Oscar Health, Inc.
Q2 2021 Earnings Call Transcript

Published:

  • Operator:
    Good afternoon. My name is Joanna, and I will be your conference operator today. At this time, I would like to welcome everyone to Oscar Health's Second Quarter 2021 Earnings Call. All participants are in a listen-only mode. After the speaker’s presentation, there will be a question-and-answer session. Thank you. I would now like to turn the call over to Cornelia Miller, Vice President of Corporate Development and Investor Relations to begin the conference.
  • Cornelia Miller:
    Thank you, Joanna, and good afternoon, everyone. Thank you for joining us for our second quarter earnings call, where we'll discuss our financial results, the momentum in our business, and updated guidance for 2021. Mario Schlosser, Oscar's Co-Founder and Chief Executive Officer; and Scott Blackley, Oscar's Chief Financial Officer will host this afternoon's call, which can also be accessed through our Investor Relations website at ir.hioscar.com. Full details of our results and additional management commentary are available in our earnings release, which can be found on our Investor Relations website at ir.hioscar.com. Any remarks that Oscar makes about the future constitute forward-looking statements within the meaning of safe harbor provisions under the Private Securities Litigation Reform Act of 1995. Actual results may differ materially from those indicated by those forward-looking statements as a result of various important factors, including those discussed in our quarterly report on Form 10-Q for the quarterly period ended June 30, 2021 filed with the Securities and Exchange Commission and our other filings with the SEC. Such forward-looking statements are based on current expectations as of today. Oscar anticipates that subsequent events and developments may cause estimates to change. While the company may elect to update these forward-looking statements at some point in the future, we specifically disclaim any obligation to do so. The call will also refer to certain non-GAAP measures. A reconciliation of these measures to the most directly comparable GAAP measures can be found in our second quarter 2021 press release, which is available on the company's Investor Relations site at ir.hioscar.com. With that, I would like to turn the call over to our CEO, Mario Schlosser.
  • Mario Schlosser:
    Good afternoon, everybody, and thank you for joining us wherever you may be listening in from. We are excited to have you on. Our results this quarter show that our member-centric tech-first approach to health care continue to generate value for our members and clients. Today, we reported that direct and assumed policy premiums increased 45% year-over-year to $841 million for the second quarter of 2021 ahead of our expectations. We have positive momentum in the business, and we are executing on our plan. Before I jump into the details, I would like to start by reminding our core thesis, how we benefit from being a health insurance business and a tech platform business. We believe that the tech and services we build for our insurance business give us battle-tested assets and insights that we can bring to the broader health care markets, at a time when those markets are undoubtedly in a period of transformation. A profitable insurance company would provide support to continue to expand our tech offerings and earnings, and the synergies of insurance and tech company allow us to access a growing portion of the $4 trillion health care marketplace and deliver an increasing profitable business over time. Now, let me expand on our approach. We combine a human touch with technology and data to make complex decisions simple and intuitive for our members and for our providers. We are the only ones in health care that run this on a full stack platform that we control and build end-to-end. And we have purposely built this since day one, always with the ambition of powering as much of the health care ecosystem as possible. Now, our core systems impact is compounding. The systems that has helped members find the right doctor, access virtual care, and power care for use in the system in navigating the health care system. We can constantly test retest and optimize what we do. And all this makes us one, a better managed care company; and two, a better technology partner to our +Oscar clients. Our strategy continues to be focused on reaching profitability as our insurance and +Oscar businesses scale. We're driving towards this by
  • Scott Blackley:
    Thank you, Mario and hello, everyone. Today I'm going to walk through how the momentum we see in our businesses is showing up in our results, how we are thinking about our MLR trends and then close with updates on guidance. I'll start with the discussion on our membership. We ended the second quarter with approximately 563,000 members, an increase of 35% year-over-year, driven by growth in our individual Medicare Advantage and Cigna + Oscar books of business. Membership growth exceeded our expectations as consumers continue to select Oscar's innovative plans during the ACA special enrollment period. As Mario mentioned, from the start of SEP in January through June 30, we've enrolled around 120,000 new members. Second quarter direct and assumed policy premiums increased 45% year-over-year to $841 million, driven by higher membership as well as business mix shifts towards higher premium silver plans and modest rate increases. Premiums before ceded reinsurance were $724 million in the quarter, up 84% year-over-year, driven by both higher premiums and lower risk adjustments year-over-year both of which exceeded our expectations. We recognized $34 million of favorable risk adjustments relative to our expectations for the 2020 plan year, driven by outperformance in our value capture activities. Based on a favorable report from lately, we also recognized an incremental $34 million of risk adjustment benefits related to 2021, as we have been increasing our mix of silver plans and we are seeing that result in a lower transfer estimate for members with higher acuity. We recognized a comparable offset in medical claims and IBNR adjustments. So net-net that's had an immaterial impact on the quarter. Premiums earned at $528 million increased 364% year-over-year in the second quarter, as we further reduced our use of quota share reinsurance from 76% in the second quarter of 2020 to 33% in Q2 2021. As a reminder, given our recent IPO and the momentum in our businesses, we chose to decrease our utilization of quota share reinsurance this year. We would expect quota share to stay at approximately the Q2 levels over the balance of the year. Our medical loss ratio was 82.4% in the quarter. In terms of drivers, COVID-related medical costs declined slower than we expected and were roughly $35 million in the quarter, contributing 500 basis points to the MLR. We had approximately $34 million of favorable prior period development related to the prior year, a benefit of 400 basis points in the quarter. The prior year period development impacted our risk adjustment, which is included in the denominator of the MLR calculation. So while the dollars of the COVID-related medical costs, MPPD item basically offset, the COVID costs had a larger impact to MLR. In the comparable period last year, we experienced a very significant decline in medical costs due to a COVID-related drop in utilization. So the year-over-year trends are understandably unfavorable. Let me offer more color on COVID and the overall utilization environment. COVID costs declined from Q1 2021 levels, but not as quickly as anticipated. Direct COVID testing and treatment costs were higher than expected, partially offset by lower vaccine administration costs. That said, cost peaked in April and declined through June. As Mario mentioned, in our current data, we are seeing an increase in our members, who are currently receiving care for COVID. And at the same time in those geographies, we are seeing a decrease in non-COVID utilization. In terms of non-COVID utilization, during the quarter, we saw a resumption of routine care. A portion of which we believe is a pull forward of demand from the second half of 2021, as opposed to a catch-up from deferred care in 2020. Specifically, we saw increases in professional visits, largely in the routine and preventative visit categories. We're managing this increase through our strong utilization management capabilities and intervening based on real-time data, through our concierge teams. Our second quarter insurance company administrative expense ratio of 19.8%, improved 330 basis points year-over-year. The meaningful year-over-year improvement in the insurance company administrative expense ratio was driven primarily by increased operating leverage from our significant net premium growth of 85% year-over-year and tech efficiencies, as well as the removal of the health insurance fee. Our tech-enabled model has helped us scale and grow premiums at a faster rate than administrative costs. Our overall combined ratio, the sum of our medical loss ratio and insurance company administrative expense ratio was 102.2% in the quarter. On a year-to-date basis, this metric was 98.5%, continuing to reflect the consolidated profit across our insurance companies. Our adjusted EBITDA loss of $50 million, increased by $22 million year-over-year, on a year-to-date basis, the loss was $77 million, a $38 million improvement year-over-year. COVID-related care is a headwind to our year-over-year results. Even with these headwinds, we delivered year-to-date improvement in adjusted EBITDA year-over-year, demonstrating that we are leveraging our tech-enabled model to fuel strong premium growth and capture administrative efficiency. Turning to the balance sheet. We ended the quarter with over $3 billion in total company cash, including $1.1 billion of cash and investments at the parent and another $2 billion of cash and investments at our insurance subsidiaries. Now, let me turn to our updated 2021 guidance. We now expect direct and assumed policy premiums for the full year 2021 will be approximately $3.2 billion to $3.3 billion. Compared to our prior guidance, that is an increase of $125 million at the midpoint largely driven by membership increases from the special enrollment period. We now expect our MLR will be in the range of 85% to 87% for the full year, an increase of 100 basis points at the midpoint from guidance. For the first six months, our MLR was 78.7% and our guidance continues to assume a seasonal ramp in the second half of 2021. We previously expected around 3.5 points of direct COVID costs for the full year and our updated guidance now assumes roughly twice that amount. We are also assuming that non-COVID utilization will be around baseline in the second half of the year. We project our insurance company administrative expense ratio will be between 21% and 22%, an improvement of 150 basis points from prior guidance at the midpoint, as increased revenue and scale are driving higher leverage in this metric. Based on the above changes, we are maintaining our insurance company combined ratio guidance of between 107% and 109%. Finally, we are maintaining our 2021 full year adjusted EBITDA loss range of $380 million to $350 million, which is a meaningful improvement from 2020. And with that, let me turn the call back over to Mario.
  • Mario Schlosser:
    Thank you, Scott. I want to thank everybody for joining our call. And I just want to share two points before we close. First, I want to thank the incredible Oscar team. We talked a lot about our differentiation and via technology services parts, we are truly powered by people. Without this team, we wouldn't be able to keep raising the bar for our members and for the industry. And second, I want to reinforce our priorities and the path forward. One, we remain dedicated to growing our insurance business, while at the same time, re-costs . Two we are driving forward the expansion of the +Oscar platform with active partner conversations, while simultaneously demonstrating real ROI for our current partners. Three, we are fully committed to becoming profitable as our businesses are reaching scale. Now with that, we'll turn it over to the operator for the Q&A portion of the call.
  • Operator:
    Thank you so much. Your first question is from the line of Ricky Goldwasser from Morgan Stanley. Your line is open.
  • Ricky Goldwasser:
    Yes. Hi. Good afternoon and thank you for all the details. So a few questions here. First just wanted to get a better sense on the MLR and sort of the underlying assumptions, as you think about second half of the year. I mean, clearly, you gave us a lot of data on what impacted MLR in 2Q, but it sounded like you're seeing towards the end of the quarter, you saw a decrease in non-COVID utilization, but you are raising MLR and you're kind of like modeling higher MLR for the second half of the year. So just sort of trying to understand what's embedded in that new guidance around just core utilization versus baseline underlying COVID assumptions. And also, what type of security are you seeing among the new members, essentially the one that's kind of like onboarded during the extended open enrollment season?
  • Scott Blackley:
    Okay, Ricky. So I'm going to try to tick through those things. I'm not going to skip through. So if I missed any of your questions, please go ahead and come back to me on any of those. So let's start with overall what we're seeing in utilization. So I mentioned in my talking points that, we saw through the quarter, we saw COVID utilization declining. On the other hand, for non-COVID utilization, we did see that above baseline and accelerating. Big picture, I would say that, non-COVID utilization again it was slightly above the baseline, which looks like to us that it's been by pent-up demand. We saw specifically that there was elevated professional utilization. PCP visits were up preventative care and diagnostics. The nature of those gives us increased confidence that that's really pent-up demand and potentially pull forward demand. And as another data point we saw that inpatient costs were also elevated in the quarter and that was driven more by increased medical admissions versus surgical admissions that were roughly at baseline. So overall, it feels like in the second quarter we were seeing a bit of pull-forward demand on non-COVID utilization. Now with respect to guidance, I just want to go back and make a couple of points. The first is that in our full year guidance, we are assuming that the cost that we will incur for COVID are basically the same as what we experienced in the first half of 2021. With respect to non-COVID utilization, we're assuming that non-COVID utilization is basically at baseline in the second half of 2021. So I would just comment that with respect to that guidance, we're not assuming that the increase in COVID care that we've modeled is going to result in a decrease in non-COVID utilization. As Mario and I both talked about we have seen evidence that that's occurring. We're in markets where there are spikes. That gives us confidence that the guidance that we're putting out there is balanced.
  • Mario Schlosser:
    Yes, Ricky maybe I'll add one more thing on the question of what the nature of the demand is. One question that I think a lot of people have had in their minds is there's, sort of, a not just pent-up demand on things like preventive and so on which we are seeing. But is there also a higher acuity from deferred care from last year kind of flowing into this year. For example, are they more cancer diagnosis or are they have higher severity and things like that. Our members who lastly perhaps were less well-managed as a result of leaving the house now coming to high acuity levels. So that we're not seeing. Examples of that would be that the new cancer diagnoses have about the same likelihood to be metastatic as in 2019-2020. So that would be inconsistent with the numbers misdiagnosis last year and are showed up more over this year. If we look at numbers with certain chronic conditions; diabetes, autoimmune disorders their rates of emergency room and inpatient usage are about the same as 2019-2020 and now also 2021. And so that seems also indicate of the fact that -- utilization isn't driven by that. The utilization in the second quarter wasn't driven by that. Yes. And the other question I think was to SEP versus OE right? Scott, do you want to take that?
  • Scott Blackley:
    Yes. Yes. With respect to OE versus SEP, basically what we're seeing is that the SEP population this year has about the same co-morbidity as the OE population. We've seen generally an increase in morbidity in our total book and that is being driven by an increase in our membership mix. We've been working to drive towards a more silver heavy population. And what we're seeing now is that we've got a slightly older population, we've got a higher mix of silver plans and that in turn is causing a bit of a higher overall morbidity. The consequence of that is kind of twofold. One, we're seeing a stronger risk adjustment so there's lower transfer for us. It also gives us the opportunity to use our customer engagement engine to really help manage those more acute members. And we think over time that delivers a higher amount of profit to the company.
  • Ricky Goldwasser:
    Okay. And then one question on +Oscar. Mario it sounds like you're having conversations with providers and payers. And it seems that now also sort of interaction and member satisfaction is becoming increasingly more important for star ratings. So, maybe can you share with us sort of the profile of payers that are engaging with you on +Oscar?
  • Mario Schlosser:
    So, let me point you to mostly our history there, it's pretty indicative of what we're doing at the moment. We -- Health First is a good example for a regional payer owned in that case actually by a provider where we're taking 37,000 MA members, 20,000 IP members, bringing them on to our platform. And so that is sort of like one archetype of clients we're talking to. There is another archetype which is providers who are either already taking risk essentially perform want to go deeper into risk, want to potentially even at their own insurance company. And then as the other essentially push we have been making both inbound and outbound. So that -- the example there obviously is ACH in South Florida, also Montefiore in the Bronx. And that can actually go all the way towards as I mentioned before us slicing into our stack and taking just a piece of it. And for example helping a physician group is already taking risk to also start paying claims. And that's obviously part of our core offering on the admin side. And we think we do this very efficiently 95% utilization rate and so on. And so we can sort of like slice that into somebody's only doing the activity if it wants to do it better. So, those are the archetypes and that's being done on upward conversations.
  • Ricky Goldwasser:
    Thank you.
  • Operator:
    Your next question is coming from the line of Stephen Baxter from Wells Fargo. Your line is open.
  • Stephen Baxter:
    Hey, good afternoon. Thanks for the question. So, I wanted to come back and ask about the tech stack and your claims visibility. I was hoping you could talk a little bit about your level of visibility into Q2 cost as you book the quarter how many months of good data do you feel like you have on the acute side also on the non-acute side? Just it would be great to hear a little bit more about your process here in a dynamic environment and how your technology is helping here. Thanks.
  • Mario Schlosser:
    It's a great question Steve. So, lots to unpack there for sure, but let me give you a couple of examples. The benefit we have is that we've got one big data lake where everything flows together. And so we can very easily say what is happening with our COVID surges for example and the membership? What is happening with conversations in customer service? What is happening with utilization management volume really all the way to the present day? I mean all the stuff is really available to my fingertips if I care to look at it and obviously, I do quite a bit. And so for example let me pull this up here. If I look at let's say COVID conversations in our virtual primary and urgent care, those are up from the kind of like a trough or bottoming where they were in about May, but they're still only at about half of the peak we had in December, January. Our -- kind of, all mentions of COVID in all customer service channels whether it's chat with the concierge teams, conversations with the form of the concierge teams, things like that, are also certainly up over the past 1.5 months or so. But they are still 34% off from kind of the peak we had there also earlier in the year. And so these are all numbers that we can pull up even by the minute basically. And they generally track quite well within the eventual number of positive cases we see, which then eventually come to the claim system and so on. Even on the claims side, we generally think we've got good visibility there because it's obviously, also a single source of truth, which means the claim comes in and we can see it right away as opposed to having run through the system and taking a while to kind of get processed there. So that gives us a little good backup to these leading indicators to these coincident indicators that we can generate more quickly. So that's for example, how we then see the point I made earlier, that if we look at elective surgeries, we have a nice realtime indicator it's for us the UM decisions to position and decisions we make on those surgeries every day that those decisions are down by about 33% in July versus June and they're down a bit more actually the first where we're now here in the first 10, 11 days or so on in August. And so these things look like then we can see them ripple through eventually when the claims come in. Now one last thing, I'd tell you that I'm always excited about is that, we can tie these things together it was going to be 4.5 of the facts. When I look at our cost estimation in our where the considerations can go in and generate costs with prospectively for what a particular kind of pathway of care where cost is a member or members can go when you can do this we track quite actually there as well. When you get a cost estimate and then a couple of weeks later we -- or whenever you actually go and get the service, we put all that together automatically in the back end and we can then see how close we are generally within those -- within kind of like 10% or so. And that's also a really high number. That's kind of like about 50% 60% or so or actually even higher now. Recently of the estimates, we gave kind of way in advance of, how much it would actually later on cost you do this now for about 93% of all utilization that generally happens in the ecosystem. We can create these kind of prospective cost. And that goes right into a claims system as well and prices that we assign. So that's the kind of I think good stuff, we're doing there.
  • Stephen Baxter:
    Very interesting. Thank you for all the detail there. And just on philosophically as you guys are approaching the individual market in 2022, it would be great to get a little bit of insight into how you are thinking about, I guess you already would have priced that market. I guess any way you'd characterize, kind of what you're assuming around baseline levels? Like kind of extended COVID cost into 2022? Any thoughts around pent-up demand or potential for pent-up demand? Just be great to get a sense of how you guys thought about approaching the pricing for next year. Thanks
  • Scott Blackley:
    Yes. Thanks for the questions. And I'll just comment that if you were here with me, you would see that Mario literally pulled up the dashboards and was giving that answer using real-time data. It's really one of the most amazing parts of this company that I've seen, since I've joined. Coming to your 2022 question, I will just say this. So first of all, we did the majority of our filings as you mentioned in May and June. When we went about putting those filings in we really tried to be thoughtful about the environment. We balanced at that time, all of the information that we were seeing in our real-time data. We've always talked about that we try to balance both risk and our ability to grow. And we think that we achieved that in the pricing that we put in. So at this point even given what we've seen thus far with COVID and with utilization trends, we feel like we've put that into a good spot with pricing. We have had a chance to refile in a couple of locations where it was appropriate to do so and we think we're well positioned to move forward. Next question.
  • Operator:
    Thank you so much. Your next question is from the line of Kevin Fischbeck from Bank of America. Your line is open.
  • Kevin Fischbeck:
    All right. Great. One of your peers provided a pretty wide range for MLR in the back half of the year and what they attributed to was I guess less about utilization, but more about the uncertainty around the risk scoring for the back half of the year that in particular the SEP enrollment has fewer days to just get coded and get that risk score together. Are you at all worried, or is that a concern issue for you around the risk capture in the back half of the year?
  • Scott Blackley:
    Look I think -- the risk capture is always one of the factors that you have to deal with when you're in an SEP environment. I think we've got some experiences we talked about the numbers that we're seeing come in look a lot like what we saw as part of the OE process. So, we've tried to give reasonable accommodation for that in the MLR guidance that we've provided.
  • Mario Schlosser:
    And I mean even in prior years we obviously have a decent chunk of come from special roaming as well. And so we have experience with Oscar members when they come in. I point you there I think to the PPD, we had on risk adjustments for last year in the second quarter, but that came in better than expected. And I think as Scott said that that was due to our improvements in systems build-outs and doing that well. And so I think that the machinery will work just as well as here.
  • Kevin Fischbeck:
    Okay. And then obviously G&A coming in well. I guess oftentimes we see this though I guess as registered companies be MLR is a little bit high G&A is a little bit lower either because of lower bonus accruals or companies kind of act to take out costs when G&A is higher. It sounded to me like you that wasn't the driver. But I just want to make sure that -- like how are you characterizing the G&A lever? Is it really more about just revenue coming a little bit better and getting efficiencies faster, or is this kind of anything kind of onetime in here that we shouldn't think about you being on a different trajectory into next year we should be kind of back to our original 2022 assumptions on G&A?
  • Scott Blackley:
    No we are -- there's nothing unusual in the trend there. There's no reversal of bonus accruals or anything like that. What you're seeing is really the leverage that we're generating in our business. And we are basically seeing our expenses grow at the same pace on the variable expenses with revenues and then we're getting some leverage on fixed costs. So, we think that that demonstrates the traction in that business and one of the reasons why we feel optimistic about the trajectory of our ability to target getting that business to profitability in 2023 in the insurance company.
  • Kevin Fischbeck:
    Okay. Great. And then maybe the last question. I guess when we think about the MA business I guess it's now -- has been a lot of talk there about that. I mean how are you thinking about your cost trend there and how you thought about pricing for that next year?
  • Mario Schlosser:
    So I'd say, it goes to the same cautious pricing process and plan design process we have internally for the IFP business and for the small employer business, which I think we've now demonstrated that we can land that in the way that we need to. I would also say that we have not obviously -- people have not seen the bid publicly, so I don't want to talk too much about that. But again this year, we try to really be thoughtful in the same way and we're launching now these diabetic plans, for example, in IFP also about MA benefits and plan design. So proud of the work we've done there. On the -- for MA growth, our focus next year is on really two different things. We first of all think that there's a lot of runway in the existing footprint we have, right? These are for the most power plans, we built together with other health systems that are co-branded oftentimes and that we're excited about growing more. And then the other part is really taking over the big book from Health First and flipping that over for 1/1/2022, where they actually get the pricing in that case. But so, we're kind of same pricing process therefore similar assumptions. But obviously, sort of like tailored for the population we have there which is different higher vaccination than individual and so on than you've seen in individual markets. But that is all kind of incorporated there.
  • Kevin Fischbeck:
    Okay, great. Thanks.
  • Operator:
    Your next question is coming from the line of Jailendra Singh from Credit Suisse. Your line is open.
  • Jailendra Singh:
    Thank you. Thanks for taking questions. I was wondering if you could spend some time on the competitive landscape for the individual market next year. We have some large insurers trying to enter the new entering market in next year and some other insurers expanding. Just wondering how this impacts your thought process around growth and expansion opportunities. And related to that, I mean three states and 146 counties you're entering. How did you go about picking those markets? What are the criteria demographics or landscape? Just curious like any thought process there.
  • Mario Schlosser:
    Yes. Good question, Jailendra. So, I'd say for a number of years, we tell now that the individual market is competitive and that's -- in my view it shifted away a bit from just kind of premium level more towards can you save members money on total cost of care. And I think that shift has been generally a tailwind for us, because that's what we do in diabetic plan designs, care plan designs, we -- I always like to remind people of the fact that I think we were the first ones to put the bronze and also silver plans out there. And these kind of plan design moving the moments have really helped us continue to gain share with not always -- or not generally pricing the lowest, right? We -- our group premiums as compared to a year ago by 45%, even though we were only the lowest price than 10% or so of the markets we are in. And so it's that same approach, a very disciplined approach on pricing that we've historically done now. We balance growth and profitability, and we'll follow that again for next year. But obviously pricing isn't public anyway yet. And so I don't want to talk too much about where we did lots and so on. But that same playbook we may follow there. And from that point of view, I don't see too much of that difference in the markets right now. The pricing we've seen nationwide, I would say, has not surprised us in either direction. And I think the balance, we've been driving towards there, it's going to be the right one. And we picked these new counties also with an eye towards where do we have a right to win, where can we get the right provider partners, and we work with more closely. That could by the way eventually turn the +Oscar clients. That has happened to us a couple of times. So we had our footprint in IFP in a particular county and then we build out overall to -- up from relationship over time, in case in Orlando and in Broward County as well. And so it comes with the right provider partnership. We think counties where we think the offering we have will resonate for the population is such that we can really make sure we can do well at those two culturally competent care, for example, I think we brought some really good Spanish speaking concierge team solutions into Florida back in the day, so that it helped us grow there early on. And we pick counties where we think that the pricing makes sense and we can really again deliver a margin there over time. So that's how we picked it. And I think that's exactly what we see into next year as well. And our -- can't wait to get into open enrollment and start getting going there. You see a broker restarting in September actually at the end of -- beginning of September, end of August hopefully in person.
  • Jailendra Singh:
    Okay. That's helpful. And just my follow-up, going back to this comment around non-COVID utilization being pull-forward in Q2, how much of that was like self-imposed given you have this higher member engagement and care management model? And as people were getting vaccinated, I mean you are pushing for them to see the doctor and just if that was one of the driver, or was it even beyond just -- which might have come across from your higher care management model?
  • Mario Schlosser:
    Yeah, so that is really always and what we do in the outreach is always intended to get members to utilize the channels of care that are going to be better for them and more efficient overall, more effective overall. So yes we see higher drug utilization members who are attributed to our virtual primary care physicians. Adherence goes from 65% level or so to drugs, which is like standards in the U.S. healthcare system to 85% medication adherence when you are a member who's a attributed to an Oscar Virtual Primary Care physician, because for two reasons really, one is we can remind you we had summer engagement and there's nice automated follow-ups and all that kind of stuff. But the second reason is that we make those drugs free in our plan designs. So the Oscar prescribes a drug and it becomes free and that is self-priced medication as well. So that does mean that medication costs or medication utilization ends up being higher and that is an increase -- that ends up being increased utilization. But even with that, we see in those Virtual Primary Care designs that the benefit we're getting lower ER utilization, better specialist care routing and so on actually outweigh that. And so that's how we generally think about our campaigns. We don't always get that right, which means we just constantly turn through ideas there of how to tweak these campaigns on how to get them right, and so that's I think, how you can think about that. The above -- based on utilization there in the second quarter, therefore, I would not say it's driven by that. But as Scott said, it's more of that pent-up demand we saw right, it's consumer and preventative concentrating some of the things people do when they first get coverage. So see that as an SAP when they come in there's a little bit of increase there, shouldn’t be an OE. But as we said, we are just prescribing at the -- if you just look at the entire first half, the entire first half of the year is below baseline. And so what we prescribe -- we're prescribing forward into the second half of the year at baseline, which we feel like is again the prudent thing to do in the development that we could have in the second half of the year.
  • Jailendra Singh:
    Great. Thanks a lot.
  • Operator:
    Thank you. Your next question is coming from the line of Josh Raskin from Nephron. Your line is open.
  • Josh Raskin:
    Hi. Thanks. Good evening. First question just -- and I should know this I think is that the MLR on the ceded premium seems to move in the opposite direction of the retained premiums. And I'm just still trying to reconcile, why that's happening, why would the MLR be dead or on the ceded premiums, or am I just completely reading that wrong?
  • Scott Blackley:
    Yeah. Josh. Well, you can certainly pull up with Cornelia, who can walk you through all that, math but when you look at the way that the MLR for ceded works there is puts and takes related to prior period development that come in and impact that. So it can skew it and make it look different than the rest of the MLR for the non-ceded book.
  • Josh Raskin:
    All right, I'll follow-up with Cornelia. And then, on the +Oscar pipeline, I'd just be curious you mentioned it's as robust as you've ever seen it. Where are you seeing the most interest? And how should we think about the cadence or the timing of potential announcements? And I know these are long sales cycles and that sort of thing. But I'm just curious, kind of where you feel like that pipeline is moving.
  • Mario Schlosser:
    Definitely it continues to be a shift of provider saying "I want to reinvigorate my insurance company I already have one" or "I want to start an insurance company." That is clearly one area. But we're actually also, now seeing demand for co-admin systems and slicing kind of into our service there. And that can actually either be as a business process, as a service what can even be as a software as a service offering where we can actually do both there. And so it's -- that's really -- it's a pretty equal fit at the moment between these three different ways of going about this. And so, yes, when we have a deal to announce there, we'll be excited to do that.
  • Josh Raskin:
    Understood. Thanks.
  • Operator:
    Thank you. Speakers, I am not seeing any other questions, at this time. I'd like to thank everyone for joining the conference today. Ladies and gentlemen, this concludes today's conference call. Thank you all for joining. You may now disconnect.