Universal Health Services, Inc. (UHS) Earnings Call Transcript & Summary

All right. Good morning. My name is Justin Lake. I cover health care services here at Wolfe Research. Very pleased to kick off the last day of our Wolfe Healthcare Conference with UHS. We've got the company's CFO, Steve Filton here. We've got the company's new VP of IR, Darren Lehrich as well. Before we jump into the Q&A, Steve, maybe you give us a couple of minutes of your reflections on the year-to-date positive-negative surprises and how you view the company's positioning into 2026?

Sure. So obviously, our third quarter earnings release just a few weeks ago. And I think from our perspective, things have been going largely as expected. I mean there were a couple of sort of nonrecurring items in the quarter. The biggest positive was the $90 million roughly of annual DPP payments from Washington, D.C. that we've been waiting for, waiting for CMS to approve for some time, offset by a couple of onetime expenses, malpractice expense and legal settlement, et cetera. But I think in terms of the cadence and trajectory of our 2 businesses, largely performing as expected and as we expect going into 2026, I think -- we think that we are truly now in sort of a post-COVID environment in which the businesses are operating in accordance with models that are much more historically normative maybe than they've been for the last several years; revenue -- same-store revenue growth, mid-single digits. I think on the acute side, something in the 5.5%, 6% range, split pretty evenly between price and volume. On the behavioral side, maybe 6%, 7%, skewed a little bit more towards pricing. We've struggled, I think, in the last few years with our behavioral volumes or getting our behavioral volumes to the targets that we anticipated they could be at or should be at. We're now talking about adjusted patient day same-store growth in the 2% to 3% range, probably in the nearest quarters closer to the 2%, but longer term, maybe closer to the 3%. So a little too early. I'm always hesitant when I appear anywhere in the fourth quarter before the holidays to really give any updates on trends because they can change pretty dramatically around the holidays. But again, I think the model or the metrics that I outlined, we certainly feel are achievable and will form, I think, the basis for our 2026 guidance, which we will, as we always do, give in detail at the end of February when we do our Q4 earnings.

Thanks, Steve. Before we get to the fundamentals, of course, we'll go through all of the DC-related changes. So maybe first off, you're the only company in, at least the hospital space that's talked about or put numbers around the potential impact of the exchange subsidy expiration. I think you've talked about $100 million in that ballpark of a headwind if that goes away. Maybe you could break that down for us, for instance, you've talked about 6% of adjusted admissions in the acute business being on the exchanges. What kind of a decline do you kind of build into that $100 million? Maybe you could bootstrap it for us in terms of what kind of decline in exchange admissions do you see? How many of them go to commercial, for instance? Let's start there.

Yes. So I think the biggest variable and assumption that companies have to make in estimating the impact, as you described it is how many of those patients who currently have exchange coverage would lose that coverage if the subsidies are not renewed in some form or fashion. As you suggest, and we've said our most recent data point is about 6%, 6.5% of our acute adjusted admissions are exchange patients. We have not given a number on the behavioral side. We think it's a much smaller, more negligible number largely because from the beginning of the ACA, the exchange products have always or have generally borne a pretty significant co-pay and deductible load that renders them a lot less relevant for behavioral admissions and behavioral coverage. So then the question is, so if those people were to lose their coverage, what's going to happen to them? Some of them might be able to go on Medicaid or get Medicaid. Some of them might be able -- some of them are working and might just return to their employer plans. Some of them might be able to afford lesser metal plans going from a gold to a bronze or whatever. And the truth of the matter is nobody exactly knows how it's going to play out. We've never really experienced this before. We've assumed, and I think that seems relatively consistent with what some of our peers have sort of talked about is that roughly 1/3 of people would lose their coverage. And as we've sort of looked into this and delved into the details on that exchange population over the last couple of years, I think we've largely determined that their utilization behavior is reflective of or more sort of, I'll call it, sympathetic with a Medicaid population than with a, I'll call it commercial or Medicare population, meaning that their utilization is very ER focused. They tend to use the ER when they have health care needs. They don't necessarily have private physicians. They don't do -- we don't do a lot of elective procedures in that population, et cetera. So we take that 1/3 of the business that we think would lose their coverage. We evaluate to what degree they use the emergency room, and we assume that they'll continue to use the emergency room and emergency procedures at the same rate and the same cadence, but now we won't get paid for them. And that's how we generally get to that number. But I think we've stressed, and I think honestly, as you suggest, we're the only company that I think has really put that number out there. I think the first time we put it out, that was at your conference last year. And I stressed at the time that it really was a guesstimate. And I'll say that again. I mean I think we gave the number, and I wanted to give it last year because people were estimating numbers that were well beyond anything we were sort of contemplating, and I felt it was helpful to ring-fence it. And I still think that's true. I couldn't tell you that $100 million is a precise estimate, but I think it's a decent ballpark based on what we know.

Got it. And I appreciate the comments on the shift in utilization to the ER and Medicaid versus -- is there a rule of thumb that you would think about if an exchange person is using 1 unit of hospital volume. Does it go down by -- I think HCA was talking about something in the neighborhood of goes down by about 1/3 to 1/2. Is that a reasonable ballpark? Do you think it goes down more?

I just want to be clear, the point they're making is they're not -- they don't utilize the system as much as a Medicare commercial patient...

Meaning that the exchange patient when they lose their coverage will utilize less, like you said, it will be more ER related. We're just trying to think about what is the step down in terms of maybe a percentage basis that you think that...

And again, speculating here. But I think our view is people use the emergency room when they have to. And to the degree that whatever pace they're currently using the emergency room, they will continue to do so. Now it's true that some people use the emergency room for nonemergent procedures. And again, I think that's largely true in the Medicaid and exchange population. But the assumptions we've made is that they'll continue to use emergency procedures at the same rate that they had been.

Got it. And then the percentage of revenue that's tied to the exchanges, you said 6%, 6.5% of volume. Is it a little bit higher on the percentage of revenue?

Yes, not for us. And I know this is a little bit of a difference between what we say and what some of our peers say. So we find that our average exchange reimbursement is closer to Medicare. And I think what some of our peers have said is that their exchange reimbursement is somewhat better and maybe closer to commercial or at least somewhere between Medicare and commercial. I think, and again, it's worth noting that the biggest chunk of this exchange population resides in a couple of states, it's in Florida and Texas. Our 2 biggest public peers, Tenet and HCA tend to have, on a relative basis, a bigger footprint in Florida and Texas than we do. And my guess is they probably have more negotiating leverage in those states, which is probably why their rates are somewhat better. But yes, so our view is that 6.5% -- 6% to 6.5% of volume is also reflective of their revenue contribution as well.

And just to put a bow on this, you talked about 1/3 of members -- 1/3 of the exchange membership loses coverage. You mentioned that some could go to commercial or Medicaid. Any thoughts on how much of that 1/3 does go to commercial and Medicaid that you've assumed in that 100?

Yes. So again, I think the way we've tried to do the projections is sort of assume that if 1/3 just loses their coverage and the remaining 2/3 have some sort of similar coverage. And again, now those are broad assumptions, and we acknowledge that. But I think we're, in my mind, sort of capturing the heart of the impact and the way we've done this.

And then moving over to the provider tax benefits. You talked about -- you talked about that picking up a bit. I think you talked about something in the $1.3 billion range in the current run rate. That was up about $140 million versus the previous. And again, you're the only company giving us an estimate on what you think the 5-year impact will be in terms of the cuts. We all appreciate that. I thought it was interesting that while the benefit was up $140 million, the cut was up a little bit less than half that, right? I'll call it half that, at about $65 million, $70 million. I'm curious in terms of why the cuts wouldn't go up by a similar amount, right, given this is going to transition off. And I -- the 2 things I can think about is, one, I know you've given us a 5-year window. And maybe there are additional cuts beyond that $65 million, $70 million incremental that go outside the 5-year window? Or you think that you're keeping some of it? And if so, I'd love to understand kind of the mechanics there.

Yes. So this is obviously a complicated issue. So I think it's helpful to sort of kind of frame it for everybody. So as you suggest, in our third quarter 10-Q, we have reestimated what we think our net benefit will be from Medicaid supplemental payments in 2025. That's the $1.3 billion that you alluded to. And we also give an estimate, which we've updated-- based on the OB3 cuts or proposed cuts to these Medicaid supplemental payments, which would begin in 2028 and then go over 10 years, we give an estimate that what would those cuts amount to in 2032, 5 years into the cuts. That's the $420 million to $470 million that you alluded to. The cuts continue -- well, first of all, we assume that the cuts will occur ratably, so they'll begin in 2028, and they'll largely sort of progress ratably getting to, we'll call it this $450 million number, $445 million number in 2032. We don't project beyond that, because we don't really project anything beyond 2032. But the cuts do -- they decline, but they will go on beyond that. The issue of why -- when there's additional benefit, why the cuts are not in the exact same proportion is really a function of the cuts by state are not proportional. They really -- the cuts are really driven by cuts to the provider tax rate in the state and cuts to the average reimbursement in the rate. And so depending on what the provider tax rates are in the state, depending on whether the state is an expansion state or not an expansion state, depending on what the average reimbursement is under the state's program, the cuts are varying. And so yes, I mean, any time there's going to be an addition to this, we'll likely update the impact of the cuts, but it's not going to always be exactly proportional. Again, I would suggest that over time, the cuts are about 1/3 of our total in total. And I think if you want to think about it and model sort of ups or downs in the overall benefit, I think the overall cut would be in that sort of 1/3.

Got it. That's helpful, Steve. And then CMS put out a letter on provider taxes last Friday. I think everybody who's read it in this room and on this webcast has probably gotten multiple headaches, and I don't know that anybody has been able to figure out exactly what CMS was trying to say here. But there was a phrase there that a lot of people have kind of latched on to around provider tax programs that are enacted and imposed, right? I know you have a -- that's kind of the language they're talking about in terms of the -- of grandfathering, right, for these provider tax programs. So I know you guys have one of the smartest people on the street that I've talked to at least on the provider tax side. I know you've run this by them. How are you interpreting that? And how do you think that there is an impact at this point to your estimates on provider taxes?

Yes. So first of all, I think your description of the proposed rule is a fair one. It was not all that easy to interpret. It's not all that clear exactly what CMS is sort of trying to get at. I think our initial read is that of the $1.3 billion that we expect to get in Medicaid supplemental payments this year, we don't believe that this proposed rule will have any material impact on that. Again, I would stress the idea that it is a proposed rule and they will go through the rule-making process. And we would hope that in the final rule, some of the questions and some of the nuances that you've alluded to will be clarified. And I really don't know what the clarification will be. But to your point, so -- if a state has passed or submitted a revised plan that has been approved by CMS, the notion, I think, from a provider perspective is that, that is a plan that would be grandfathered. Whether this imposed, enacted sort of language nuance means just passing it is not good enough, they have to have collected the tax or build the tax or we don't know. We don't really know what -- whether in fact, CMS really is trying to kind of parse the language and say that impose is something different than enacted. Again, the notion is, I think, that, that will be clarified in the final rule, I think what -- again, we would say at the moment is the $1.3 billion that we're projecting for this year, we don't think would be materially affected by the rule.

Got it. What did your provider tax guru, so to speak, have a view of what that might mean for stuff that's been proposed, but not either collected or approved by CMS? I'm thinking, think of Florida, for instance.

Yes. So I think it's more in the new programs, which, for the most part, any program that's been proposed that is sort of pending CMS approval, we've not included in our estimate in that $1.3 billion. Florida, probably maybe the most notable among those. And yes, so I think there would be an argument about whether the Florida program, which I think a few months ago, people assumed would be grandfathered. And I think we still assume may will be grandfathered is going to meet the CMS definition. And again, my sense is that CMS will have to clarify that in the final rule.

Got it. And just to put a bow on this, the -- in terms of provider tax programs that are out there, I think Florida, Georgia, Virginia, et cetera, can you give us an estimate of what that could be for the company outside of the $1.3 billion?

Yes. So if...

Everything proposed...

No, I understand. I was going to say if it doesn't give you a headache and you want to read through our 10-Q disclosure on the DPPs, I think we give a lot of that detail. We talk about Florida and have talked about Florida. Georgia is a moot point for us. Georgia, we only have behavioral operations in Georgia and the Georgia DPP program does not include behavioral. So yes, I mean, I think we land -- the state, I think, that we often get asked about and that we don't provide a tremendous amount of detail is California. California has proposed an expansion of their DPP program. California has historically struggled with CMS approval. CMS has issues, structural and I think technical issues with California -- programs in California and New York, New York is relevant for us. And so we've never tried to quantify what the benefit would be if the California rule is adopted or if it's changed and modified. But I think we do clarify the potential benefit from Florida in the kind of $45 million to $50 million. I think California could be a significant additional benefit, but we sort of prefer to wait and see how that gets resolved between California and CMS.

And if I just press you one time on that, if California -- would California be closer to Florida or closer to D.C.?

Yes. And I'll be absolutely honest, Justin, I mean honest -- not -- they feel pressed, I think the reason we've not tried to quantify it is we really don't know. We think the program would have to be modified to meet the CMS requirements. And obviously, not knowing how the program will be modified, it's very difficult for us to provide any estimate.

Got it. Appreciate that. So we'll close out that section of our Q&A here, and we'll talk a little bit about fundamentals. The -- you gave us a little bit of a view of how the quarter is progressing, right? Specifically, you talked about, I think, behavioral patient days were up about 1.3% in the third quarter. You thought you could get to the low end of that 2%. You've seen October kind of past by now, you're halfway through November. How is that kind of trajectory look versus that 2% target with the caveat that we know the holidays could throw a wrench in there?

Yes. I think the point that we tried to make on the call was that we thought that 2% to 3% same-store adjusted patient day growth in behavioral was a reasonable target in, I'll call it, the short to intermediate term, getting to the higher end sort of the more time passed. I don't think we were in any way sort of trying to guarantee or sort of strongly suggest that we can get to the 2% in Q4, we might. But as I said, I'm always often hesitant to give sort of intra-quarter updates, but always in the fourth quarter hesitant to do it because so much can change around the holidays, for better for worse. And obviously, it's the same every year, meaning we see these dips particularly in kind of elective and child and adolescent admissions on the behavioral side. But it can vary year-to-year based on timing, I think based on some other factors. So we'll see. But again, I'll just reiterate that we're comfortable that, that 2% to 3% behavioral range that we gave is reasonable and not an overly optimistic projection in the, I'll call it, near to intermediate term.

Okay. And some of that more positive view and kind of confidence seem to come from a tick up on the employment side, you talked about on the third quarter call. I know you've given us some interesting metrics there in terms of turnover over time, new starts. How would -- maybe you could put some numbers around that employment change and how you think it benefits you?

Yes. So I think we've really talked about the upside potential in behavioral volumes. And again, we're not -- you've sort of run through the numbers. We're not talking about seismic increases in our ability to meet demand. We're talking about going from 1.3% in Q3 to something closer to 2% in the next couple of quarters. But we think the biggest potential for continuing that growth is making additional improvements and incremental improvements in our staffing. We still say that somewhere between 1/4 and 1/3 of our behavioral facilities struggle with filling all of their labor vacancies, that can be nurses in some cases, it can be therapists. And in many cases, it can be nondegree professionals, the people that we call mental health technicians or aides, who are sort of critical to the behavioral clinical process and care process. And we continue to make progress. We measure that both in terms of hires. We measure it in terms of reducing turnover. And those metrics have been improving. But to be fair, they've been improving in a pretty incremental way, not as quickly as we'd like. It's challenging. But I think we're making progress, and I think we'll continue to make progress. And then the other area that we've talked about, particularly, I think, in the last few quarters is, I think we've seen an uptick in the shift of patients from the inpatient setting to the outpatient setting in behavioral. This is not a new phenomenon. Obviously, we've seen that phenomenon in the -- on the acute side of the business literally for decades. But we're really seeing more and more payers, more and more government entities, really trying to move patients where they can to a more efficient setting of care, whether that's partial hospitalization or what we call intensive outpatient. And these are not sort of individual therapy sessions. People often, when we talk about outpatient behavioral care, people often think about the 50-minute session that an individual might have with a therapist once a week or twice a week. And while that's obviously an important component of behavioral care, it's not really the core of our business. So when we talk about outpatient care, a lot of our focus is on, again, inpatient -- intensive outpatient, partial hospitalization. These are people who are getting 4, 5, 6 hours of therapy a day. They're just not staying over in a facility, and they may be getting that care 5 days a week, 3 days a week, but that's -- so in any event, we're doing a number of things to build up our sort of internal referral network. So as we discharge patients, we capture as much of that as is appropriate for patients who need continuing care and a lot of patients who do get discharged from an inpatient facility do need continuing care. And we're also really starting an initiative of standing up these freestanding outpatient facilities, which really have not been a significant focus of ours historically because we do know that there are patients who require that sort of care, but don't want to receive it on a hospital campus and in a facility associated with a the hospital because they have fears, whether they're founded or unfounded that they're going to be sort of swept up in that sort of inpatient dynamic. So we've been talking about standing up 10 or 12 of these freestanding facilities a year for the next several years.

Got it. And what percentage of your behavioral revenue is outpatient right now?

Yes. So today, it's a pretty small percentage. I'm going to say it's in that 10% to 15% range. And again, as you know better than I do, you're more familiar with payer commentary, but a lot of payers have commented in the last few quarters that some of their increase in medical loss ratios and utilization is really coming from behavioral care. And a lot of times, they're citing outpatient and the growth in outpatient. And we just feel like historically, we haven't captured our fair share of that. And we have a lot of advantages that should allow us to capture more of it. We tend to be an in-network provider with most providers in most of our markets. We tend to have strong referral relationships with hospital emergency rooms and military bases and school systems and all those components should allow us to garner a significant share of that outpatient business, which historically, we just have not really focused on.

And what's the growth rate of that 10% to 15% of the business running versus the inpatient side?

Yes. So I don't give the exact numbers, but I mean, we made the point in Q3 that of our 1.3% adjusted patient day growth, clearly, outpatient grew faster than inpatient. And I would expect that will be the trend for the foreseeable future that we're going to see faster growth in outpatient than we do in inpatient.

Got it. Maybe we wrap up with a discussion of AI and how that's impacted your business, specifically on the revenue side. I think in the third quarter, you mentioned that maybe a 1% to 2% pickup in pricing came from better collections, right, on the acute side. And you mentioned the payers, right? They're all complaining that sepsis is up 50% or 100%, right, and that hospitals are utilizing AI to code better, right? No one is saying that the coding isn't correct, right? No one is saying it's fraud or anything. Just you guys are doing a better job of coding. So curious how that's impacted your business. Would you kind of attribute that 1% to 2% to AI? And what are kind of the initiatives out there?

Yes. So I would say beginning a number of years ago, we really began to focus particularly in our acute segment on improving our revenue cycle performance, basically our billing and collections. And to be fair, again, part of that initiative was really in response to the fact that we felt like payers were getting more aggressive, and they were using -- and when we think about it, when we think about these improvements, we think about it as people, process and technology, and I think I feel like we've been improving all 3. But again, in response, we think that payers have been using technology to generate denials, to generate patient status changes, whether a patient is an inpatient or an observation patient on the acute side, that sort of thing, in managing length of stay on the behavioral side. And so we're responding to that. And so yes, we talked about, I'll call it, pricing improvement or revenue per adjusted admission in acute care being at 5% in the last quarter, whereas we think kind of a sustainable number is maybe in the 3% range. And I think a lot of that gap or a lot of the difference is some of those improvements that we've made. Some of it's in technology, making sure that when a bill goes out, it's clean, it's complete, that reduces delays, et cetera. It reduces the ability of the payer to sort of bounce claims, even if it's a temporary sort of kind of thing. We believe that payers have been using AI for years to generate denials. And now we're using AI to generate denial appeals, et cetera, making it more efficient, making it more complete. So yes, I mean, I think -- and we'll continue to do that. I think in terms of some of the coding things, et cetera, honestly, I feel like we've been doing that for years in terms of having outside parties review our coding and doing some of that work that maybe the not-for-profits are just sort of getting into.

Got it. Do you think that -- I don't want to put a number around it in the quarter, it was 1% to 2%. But do you think that kind of tailwind is durable? Or do you think it's kind of going to bounce around and you're going to see -- I know there's all kinds of accruals around collections and when you recognize them. Do you think it's going to bounce around? Or do you feel like this is something that's durable and you feel like could be a tailwind for a year or 2?

Yes. So honestly, I think we've been getting that benefit. If you go back and look, our acute pricing has been strong now for quite a while. So I think we've been getting that benefit, which is why I think as we sort of suggest going forward, pricing probably moderates at some point closer to that sort of 3% range. But obviously, we're going to continue all this focus on all those initiatives. And honestly, we're shifting to do this now on the behavioral side, where I don't think there's quite as much opportunity because I think billing and collection on the behavioral side is not as complicated as it is on the acute side. But I think there are opportunities. We're just -- we're not nearly as centralized and I think as efficient on the behavioral side as we are on the acute side, and we'll get there in the next year or 2.

Got it. Steve, Darren, I want to thank you for your time. I appreciate you being here with us today, and thanks, everybody, for joining us.

Thank you.

Thanks, Justin.