Insulet Corporation (PODD) Earnings Call Transcript & Summary

My name is Jeff Johnson. I'm the senior medical technology analyst at Baird, and our next presentation this afternoon is from Insulet Corporation, a leader in the $3 billion global insulin pump market. With us today from Insulet, we're happy to have Chief Executive Officer, Jim Hollingshead; Chief Financial Officer, Wayde McMillan; and Vice President of Investor Relations, Deb Gordon. Jim, I'm going to turn it over to you if you have a few minutes of any prepared remarks you'd like to make, and then we'll move right into Q&A.

Yes, great. Thanks, Jeff, and hello, everybody. Good afternoon. Thanks for being with us today. We're already on the clock here so we're at 27 minutes and counting, we can see up here, so I'll be brief. Insulet was founded over 20 years ago with a really clear mission, which is to simplify and improve the lives for people with diabetes. And diabetes, as I think most of you know or all of know is a growing global epidemic, so it's a market with massive unmet medical need. And with regard to how medical technology is used in the treatment of diabetes, very, very underpenetrated. Our core product offering is based on the Omnipod product platform, which is a tubeless wearable disposable highly discrete insulin pump. And we've been competing very successfully with a product called Omnipod DASH globally, and in the U.S., we have the patient preferred form factor. We're really excited here today because we're just a couple of weeks, a few weeks into our full market release of our latest product platform on the Omnipod platform, the latest product offering, the Omnipod 5, which brings all of the advantages of our Omnipod platform with patient preferred form factor, channel access through the pharmacy, great economics for the payer and for the patient but also now introduces, for us, the first time in automated insulin delivery algorithm, which has fantastic clinical benefits in terms of improving time and range, managing highs and lows for the patient and does that all with that seamless form factor that's fantastic. We're early days, the launch is going really well. I'm sure Jeff will have a lot of questions about it. We're really excited to be here for you today.

Yes. Thanks, Jim. That's a great overview. And we'll start with Omnipod 5. I guess, no surprise that we'll start there, but I'm not going to let you give me the easy answer just like, oh, it's great demand and strong demand in that. What's been maybe the 1 or 2 surprises so far with the introduction of Omnipod 5 over the last few months?

I do think -- well, demand, I'm going to answer that anyway, right? Demand has been very high for us. And I think that we've been a bit surprised because we're at a high case with all of the patients that are coming on to Omnipod 5. So our target market is patients who use multiple daily injections, because what we're trying to do with the Omnipod platform and have all along is we've tried to remove the burden of caring for your diabetes, living with diabetes by doing finger sticks and needles. And so that's still our target market. The demand for Omnipod 5 with those patients is very high. We're also seeing very high demand from our own loyalist patients who are on Omnipod DASH. And so conversions from DASH are high. And we created that demand ourselves because for the last 2 quarters, as we've been in limited market release, we've been saying to people, the fastest way to get on Omnipod 5 when it comes to full market release is to be on DASH because it's easy to convert. So that demand is high. And we're also seeing a lot of demand coming from patients who are already using a competitor's AID-tubed pump. So our demand case is high. If there's a problem that creates and we've been talking about it, I borrowed this phrase from Wayde, it's a champagne problem, we're not providing the level of service that we want. So demand is so high that we're having trouble answering the phones in a timely fashion. And so -- and we're aware of that. We're not happy with it. Any of you looking at it will probably notice, you might see that on social media. We're investing now in the capacity. We already have planned to add capacity to our customer care and product support. We're adding that now. We're investing even faster to accelerate that. And that's maybe the one, if you're fishing for something negative, that might be the one negative. But it's a good problem to have because demand is so consistently high for us that we're having to invest to provide the service that our patients want.

Yes. Well, that makes a ton of sense. So Wayde, maybe we'll circle back and talk about the SG&A implications of that in a few minutes. But the source of excitement there, you guys are in a unique position with the recurring revenue model stream that you can do more DTC advertising than a lot of your capital-based competitors, which seems to draw in patients, I know. But the dichotomy between then where is that excitement is that the doctors who know that this product is out there now and they're driving that excitement? Is that the patients who might be seeing some of the DTCs, the caregivers, the parents who have young kids on Omnipod? Just what's the source of that excitement? Is it truly both patients and docs?

We think it's across the board. We see it across the board. And so Omnipod 5 is one of the most anticipated product launches in our space certainly for a long time. And so there has been pent-up demand and people waiting for it. And that's true of parents of kids with diabetes. It's true for patients who are treating their diabetes with Omnipod DASH or maybe with a tube pump, and it's true of physicians. There are a number of physicians that have been waiting for this launch. And so the excitement is, I think, quite high across the board.

Yes, interesting. And when you talk about some of the competitive converts, I think historically, you guys have been 80-20, 80% MDI, 20% competitive converts on that. Would that percentage shift at all, do you think? Are there so many MDIs coming in that even if you take some more competitive converts, the 80-20 split kind of holds?

We've seen that ratio to be pretty robust. As we said at the end of Q2, Wayde, you might want to comment on that.

Yes, sure. Happy to, Jim. As Jim said, at the end of Q2 when we're still in limited market release, we set record new customer starts for the quarter. Omnipod 5, even though in limited market release, was 25% of our new customer starts. So we know Omnipod 5 helped us achieve that record new customer start level. And as Jim just said, our metric held so 80% from MDI, 20% from competitive pumps. So what that means is we took the most competitive pump share ever in Q2 because the metric held and it was a record new customer start for us. So as Jim said earlier, we're seeing it come from competitive switches from MDI, consistently 80% from MDI and obviously, from our own Omnipod products. And so for us, we're waiting to see, Jeff, actually. It's going to be really interesting to watch that metric over time. What we know is we've competed really well on form factor, as Jim said in the upfront. And we've grown our business very well over the last 5 years while the 2 pumps have had AID. Now that we bring AID and have leveled the playing field, it's going to be really interesting how many of those people chose to go on the competitive pumps because we didn't have AID and now they have the option to switch over. And so it's going to be a real race actually for MDI users, and we're curious to see how that metric shakes out.

Do you have the visibility in the pharmacy channel to know where your patients are coming from and be able to track that close enough?

Yes, great question. We do. And so when our customers come on to Omnipod 5, they go through a registration process. And most of the time, almost all the time, they tell us information about where they're coming from and what type of diabetes they have. And so we've got great data insights with Omnipod 5.

Yes, okay. Interesting. All right. Moving on, just kind of thinking about the O5 algorithm. Obviously, there's been discussion out there. It does take a couple of pod changes for that to settle in, things like that. Are physicians starting to understand that, obviously, comfortable with that? Any kind of extra handholding you have to do with patients or physicians to get them through those first couple of pod changes?

Yes, that's a great question, Jeff. Let me talk just for a second about how the Omnipod 5 algorithm works. So what the Omnipod 5 algorithm is designed to do is to track your blood glucose and your trend. And so you put the Omnipod 5 on, it works out of the box and it delivers as designed and as approved, really substantially improved time and range and managing highs and lows really effectively. But the other thing it does is it learns and so there's a period where it's personalizing care for each individual patient. And that algorithm is watching how -- for a patient that puts that device on, the algorithm is watching how they respond to micro-dosing every 5 minutes. So it's looking at a trend, it's predicting where you will be, where your blood glucose will be in 60 minutes. And then it's -- if you need insulin, it's providing a small dose of insulin to nudge you down in trend. If you don't need insulin, it doesn't give you anything. And that -- over time, as that algorithm learns how your body responds to insulin, it tightens that up. And that takes about a -- maybe the second pod, maybe the third pod, so that would be day 4 or day 7 that the algorithm has really tightened you up in range. And for -- if you then take that to patients, for a patient coming off of MDI, that's a very seamless experience because a patient on MDI might be doing finger sticks for BG and is doing injections. And so now what you've done is taken all of that burden of those multiple moves off of that patient and the Omnipod 5 is managing them. And mostly, they don't think about the fact that you might -- they might start a little bit high and then tighten up because the algorithm is designed to be a little bit conservative at the start. You don't want to be aggressive with new insulin therapy. So it's designed to be a little bit conservative and then learn over the course of 2 pods or so and tighten up the algorithm. For a patient that's coming off of an AID system, they will first have probably forgotten the experience that they've had in getting on that AID system. But they've had to go through many hours of training and then they've worked with their physician or with their diabetes educator to titrate and do a lot of stuff to get there. When they go on Omnipod 5, that first pod will probably look to them like they're running a little bit high because they are probably running just a little bit high. And their expectation is they have to do something. They don't. And so that's been something that we've learned at the end of LMR and into full market release here that what we need to do is manage the expectations of those patients. That's not our target market. Our target market is MDI. And so now what we're doing is we're adding a lot of things to our marketing, a lot of things. We're doing webinars to help people understand. We're also providing tips and tricks because you can actually tweak the Omnipod 5 if you take it out of closed loop and adjust it if you want to. But the message is you don't need to. The Omnipod 5 is going to learn you and is going to personalize to you. And then if you take that to the physician set, it's the same sort of thing. Physicians -- not all physician practices but a number of physician practices who've been writing prescriptions for AID systems have an expectation for the first several days. They'll have to be hands on with the patient. And to your question, we are already seeing now, as physician practices who have that expectation, once they put 10 or 12 patients on an Omnipod 5, we're getting feedback from them saying, oh, okay. Now I see. I actually just have to convey to the patient. Listen, give it a pod or 2. You're going to tighten up. And that's the whole point of how we designed it, right? So it's a seamless experience for our target market, which is MDI. And it's really just a matter of helping the market, both patients who might be coming from an AID pump or physicians who might be used to doing all that upfront work with an AID tube pump, get them to understand that they don't need to do that work, and we're seeing that learning curve happen already.

All right. That's helpful. Wayde, I want to go back to one more number question on O5 and then we'll move on to a couple of other things. But on Omnipod 5, I think -- maybe it's not even Omnipod 5 plus DASH. 65% of your installed base currently is in the pharmacy channel in the U.S. When you talk to Dexcom, they talk about they think they're going to settle out somewhere around 75% pharmacy, 25% DME. There's always some payers, Kaiser and some others, who are never going to move CGM away from DME and what have you. But I think your new patient starts have been much higher than that in pharmacy. Is that right? I mean, where could that 65% number eventually settle out for the entire installed base? Can it get above Dexcom's kind of 75%? Can it go 80%, 90%?

Yes, I think you're right, Jeff. We're curious to see where it shakes out ourself. Probably the most direct proxy for it is where you get access. So where we have access in the pharmacy on DASH, we got up to 90% coverage. So you would assume as a proxy that we'd be able to get 90% of our volume through the pharmacy if we have access there. That's going to take some time. You mentioned 65% of our volume through the pharmacy now, and that's really almost all DASH because we've been on the DASH product and just limited market release on Omnipod 5 through Q2. Having said that, we do think Omnipod 5 will be a bigger catalyst to pull more people into the pharmacy channel, probably faster than we did with Omnipod DASH. So I think you're right, we're going to see levels that settle out probably around where access settles out. And it's our expectation that many people want to take advantage of the pharmacy channel. As you know, we've got a low co-pay. Almost all of our customers pay less than $50. The average co-pay is less than $50. And so that's a really low option for payment for people. In fact, it's about the average cost for people who are on multiple daily injections. And so we think pricing Omnipod 5 at parity with DASH and at about the same price they would pay for multiple daily injections is just a massive amount of additional value for them. And so we -- that was one of the hurdles that we knew we had to get over. We had to improve the cost position by both removing that upfront high payment that our competitive pumps charge in the upfront and we removed that. And so we're a pay-as-you-pod model in the pharmacy channel. We've had great success partnering with payers on it because payers now, they don't carry the risk of spending thousands upfront for a tubed pump. They just pay for pods as customers are using them. And so we've had great success ramping up access for DASH. We assume we'll have the same success for Omnipod 5.

Yes. And just to be clear, I mean, unlike Dexcom and again, fantastic company that had fantastic success over the years on the CGM side. But when they move a patient from DME to pharmacy, it's about a 50%, 60% price haircut for them. For you guys, you still get that price premium in the pharmacy channel over DME.

Yes, that's correct. But one thing to keep in mind is in the pharmacy channel, we give the PDM for free. So the value proposition for payers is strong because we don't charge that upfront fee, that upfront code for the PDM. We give the PDM at no charge. And so we actually carry more risk in the beginning because we've now trained the patient. We've given a PDM for free and we're only getting paid pod for pod. And so as you mentioned, what we did then is raise the price on the pod slightly to compensate for that. So over time, then we make up for that no-charge PDM upfront. So, so far, it's been a great partnership with payers. We've had great luck ramping up DASH. Customers have great experience in the pharmacy channel. It's a lot easier process for them. Prescribers have a lot easier process in the pharmacy channel. Both for type 1s and type 2s, we're prescribing DASH. And now for type 1s with Omnipod 5. So the pharmacy channel has been really great for us with DASH, and we're really excited to see what we'll do with Omnipod 5 as well.

Yes. And those PDM comments are maybe a good segue into the next topic I wanted to cover, which was kind of the cloud-based efforts you guys are putting in. And I think one of the things I'm intrigued by and I think had underappreciated until last quarter's call is putting that SIM card in the PDM for your 6-year-old type 1 kid at school who doesn't have a cellphone. Now mom and dad can track and see those real up-to-date numbers, things like that. But what's the appeal been? Or what's the response been from patients who now have a device that has a SIM card in it? They can always be uploading to the cloud and share or any other kind of details they can get out of the cloud from that setup?

Yes. The cloud connectivity creates a lot of benefits for our patients and our stakeholders right now. And then it will create even more benefits over time. So we're really excited about that. So what it does right now, as you said, Jeff, there's a use case where the parent can be tracking the kid's therapy overnight. It also enables seamless virtual setup. And so the fact that a patient can open their box, open their pod, connect their G6 and connect their pod and do all of that through a virtualized setup online, that's enabled by the cloud. That's a huge benefit and has been working really well as we've been driving the launch. It's also a benefit to physician practices. And so right now, if you're -- typically, if you're on a tubed pump, you're going to take in your -- or if you're on a BGM or whatever, you're going to be going into your physician's practice and spending the first 15 or 20 minutes connecting your device or downloading your data into a practice management or a patient management piece of software. We have cloud-to-cloud connectivity with our partner, Glooko. And so that data is in the cloud all the time. So it streamlines workflow for physician practices pretty dramatically. And it also means that those patients can be looked at remotely by a physician practice or a diabetes educator or a PA. So that's already cloud benefits just in the ecosystem and how we're managing care. Over time, it will create some other benefits. So an easy one to talk about is a cloud-connected platform can do things like over-the-air upgrades. And so as we have new features or if we do things -- when we do things like CGM of choice, we'll be able to allow patients to choose their CGM. We have to obviously complete our integration with the Abbott products, but we'll be able to do CGM of choice. And then something we haven't actually talked about very much, but we have real-time data feed on all the patients on Omnipod 5. Now we protect that data very carefully. We're very focused on data privacy and security. But because we're going to have therapy usage data over time, that's an asset that we'll build. And we will continue to find really useful things to do with that data. And that would include things like constantly improving the patient experience on therapy, constantly helping physicians to streamline their workflows, and over time, tying that usage data to claims data and building out real-world evidence to drive adoption and to drive things like population health management for payers and systems and national payers. And those -- so there's sort of waves of benefits that we can build. It's completely unique as a set of assets. Nobody can do it. Even if you're taking -- if you're plugging in devices, the only data you'd be getting, a competitor would be getting is the patients that are highly engaged enough to share their data in the platform. Those are going to be a subset of patients. We can see all the patients on the Omnipod 5 platform. And to build out that capability for anybody to try to follow us, that's a huge moat for us, right? And so it's a capability. It's an asset that works right now for the patients and for providers, and it's something we're going to build on over time. And I think we had some experience -- I think many of you know, I came from ResMed. We had a lot of experience with this at ResMed. We found very useful things to do with the data. I joined the Board from ResMed with that experience in mind. But Insulet now, I think, especially with the virtualized care we can do on the front end, we're well ahead of what ResMed can do on that virtualized startup and care with patients. And we'll continue to build on that asset. I think it will be a really exciting and really differentiated part of our offerings.

Yes. One follow-up question on that. We're down to 8 minutes. I want to keep moving.

Sorry.

No, no, no. I just -- I think one thing that's going to be interesting is we see real-world data from Control-IQ, even 780G data over in Europe, real-world data and it looks good, right? I mean, mid-70% time in range, individual patients up in the 80%, 90% range. You guys are going to have such a diverse population because all your patients are going to be in that. Is there a chance we could see a real-world data update from you guys where your time in range is like 70% because you have some outliers there. And all of a sudden, we have to say, well, wait a minute, that's not quite as good as what we're seeing in Control-IQ, but that's because this is a more representative data set versus Control-IQ, which is those biased higher involved or higher -- the patients who are really taking more time to control their diabetes.

I think that's a really insightful question. And what I would say is, I do think there's a chance you would see that because it will be the whole -- as you're saying, Jeff, it will be the whole patient population, all the users, right? And so you would see -- you would expect to see a broader range. But the benefit of that is that we should be able to use that data to do things like risk stratification compares. And we should be able to understand which types of patients need more handholding. And so there's a huge potential benefit when you start to see a broader spread of response to therapies where you can focus efforts. If you back up for a second, chronic diseases are very labor intensive for providers and very expensive for health systems. And so if you can look at that patient population and say, these patients are automatically adherent, you don't have to hold their hands, they're doing great. This is the type of patient that needs your focus on care. What we learned in the sleep market and what I'm sure will be true in the diabetes market is that if you can focus your resources there, you'll coach the patients that need coaching. You won't spend your resources coaching to patients that don't need so much help and you lift the whole tide that way.

Okay, okay. And I've covered the diabetes space now, I don't know, 6, 7 years. One thing that still confuses the heck out of me. So you guys are going to get CE Mark, you think, for O5 pretty soon, sometime this year. Commercial launch somewhere in the middle of next year in select markets. Now is the timing from CE Mark to launch because you just have to ramp up manufacturing in that? Or are there other hurdles that you have to cross on a country-by-country basis to get the product out there?

The hurdle now -- so the way we were thinking about European launch is we were trying to time the build-out of our cloud capability in Europe with the CE Mark because we thought they would take about the same amount of time. The CE Mark turned out to move faster than we thought. So the long pole in the tent now is to build out the cloud capability. And cloud is intrinsic to the Omnipod 5 offer, especially for what I was just talking about, especially for the virtual setup on the front end and for cloud-to-cloud integration for patient software management in Europe. So we have to complete our cloud infrastructure in Europe. That is a pan-European build because you have to comport with things like GDPR, which is the European -- the pan-European set of privacy laws. But then there are specific things in certain countries you have to accomplish. And so for example, hosting patient data in France has a whole set of very specific French requirements. The hospital systems in the Nordics usually have very high hurdles on how we're going to do data privacy and security. So it tends to be customized by market. So what we'll do is we're working really hard right now to accelerate our build-out of the cloud. We expect the CE Mark very soon and then we'll focus on markets and we'll set ourselves up to win in appropriate markets as we sequence it.

Okay. That is helpful. All right, Wayde, we're going to go over to you on some gross margin stuff now. So I think your guidance for this year should be somewhere around 65% or just below that, but definitely a little bit below that over the back half of this year, I think, is how you've set it up. Is that...

Yes, that's correct. Guidance for this year, 64% to 65%, and that's a step down. We did touch 70% in Q1, which was a target for us for some time. And clearly, in this inflationary environment, we've got some challenges. We also have some micro challenges with our business as we ramp Omnipod 5 up, some headwinds there. So yes, so the second half will be a step down from what we saw in Q1 and it settles us out in that 64% to 65% for the year.

Yes. And should that then second half range, I think in my model, I'm at 63.5% or so, I forget exactly, is that a good stepping-off point for next year? I know you're not guiding to '23.

Yes, it's a really -- we get this question a lot because it's a really tough thing to estimate right now. So we're -- as you said, we're not guiding to gross margins for next year. I think it's fair to say, if you're in that range for the full year guidance or the second half, it's probably a fair place to be without us providing guidance at this point. And the things that we're monitoring before we set guidance is how impactful is the inflationary environment. Also how fast does Omnipod 5 ramp? One of the biggest tailwinds for gross margin improvement for us is scale. The faster we scale, the more volume we drive, the better our cost structure is and the more we absorb fixed overhead and improve our costs over time. Also, our operations team is doing a great job. These are -- on new products, you typically have more opportunity to drive cost improvements. And so we're at the front end of a lot of those programs. So a lot of tailwinds. Obviously, some headwinds to gross margin at this time. And so we're just going to continue to monitor it through the end of the year here, and then we'll set guidance for 2023.

Yes. And then on the PDM accrual cost, that came up on the warranty expense side last quarter. Have you been able to go back to the -- is it -- I don't even know who supplies your handpiece for the PDM. But have you been able to go back to them, have any discussions on replacements there or the ability to replace the battery, number one? And number two, now that O5, everybody is going to get a new PDM on O5, right? Is there a way to ensure that maybe you have a longer-lasting battery in there, so we're not 2 years into O5 and get coming up with the same issue?

Yes. So important issue for us. As we disclosed in our Q2 call, we increased our warranty reserve cost for faster PDM replacements related to the battery on DASH. So we're having more experience to where customers are wanting to replace their battery or need to replace their battery on DASH PDMs much higher than we had experienced before. And so the team is working on that, root causing, as you said, working with the manufacturer of the batteries, all through the whole value process to really understand it. This is something that really just started for us in the middle of Q2 and started to ramp up. And so teams are working on it. We don't have an update at this point regarding that, but obviously important to us because we want to service our customers the best we can. And one of the reasons we increased the warranty reserve is whenever customers are having an issue, then we're replacing the PDMs at this point. So no other update other than that at this point. But having said that, Omnipod 5 is a different PDM and a different battery.

Okay, all right. So I'm going to skip 1 or 2 other margin and cash flow questions here just because, Jim, I want to get to this one before -- in the last 1.5 minutes. But AID as a standard of care in newly diagnosed T1 patients, I mean, it seems like we're making some progress towards moving in that direction. One, do you think that increasingly becomes the way these T1 kids are treated when they go in polydipsia, polyuria, polyphagia go into the office, it's a classic. They got T1, you don't even have to test them. Do they leave with an Omnipod 5? And talk about your competitive positioning there because of form factor because of the 2- to 6-year-old indication, all the benefits you have there?

Yes. Thanks, Jeff. So just to hammer on that point a bit, we did just receive our FDA clearance down to age 2, so we now cover all feeds down to age 2. Omnipod 5 is a fantastic form factor for kids because they can put it on, wear it, forget they have it on, sleep with it. They're not going to get tangled up in a tube. They don't have to do anything. And so we've historically, even with DASH before having AID, we've been very, very strong with parents who want to provide that kind of care for their kids. And now with AID, one of the most moving things for me with Omnipod 5 is we constantly get messages from parents saying things like, it's the first time I've been able to sleep through the night for 6 years because now I know my kid is not going to go low overnight. So it's a sweet spot market for us. In terms of adoption, to me, we're speculating now. But to me, it just makes sense that, that would become standard of care for a child who's diagnosed with Type 1 that they would just go home with an Omnipod 5 because the form factor is fantastic. The ease of use is great. The parents have reassurance that we're going to provide really great time and range and guard against lows, which is a terrific thing about our algorithm. And there's no barrier to it, right? I would go further and say one of the things that we talk about a lot is the adoption of pumps. Pumps have been around for 30 years. And the adoption of pumps have stalled out until Omnipod 5 at about 30% of type 1 patients. One of the reasons for that is that physicians are constantly making the judgment about whether or not a patient can handle the complexity of a tubed pump. And so often a patient isn't offered a pump because the physician is making a judgment about will they be able to do it? Will they stick with it? It's a huge capital expense. And I think that those barriers are coming down, and I think Omnipod 5 is bringing them down. And so we had an ADA this year. We had one of our KOLs make the strong pitch that there is now no reason to not offer every patient an AID pump because Omnipod 5 is so easy to use. So I think it will become increasingly standard of care. And I think with kids being diagnosed, it should be standard of care.

Understood. All right. Well, our time is up, so on that, we will end. Please thank me -- or don't thank me. Please join me in thanking Jim and Wayde for a wonderful overview here of Insulet. And as a reminder, the next presentation is set to begin at 12:50 p.m., Include Bio-Techne in this room, Bridge Biopharma in, I don't know, some A room of some sort, Frequency Therapeutics in...