians that are affiliated but I think a smaller number that as far as the physicians that are directly employed versus just affiliated.

Correct.

Maybe just talk through the strategy of when it's right to directly employ physicians versus you just having that contractually affiliated relationship. And what are the pros and cons on both sides of that?

Yes, absolutely. So by way of background, it's around 11,000 affiliated physicians in the network, contracted on either our paper or one of our medical group's paper. But in comparison, in contrast, like you said, only 30 -- around 30 unoperated -- primary and multispecialty clinics. So as you said, definitely a larger emphasis today on kind of the affiliate model and enabling existing kind of physician groups and physician practices to participate in and succeed at value-based care. That's been by design. It's an asset-light -- I mean, these are things I'm sure folks have said before, but it's a quicker way to enable value-based care in a place where that might not have been the case before. CMS, for example, has stated that they want all Medicare -- original Medicare patients in an accountable care arrangement by the end of the decade, 2030. It seems infeasible to me if that is the true goal to build enough or to acquire enough or to employ enough physicians under one entity or a group of entities even to accomplish that goal as quickly as CMS might want it to happen. Working with the existing infrastructure, working within the bounds and the trust that physicians in the community have built with patients over sometimes decades of their lives, following them from job to job, following them into Medicare eligibility is something that we strongly believe in and enabling those doctors who are already ingrained in the community who understand the specific levers that it takes to change patient behavior and then giving them the tools to participate in succeeding value-based care, we think, is sometimes more effective than going in, employing a doctor and expecting that they can build these relationships with patients from the ground up within a very short period of time in terms of expecting ROI on that investment. So I think that's a big part of why we like to partner with physicians, especially those who have already been ingrained in the community because sometimes they don't have the ability or the knowledge or the time, frankly, to engage successfully in value-based care. And that's something -- that's where we think we excel at doing that. The reason that we also have clinics as part of the model is because it's important to supplement supply of health care when there is excess demand for that care in a particular region. So what we find is if we're able to partner with physicians who are deeply ingrained in that community and can -- and serve those patient populations, that's wonderful. Let's do that. if there's too much demand, for example, and not enough physicians there to fill the demand, we might have to build new centers, and we have done that. We've built the model centers. We may have to acquire practices and make them more efficient. That's something we've done as well. And so at the end of the day, the operating model, technology platform is agnostic to whether we own the clinic or not. It really doesn't care. It's not like there's a different clinical pathway if we own the clinic versus it's a physician that we don't that's an entrepreneur and has a private practice. The recognitions are the same. The operating models are the same. The economics are a little different, obviously. But at the end of the day, we're going to do what is necessary to deliver a coordinated high-quality experience to patients.

Okay. Also you remind us, too, just within the 11,000 total physicians, mostly affiliated, what's the current mix related to PCPs versus specialists? And is that ratio a very different number within 30 clinics that are directly owned by you guys? And also, how is that ratio of PCP versus specialist kind of evolved over time and going forward for that matter?

Yes. that's a good question. Technically, we haven't disclosed kind of that ratio, but I would say it's around 20%, 30% of PCPs, so call it, a couple of thousand, 3,000 PCPs out of that 11,000. The rest are specialists, again, all contracted in our paper. It's not piggybacked off of insurance contracts or anything like that. Of course, that ratio is definitely different when we look at our employed model clinics. In our employee model, it's primarily primary care physicians. It's over 50%. We do have some specialties. We own a CLIA-certified lab as well. I mean, really, it's all about -- it's not that we want to be a major lab player. It's about fulfilling demand for these kinds of services in places where, historically, they just haven't existed. So absolutely, the mix looks different in our -- their employee model because we focused specifically on -- more on primary care.

Okay. Just to build on that a little bit further. So then within the employee model then, as far as the the specialties that you're focused on, I just want to give a little more color on which areas in particular that you're really trying to drive value-based care initiatives?

Yes. I mean it's still primary -- primary care, as we said. We own -- it's a lot of kind of high demand specialists who traditionally may not want to work or -- just by whatever circumstance just are not in a particular region where we have a great density of patients that we feel the need to augment that demand and supply imbalance. So for example, dermatologists don't -- we're in LA, but typically, dermatologists don't tend to go too far East. For example, downtown or Beverly Hills, it's a stereotype to say that. But it is more difficult in terms of access in kind of more rural regions or more low-income regions to high-quality dermatologists. And so for example, we would employ -- we do employ dermatologists -- several dermatologists for that reason to fulfill kind of demand for medically appropriate dermatology services in certain regions that traditionally wouldn't have someone in that community to serve them. Cardiology is another example, where it's very difficult. OB-GYN is another example where often culturally sensitive -- OB-GYNs are harder to come by in certain regions, and not everyone has access to the same level of care as they might living in a kind of more densely populated region. So that's also -- these are just some of the specialties that we employ. But primarily, it's primary care physicians.

Okay. Got it. Okay. So on that first question, when you provided some of the color on the business, you talked about some of the states already where you're looking to expand outside of California. So we don't have to reroute all those off. But the -- yes, I think just given how competitive value-based care has become really across the entire health care continuum across almost all geographies, it's probably more challenging than ever to recruit physicians to some of these new markets. So what's the strategy when you are entering these new markets? Are you still trying to do more through the affiliation? Or do you want to have the anchor locked in employed physicians just to give you more control over -- your cost containment and everything else when your kind of thinking about entering the new markets?

Yes, totally. Now that's a great question and one we've thought about a lot because, in some sense, Steve, it really does feel extremely competitive, right? All the companies here at this great conference, a lot of folks are really talking about investing in billions of dollars in value-based care not only in primary care, but now we're hearing specialties as well. And then that's kind of a great area of investment for a lot of folks. But on the ground, it's interesting because it often feels -- still but there's an amazing amount of white space. I mean hundreds of thousands -- millions of completely unmanaged fee-for-service construct members that -- really, it's not like we're fighting with some of our peers to get. I mean they're just out there, and it's our ability to go and implement our clinical model repeatably and operating model as well repeatably in some of these areas that will allow us to kind of win or penetrate into these markets. So in some sense, yes, it feels like it's very invested in. But if you look at the broad landscape, I truly do feel like there's -- I mean the numbers bared it out as well. I mean, the percentage of patients in an ACO, for example, is nowhere near even 50%, definitely not 100%. So there's still a lot of white space out there, I think. The key is going to be who is able to scale the model effectively? Who is able to provide a value prop to providers that gets them paid more and allows them to spend more time with their patients? And bringing that model to those places, and that's something I think we're very focused on. For us, it's also about depth over breadth. I mean, we'd rather have a real, very coordinated system that we've created in just to be -- as things really quick, but at of Texas, New York versus having a more shallow presence in 30 states, for example, because I think that's where we can really start to affect clinical outcomes in a meaningful way.

Okay. Got it. Okay. Even though the company has been around for a long time on a relative basis, pretty strong market share and presence in the state of California and just the overall model, I think only more recently, over the past couple of years, have you really started to highlight the internal technology, kind of label yourselves as a tech-enabled BBC provider. And for a little while there, that's worth probably 3x the market cap versus what the company was getting previously. But maybe just talk about the technology component of what you're trying to do. I think, from meeting with you guys previously, I think a lot of that -- almost all of it was developed in-house but just talk about how that could be a point of differentiation on controlling costs versus some of the other companies that are out there?

Absolutely. Yes. I mean I'm a technologist by background. I'm not a clinician. I've to admit -- I did my undergrad and graduate studies in -- computer science and building machine learning models. I used to do that at so -- in -- on the buy side as well. So a lot of my background is in risk management, in alpha generation and machine learning. And I'm probably the first to admit that tech enablement is not the [ NOBL ] in health care. And it will not be, at least in the foreseeable future. We do build, like you said, all of our technology in-house. It's a recent thing because I was the first software engineer ever employed by Apollo. And we now have a team of close to 50 engineers, data scientists, machine learning experts in-house onshore. But all that has been for the aim of enabling physicians and not just to say that we have a shiny kind of technology platform. And the results bear that out, I think. From 2019 until I mentioned earlier, until 2023, we've really driven scalability. We've driven revenue growth. We've driven real earnings by being efficient on the G&A side as well as by helping physicians with maybe optimizing some of their care workflows. And so I don't want to go too much into it -- because of given time, but the technology platform is really kind of multifaceted. We built what the business has needed and not what we think philosophically AI-based or machine learning, that -- even though that is kind of what my background is. For example, we built payer tools. We automate 90% of our claims processing that's really saved cost in terms of -- that there have been estimates out there from payers, hospital systems, $10 a claim. You process millions of claims a year, many millions of claims a year. So automating a lot of that has really cut down on costs. We automate 60% or more of our prior authorization requests. I might remind the audience that we are delegated for all of these payer-related services in California. I think that's rare kind of the unique to the state of California. Outside, the payer retains kind of control of prior auth claims -- claims payment, credentialing, contracting and so forth. That's all something that we do in California for same percentage of premium. And so for prior auth, for example, it used to be that even the most routine obviously medically necessary requests needed to be reviewed by a nurse or a coordinator. Now that obvious ones are automatically approved, the patient can actually, in that same visit, get that approval and then go do that procedure, the follow-up procedure right away. This is a big deal. I mean one example I gave in dermatology, again, oftentimes, for a biopsy, for example, a patient will need to go get an initial consult with the dermatologist, get a referral for the biopsy, get it approved and go back to the dermatology office to actually get the biopsy performed. It's an extra visit for absolutely no reason. What we can do, if we can automatically approve that authorization as we approved auth, in that same visit, the patient doesn't have to come back. They billed once for the visit and the biopsy altogether, and the patient saves a lot of time as well and get the results back faster, higher patient satisfaction. The derm doesn't feel like they wasted time having to see the patient twice for no reason. It's simple things like that where we can really enable workflows at least on the payer-related side. Then obviously, on the provider facing side, we built -- we have an NCQA-certified HEDIS engine to calculate gaps in care. We have HCC suspecting algorithms. We've built models to predict different risks profiles of patients and stratify patients by risk. And all that flows into clinically guided care pathways for each of those risk stratification buckets that our team of clinical coordinators and NPEs and social workers at the corporate level. We have hundreds of them -- use that platform every day in order to identify and then act on parts of the population that are higher risk. All that together, it's a couple of basis points here, 50 basis points there. All of it leads to kind of these industry-leading clinical outcomes in terms of inpatient bed days, in terms of avoidable readmissions, in terms of ER utilization and so on and so forth that drive the profitability of the business.

Okay. All right. Just jumping around here a little bit. This next question, not as close to this as I would be is I don't officially cover the company right now, but just given that 80% of your revenues in '22 were comprised of capitation payments, how much disclosure have you given around just your mix of the individual payers, kind of where your strongest relationships are, where most of the revenue is coming from, unless you're not giving any color on that, but I just wanted to just to -- peel you on your back on that a little bit.

Yes. No, totally. I mean we have -- I don't think we call each pay payer out by name, but we do have a pie chart in our Investor Day 10-K around the kind of distribution of payer revenues. The biggest guy is 50-ish percent of revenue. I think it's no surprise that, that is CMS in terms of Medicare payments. But other than CMS, there's not a single payer, I believe, that has more than 20% of our revenue. In a way, we played Switzerland with a lot of our payer partners. We want to have a payer-agnostic relationship with the patients that have served us well, for example, in times where especially in the payer space in MA has been very competitive in terms of benefit design and rebates and so on and so forth. And so for example, in Southern California, this is local market dynamics, but we saw payer A, not to be named, lose a lot of membership to payer B, who had very strong kind of supplementary benefits that year. But at the end of the day, we got a lot of questions about, hey, does this affect you? Payer A stock price was affected materially by this. At the end of the day, it didn't really matter because we had contracts of both. We were payer-agnostic. We continue the continuity of care for that patient, regardless of the benefit package that they were choosing relative to the formulary that or broker relationship that they had, that particular year in AEP. We are able to continue to deliver them care, same PCP, maintaining that relationship. And actually, in subsequent years, some of those benefits were peeled back. So the patients move back to the original plan all good, right? Same continuity of care. It's really that longevity of that patient provider relationship that we can engender in this payer-agnostic model that yields some of the outcomes of value-based care. Value-based care is not a game for 2, 3 -- 1, 2, 3 years. It's an investment you make now in a patient's health that may not materialize in improved MLR, improved clinical outcomes for many years to come. And if you're investing now and that patient is no longer in your risk ecosystem 3 years later, it's almost as if those investments you've made are now going to accrue -- those benefits are going to accrue to someone else, the person -- the new risk-bearing ecosystem that they're in, even though you might have been the one to make those investments into their health preventive care and all of that in the first place. And so I think a key part is not only providing the key preventive care that we're talking about that technology enables but also keeping that member in your system so that, when the benefits accrue, they accrue to you and not to someone else. And so big part of our payer-agnosticity is to keep that member in our risk-bearing ecosystem so that, that actually happens.

Okay. And within that 50% mix that is CMS is your largest player. I think you got a pretty good mix of Medicaid within that, where it's not just Medicare, but maybe we're kind of running a long time here, but maybe just talk about -- remind everybody kind of what you expect around Medicaid redeterminations and how that impacts your overall business? Is there a way to neutralize that? I mean how are you guys thinking about the exposure on that and how that impacts the flow of the the mix of profitability as the year goes on and also into '24 as well?

Sure thing. Yes. That and the RADV stuff has been -- I think the top 2 questions we've done. It only makes sense. I think I mentioned earlier, Medicaid, 25% of revenues -- we expect 10% to 15% of the members to be probably not eligible post redetermination. That impact will be spread over '23 and '24. You can call it 10% at the midpoint but over 2 years, so maybe 5% of that 25% of revenue each year. We expect to recapture 40% to 50% in exchange of commercial products, probably kind of a subsidized exchange product potentially. And we have all lines of business. So we would probably recapture that. So the overall impact probably wouldn't be the worst in the world. Given where we stand today, we'll obviously update the group investors as if that changes. There are also tailwinds in terms of California Medicaid. As of 1/1 23, all people who qualify from a financial standpoint, regardless of immigration status in California, signed by Governor Gavin Newsom, are going to be eligible for Medicaid. That changes to everyone above 26 years old in 1/1/24. So there are slight tailwinds to Medicaid eligibility, obviously, bound puts and takes with Medicaid redetermination. In terms of '24 RADV, I know we're out of time, but I think that's just the common question I've gotten, we've -- we are actually typically under coded. I mean our -- this is public information, but our Medicare book of business is running at 0.95 to 1 RAF -- wrap score across the board. Like I mentioned, this has not been a super technology-enabled company prior to very recent times, and it takes 18 months for some of the RAF sweeps to come in. And so some of the more recent efforts you've built around getting coding more accurate haven't really shown up yet, and we're still under coding. And so I think -- look, I know everyone says they're going to be fine, but we've run the analysis on the prevalence of some of the big hitters that are going away, diabetes complication, morbid obesity, some of the CHF items, and it's really not going to be a big impact. That, coupled with the higher demographic for new members, RAF score, we think the puts and takes are going to come out on balance around equal or if not even a slight tailwind. So truly, we've done some of the analysis and not too concerned about about RADV either.

Okay great.

We'll see what happens on April 3, right?

Yes. That's right. Yes. Right. That's certainly helpful. Okay. But with that, we went over by a few minutes. So I want to thank Brandon for his time today and to the rest of the conference. Thanks.

Absolutely. Thanks, everyone. Thanks, Steve. I appreciate it.