環球健康 (UHS) 2018 Q4 法說會逐字稿

Good morning. My name is Emily, and I will be your conference operator today. At this time, I would like to welcome everyone to the fourth quarter 2018 conference call. (Operator Instructions)

Steve Filton, Chief Financial Officer, you may begin your conference.

Thank you, Emily. Good morning. Alan Miller, our CEO, is also joining us this morning. Welcome to this review of Universal Health Services results for the full year and fourth quarter ended December 31, 2018.

During this conference call, Alan and I will be using words such as believes, expects, anticipates, estimates and similar words that represent forecasts, projections and forward-looking statements. For anyone not familiar with the risks and uncertainties inherent in these forward-looking statements, I recommend a careful reading of the section on Risk Factors and Forward-looking Statements and Risk Factors in our Form 10-K for the year ended December 31, 2018.

We would like to highlight just a couple of developments and business trends before opening the call up to your questions. As discussed in our press release last night, the company recorded net income attributable to UHS per diluted share of $8.31 for the year and $1.70 for the quarter. After adjusting each period as indicated on the Supplemental Schedule included with last night's earnings release, adjusted net income attributable to UHS increased to $220.1 million or $2.37 per diluted share for the quarter ended December 31, 2018, as compared to $189.6 million or $2 per diluted share during the fourth quarter of 2017.

As reflected on the Supplemental Schedule, our adjusted net income attributable to UHS during the fourth quarter of 2018 excluded a pretax increase of $31.9 million in the Department of Justice Reserve and a pretax provision for asset impairment of $49.3 million, which reduced the carrying value of a trade name intangible asset recorded in connection with our 2015 acquisition of Foundations Recovery Network.

On a same-facility basis, in our acute care division, net revenues increased 4.7% during the fourth quarter of 2018. Excluding our health plan, same-facility revenues increased 6.1%. The increased revenues resulted primarily from a 2.2% increase in adjusted admissions and a 4.2% increase in revenue per adjusted admission. On a same-facility basis, net revenues in our behavioral health division increased 2% during the fourth quarter of 2018. Adjusted admissions to our behavioral health facilities owned for more than a year increased 4.5%, while adjusted patient days increased 1.2% during the fourth quarter of 2018 as compared to the fourth quarter of 2017.

Revenue per adjusted patient day rose 1.1% during the fourth quarter of 2018 over the comparable prior year quarter. Our cash generated from operating activities was $1.341 billion during 2018 as compared to $1.183 billion during 2017. Our accounts receivable days declined to 50 days during the fourth quarter of 2018 as compared to 52 days in 2017. At December 31, 2018, our ratio of debt-to-total capitalization declined to 42.6% as compared to 44.7% at December 31, 2017. We spent $144 million in capital expenditures during the fourth quarter of 2018 and $665 million during the full year of 2018.

In 2018, we completed and opened 234 new acute care beds and 734 new behavioral health beds, including de novo facilities. Our behavioral health integrations joint venture pipeline continues to be very strong. Today, we are announcing our latest joint venture, a partnership with Southeast Health to build a new 102-bed behavioral health hospital in Southeast Missouri.

During 2019, we expect to spend approximately $675 million to $725 million on capital expenditures, which includes expenditures for capital equipment, renovations, new projects at existing hospitals and construction of new facilities.

In conjunction with our share repurchase program that commenced in 2014, during the fourth quarter of 2018, we repurchased approximately 1.22 million shares of our stock at a cost of approximately $149 million or $122 per share. During the 12 months ended December 31, 2018, we have repurchased approximately 3.32 million shares at an aggregate cost of approximately $401 million or $121 per share.

Last night's press release included our 2019 operating results forecast for the year ended December 31, 2019. Our estimated range of earnings before interest, taxes, depreciation and amortization, net of controlling interest, is $1.826 billion to $1.909 billion. Our estimated range of adjusted net income attributable to UHS for the year ended December 31, 2019, is $9.70 to $10.40 per diluted share. The adjusted EPS guidance range represents an increase of approximately 2% to 9% over the adjusted net income attributable to UHS of $9.53 per diluted share for the year ended December 31, 2018, as calculated on the Supplemental Schedule. During 2019, our net revenues are estimated to be approximately $11.21 billion to $11.36 billion, representing an increase of 4.1% to 5.5% over our 2018 net revenues.

Alan and I will be pleased to answer your questions at this time.

(Operator Instructions) Your first question comes from the line of Matt Borsch from BMO Capital Markets.

So I was hoping that maybe you could touch on 2 things, just helping us understand how much the outlook in 2019 is going to be driven by the acute side versus the behavioral side. And then maybe just in the quarter, if you could comment on -- the metrics were good -- were strong, but the revenue per adjusted admit on the behavioral side, I was hoping you could just comment on that one.

Okay. I think, generally, Matt, the approach that we took for our 2019 guidance was that the business trends in each segment would, for the most part, continue as they've been. So I think on the acute side, we took the position that the guidance, or at least the midpoint of the guidance, was based on something close to 5% to 6% revenue growth and 6% to 7% EBITDA growth. On the behavioral side, much more modest revenue growth, generally more like 2% to 3% and sort of flattish EBITDA. Although, particularly on the behavioral side, there are a number of headwinds that we faced in 2018 that will reverse themselves in 2019. Those include the start-up facilities that we had opened in 2018. It includes the regulatory sort of challenge facilities that we had in the beginning of the year, that continued to be a drag for part of the year. It includes the end-of-the-year challenges we had from the Florida hurricanes and the California fires, et cetera. So I think from an EBITDA perspective, I think I'll call it from a core -- on a core behavioral basis, we expect EBITDA to be sort of flattish to maybe up 1% on a kind of all-in total basis. I think we're expecting EBITDA growth of sort of 3%, 3.5% at the midpoint. As far as your question about revenue per admission, I'm not sure if that was an acute question or a behavioral question.

Behavioral. It was on the behavioral side because that was the only one that was -- the others were strong, and it was just that one that sort of stuck out because it was a decline.

Sure. So what we always, and traditionally encourage people to do on the behavioral side is really look at revenue per patient day rather than revenue per admission because the revenue per admission is distorted by the length-of-stay change. And in fact, in Q4, we had a relatively measurable decline in length of stay, a little over 3%. So even on a patient day basis, on an adjusted patient day basis, our revenue per adjusted patient day was only up a little over 1%, which is certainly less than it's been. I think it's a function of a couple of things. One is the continued growth in our managed Medicaid business, which tends to have a lower revenue per day and reimbursement than our Medicare commercial business; and also the challenges we were facing in our addiction treatment business, where we continue to move from out-of-network rates to in-network rates. And that had a bit of a dampening effect. I think over time, we think that that revenue per behavioral day should grow in the 2%, 2.5% range. And for the most part, over the last year or so, it has gravitated either to that range or slightly above it.

Your next question comes from Justin Lake with Wolfe Research.

First question, just on behavioral length of stay. Steve, can you give us an update on what's going on with this continued shift from Medicaid fee-for-service to Medicaid managed care, the magnitude of it and how much of the decline you think is coming from this?

Sure, Justin. I mean, we've said for some time, I think, certainly since about the beginning of 2017, so really for the last 2 years, that the reduction in length of stay during this period has been driven primarily by a continued shift of traditional Medicaid patients to managed Medicaid patients. And that remains the case. The majority of our total Medicaid patients are now in managed programs. In Q4, I believe, 65% to 70% of our Medicaid patient days were represented by managed Medicaid days. And that number has grown from, I would say, 50% within the last 18 months or so. So it's been a fairly dramatic shift. A number of large states or states that are large from our perspective, in terms of behavioral presence, have more recently gone to a managed system, states like Florida and Kentucky and Illinois. And we've felt the impact of that, and I think that continues to be the major driver of the length-of-stay contraction. Going forward, and I think we talked about this last quarter, I certainly have talked about it over the last quarter at conferences, et cetera, I think our point of view is that over the intermediate term, the next 12 or 24 months, we're still presuming that that process has to play itself out. And as a consequence, I think our guidance and our projections for the next couple of years presume that length-of-stay continues to decline by 1% or 2% a year. Those numbers can bounce around each quarter, but I think that's our point of view over the longer term.

So Steve, that's really helpful. So 50% goes to 65% to 70%. Has that been a steady sloping line? Has it leveled out at all?

It really has not leveled out in the last couple of years. I mean, it's been an increasing trend in the last couple of years. It's not an absolute sort of straight line, if you will, Justin. So I think length-of-stay decline bounces around. But if you kind of plotted it as points on the regression graph, I think you'd see a fairly steady decline over the last couple of years that, on average, is around a 1% to 2% decline. And again, that's the trend that we would project would continue for another year or 2.

And Steve, just last question, I guess, what I was asking is more just the last couple of quarters, right? Obviously, this would annualize at some point once it steadies for the last couple of quarters. And then if you're at 65%, 70% now, can you run us off a list of stance or maybe we could follow up just -- is there 2, 3, 4 states that kind of the bolus of where they still haven't gone to Medicaid managed care and on behavioral that we should be kind of looking at? That kind of the key last states that would be transferring over to this?

Yes. I mean, it's a perfectly reasonable question, Justin. And as you might imagine, I get asked that all the time. And I think we've been fairly candid in conceding that it's been difficult for us to predict with great precision how this shift was going to take place over the last couple of years and how it will take place over the next couple. We certainly know the states that are moving to a more managed sort of approach. Although I will make the point that in many states, the changeover for behavioral care is not always gone at the same time as the rest of the medical services population. So we can't always tie those together. And we don't, I think, have the same sort of insight and kind of real-time data that the payers have, quite frankly. So the best that we can do like I said, is to sort of largely kind of track where we are. Obviously, we know that no more than 100% of the population can shift, although I'm not sure we're convinced that it will be that high and presume that it'll continue to occur at about the same rate. But we can see it, and we wish this were not the case, that we were better able to predict with greater precision the trajectory of how that will go.

Your next question comes from A.J. Rice with Credit Suisse.

First, just to clean up a couple of things on the guidance. Do you have a bed growth number on the psych side? And would you highlight on the acute side any unusual items that are beyond just sort of the expectations for same-store revenue and EBITDA growth, maybe supplemental payment changes or -- I know the health plan has been a drag. You can -- do you expect that to be a drag above and beyond what the acute business does? Anything that's unusual that we might want to factor into our modeling on the acute side.

Sure, A.J. So I think your question surrounded mostly the acute care business. As I said, same-store revenue growth on the acute side is sort of 5%, 6% in the model. We acknowledge that that's kind of on the high end of what seems to be sort of industry averages. Although there's not a great many good public company comparisons for our acute care business any longer. I will say that most of that number is based on our historical trends over the last couple of years. We -- while I think that's a strong number, it's reflective of the sorts of experience we've had over the last few years and the strong performance, particularly in a number of our better kind of stronger franchise markets like Las Vegas, Southern California, D.C., et cetera. I will also say, however, that I think that number is inflated, particularly in 2019 by some of the increased capital spending that we've seen over the last few years. I mean, our overall capital spend has gone from roughly $350 million 3 or 4 years ago to closer to $700 million in '18 and again in '19. Some of that has been real big-ticket items like the new hospital in Henderson, but a lot of that is just continuing to enhance our franchises. We've added beds to our Spring Valley Hospital in Las Vegas. We've added beds to Henderson even though it's only a couple years old. We've added beds to Summerlin in Las Vegas. We've added a very large new emergency room project to our hospital in Manatee, Florida or Bradenton, Florida. We've added emergency room capacity and beds to our hospital in Denison, which is in the North Dallas market. So I do think that some of that acute care revenue growth is embedded in our guidance as well. As far as the health plan goes, as we've talked about over the last several years, the health plan has been steadily improving. The guidance for next year continues to expect it to improve. I would say it's a kind of smaller incremental improvement next year, maybe in the $8 million to $10 million range. And then the offset to all that is, as you mentioned, is the supplemental payments. I mean, we've got to schedule in the 10-K that projects this or it lays out the supplemental payments over the last several years as projected -- the 2019 number. And in 2019, we're expecting a measurable decline in supplemental payments. All that, of course, is included in the guide.

Okay. And do you have a bed growth target for the psych business?

Yes. I mean, we've been talking about, for the last several years, bed growth in the kind of 600 to 800 range a year, which includes, I think as I said in my prepared remarks, additions to existing facilities as well as de novos. It's sometimes a tough number to really get again down precisely because there's always a lot of -- a lot of it depends on local zoning and regulatory clearance and those sorts of things. But I think our point of view is that we ought to be in -- we were at the high end of that range in 2018, and we certainly ought to be in that range again in 2019.

Okay. Maybe just one last one on the capital structure. You're down to like 2.3x debt-to-EBITDA. That's low for the industry. It's low for you guys historically, even though you tend to be more conservative than the industry. I know you got the DOJ settlement. It sounds like that's getting close. You took another accrual for that, so if that's a cash outflow. But any thoughts on share repurchase? I know you've been at sort of $400 million, it sounds like, for the last couple of years or so. Any thought about stepping that up?

Yes. So I think you're -- you've almost asked and answered your question. That is correct that our share repurchase has averaged about in that $400 million range pretty consistently for the last several years. It's what we had embedded in our guidance for 2019. I think it's possible that that number accelerates either as a result of a settlement of the DOJ case and/or how we think about other external opportunities that there may be out there outside of CapEx from an M&A perspective. Those are often more difficult to predict. But at least, sort of what we have in our guidance is fairly similar to what we've run in the last few years.

Your next question comes from the line of Steve Tanal with Goldman Sachs.

Just looking in the queue on the Medicaid addition supplemental benefits. I know it's not an item you guys typically guide to. But the Q had a 1/5 -- like sort of implies $50 million for Q4. It looks like it came in sort of $67 million. Is it fair to sort of say that was upside versus the plan, without it, you might have been below the line in the guidance? Or how should we think about the timing there as well? The Q was filed in November, like, was that a really late in the quarter sort of surprise or...

Yes. So it's a good question, Steve. And to be fair, I know you posted it this morning. We've been looking at it. As I responded to you and look at everyone, the increased supplemental payments in Q4 were in our guidance. They're about $15 million or $16 million in Texas. And in my mind -- in our minds, I think largely an offset to the nonrecurring benefits we had in last year's Q4, the flu impact and the California UPL. I'm not exactly sure why the schedule didn't seem to reflect that in Q3. We're taking a look at that, and we'll let people know after we've had some time to...

All right. That's helpful. And then just thinking about that sort of program holistically, for the full year, net supplemental payments stepped up $22 million. Is it fair to assume that was driven primarily by a mix shift toward Medicaid? Or is there other sort of discrete factors that that would be overlooking?

Yes. I mean, I think it's not appropriate to just sort of make that assumption, although it seems somewhat intuitive or logical. The states themselves, and Texas, in particular, I think tweaks and retweaks and changes their supplemental programs quite a bit to respond to sort of various needs in various constituencies. So oftentimes, it's the underlying formula itself rather than changes in sort of the nature of our business per se. And I think in the case of Texas, over the last several years, it's been much more of that, that they've been changing the formula than it has been changes in our underlying business.

Got it. And is that kind of a similar sort of story for '19? In the [K] you spell out, you expect that to step down $30 million. So that's not really a reflection of how you're modeling payer mix shift, I guess, or is it, to some extent...

No, no, and that's a good question and a good example of the fact that the reduction in particularly in Texas UPL, which is probably about 2/3 of that overall reduction is again absolutely related to a change they've made in their, I'll call it, their formula or their approach rather than any change in our business.

Got it. Okay, that's helpful. And then, I guess, just bigger picture on the behavioral segment. Just would want to check in on kind of the long run thought process for the algorithm. Are you still kind of thinking that 5% same-store revenue growth is achievable? So when and how are you guys thinking about kind of same-store EBITDA growth that's achievable longer term in the business?

So I think our view of the fundamentals in the behavioral business have really changed very little over the last several years. Although certainly, we understand and can see that the business itself has been under more significant operating pressures than we've seen in some time. And I think our point of view is that the behavioral business same-store revenue growth had been averaging for many years that sort of 5% to 7%. Certainly, it had been averaging that in the first half of this decade, the 2010 decade. And then that growth slowed pretty considerably around late '15 or early '16. A lot of that slowdown was, we believe, attributable to a labor shortage. We still can see that we're in a pretty tight labor market, but we've made some improvements there and I think solved some of those issues. We've also struggled with a length-of-stay issue that I discussed with Justin before. But I think we ultimately believe, and I think, are heartened by, for instance, the 4.5% same-store adjusted admission growth in the quarter. We're heartened by the idea that, and I think maintained this all along, that the underlying demand for behavioral services continues to be strong throughout our portfolio, throughout our service lines, et cetera. And that continues to be our view. And our whole focus is just on doing the things we have to do to be able to sort of solve the issues, whether they're labor shortages or length-of-stay issues that will allow us to get to those levels of historical benefit from that underlying demand. I think the one change we've made in our guidance for 2019 that we -- that is different than what we did in '17 and '18 is, we're sort of no longer projecting or predicting an exact timeframe in which we'll get back and restore that 5% growth. I think we, as I said in my remarks earlier, are projecting that in our guidance at least, that the behavioral business continues at about the same pace it's running right now. And at whatever point, and we're working quite on a very focused basis to get there. But on whatever point it improves, we'll adjust our guidance. We'll revise our going forward sort of projections, but we're not going to kind of play this game of projecting and then reprojecting at the moment. Because I think -- we think that's been a little too difficult to do.

Got it. And maybe just the last question on behavioral and or in general for me. The write-down or impairment charge for Foundations Recovery Network, there's 3 sort of pieces to that. One was the wildfire impact on a facility. Honestly, not sure how large that was relative to the total write-down. But you did call out kind of tougher expectations for reimbursement and perhaps competitive dynamics, indicating fewer de novos. Is that a localized issue? Or is your outlook on such -- sort of changing a bit in that? Help us think about that.

Sure. So we've talked, certainly for the last few quarters, about changes in the addiction treatment business model that certainly the Foundations Recovery Network represented. That was an addiction treatment model that really relied heavily on direct-to-consumer marketing, either through the media, television, radio or through the Internet. There have been a number of changes in that sort of marketing, particularly in the Internet and some of the search engine logic that's made patient capture more difficult for providers over the last several years also. And I think many providers have acknowledged this that, that business has moved over the last several years from an out-of-network model to more of an in-network model, and that certainly results in lower reimbursement. Foundations also relied more heavily on a travel-to-treatment model in which patients would often travel longer distances, sort of outside of their home markets, to get what they consider to be sort of the best treatment available. I think payers are -- have been more restrictive about those kinds of decisions, et cetera. So all those things have, I think, affected the growth trajectory of that business. And then as you point out -- and certainly, we've acknowledged that and talked about that for some time. And then in the fourth quarter, I think our most profitable facility in the Foundations portfolio or addiction treatment facility in Malibu, California was closed as a result of the wildfires there. And there doesn't seem to be any path to reopening that anytime in the either near or intermediate future. So that was sort of the triggering event, although certainly not the primary one to the write-down.

Awesome. All right. Maybe just an update on behavioral leadership changes, well, how is that search going? And then, sorry for all the questions. I'll leave it there.

Yes. I mean, as we -- well, as I sort of discussed in the conferences that I've attended since Debbie left the company, we've talked about the fact that we would undergo an aggressive and comprehensive search to replace her. And we've done that. We've -- it's still relatively early in the process, but we've been pleased at how the search is going. We think there are a number of good, solid, viable candidates who we're exploring and we'll continue to do so. But we're quite pleased with the way that our behavioral team has sort of stepped up in the interim. Lots of people kind of filling in and taking on additional responsibilities in the interim and feeling very comfortable that how the business is being run at the moment. And so when we obviously have an announcement of a new person, we'll make that. But we're very comfortable in the interim that things are progressing as we would have hoped and expected in the interim.

I've mentioned also that Mark has now taken a very active role of replacing Debbie, and it's working out very well.

Your next question comes from the line of Josh Raskin with Nephron Research.

Steve, a quick -- I guess, the first one just a clarification on the DOJ settlement accrual. Would you characterize that as any change in progress? Or is that just sort of latest proposal from UHS at this point?

Well, I think it's both, Josh. I mean, as we've said for a while now, we've adjusted our reserves periodically, pretty much lately every quarter to reflect whatever our latest offer is. So that's certainly what the reserve reflects or something very close to our latest offer. But I also think it's worth noting that the gap between us and our offer and the government's demand, has narrowed quite considerably. And I think we view ourselves, certainly on the monetary issue, to be close to agreement on a final number with the government. There obviously are other issues to be negotiated along with that, including release terms and a compliance agreement and the end of kind of all -- the spectrum of investigations. But we would hope that the monetary piece of this is sort of the most difficult and that once we can agree on that, the other items will fall into place in short order. Although that's always difficult to predict with the government. So I think it's difficult for us to project a -- any sort of precise timeframe here, but we're certainly optimistic that on the core monetary issue, we're close to a settlement with the government.

Got you. All right. Definitely a change in tone then, I guess. And then just on the acute care side, I'm curious, are you seeing any changes in trends around CapEx spending by competitors in any of your, let's call it the larger markets, either new capacity whether that's inpatient or outpatient? Are you seeing others sort of step up their spending as well?

That's sometimes hard to say. I do believe, particularly in our markets that I think have shown relatively robust economic growth markets. Again, as I mentioned earlier, like Las Vegas, like Riverside County, California, like the District of Columbia, some of our Florida markets, the North Dallas market, because I think they're growing markets because we've done well in those markets, I think that, for the most part, we're seeing our peers in those markets investing as well. But I think one of the reasons why we tend to be focused on making sure that we're maintaining our franchises in those markets and hopefully enhancing our market share positions is the idea that we want to make sure that we're not overtaken by our peers. So I don't know. And obviously, we don't really have access to a lot of objective data about exactly how much is being spent in each of these markets. But my anecdotal sort of notion is that many of our peers, at least those that can afford to, are investing in the better markets. But I think we feel that we're maintaining our competitive position in -- at least in every one of these markets.

Your next question comes from the line of Sarah James with Piper Jaffray.

You've talked about the behavioral model yielding flat EBITDA near for top line growth 2% to 3.5% and more like 6% to 7% EBITDA achievable on a 5% top line year. I'm wondering if there are things that you can do on the supply side or other areas that would allow you to achieve EBITDA growth even if top line is growing less than 5%.

So Sarah, I mean, as CFO, I always sort of have the position that we can always drive more efficiencies than we have. And that's sort of the message I deliver to operators. But the reality is, if you look at our behavioral business over the last several years is that even on relatively modest growth, we've maintained margins in the mid-20s. I think it would be unrealistic to expect quite candidly that until we can engineer or restore that, that sort of historical level of revenue growth at around the 5%, that at 2% or 3% growth, it's possible to really have any sort of measurable, either margin expansion or EBITDA growth, et cetera. We certainly strive for that, and we strive to be as efficient as possible. But the nature of the operating model is that most of our costs are fixed and semi-fixed. So where you really generate the leverage in this model is through, at least, modest revenue growth. And until you get that, it's hard. Again, I'm going to always say, there are pockets of opportunity for greater efficiency. But I wouldn't say that there are -- there's any low-hanging fruit out there in terms of driving greater efficiencies.

Got it. That's understandable. What about on the acute side? Are there initiatives underway with supply costs or scheduling that you could look to drive some improved leverage there or margin expansion on the acute side?

Yes. I think the acute side is a different question. I think that there's an acknowledgment on our part and on the part of our operators as well as on, I think, observers in the industry that there is a fair amount of duplication and some excess costs broadly in the acute industry. This is not specific to UHS. I think that the general view is that the real way to drive improvement there is through changes in the payment model, and those are certainly occurring slowly. And we're moving, again, incrementally away from the traditional fee-for-service reimbursement model to more of a what's called fee-for-value model. But I think, I would describe as more of a risk-based model. We disclose in our 10-K, for instance, that we've agreed to participate in a number of additional bundled payment projects for Medicare. I think those will drive incentives and encourage sort of throughout the continuum, more effective behavior and sort of a ringing out of costs. And we're very focused on that and have talked about it for a number of years. So as the payment model changes, I do think that there's an opportunity on the acute side to drive more efficiencies and less variability in the system. And that should be helpful in, at least in part, in growing profitability and EBITDA margins.

Your next question comes from the line of Kevin Fischbeck with BoA.

Wanted to ask about the psych revenue per patient day. It sounds like one of the factors is the Medicaid mix shift, which is likely to continue, it sounds like, for the next year or 2. But the other factor you mentioned was going back in-network with some of the addiction centers. I guess, I don't remember really hearing a lot about that in the past. Is that something that's just happening now? When did that anniversary? Or is that going to be a longer-term headwind to pricing?

We certainly have talked about that dynamic the last couple of quarters. I don't know, honestly, that it has had a measurable impact on pricing before the fourth quarter. And so my sense -- and largely because the Foundations contribution to our overall behavioral performance is still relatively small. I don't know that over time, it will have a real dampening effect on our revenue per day, which is why I said earlier, Kevin, that I think our longer-term view is that behavioral pricing should be in that kind of 2%, 2.5% a day range. And the reality is over most of the last couple of years, we've been hitting that range, if not somewhat exceeding it. And I think as we looked at some of the factors affecting Q4, they were a bit anomalous and don't really expect them to continue at the same level of magnitude.

So would you say that -- so that 2019 is going to be below that number in your guidance?

Below? Well...

[2% to 2.5%.]

No, no. I think that effectively, what we're really projecting -- when we're projecting 2% to 3% revenue growth in 2019, we're essentially saying that that will largely come from pricing and that volumes are projected to be relatively flat in our guidance. Our hope would be we can exceed that, but that's what our guidance implies.

Okay, great. And then just circling back on the labor cost issues and the behavioral business, you talked about making progress there. So where do you think you are in that progress? Are you halfway through 2/3 of the way through? How is that shaping up?

It's always a difficult question to answer when posed that way, Kevin, in the sense that it sort of suggests, it's sort of a linear process that we have x amount of openings at a point in time, and then we're able to fill 50% of them or 75% or whatever. And the reality of it is that we fill openings and we hire and we train people and then people leave, and et cetera. It's a very fluid kind of a dynamic. And particularly in a tight labor market, which I think the current labor market is appropriately characterized as. So I think we've made a lot of progress since we began to talk about this issue in late 2015, early '16. But we also acknowledge that it remains a tight labor market, and there are still facilities where we have vacancies. And in some cases, they're sort of a chronic problem. But certainly, we don't have the level of closed beds and closed units that we had 2 or 3 years ago. And that tends to occur now on a much more sort of one-off basis. But again, I think that providers, in general, and behavioral providers in particular, are going to be facing and -- the issue of the labor shortage at both the nursing and the psychiatrist level will continue to be an issue for the foreseeable future as long as the labor market remains as tight as it is right now.

Okay. And just last question. The new head of the psych business will you bring, whoever that is, is there any change in direction or emphasis that you would expect the new person to bring in? And does having a new head of that business in any way help a DOJ resolution?

Yes. I think we have a point of view that we, and I think folks who listen to our conference calls, et cetera, certainly have a good appreciation of this over the last several years, have faced some difficult operating challenges in the business. We've touched on those already in the call, the labor shortages, the pressure from our managed Medicaid payers, increased competition, et cetera. And I think it's been difficult for all of our operators, including the head of the business segment, to kind of take a step back and think about how to grow this business over the kind of longer term. We believe very firmly that there's a very significant role for behavioral care in the future health care landscape. We also believe there's a growing demand for behavioral care. And I think all that is validated by much of the literature, et cetera, that's being written about how to effectively deliver health care in the future, et cetera. And so I think we're hoping that with a new person sitting in that lead chair that they'll have a bit more time to reflect on some of those longer-term issues and the longer-term growth opportunities in the business. And in the meantime, we remain, and I think Alan's comment is we remain at every level of the organization, both he and I and Mark, as well as a very capable staff of senior and mid-management behavioral leaders are very focused on sort of making the trains run on time and solving and addressing all those operating issues. But we would hope that a new person would really be able to do some things from a longer-term perspective that maybe we've neglected for the last year or 2.

Okay. And is there any implications for the settlement? Is it having a new person in there clear the air at all may change the timing in your view?

No. I don't think that personnel change has anything to do with the DOJ investigation or settlement.

Your next question comes from the line of Steven Valiquette with Barclays.

So on this whole behavioral managed Medicaid issue regarding the shorter length of stay and the leverage you can maybe pull to try to offset this or mitigate the impact. I'm just curious about on the revenue side, perhaps renegotiating contractual terms and diving a little bit deeper. Is there any color on the notion where under value-based care, you could receive either bonus payments or just some sort of better compensation for having shorter length of stay, whether it's versus peers or some other metric? And shouldn't that kind of be the end goal to some degree? But just curious to get your thoughts on this concept of being rewarded for having shorter length of stay in behavioral.

So it's a good question, Steve, and I think that there are payers who would make that argument. I think, unfortunately, we have a point of view that length of stay has been viewed on the behavioral side of the business as a proxy for some sort of quality of care metric. And I think we feel that's a fundamentally flawed approach. At the end of the day, we have a point of view that length of stay is really a clinical determination that should be made by clinicians based on the clinical needs of a patient rather than the financial outcomes. And so I think we're reluctant to sort of promote a system that encourages anybody, providers, et cetera, to really drive lower length of stay just to achieve a better financial result. Because at the end of the day, I think we're concerned that that shortchanges the clinical needs of the patient. So while I think there would be payers that would welcome that, I think that fundamentally, we prefer other measures of quality that we believe exist and we believe are perfectly appropriate in terms of quality, rewards and quality, bonus payments. We think length of stay is the wrong measure for that.

Your next question comes from the line of Ralph Giacobbe with Citibank.

You mentioned a strong JV pipeline and a new build. I guess, can you just help frame the opportunity, Steve, there and whether sort of contribution there, sort of an accelerant to growth or just sort of needed to get back to sort of the baseline that you talked about in that mid-single-digit. And then you had the write-down on the addiction treatment center. Is that still an area of focus or opportunity? Or what's your interest maybe more broadly and generally around expanding service lines at this point?

Yes. So I'm sorry, Ralph, can you just remind me what the first half of your question was?

Sure. The -- just the JV pipeline.

Yes, the JV pipeline, I'm sorry. So look, we've talked about this a lot. I think we think that the -- broadly, the JV opportunity is a very significant opportunity. Somewhere around 50% or 55% of all the inpatient behavioral beds in the U.S. are today, operated by acute care hospitals. And so to the degree that we can penetrate that market in some way by helping to manage those businesses, by leasing those beds, by partnering with those acute care hospitals, by building new facilities with those acute care hospitals as we've done in many of these instances, that's probably the single biggest domestic growth opportunity we have in the behavioral business. We've also acknowledged that despite our focused efforts, it's a relatively slow-developing opportunity and will continue to be so. Now we'll continue to focus on it, and we'll continue to do those transactions that make economic sense. But it's hard sometimes to make the acute hospitals want to go or need to go faster than they're going at the current time. In the short run, and I think I kind of touched on this before, the sum of those projects can be a little bit of a drag. So in 2018, we opened the joint venture, the Lancaster new beds. We opened a new hospital in Washington State with Providence Health Care. That's a bit of a drag. And in 2019, actually, some of the improvement in behavioral will come from the continued ramp-up and growth in those facilities. And there's not a lot of brand-new facilities coming on in '19, so there's not much of a drag. But ultimately, over time, and again, when I say over time, in this case, I'm really talking about a time frame of 4, 5, 7 years. I think we think it's a very significant growth opportunity and one that positions us, I think even more importantly, as a partner with not-for-profit acute care hospitals around the country in a way that we may be able to lever in other service areas. As far as your question about the addiction treatment business, I mean, in general, I think we are taking a pause on expanding, again, what I'll sort of call this new style model that characterize Foundations Recovery Network of direct-to-consumer marketing, travel-for-treatment, that sort of issue. In general, we acknowledge that addiction illness, particularly opioid addiction, but quite frankly, addiction illness of all sorts, continues to be a growing phenomenon in the country and needs to be treated. It can be treated in many ways. In our old model, which is sort of not a direct-to-consumer marketing, but a referral source marketing kind of an aspect. So we'll continue to take advantage of that. In terms of other service lines, as best as I can tell, Ralph, we have probably the broadest service line offerings of any inpatient behavioral provider in the country that spans general psychiatric treatment, general psych treatment for the elderly for diseases like Alzheimer's and dementia, autism, eating disorders, all kinds of sort of niche behavioral treatments. So we're certainly open to expansion, but there aren't a whole lot of behavioral illnesses that we don't already treat somewhere in our portfolio.

Yes. That makes sense. I was asking more about sort of the -- if that gave you a pause on -- but it doesn't sound like it does. And then just a follow-up question. You typically break out the performance of some of your bigger markets on the acute care side. I know behavioral isn't as concentrated, but can you give us a sense, even by market or maybe even as you look at the broad portfolio in terms of -- what percentage is really underperforming? I guess I'm really just trying to get a sense of whether it's a small percentage of facilities really driving the softness or if it really is sort of a broader pressure that you're seeing across all your markets.

Yes. I mean, the math is such, Ralph, that the reality is our 2 business segments are about the same size from a revenue perspective. But obviously, we have a much smaller number of acute care facilities. And those facilities tend to be more concentrated than they are on the behavioral side. So we certainly have talked always about the Las Vegas market, and it's hard -- I don't know that there's another -- I'm certain that there isn't another public acute care company that has sort of a market presence comparable or market contribution comparable to that. But on the behavioral side, we generate roughly the same amount of revenue, but we have a much larger number of 200-plus domestic facilities. It's really impossible for any one facility or really even any one market to have the same sort of measurable and material impact that a Las Vegas does on the acute side, which is why we really never talk about kind of individual markets for the most part on the behavioral side. I think sort of your question about sort of what percentage of the hospitals are underperforming, et cetera. My general sense is that in a portfolio of 200-plus hospitals, it's always going to be sort of like a bell curve where there's going to be a small number of outperformers or a small number of underperformers, and the vast majority of hospitals are going to be in that large middle. And in that large middle, I think we have a point of view that the issues that we've discussed over the last few years, labor shortages, managed Medicaid, length-of-stay pressure, increased competition, are issues that are being felt by a relatively wide array of facilities and are not particularly focused on a specific market.

Your next question comes from the line of Pito Chickering with Deutsche Bank.

A few quick ones here, following up on A.J.'s acute bed question, you grew beds fourth quarter by 2.4%. What percent of your 5% to 6% revenue guide in acute business comes from bed additions? And as you look at the supply versus demand in your markets. How confident are you that is bed additions continue over the next few years?

So I tend to think about the capital investment in the acute business not so much on a bed basis. I think that's a relatively kind of dated way of looking at the business, which is not to say, again, we certainly have added beds, and I think that's a reflection of our ability to want to meet demand. But we've also are spending a lot of money to increase emergency room capacity and to increase surgical capacity at many of our hospitals and other service lines as well, particularly in what I would describe as sort of the high-end service lines like cardiology and orthopedics and neurosurgery. At the end of the day -- and again, to be fair, Peter, I don't know that I have a precise number because I think it's difficult to really parse it to that level. But what I was trying to say earlier is that of our 5% or 6% acute care revenue growth, some of it, maybe 1% or 2%, I do believe, is really being driven by this increased capital spending not just on beds, but again, on the other items that I talked about. But it's very difficult to say when you add a bed or you add 5 more ER days or you add another operating room, exactly what the contribution of the incremental investment is because these things -- this capacity effectively becomes fungible.

Okay, fair enough. On the behavioral side of the business, you guys did a great job with admissions to offset length-of-stay pressures. How sustainable do you think the 5% admission growth is? And is there any margin impact from having more patients stay for shorter period of time?

Yes. Look, I think we've conceded that a shorter length of stay is a bit more of an operational challenge, which I think is probably intuitive to people who think about it that turning over patients more quickly requires a bit more of an effort, et cetera. But we also acknowledge that, to some degree, that's the way the business is headed. And so we'll deal with that. I think we have a point of view, and again, sort of harkening back to an earlier exchange I had, that 5% revenue growth, at some point, is -- restoring that number is not unrealistic. It continues to be, in our minds, very achievable. We think of that as sort of 2.5% to 3% volume growth and 2.5% to 3% admission -- pricing growth rather. Depending on what happens to length of stay, that will drive sort of what the required admission growth will have to be to get to that level. But again, I think we have a point of view that in the relatively near or intermediate term, 2.5% patient day growth -- 2%, 2.5% patient day growth should not be unrealistic and would be consistent with what we've run for an extended period of time historically.

All right. And then last question on the behavioral leadership. Have you had any increased turnover at the divisional or other management levels since Debbie left?

We've lost a couple of people to Acadia since Debbie left. I'll make the point that in the normal course with UHS and Acadia being the largest certainly for-profit behavioral providers in the country, there's always a flow of personnel at various levels of hospital and regional levels back and forth between the 2 companies. So it's hard for me to say that that's terribly unusual. But it's only been a couple of people.

We just hired somebody from Acadia yesterday.

Your next question comes from the line of Whit Mayo from UBS.

I wanted to go back to the supplemental program question and just rip in the headwinds that you've called out in your 10-K. Just to be clear, there are also some tailwinds that you have coming with federal DSH, how much favorable IPPS update that should more than offset those headwinds, correct? And is there a number that you could size for what you think your Medicare DSH is this year?

Yes. I mean, the impact of the Medicare DSH, and thanks for reminding me, Whit and others, I think we've talked about it on previous calls. But it's probably in that $18 million to $20 million annual range for us. It begins in October of '18. But over the federal fiscal year, it's about an $18 million to $20 million benefit. That's embedded in our guidance as well.

Right. And then you've got an incremental pickup with your IPPS update that could be one another $10 million, $15 million or so?

Correct.

And coming into 2018, when I go back and look at your 10-K, you predicted that you would see a $30 million headwind to all of these state programs, and it actually came in much higher. I think you said 156. You recorded over 200. Before that, you expected 145 coming into 2017. It came $40 million higher than your 10-K disclosure. So I guess, I wanted to understand how you come up with this forecast because it almost always has an upward bias. And I don't think investors probably appreciate how fluid some of the calculations are inside these programs.

Yes. And it's a reasonable point, Whit, and I think it gets back to what, I think it was the conversation that I was having with Steve Tanal earlier in the call. What we're able to do at the time, we give our guidance is based on what the states' current model is we project what we think our impact or benefit is going to be. But often, that model changes. Often, they sort of have interpretive changes and whatever. And there are some underlying changes to our or some changes to our underlying business that are difficult to project. But yes, I think it's much more sort of the states changing their models that we often cannot predict. And so I do think we take a little bit of a conservative approach when we project that at the beginning of the year.

Yes. I'd agree. And can we go back to some of the 2018 behavioral headwinds? I just want to make sure that we're all in the same page with the numbers for hurricanes, the regulatory challenges, wildfires. Is there any way to maybe size each individual bucket so that we're all thinking the same numbers?

Yes. I mean, the things that I would call out, and at least I called them out sort of by item earlier, I mean, the continued improvement at the Lancaster and Spokane de novos, the Gulfport acquisition that was done at the very end of last year, the turnaround in those 3 facilities is probably an $8 million or $9 million benefit going into '19. Having a full year of the Danshell acquisition in the U.K. is probably another $4 million or $5 million in '19. The hurricane -- Florida hurricane and California fire impact is probably $7 million to $9 million drag in the second half of '18. And then the regulatory facility challenges that we had early in '18 were probably another $5 million or $6 million. I think all those things have been clearly discussed and delineated. People want to go back and check the numbers, but that kind of my recounting of it, at least, Whit.

Your next question comes from the line of Frank Morgan from RBC Capital Markets.

Most of them have been asked, but just, Steve, you mentioned CapEx in your guidance. Could you tell us what your implied cash flow from ops would be on that guidance for 2019?

I mean, I think our free cash flow in '18, Frank, was sort of close to $800 million. And basically, I think the free cash flow guidance for '19 is sort of [akin] to that with a slight growth in EBITDA.

Okay. And then finally, just any color around surgical volumes, inpatient, outpatient as well as ED business. And I'll hop.

As it is in most periods, I think surgical volumes have grown pretty consistently with that 2% or with whatever the admission growth is. So in Q4, that 2% admission growth sort of would imply, and I think is what we ran, kind of 2% to 3% surgical volume growth, both on the in and outpatient side.

Your next question comes from the line of Gary Taylor from JPMorgan.

Mine will be really quick because I think you've answered everything. I just wanted to go back and clarify one thing, Steve, and just make sure I understand it. So when you were talking about how you built up guidance for the year and you talked about behavioral, and I think you're suggesting that you still kind of view that business is running 2% to 3% top line with flattish EBITDA. But because of a number of idiosyncratic factors, a number which you were just discussing with Whit, the actual behavioral EBITDA growth guidance is plus 3% to 3.5% for 2019. Is that all correct?

I think that is all correct, Gary. Okay. We'd like to thank everybody for your time and look forward to talking to everybody again after the first quarter.

This concludes today's conference call. You may now disconnect.