Universal Health Services, Inc. ($UHS)

Hi. Good morning. Welcome back to the Barclays Global Healthcare Conference. I'm pleased to welcome Universal Health Services. And joining me on stage here is Marc Miller, CEO; Steve Filton, CFO; and Darren Lehrich, Vice President of Investor Relations. Welcome.

Let's start with the acquisition of Talkspace that you announced yesterday. You characterized Talkspace as an accelerant to UHS' outpatient behavioral strategy. From your seat, what is the key strategic gap this deal fills? And why was now the right time to buy such an asset?

So we have been talking for the last few years about our intent to continue to grow outpatient services. We do a lot of outpatient now. But as compared to our inpatient portfolio on the BH side, not nearly comparable. So we've been trying to grow that. It's been a little bit slower than we would have liked. So in lieu of that or in addition to, we've looked to really partner with different companies and have been quite interested in who's been doing what in the market and certainly the better players. We had occasion to talk with these folks for a while about maybe different partnerships working together. They decided to go into a process. And so the timing worked out well in that regard. We just think that this company complements a lot of what we want to do and really fills out the continuum for us because the types of intensive outpatient that we do, partial hospitalization programs, those are a lot of step-down programs from our inpatient. What Talkspace does with their front-end lower acuity types of solutions is totally additive to what we have now. So we really think that it couldn't be a better fit. They're a very high-quality outfit, and they are really tech-enabled. The tech that they have already deployed, what they've been doing for a number of years is impressive. And the things that they're talking about doing in the coming years, equally, if not more impressive. So we're very excited about the opportunity there.

Right. And how do you plan to integrate and leverage that platform across the UHS system to drive that growth?

So again, I mean, there are a number of things, but I mean, just basically, they do things that we don't do and we do things that they don't do. So we think that there's going to be an opportunity on both sides where, for example, patients that they're touching on the front end who don't have a need, or many or most, who don't have a need for inpatient, some percentage will have a need for either more intensive outpatient or ultimately maybe inpatient. We had no purview to those patients previous to now. So now we'll have a direct connection to possibly getting those patients into the UHS stratosphere. Conversely, they would never have seen any of the UHS patients. We're going to be able to directly introduce them to some of those as well. So we think that's going to be additive. But on top of that, it's going to be some of the things that they're developing, new relationships that they're already working on that we think we're going to be able to take advantage of.

Right. And understanding the different sort of modalities, how do the demographics compare of their customer base versus your customer base? Is this something that you see expanding the potential reach of your platform?

I think so. I mean, we focused a lot on their younger population. They have a younger population. We touch that population, but that's only one part of our overall. So I mean, they're certainly skewed younger. They're also skewed towards a commercial base. We touch a commercial base, but we treat everybody. So there will be some effect there as well.

Right. And you've highlighted Talkspace is, I think, 6,000 therapists and view that many of them have additional capacity, what's the plan to scale the volume while maintaining that clinical quality, especially given I think their model is largely independent contractors versus W-2s?

Yes. I mean, again, they are able to attract a great number of therapists all around the country because of the tech and the platform that they've created. And so we see it as a huge opportunity for us to tap into some of these folks that maybe have never thought to work anywhere else. They might have their own small practice. They work with a Talkspace type of a platform. And this might not be another opportunity for them to do other types of work that they've not seen previously. So we're not 100% sure what that will mean. But I think, again, it's greater opportunity.

Great. And moving beyond just outpatient behavioral, it appears you're leaning into the AI capability building through internal hiring, working with some external vendors as well as investments, including your stake in Hippocratic AI. Can you help us understand the key pillars of your strategy with artificial intelligence and how your approach may differ from -- in the behavioral business versus the acute business?

Yes. I mean, honestly, our approach is similar depending on almost regardless of which business. It's identifying what the needs are and trying to find the best fit to address those needs. So Hippocratic, you mentioned, is a company that we were an initial investor. There's a few health care investors for Hippocratic AI. We think that they have a terrific platform, and they continue to grow in a whole host of ways, mostly nonclinical facing. So it's a lot of back office. There are patient-facing items, but it's not in the operating room or things like that. We think that's the right place to focus on the acute care side for the time being. But we're not limited to Hippocratic AI. We've got many relationships with different vendors who are utilizing different forms of AI for revenue cycle, finance, operations, really everything going on, on both sides of our business, we're looking to see if there are AI solutions that can help us into the future. And we're involved with some now, and we're looking at new ones going forward.

Right. So it sounds like there's some kind of tiered investments going on here. Is there a framework for how we should think about the timing and cadence of the benefits here in terms of which segment it accrues to and how it will materialize in the P&L?

I'll let you answer that.

Yes. I mean I think one of the challenges we have, Andrew, is that there are multiple opportunities. And we have an infrastructure, sort of an AI steering infrastructure, if you will, that tries to prioritize and identify the opportunities that are most impactful, can be realized quickly. And as Marc suggested, the different ways we can do that with a vendor relationship, developing it on our own, working with Hippocratic AI to do so. I think it's -- we're in the very early innings of this process, and I think it's difficult to quantify what the ultimate benefit could be. But I think the way we look at it is we haven't spent a tremendous amount of money. We've invested $20-some-odd million in Hippocratic AI, which we think is actually -- we're going to earn a significant return on. We've entered into these relationships with vendors that I think, quite frankly, are more efficient than where we've been doing the tasks manually ourselves. And we've invested -- we've shifted some of our internal IT investment and IT expenditures more towards an AI focused. But I don't think we're ultimately incrementally increasing our IT spend dramatically. So our investment, not that great. I think our returns are starting to build, and they'll continue to build incrementally. I think in an industry like health care and in hospitals, the development of AI is going to be just a significant net positive because it doesn't threaten our business at all. It doesn't threaten our demand, but it allows us to deliver a lot of the routine recurring functions that are a big part of what we do in a much more efficient way.

Right. And I think some of those functions, RCM fits into that category. revenue cycle management. That's become an increasingly strategic priority for hospitals. Which capabilities have driven the greatest benefit in RCM? And where do you see the greatest opportunity from here?

So we've talked about a couple that we've cited. I mean one is, again, using a third-party vendor to do emergency room coding, which is something that we used to do on our own. They use an AI application to do so. We tested it extensively in sort of a parallel fashion where we would code claims manually and compare it to how they coded and found out that they were coding more accurately and quite frankly, obtaining greater reimbursement because they were coding more accurately. The payers have noted that, that broadly, the providers, I think, have gotten better and more accurate in their coding, not just in ER, but in inpatient as well. So that's an example where I think we've got an uplift of tens of millions of dollars annually just by coding more accurately, just by using a vendor who's using AI technology. Another example that we've cited probably before is using AI to appeal denials. We believe that the insurance companies have been using AI to generate their denials for several years now. And then they were doing so in a very sort of efficient nonmanual labor-intensive way. We were replying to these denial letters with a nurse or other clinical professional having to go through the denial letter, having to go through the medical records, draft the denial appeals letter, et cetera. Now we're using a vendor to generate those letters and having the nurse spend 5 minutes to review each one just to make sure they're accurate and complete, again, just gaining that sort of efficiency. And like I said, I think we're just scratching the surface on these sort of opportunities. We'll continue to find more, and I think they'll continue to make us more efficient and more productive.

Great. Sticking with the hospital segment. We're now a few months into the year following the expiration of enhanced ACA subsidies. Is there anything you can share on how exchange volumes or patient behavior is tracking relative to your expectations? And related to this, what impact do you expect the shift towards bronze tier ACA coverage to have on utilization and acuity trends this year?

So just to remind everybody, we have framed an estimated impact from the lapsing of the HIC subsidies at about $75 million, primarily in our acute division. And it's really based on a couple of very high-level assumptions. One is that about 25% to 30% of the HICs volume that we currently have will lose their subsidy or lose their exchange coverage as a result of the subsidies lapsing. In the first couple of months of the year, we've seen a reduction, not quite at that 25% to 30% level. I don't think we're surprised that we haven't seen us sort of reach that 25% to 30% level immediately because I think we'll find that as time passes, some patients will not pay their premiums and that sort of unsubscription level, if you will, will creep up. So I think that as an industry, we still have to experience more. We have to see how the next couple of months play out. We have to see -- you suggested another nuance that people may not completely lose their coverage, but they may shift from a gold tier to a bronze tier, and that could create some uncompensated pressure on us if they have a higher co-pay and deductible. We've tried to factor all that into our estimate. And again, I think based on the first couple of months of experience, we're feeling reasonably comfortable with our estimate. It's certainly not expressing 100% confidence that we got the point exactly right. But I think feel like we certainly have in the ballpark on what the impact is likely to be.

Great. Sticking with the commercial coverage more broadly, there's been mixed data points on the strength of the U.S. employment market. What trends are you seeing across your core markets such as Las Vegas? And how are those showing up in volumes and payer mix so far?

Yes. I think employment trends remain pretty stable, and that's encouraging. There's been some attention to the Las Vegas market in the last year because tourist volume in the market is down. But I think one of the more encouraging things for us in that market is that the employment trends are actually quite strong, which I think suggests that the gaming properties themselves are expecting an improved year and improved tourist volume in 2026. And I think we find whether you want to look at it from the perspective of employment trends or unemployment trends, that tends to have a bigger impact on our business than, let's say, tourist volume, that sort of thing. So it was a bit softer year in Nevada and Las Vegas, I guess, specifically this year for us. I think that the Las Vegas market performed probably closer to the divisional average, the divisional acute average than it has in the past where it's generally been an outperformer. But it has not been a drag. It's not -- and I think that's our expectation for '26 at a minimum is, again, it performs at the division level. And if the employment trends and the employment data is accurate, I think it suggests that the economy in Vegas may be on the upswing.

Great. Let's turn to the fundamentals in the behavioral business. Same-store patient day growth improved throughout 2025, but finished at 1.5% growth and short of the 2% to 3% target. You've consistently pointed to labor as a constraining factor, but the labor backdrop does seem relatively stable. So what needs to go right operationally for that acceleration to materialize?

Well, we did disclose in our 10-K, and I think we discussed on our fourth quarter call that we increased headcount or full-time equivalent count in the behavioral segment by a little over 4% in 2025, and that was certainly in excess of how much our volumes grew at a little over 1%. And I think that was a conscious and intentional investment, if you will, in labor to position us better to be able to grow closer to that target. I think the other thing that gives us confidence is that we have been investing in outpatient growth, which I think clearly, outpatient demand in behavioral is growing faster than inpatient. And that's why we've been focused on it, and I think we'll benefit from that. Obviously, the Talkspace acquisition will help that. I'm not sure it will have a material impact in 2026. But over the next several years, I think growing our outpatient at a faster rate than inpatient has been growing will help us achieve what has historically been a very achievable volume target of sort of 2% to 3% adjusted patient day growth.

Right. And today, I think your outpatient behavioral revenue is about 10% of the segment. You're clearly leaning into that category. As you scale that footprint, how do you ensure that new outpatient capacity is incremental to the system rather than cannibalistic?

Look, I think the truth is that you need to treat patients where they want to be treated and where payers want to treat them. If you think about the acute business, there's been this shift over time out of the hospital and into outpatient settings, whether that's ambulatory surgery or outpatient imaging or freestanding EDs. And we've shifted along with that. And some of that, to your point, is cannibalized business. Some of that is business that was performed on an inpatient basis. But there's no really resisting that. We've got to deliver the care where patients want to be treated and where payers are sort of demanding they be treated, and we'll do that on the behavioral side as well. And we'll do that by focusing and dedicating people to growing that outpatient business. We'll do it by building out these freestanding outpatient facilities that we really never had a presence in before. And I think now we'll do it by taking advantage of the synergies with Talkspace over the next several years that will really kind of -- you used the term or repeated the term that we used in our press release that I think Talkspace is really going to help us accelerate that outpatient growth that is already underway.

Right. And as you look at those different avenues of investments, like where do you see the strongest returns between step-in and step-out services, particularly in the context of that Talkspace acquisition?

Yes. Again, I don't think it's kind of a linear sort of thing where we're choosing one option. I think broadly, margins in the outpatient space tend to be better than the inpatient space because I think, quite frankly, the payer mix tends to be better in outpatient. It tends to be more commercially and Medicare-centric and less Medicaid-centric. So I think, obviously, we'll make judgments about where the investments make the most sense. But investing in these freestanding centers, which require minimal capital, investing or continuing to invest in the Talkspace technology, which Marc alluded to before, is we find quite advanced and quite sophisticated is something we'll continue to do. Again, I think this is all about creating this end-to-end seamless continuum. So a patient can be treated anywhere along what I'll call a UHS Talkspace continuum in the future, where it's most appropriate, where the payers are willing to pay, where they're going to be most satisfied. One of the things that interests us in Talkspace is they have very high levels of patient satisfaction as well as therapist satisfaction. And so building out outpatient should be economically and financially beneficial to us, but it should be really quite beneficial from an outcome and a satisfaction perspective for the patients and for those delivering the therapy.

Great. And on the pricing side, behavioral pricing, especially on the inpatient side, has been particularly strong in recent years. What are the primary drivers of that strength? And then I think you've more recently noted that started to moderate. So what's driving that moderation? Is it state budget dynamics, payer behavior, some combination of the 2, changes to the commercial side?

I really think it's just an outgrowth of the success that we've had in the last several years. We've talked a lot about one of the dynamics in the last several years is that because of a lack of capacity, not just in our own hospitals, but in the behavioral space broadly, particularly in terms of inpatient capacity that we've been able to go back to payers, particularly managed Medicaid payers for -- from whom we didn't necessarily have regular increases over the last several years and negotiate what we believe to be just fairer rates, not particularly outsized increases, et cetera. But some of those increases that we got in '23 and '24 and '25 have now started to anniversary. And so as we anticipate what the year-over-year growth is in '26, it's not as great as -- we're not expecting it to be as great as it has been over the last several years.

Right. And Medicaid supplemental payments, I think, are a component, right, within that. Can you provide an update on the status of some of the larger pending programs such as Florida? To your knowledge, is the holdup related to the structure of these programs or some sort of administrative backlog at CMS?

Yes. So look, I think it's worth noting that CMS clearly is willing to approve new programs. We have acknowledged that at least 2 new programs CMS has approved in the first quarter that will record the benefit in the first quarter, one in Ohio and one in Nevada. So I think broadly -- and they've approved other new programs as well. So CMS continues to approve new programs. The Florida program, as your question suggests, has been delayed. What the state of Florida says is that there are some structural issues that CMS has objected to or at least questioned. The state of Florida suggests that they're in what they, I think, describe as sort of constructive engagement with CMS and continue to imply that they're expecting approval. We don't know what the timing of that is. We don't know if that's weeks away or months away. And we've taken a conservative approach in terms of recording and not including in our guidance that Florida benefit until it is approved. But based on the feedback that we're getting from the state of Florida, still expect it to be approved.

Right. And policies in the One Big Beautiful Bill are set to phase down Medicaid supplemental payments starting in 2028. Marc, how are you positioning the organization to absorb those changes? And what sort of advocacy efforts are you making in D.C. to delay or adjust those policies?

Yes. So we've been spending a lot of time in D.C. I honestly think that it's unknown what's going to happen, come '28. We've gotten a lot of positive feedback, though, from moderates on both sides of the aisle that this won't stand as it is today. And they expect something to happen legislatively, probably in the next Congress. But if not this one coming up, then the one after in some way because we'll be able to absorb whatever we have to absorb. We'll adjust our operations. We've proven that we can do that. I think the 80% of the hospitals that are not-for-profit in the U.S., a lot of them are going to have a much harder time. And the folks in D.C. recognize that. So it's a wait and see. I can't really predict what will happen. I just think that it will moderate in some way, and we'll be in the forefront there, and we'll take care of that.

Great. And Marc, it's hard to believe it's already been 5 years since you've stepped into the CEO role. How has the company changed over your tenure? And how do you view the competitive positioning of the company today versus 5, 10 years ago?

Yes. I mean, overall, the company is similar. I mean I've been President since '09. So it's been 15 years. And we're pretty consistent in our approach. I did not come into this role really changing anything because it was a continuation of what I was already doing. I think the company is on a good track. I think this latest acquisition puts us in an even stronger position for the next few years and beyond. So we're very excited about where we are, where we're going. And again, our consistent approach continues. And so we have a very positive outlook going forward.

Great. Well, with that, we're out of time here. So let's end it, and thank you so much for joining on stage, and please enjoy the rest of the conference.

Thanks, Andrew.

Thanks Andrew.