DexCom, Inc. (DXCM) Earnings Call Transcript & Summary

Welcome to the Wells Fargo Health Care Conference. [Operator Instructions] Also, please be advised that today's conference is being recorded. Thank you.

Good morning, everyone. I'm Larry Biegelsen, the medical device analyst at Wells Fargo, and it's my pleasure to host this session with DexCom. With us, we have Steve Pacelli, Executive Vice President of Strategy and Corporate Development; and Jereme Sylvain, Senior Vice President of Finance and the Chief Accounting Officer. In terms of the format, it's going to be Q&A. DexCom has been kind enough to give us a 1-hour slot. So hopefully, we can cover a lot of ground here. If people want to ask questions, I'm going to ask you just try to email me. So if I'm looking to my left, you'll know I'm just scanning my e-mail to try to get questions from people on the line. We're also going to try to do one polling question, which is -- has actually worked well so far. And so with that, Steve, and Jereme, welcome. And hopefully, you can hear me okay.

Yes.

Thanks, Larry.

Good. Thanks for being here. So I wanted to start with the upcoming investor meeting in December. And just get some color from you. I know you're not going to provide any specifics but what we can expect from that meeting, it looks like you're on track to exceed almost all of the goals you said at your 2018 meeting. So I'll hand it over to you, Steve.

Yes. I can start, and then Jereme can jump in. I think -- yes, I mean, we're certainly not going to give you specifics here on this call, but I think you're right in that we've been tracking. We laid out some kind of long-term guidance a couple of years ago, late 2018. And we've obviously -- if things keep progressing, we're on track to meet or exceed that on that 2023 time frame that we've laid out. So I think you would expect that we would kind of update the numbers, broadly speaking, and then really kind of update the -- really, the portfolio and the product road map for the next, call it, 3 to 5 years. So you should expect some discussion around the core around intensive insulin therapy of type 1 and type 2. You should expect us to talk about our inroads into non-insulin-using type 2s. And then we'll talk about the hospital, gestational and probably some broader population health initiatives that we're looking at. Mindful that we don't want to give Abbott, Medtronic and others, our full playbook. So we may be a little cautious of how much on the portfolio and the road map we present. But certainly, we'll give some updated numbers.

Okay. That's helpful. And Jereme, did you want to add anything?

No, I think that's exactly it. And we're going through those numbers and updating those now. So we'll certainly be able to provide you really a financial overview of where the company's heading. So we look forward to that, and it will obviously take into account all the performance you see to date.

Okay. The one area -- I mean, you've exceeded, you've met or exceeded. It looks like you're on track to meet or exceed everything, except maybe the international target of 30% of your core business by 2023. Obviously, you still have a few more years. But right now, international is about 21% of total sales. Any comments there on whether that 30% target is still achievable?

So I don't think we're willing to update it yet. Although one thing we've always said is we do expect over the long haul, our international business to accelerate past out of our U.S. business. Obviously, this year, the OUS markets, at least for the middle part of the year, I say that I'll call it Q2 performed a little bit less so than the U.S. business. We don't expect that to be the case. If you look at our market share outside the U.S., it's certainly lower than what we have in the U.S. And so we believe that, that's an opportunity over the long haul. And so nothing has really changed from our expectations. But in the event that we do reset any bars, you'll see it at that period. But we've always thought the international market has a much larger runway for us than the U.S. market does.

Okay. Well, why don't we transition to COVID-19 impact. When we look at the IQVIA data, I know you don't want to give any intra-quarter updates. But when we look at total sensor prescriptions and new patient starts, it looks like they're almost back to pre-COVID levels for new patients starts based on this data. Any color? Any reaction?

Well, I think what we would say is, in general, around IQVIA data. IQVIA data is really through the pharmacy. And so I think you just have to be careful. I think we talked about this maybe last time we spoke about it, but a lot of our business still resides in DME. And there's a migration of DME to pharmacies. You have to be very, very careful in just assuming that one portion of the business or one viewpoint of business is a broad-based assumption for everything. In terms of new patient starts, we won't necessarily comment about our numbers other than there's been public data out there regarding many endocrinologist interviews. A lot of folks have different assessments of inpatient, outpatients. People are looking at IQVIA, Ziten, et cetera. And what we would say is, typically, what we've seen is patient visits are still not where they were pre-COVID levels, and that's kind of what we would have expected just given some folks are seeing massive amounts and not. So it's nothing different than what we've really commented on in Q2. We don't really have a Q3 update. But I think it does give some good context around IQVIA. It's a good data point. I think it's helpful for you to see how things are going and progressing. But I always just want to make sure that you understand pharmacy is only a portion of our business. There's a large portion that still remains in DME. And so it's not a total viewpoint of the business.

Okay. And you talked about endocrinology in-person visits, which we also track, still being below pre-COVID levels. We have heard that it's easier to onboard a new patient to Libre than it is to DexCom to G6. Is that fair? And what are you doing to level the playing field? To onboard through telehealth?

Yes. I would take exception to that, but for the fact that Abbott has been very aggressive at not pursuing DME. So that they're trying to push product through the pharmacy. And so where -- it's an apples-to-apples comparison, where our patients are picking up their products, the drug store, ordering through the pharmacy channel. I would say the onboarding is identical, where that would ring true potential is in the DME world, where patients are still required to get fingerstick logs or chart notes from the doctor still jump through all those hurdles that we've talked about over the years in DME, which is, again, why we've been so aggressive in trying to push our business into the pharmacy. So I think where there's an apples-to-apples comparison, there wouldn't be a difference in onboarding for a Libre patient versus the DexCom patient. We even talked on our last call, even from a co-pay perspective, the co-pays are very similar, in fact we're often a lower co-pay than the Libre. So I think it just speaks to our continued effort to push the business in the U.S. into the pharmacy channel and away from DME. But as Jereme just mentioned, it's going to take some time because we've got that legacy business that is largely through our third-party DME distributors, it's just going to take some time to convert them from DME over to pharmacy.

Question that was e-mailed to me. Are there any sell-in and sell-out dynamics to call out with DMEs in the second half of 2020? I don't know if it's something you had commented on in the past?

Yes. Typically, no. One of the things we've always been mindful of doing, and this is just over time, we don't really have a large -- we don't want to have inventory sitting out in any of these channels for extended periods of time. And so I assume what they mean, sell-in, sell-out is really stocking levels. And we don't enjoy having those stocking levels. We keep them all within a certain amount of days, and that's that we can be nimble with our inventory levels. And that's been the way we've always really run the business. So that's the way we would continue to look at how we operate the business.

Okay. And then recent commentary from Abbott, CMS and even DME providers suggest that CMS is providing interim coverage for CGM. The type 2 patients without regard, the daily testing need or confirmed COVID infection. I know there's been a lot of controversy about this -- that CMS announcement that came out earlier this year. What's your perspective? And what impact do you expect that to have?

So we've taken a conservative review of the CMS language. And in our view, it really is, yes, there has been some laxing of the rules. But I would say our view is that if a patient has diabetes and has COVID, regardless of whether they have type 1, type 2 or even not taking insulin, they would be eligible for coverage during COVID. I would caution, though, I don't think you're going to see a huge influx due to COVID, a huge influx of non-insulin-using type 2s as a result of CMS flexing these guidelines. I just don't think that's going to really move the needle. I don't know, Jereme?

Yes. No, I was going to say that the guidance we read it to be specifically somebody who's contracted COVID, not just any type 2 or anybody that's afraid of getting COVID. If folks have taken the angle, and we've heard some obviously rumblings of folks taking a more aggressive angle. If you want to play auto roulette with Medicare, we don't recommend it. And I think a lot of the suppliers that we've spoke to also don't recommend it, but that's at least been our approach and ultimately how a lot of the folks have applied at least DexCom product into therapy.

One more on COVID. I mean patients, diabetes patients, with quarterly controlled glucose fair worse if they contract COVID. For lack of better word, maybe silver lining. What are you seeing? Are endocrinologists and diabetes patients more eager to go on CGM to control their glucose in this environment because of that?

It's a good question. It's one of the things as we look at the feedback that comes back in, a lot of the folks that are aware of CGM, this can be a catalyst to hurry up and accelerate folks to then move on to CGM. There's no question there. For folks that aren't aware of CGM, and as you get into the primary care space, where folks aren't necessarily up to speed on diabetes, they might have a few diabetics in their practice. That's more of an education thing. And as we educate them, they say, well, why would we die? So we do believe that over time, the move to telehealth, the move to COVID being an impact and who it disproportionately impacts those with diabetes. We do think it's a tailwind. It's hard to know, just given all that's going on with patients and are they coming and who's on telehealth and do PCPs have a telehealth practice? Or are they just slower? There's a lot of ins and outs. But the feedback we've gotten over time, especially as those have realized what our product do is that this is a perfect product for telehealth and telemedicine. And we believe that over time, it could be a catalyst. It's going to be hard to tell. I think it's going to take time to ultimately tell how much of a catalyst it could potentially be.

So I wanted to transition to G7. Obviously, critically important product for you guys. So we understand the first-line is up and producing units, and you're setting up incremental lines to ensure a smooth launch. Could you provide more color on where you are with G7 manufacturing scale-up and the clinical work?

Yes. So we're not going to give you a specific update. On our Q2 call what we said is that due to COVID, we expected to push out the clinical trial, just access to the clinics by around 6 months' time frame. So can't give you an update on this call. We'll give you an update on the Q3 call as to where we are with respect to kind of pivotal trial. And remember there's a pivotal trial, if you will, in the U.S. and in Europe. So we'll probably try to give you some color on both on the Q3 call. Yes, we can produce G7 sensors, but I wouldn't suggest by any means that we're at anywhere near close to scale. I would actually characterize what we're producing today as more of a pilot-line, producing product for our clinical efforts and not for mass production from a consumer perspective. So that will continue over the course into building -- bringing in the equipment, building the equipment, building the lines we'll continue through the balance of this year and really throughout the balance of next year.

Steve, is there any reason the G7 pivotal trial will look different from the G6 pivotal trial, which is obviously in the public domain?

I think at the end of the day, we have to meet the iCGM standard, right? So you have to power the trial appropriately to achieve the statistical metrics that you need to obtain an iCGM. It's actually, I would suggest that it's a more complex trial than the G6. But in terms of numbers of patients, it's roughly -- there's going to be more patients but not in order of magnitude more patients. So by and large, yes, you're still running patients in the clinic, multiple days during the course of the sensor where you're manipulating blood sugar levels, while they're in the clinic. So in that respect, it's very similar.

And you talked about COGS for the sensor being below $15. Eventually, I guess, at scale for G7. Is that accurate? And how does that compare to G6?

Yes. So we've talked about getting it down to right around that range at scale. And so we haven't really talked about specifically comparing it to G6, G7. But what we would say is, as you get it down -- as we ramp upscale, I think we said it in the context that you shouldn't expect to see our COGS to be an issue to compete with our competitors. We believe we can get it down into a similar cost profile as our competitors. We won't comment on where G6 and G7 are right today. But I think in that question, it came down, I think the other question that came up is, well, where can you get G6 over time? And while we didn't give it a specific price as where we are today, a good chunk of the cost of G6 today is fixed overhead, and that fixed overhead is depreciation. And when we get to a fully depreciated line, the incremental cost on G6 can also be pretty low. And so I think that's where we've talked about being able to compete in COGS with really anybody in building that infrastructure to do so. So I think that builds the confidence at least. When you get into lower-cost markets, are we able to compete? We believe we can, and certainly, we believe we can compete with both G6 and G7.

So just to follow-up on that. It sounds like you're planning to continue to offer G6 after you launch G7 by segmenting the market with a higher and lower-priced offering. Is that what I'm hearing?

It may not be in the same market bracket. Maybe a lower-priced offering or a different priced offering in markets where reimbursement is lower and at G7, where reimbursement is higher. It may be for different use cases within those markets as we start to think about which use case we're targeting, whether it's type 2 and in a type 2 environment or type 2 intensive. We're working through those types of items. But I do think you should expect to see G6 lines continue to run and to continue to supply for various groups of folks for foreseeable future. It's a lot of investment we've made. It's a great product. And we can use it to attack certain markets that maybe we're not attacking today.

And I know pricing have got a lot of attention 1 or 2 years ago, doesn't seem to be getting a lot of attention right now. But you've talked about a 10% annual price erosion for a couple of years, I think, at least this year. Is there a level where you think the price erosion plateaus?

I think it's when the transition -- so price erosion if you think about it today, majority of price erosion is actually channel shift as opposed to same store, same SKU type sales. And so if you're in the pharmacy yesterday and you're in the pharmacy today, the pricing erosion isn't quite as there. Same story on the DME space. If you're in DME yesterday and you're in DME today. The issue is, is we have this migration over time of folks in DME into pharmacy, where there is a pretty dramatic price difference. And so where that plateaus is when that migration shift stops. Steve had mentioned, alluded to it earlier, we want folks to go through a pharmacy. We believe that's where you get those barriers to adoption down and that ease of use down, where you get the volumes that will offset it. And then our OpEx associated with serving the pharmacy gets dramatically lower. I think that's where we'll continue to push. So when that migration takes place, we'll certainly continue to see it. We've talked about it being this year. We've talked about it even being next year. We haven't really commented beyond 2021 in terms of pricing. It really depends on how many folks we're able to migrate over time.

I mean, Jereme, when you talk about 2021, it's about the same level as 2020?

We talked about the same dollar value, so around that $150 million. So really it's about taking the existing base and moving them. And as you move that existing base, each person comes with an incremental price set.

Okay. And sticking with G7, is the U.S. going to be the first market? Or could it be international given that it seems like a CE Mark are easier to get?

That's still to be determined, and I would also suggest it's not just a function of regulatory, but a function of our capacity and our scale-up. Because remember, the hole we put ourselves into with G6, trying to launch G6 for the first time, it was -- tried to do a global product launch. And we underestimated the demand. And we were living most of last year in kind of a hand to mouth or even oftentimes a backorder situation. We're not going to do that with G7. We're going to be much more thoughtful in how we roll it out. So the U.S. being our biggest market, right? If we're not ready to go in the U.S., we would probably consider launching it somewhere else, but we're still working through that as we work through the clinical regulatory and scale-up process.

Okay. So let's transition to type 2. And I think first question here. I've got a couple of those -- a couple of type 2 questions e-mailed to me already. But just at a high level, I think people were surprised by the 20% figure you provided on the Q2 call for type 2 patients as a percent of your U.S. installed base. Where was that before Medicare reimbursement came into effect in 2019? And how much of growth is being driven by Medicare versus commercial?

I would say prior to Medicare and prior to the German reimbursement decision, if you remember that in Germany, out of the gate, we received type 1 and intensive type 2s, I would say, it was very minimal, right? And certainly, less than 5%, probably mostly cash pay. So what we've seen, particularly with Medicare and Medicare has been obviously the key driver for our intensive type 2 business here in the U.S. But then what we've seen is stemming out of Medicare, you saw United, you saw Aetna, there's some others. What we expect now is sort of an acceleration over the next, call it, 12, maybe 18 months that we expect most of the larger payers to fall in line and actually adopt intensive type 2 coverage along with their type 1 coverage. So it is a huge -- it basically -- actually, probably more than depends on whose numbers you believe, but probably more than double the size of the addressable market just in our core intensive business in the U.S. and in Europe. So pretty exciting stuff. I would tell you, though, that the biggest challenge we have with the intensive type 2 is that many of them are not seen by endocrinology. So they're not -- they don't have the education and the access that our -- most of our type 1s have, who are being seen by an endocrinologist who are very well versed in CGM as a category, as a therapy. We're working through our direct-to-consumer campaigns really to target. That's the effort that we're putting forth on these non -- or these intensive type 2s. Just to educate them because the primary care or internal medicine docs, they don't necessarily know. They don't have experience with CGM. So we're having to really educate them through the DTC campaigns, have the consumer get to the doctor to drive the prescriptions.

That's helpful. So one question that was e-mailed to me that you touched upon, Steve. It's could the pace of coverage for type 2s, could we see more private payers come on before year-end?

Yes, it's possible. Many of them, remember, payers work in like cycles, right? So there may be things that we have in the works that we are expecting someone to adopt something, but it may come into play January 1, right? Even though we're doing the work now, they often put out revised coverage policies in buckets, as you guys know. So if there's anything material, obviously, we would talk about it on the Q3 call.

Yes, but I think it's worth noting, I mean, 2 of the 5 came in this year in 2020. And so on top of Medicare, you get Unitedhealthcare and Aetna who have moved into type 2 intensive coverage. So you're starting to see that commercial mind shift take place or that private payer mind shift take place. So I think that's certainly encouraging. We believe it's a matter of time, not if and so I think the timing, I think, to your point, is the big question. But of course, we have our access folks working with them daily to look at these items.

And the 20% figure you gave on the Q2 call, was that U.S. installed base or global? Steve, it sounds like it was global based on your German comment a minute ago.

I would say global, but predominantly U.S. The uptake in Germany with intensive type 2 hasn't been nearly what we've seen in the U.S. base.

So does that mean -- so 20% applies to worldwide and the U.S.? Or is the U.S. now even higher than that?

The comment was worldwide. And I think that's where we left it.

Stick with that.

Okay. But U.S. should be too different. It doesn't sound like or it could be actually a touch higher, not dramatically different?

Yes. I think what we had said is 20% worldwide. But we don't really get down into patient numbers, U.S., OUS, I think you can take Steve's comment and take it where you want. But I think 20% worldwide is the number we publicly disclosed.

Okay. Fair enough. All right. And I mean, type 2 penetration in the U.S., we got to about 25 -- for CGM, we got to about 25%. We kind of updated our numbers. Based on that comment, we got to about CGM, Q2, in type 2 insulin intensive about 25%, type 1, about, call it, 55%. Are we close, Steve? Or Jereme?

Yes, I was going to say, what we've said on the call is we saw T1 passing 40%. I don't think we had it above 50%, at least that's from our data. And we publicly said that T2 intensive penetration is in the teens. And so I think it might be a little higher than we are, but I don't think it's wildly enough.

Okay. And so what's next with the type 2 non-insulin intensive population? Obviously, that's where there's a lot of focus. We all saw the level 2 program. I guess, maybe what do you -- when do you expect United to roll level 2 out more broadly?

Yes. So that's a very broad question, right? So we've got multiple shots on goal for the non-insulin taking type 2 patient population. There's going to be a subset, some of the work we're doing, United is a perfect example, the work we're doing with Intermountain Healthcare, with Onduo, with WellDoc with others. Very much a kind of B2B type offering where we're empowering others really to incorporate sensors into these programmatic experiences where they're trying to take other data points and analyze it. So health, fitness tracking data, meal data, that kind of thing. And even in some of these scenarios, they're adding either virtual or actually human coaching. So that's kind of one avenue that we're pursuing is really trying to empower others with our sensor technology to deploy their programs. Level 2 is obviously one of the most exciting, right? Because not only is United offering it to hundreds of thousands of people, United has the balance sheet to really -- when they approach an employer, right now, they're only rolling it out to their fully insured programs. So they're actually taking the risk in the program now. I think we would expect, over time, when they generate the claims data, they know the risk that they're able to bear. You'd see it roll out to a larger, even the self-insured population over time. But United is in a unique position because they hold all the data, right? Not just our CGM data. They have an underlying data analytics engine that looks at multiple data points. They actually have live coaches, they have the Optum subsidiary that actually already do proactive outreach to their Medicare base. So they're uniquely positioned, plus they have the claims data, right? So when it comes down to the economic analysis, they're better suited than anybody to understand the cost savings that we can generate when you use CGM in the non-intensive type 2 population. Intermountain is similarly situated though even you saw the data that we released earlier this year. Small end, right, it was 100 patients. But the cost savings that they extrapolated for patients using CGM was dramatic, tracking $5,000 a year per member per year is huge cost savings, and that was just with the addition of CGM in the Intermountain study. There was no coaching. There was no other intervention. It was literally just half the population got fingersticks, half are on CGM for 6 months, and the cost savings were dramatic. So that's kind of the one big avenue we're pursuing on the kind of the population health side and type 2 is really through a B2B channel. But I think you'll also see us continue to push. And we're exploring what it's going to look like as this thing really broadens. And as it becomes abundantly clear, the benefit that we provide non-insulin-using type 2s with our product, you're going to see some B2C consumer efforts come to life over the next 12 to 18, 24 months, where we're really going to target the consumer directly. And I can't give you too much in the way of details there. But I think you're going to see us take multiple shots on goal there because we recognize that this is a huge opportunity. People often ask, what is the use case going to be? Meaning, is it going to be real time, all the time? Intermittent? I think there's not going to be a single answer there. Some of our market research suggests that patients actually really want to wear it in real-time all the time. In fact, the commentary we're seeing back from this patient population is very similar to what we saw back in the early days with type 1s where they feel naked without it. They become reliant on the technology, even though they're not taking insulin, it's something that just helps them on their day-to-day management of their diabetes. And when you take the product away from them, they really do feel like they're missing something. So this is not going to be this year, kind of huge revenue contribution. But over the next, I would tell you, 3 to 5 years, the non-intensive type 2 businesses is a huge, huge opportunity for us.

Yes. And Larry, just -- you asked the question, how does UNH, what about Level 2? And how do we get a little more insight as to how that rolls out? One of the great things, I think Steve commented on it. This is what they've done is they've put it on their own book of business. And so this is fully insured UNH patients. This is -- they are basically saying, we believe in this so much so that we will take all the risk. I think over time, what you'll find is those self-insured companies that are using the UNH formula, it's about getting them to an ultimately opt-in and using the data to ultimately adopt them. We believe UNH is doing this kind of pilot rollout to prove you can reduce costs. And if I was them, I would use that to then attract a bunch of people to my formulary, my programs, and I would use it as a differentiator.

That's helpful. So the B2C for type 2 non-insulin intensive was new to me. I don't know if it's new in general. But I did have questions e-mail to me on DTC. But I guess is that -- you said 12 to 18 months B2C, it sounds like sometimes -- yes, consumer outreach, obviously, is that new? Is that something that you started...

I would say it's evolving. And really, again, it comes back to reimbursement, right? In the absence of broader reimbursement for that patient population, it's a little bit tougher. Because you'd be asking people to come out-of-pocket for the product. And that's -- particularly for someone wanting to wear it full time, year round, that could be a pretty expensive proposition. But we're looking at a number of options in terms of different configurations of the product, different price points for the product. So again, I can't give you specifics on this call. But yes, I would tell you, it is new in the sense it's evolving. We've always been kind of studying it. I think I'm not optimistic that within the next 12 months, you're going to see the major payers adopt kind of fee-for-service type coverage for non-insulin-using type 2, meaning similar coverage that we have for type 1s where you just get to go to the drugstore and pick it up. But I do think you're going to see some coverage evolve, whether it's through these programs or otherwise.

Yes. And I think, Larry, just so you know, we're already doing B2C outside the U.S. through our e-commerce platform in both Canada and the U.K. We're rolling those out more. And so we're taking some of the learnings in and outside the U.S. trying to figure out how to apply it in other markets. And it really depends, to Steve's point, on the reimbursement landscape and how that ultimately works. But we have experience in it now, and we're going to continue to look for ways to expand it.

I've got a lot of questions e-mailed to me on this. But Steve, I think you kind of hit the nail on the head on type 2 non-insulin intensive and B2C. You typically would wouldn't want to go to a consumer until there's some kind of coverage. So I guess my question is, does that mean that 12 to 18-month comment you made that you expect more coverage? Or you know how you're going to be very targeted and you're going to in your outreach, targeting patients that might be willing to pay out of pocket? I'm trying to understand what the implications are on the payers' side?

That's great insight. It could have been both, right? Because we know there is a subset of patients who can afford it and who may be in at a stage in their type 2 diabetes who are actually willing to afford and potentially a lower cost differentiated product from our type 1 product, certainly from our intensive product. But yes, we do believe there is a subset of that patient population who would be willing to come out-of-pocket as well. So you're right, we're not going to wait until we have broad-based fee-for-service type coverage for the non-intensive type 2 to go to the consumer. That's fair.

Okay. And so a couple of questions e-mailed to me here. Sorry for looking over my shoulder. On DTC, in general, the DTC -- any update on the DTC advertising campaign and expected return on investment from this?

Yes. So we are moving into DTC. For folks that are watching TV, you might see a lot more of our ads. I certainly do. I work here so I pay more attention to it than most. But we are investing in DTC. We believe the return on investment is certainly higher than any sort of hurdle rate you're ever going to ask for us. And so we'll keep those closer to the best in terms of what that return is. But as an investor, you want us investing in this. In terms of how that ultimately plays out and why it works, Steve mentioned it, as you get to consumers beyond your heavy type 1 users and your endocrinologist visitors, a lot of folks that are in the PCP space, whether it's actually the physician, the clinician and/or the individual that happen to be watching TV. They just aren't as aware of it. And so that's where we -- it's about reeking awareness beyond those that are in the know in the space of diabetes. And we found that through those investments, whether they're in place to TV ads and whether they're through other forms of medium. We found that the funnel for opportunities does go up into direct correlation. And so we're going to continue to make those investments. We've started on how those ultimately play out through the conversion cycle, we haven't given guidance on that. It does take a little bit of time for initiation to customer to ultimately pulling on therapy. But we do know it's a good investment, and it's something we'll continue to do so as long as we continue to see the feedback that we've seen early on.