Oak Street Health Inc (OSH) 2021 Q2 法說會逐字稿

Good morning, and welcome to the Oak Street Health Fiscal Second Quarter 2021 Earnings Call. (Operator Instructions) Please be advised today's conference call is being recorded.

Hosting today's call are Mike Pykosz, Chief Executive Officer; and Tim Cook, Chief Financial Officer.

The Oak Street Health press release, webcast link and the other related materials are available on the Investor Relations section of Oak Street Health's website. These statements are made as of August 10, 2021, and reflect management's views and expectations at this time and are subject to various risks, uncertainties and assumptions.

This call contains forward-looking statements and that statements related to future, not past events. In this context, forward-looking statements often address our expected future business and financial performance and financial conditions and often contain words such as anticipate, believe, contemplate, continue, could, estimate, expect, intend, may, plan, potential, predict, project, should, target, will and would or similar expressions. Forward-looking statements by their nature address matters that are, to different degrees, uncertain. For us, particular uncertainties that could cause our actual results to be materially different than those expressed in our forward-looking statements include our ability to achieve or maintain profitability. Our reliance on the limited number of customers for a substantial portion of our revenue, our expectations and management of future growth, our market opportunity and our ability to estimate the size of our target market, the effects of increased competitions, as well as innovations by new and existing competitors, in our market, and our ability to retain our existing customers and to increase our numbers of customers.

Please refer to our annual report for the year ended December 31, 2020. We filed the Form 10-K with the Securities and Exchange Commission, where you will see a discussion of factors that could cause the company's actual results to differ materially from these statements. This call includes non-GAAP financial measures. These non-GAAP financial measures are in addition to and not as a substitute or superior to measures of financial performance referred in accordance with GAAP. There are a number of limitations related to the use of these non-GAAP financial measures. For example, other companies may calculate similar titled non-GAAP financial measures differently. Refer to the appendix of our earnings release for a reconciliation of these non-GAAP financial measures to the most directly comparable GAAP measures.

With that, I'll turn the call over to Mr. Mike Pykosz, CEO of Oak Street. Mr. Pykosz?

Thank you, operator, and thank you to everyone that is joining us this morning. Joining me on today's call is Tim Cook, our Chief Financial Officer.

I'd like to start my comments this morning by once again thanking our team members who continue to work tirelessly to support our patients and communities. Our team is continuing to adapt to changing visions around them, and we've settled into operating all aspects of our model with COVID remaining part of our lives. We are encouraged by strong results across a majority of the drivers of Oak Street performance.

We had strong revenue growth during the second quarter, driven by patient growth in the center, and corporate costs were in line with expectations, resulting in increased operating leverage for the business. We are ahead of pace on our center openings, and our growth outlook remains encouraging on both the patient and site level. Despite this performance, medical claims expense was higher than projected in the first half of the year, leading to a higher-than-expected adjusted EBIT loss in Q2. Looking forward, due to the increase in cases driven by the Delta variant as well as unknowns around the shape of the COVID recovery, we are projecting similar levels of medical costs for the remainder of the year.

However, based on the results year-to-date on patient growth and the disease burden of our patients we are capturing, we are confident that our patient incentive level economics will return to historic levels in 2022, even at this level of medical costs. When medical costs try to return in whole or in part to pre pandemic levels, we expect to see a step-up in those economics. For that reason, we remain confident in our unit economics and are planning to increase the patient center expansion in 2021, raising our new center guidance of 46 to 48 new centers.

The second quarter was the time of transitioning back to focusing on our core model for Oak Street. The quarter opened in the midst with an all-hands-on-deck approach to vaccinate our patients and communities. We've delivered over 180,000 vaccine doses. And when the vaccine was in short supply earlier in the year, we took a resource-intensive approach to ensure doses were going through residents of the communities we serve, many of which were disproportionately impacted by COVID. The second quarter continued, and we were able to vaccinate the vast majority of our patients and teams, the all-hands-on-deck approach with lockdown, and we refocused our efforts on the core of what we do, keeping patients happy, healthy and out of the hospital. In this time at Oak Street, COVID is still a factor, but is not the factor that consumes the primary focus of our teams.

As the country has returned to normal, we've been rolling out the community marketing approach that was the foundation of our sales and marketing results pre pandemic. We're in the midst of our MeetMe at Oak Street campaign, in which we are hosting flagship events at all 100 of our centers. These events are generally held in the parking lots of our centers and have themes chosen by our local teams, such as a Welcome Back America Jazz Jam for one of our Philadelphia centers, (inaudible) Avalon Park in one of our Chicago centers and a senior resource expo at a center in Dallas. We've averaged over 150 senior attendees per event so far and are on pace to host over 15,000 seniors. In addition to being a great way for us to meet older adults in our communities as we give them an opportunity to safely get out and socialize, the event also give us an opportunity to introduce or reintroduce ourselves to chambers of commerce, community groups, local politicians and other community partners, setting up the opportunity to schedule smaller events over the months and years to come.

We are excited to build back to the volume of events we had in 2019 when we conducted an average of 400 events per center.

While the (inaudible) events and the ramp of our community marketing approach will not impact our financial performance until Q3 to some extent and even worse in Q4, we did see the early returns of our average team being able to return to the community, as well as continued execution of central growth channels in our Q2 performance as evidenced by the higher than projected revenue and patient growth. We generated record revenue of $353 million in the quarter, exceeding the high end of the guidance and representing 65% growth compared to Q2 2020, bringing our growth for the first half of 2021 compared to the first half of 2020, up 56%. We expect a 47% pace of growth from Q1 2020 to Q1 2021 to represent a low point for the next several years.

The first half of the year showed a large increase in third-party medical costs compared with prior experience and was higher than our previous projections. We believe this increase was driven by the direct and indirect impacts of COVID across 3 primary categories. First, costs from admission. In the first half of the year, Oak Street experienced 15 million of direct costs from COVID admission. COVID admissions were highest in January, began to drop in February and were reduced by 97% by June from the January peak, which we believe were driven by both the efforts of our team to get our patients vaccinated as well as lower community infection rates.

Second, an increase in non-acute utilization. Non-acute utilization, including specialist business diagnostics and outpatient procedures, significantly increased in March following the vaccine roll off for older adults. Non-acute utilization was $8 (inaudible) higher in March than our average monthly non-acute utilization in the second half of 2020, and a similar amount higher than our non-acute utilization in 2019. Our April results, while not as complete as March, suggest a similar level of non-acute utilization. We booked Q2, assuming this increase in utilization will continue, leading to an increase of $24 million of medical costs from March through June. We believe this is driven in part by increased comfort with patients to access medical care following vaccination, relaxed payer standards due to public health emergency and specialists and hospital system behavior.

Third, significantly higher new patient out costs compared to what we've seen in the past. New patient medical costs were 50% higher than what we've seen historically for new patients. This increase in new patient costs grew up $20 million in higher costs for the first half of the year. Despite the increase in (inaudible) cost, we only saw a small increase in the risk score for these new patients compared to the new patients in prior years, which we believe is driven by lower engagement by the health care system in 2020, which flowed through to 2021 revenue. As a reminder, risk scores lag a year and depend on diagnosis capture during provider business, thus, lack of engagement like a double effect of reducing the incoming risk score while also likely increasing disease burden of patients.

The total result of the above is an additional $59 million in third-party medical costs for the first half of the year, driven by the lingering impact of the COVID pandemic. While we did project a portion of the above cost, the magnitude was higher than our expectations. The higher than projected COVID related costs offset improvements in other medical cost components, favorable prior period adjustments and strong performance in other business drivers. The net result was an adjusted EBITDA loss of $53.5 million for Q2.

Our increase in medical costs is concentrated in our D-SNP and MA HMO patients. Our PPO patient medical costs were essentially flat, which given that our PPO patients (inaudible) are higher income than our HMO and D-SNP patients, leads us to believe the results we are experiencing are being exacerbated by social factors and their impact on lower income older adults during the pandemic.

Because of the payment lags, we have the best data availability through April this year. We booked May and June with an equivalent PMPM medical cost we experienced in the first 4 months of the year. Looking forward, our updated guidance reflects the med cost increase observed in the first half of the year continuing throughout the year, given uncertainty around the impact the delta variant had over hospitalizations for older adults, lack of precedence around elective utilization on the tail end of a pandemic and our expectation that new patients will continue with higher disease burden compared to past experience.

Our care teams are laser-focused on continuing to elevate the care provided to our patients, and we aim to reverse the trend observed from medical costs in the second half of the year, although we've not concluded the potential for this improvement into our guidance. Based on the data collected year-to-date, we have seen that the disease burden of our patient population gets substantially increasing compared to prior years. This is true both for new patients as discussed as well as existing patients. This leads us to believe that despite capturing similar overall new burden on our patients in 2020 as we did in 2019, the challenge with caring for patients in the early days of COVID resulted in us not capturing the increased disease burden of our patients that we are observing this year.

The disease burden we are capturing on our patients today will not translate into corresponding increase in revenue until 2022. From our results to date, we expect the increase in revenue per patient in 2022 to offset the increase in medical cost per patient we have witnessed this year. Said another way, vis-à-vis the elevated medical cost witnessed in the first half of the year continue going forward into 2022 and beyond, with the increased revenue per patient associated with the increased disease burden of our patients will actually have a similar patient contribution in 2022 compared to 2019. If COVID related medical costs received causing medical costs to revert to a level more in line with what we have witnessed prior to the first half of this year, and our care model is able to further impact the cost trend, we will see significant improvement in per patient contribution compared to 2019. This, combined with a strong result on patient growth and operating costs, give us confidence in the continued strength of our center, economic and center ramp. Additionally, despite the medical cost headwinds, we are still seeing our immature centers performing ahead of our (inaudible).

For these reasons, we are raising the guidance around the number of new center openings from 38 to 42 to 46 to 48 as we are confident in the durability of our core economic model and believe the additional centers will drive increased profitability in 2023 and beyond as they mature.

In summary, we are encouraged by the performance across the majority of our results in the quarter, including strong patient revenue growth and increasing operating leverage. The lingering impact of the pandemic led to increased medical cost into lower-than-expected adjusted EBITDA, but we believe the medical cost impact is temporary in nature, and our financial performance will be boosted by the expected increase in per patient revenue in 2022. Our confidence in the future strength of our unit economics gives us the confidence to continue to increase the pace of center expansion, and we're enthusiastic to continue to build a transformative organization.

I'll now turn it over to Tim Cook, who will walk you through our financial results in more detail. Tim?

Thank you, Mike, and good morning, everyone. We produced another strong quarter with $353 million of revenue, up 65% from a year ago and exceeding the high end of our guidance range by over 10%. Patient demand for Oak Street remains high as it provided care to 122,000 total patients during the second quarter, and our at-risk patient base, which now includes our direct contracting patients grew by 54% to 88,500. At the end of the second quarter, we operated 95 centers with an increase of 9 centers compared to March 31, 2021, and 41 more centers we operated at the end of the second quarter of 2020.

Capitated revenue for the second quarter of $346.7 million represented growth of 67% year-over-year, driven by a 54% increase in our at-risk patient base, an increase of approximately 9% in our capitated rates attributable to increased premiums from higher acuity patients.

Total revenue grew 65% year-over-year to $353 million, primarily driven by the increase in our at-risk patient base. Additionally, $14.5 million of capitated revenue in the second quarter of 2021 was related to prior periods. $10.7 million of this amount pertained to our 2020 financial results, primarily related to the full year payment for 2020 risk adjustments in patient retroactivity and the 3-point million dollar balance was related to Q1 2021 patient retroactivity.

As a reminder, patient retroactivity is typical and occurs when health plans pay Oak Street retroactively for patients managed in prior periods, but not previously included in our rosters, and therefore, not previously recognized in revenue or medical claims expense.

Our medical claims expense for second quarter 2021 of $281.4 million, representing growth of 81% compared to second quarter of 2020, driven by the 54% increase in patients under capitated arrangements and an 18% increase in cost per patient. Mike already walked the key drivers of this increase, but I would add that our second quarter results included $19 million of negative prior period development, $24 million of which was related to Q1 2020 offset by $5 million of prior period favorability related to fiscal year 2020.

The negative prior period development related to Q1 2021 was due to us having relatively limited claims data when we closed the first quarter in claims volumes ultimately being greater than we estimated at that time. Upon receiving incremental data in the latter half of the second quarter, we better understood the drivers Mike walked through a minute ago.

Our cost of care, excluding depreciation and amortization, was $67 million for the second quarter, a 70% year-over-year increase driven by increases in salaries and benefits, occupancy costs, as well as higher medical supplies and patient transportation costs related to a 76% increase in the number of centers we operate and growth in our patient base.

Sales and marketing expense was $25.9 million during the second quarter, representing an increase of approximately 156% year-over-year and was driven by greater advertising spend to drive new patients to our clinics and net headcount growth. As a reminder, sales and marketing expense is artificially inflated on a year-over-year basis as it was partially depressed during Q2 2020 due to the COVID pandemic, which included a temporary suspension of community outreach activities, the furlough of our local outreach teams and other marketing initiatives. This increase also reflects an investment to support our significant year-over-year growth in new centers and new markets.

Corporate, general and administrative expense was $74.2 million in the second quarter, an increase of 139% year-over-year. Stock-based compensation expense was the largest driver of the increase, representing $39.7 million of expense in the second quarter of 2021 compared to $4.2 million in the second quarter of 2020. As a reminder, the increase in stock-based compensation is primarily driven by an accounting change related to awards issued prior to our IPO in August 2020 and is not a function of stock awards issued since our IPO.

Excluding stock-based compensation, corporate, general, administrative expense was $34.5 million in the second quarter of 2021, an increase of 28% compared to the second quarter of 2020 driven by primarily headcount costs necessary to support the continued growth of our business.

I will now discuss 3 non-GAAP financial metrics that we find useful in evaluating our financial performance: patient contribution, which we define as capitated revenue less medical claims expense grew 24% year-over-year to $65.3 million during the second quarter. Platform contribution, which we define as total revenue less the sum of medical claims expense and cost of care excluding depreciation and amortization, was $4.7 million, a 76% decrease year-over-year, driven by the previously discussed increase in medical claims expense, as well as a significant growth in our center base, and therefore, the portion of our centers, which are immature. Adjusted EBITDA, which we calculate by adding depreciation and amortization, transaction offering related costs in stock and unit-based compensation, but excluding other income to net loss, was a loss of $53.5 million in the second quarter of 2021 compared to a loss of $17.5 million in the second quarter of 2020.

We finished the second quarter with a strong balance sheet and liquidity position. As of June 30, we held approximately $1,080 billion in unrestricted cash and marketable securities. Our liquidity position will support our continued growth initiatives, primarily our de novo center-based expansion. For the second quarter of 2021, cash used by operating activities was $53 million, while our capital expenditures were $10.7 million.

Now I'll provide an update to our 2021 financial outlook. For fiscal 2021, we are increasing our guidance for total centers to a range of 125 to 127 from our prior outlook of 117 to 121 centers. Total at-risk patients to a range of 109,000 to 113,000 from our prior outlook of 107,000 to 112,000 and our revenue guidance to a range of 1.37 billion to 1.4 billion from our prior outlook of 1.3 billion to 1.34 billion.

We are reducing our adjusted EBITDA guidance loss to a loss of $240 million to $220 million. As Mike mentioned, our EBITDA guidance is seen as the continuation of the medical cost trends that we experienced in the first half of the year, and also includes losses from the incremental new center growth.

For the third quarter of 2021, we are forecasting revenue in the range of $355 million to $360 million and an adjusted EBITDA loss of $65 million to $70 million. We anticipate having 109 to 110 centers in an adverse patient count of 98,500 to 100,000, including direct contracting patients at September 30, 2021.

With that, we'll now open the call to questions. Operator?

(Operator Instructions) Your first question comes from Ricky Goldwasser from Morgan Stanley.

I'm trying to understand the higher acuity of the new patients. So how do you measure higher acuity? Are there any patterns or disease categories that you're seeing? And then also, when you think about these new members that are just onboarded, what's the vaccination rate among the new members? And is this a population that is now at a higher risk of contracting COVID that might impact the second half?

Yes, thanks, Ricky. This is Mike Pykosz. On the new patient question, I think you can look at the acuity in really 2 ways: One is the utilization of both acute and non-acute that those patients are doing. As I referenced, that's 50% higher than we see in the past. So they're obviously going to hospital a lot more. They're using a lot more medical resources. So that obviously impacts cost today. The other way to look at it -- another way we do look at it is once the disease burden (inaudible) population. And that's really driven by what are the chronic disease codes that we're capturing as we're doing our really intensive onboarding approach for those patients. And on that dimension, we're capturing a lot more patients with CHF, diabetes complications, COPD, et cetera, et cetera. So I think that it's really both on our new patients. They're using the system more resulting in higher cost today. We're capturing a much higher disease burden on them and we captured kind of an equivalent new patient period in years past, which is going to drive higher revenue next year that won't flow through until 2022.

On the vaccination status, our outage model generally pulls patients from the communities we serve, albeit usually patients who are slightly less engaged in health care than the average patient in those communities. Keep in mind, our communities served are generally moderate to lower income. The majority of our patients are people of color. And so the vaccination rates of our patients who we're pulling in generally are generally, directionally, what the communities that we serve are maybe a little bit lower. So Medicare on average is -- 80% of Medicare recipients are vaccinated. The patients who are applying new patients are (inaudible) lower than that. That said, we're very effective at overcoming vaccines -- haven't seen getting patients back and that's obviously a core part of every business, right? The patient who is vaccinated is having that conversation. So I do believe that we'll get our new patients up to kind of that same level of Medicare and hopefully higher where we can see our existing patients as well, with the understanding that the challenge I think we're facing is higher than you'd basically be at a different patient demographic.

Okay, and then my follow-up is on the marketing campaigns. If you think about the surge of the Delta variant, what is the contingency marketing plan if you have to scale back on the community events?

Yes, I mean I think the contingency will be doing something like we did really prior to the vaccines, right? We did grow in second half of 2020 and in the first part of 2021 at a fine pace. So a lot of those same channels that we leverage, that will continue to leverage. That said, I think we found ways to safely and effectively run our community-based marketing approach given the vaccine status of our team members now lowers the risk of even the communities, finding alternative venues of outside events and things of that nature. So again, I don't expect us to go back to kind of where we were in the second half of 2020 or even early 2021, where you're really trying to avoid those types of events. I just think that we are excited, and I think you find ways to run them safely. And what safely means that it differs based on where the Delta variant is, et cetera. But I think it is important that mental health and the different components there are also really important. So I think it is important, just overall that we're getting people out, socialize, kind of enjoy life but doing it in a very safe manner.

Your next question comes from Sean Wieland from Piper Sandler.

So I appreciate the breakout of the excess medical expenses, the 3 categories. And I understand how you get out of the box with the COVID and with the higher disease burden. But less certain, I don't understand really the path forward on excess utilization of non-acute care. You run in an open network. These new patients are used to going to the doctor or specialist anytime they want. How do you get that back in the box?

Yes. Sean, this is Tim. Thanks for the question. It's across a number of fronts. Our teams, as Mike mentioned, have been working very hard to execute the vaccination plan that we put in place. And while we talk about vaccinations as if they're relatively easy, a shot in the arm is actually incredibly complex and it's very distracting for our teams. So part of it, I think, is just refocusing our efforts and moving beyond that vaccination efforts to, I'd say, more core care model execution. These things are not always mutually exclusive, but the reality is there's always so much bandwidth within a center. So I do think we'd benefit from moving more -- getting back to basics on our care model, not that the vaccination efforts will entirely cease because, of course, we're going to continue to push our patients to be vaccinated and hopefully, we'll get to you as close to 100% as we can here over the coming months.

The one thing -- the additional thing I'd say is hard to put data points around this, but I think we're also seeing other market forces at play today, which is more volume flowing in to fee-for-service medicine as the system rebounds from a tough 2020, particularly in the middle part of the year. And so I think we're seeing elevated levels of utilization now (inaudible). And part of that, I think, is to Mike's point of patients coming in sicker and being heavier users of the system. I think it's also more of the system playing a little bit of catch-up. And by and large, what we've experienced in health plans is historical processes they may have in place to limit those or more active network management. A lot of those activities have either been put on hold or are dialed back. And so we're seeing this stage in the market where utilization is coming back online, and there's less management in managed care right now. And so I think the combination of us getting back to where we had been from a care model perspective, as well as the market return to normal should, over time, exactly how long is the question, but over time, serve to drive down that non-acute utilization that you mentioned.

And how many patients in total are we talking about driving these excess costs?

So we've seen -- I think as Mike mentioned, we've seen an increase in (inaudible) across all of our patients. It is most pronounced for our newer patients. You can get a sense for how many patients we've added since 12/31 to June 30. And then on top of that, obviously, that's a net number. We've obviously had a lot of patients and depending upon what you assume for Oak Street attrition, it's a reasonable number of patients that are new patients to Oak Street that are driving those costs.

Sean, the thing I'd add to both of those questions quickly is, on the non-acute costs, the factors you mentioned around kind of active special have been present for the history folks, right? And so we have a very good baseline what we observed in 2019, 2018, other years and even the second half of 2020 is all pretty consistent. In March, it really spiked. And again, that could be some kind of ongoing change in health plan or provider or patient behavior. That will be kind of part of what we do at Oak Street long term. That is, I guess, possible. And again, even if that persists, we're confident the economics will return in 2022.

I do believe that a lot of it was driven by the fact that when I (inaudible) that's exactly when all of our patients kind of came back to it, right? And so I think there was a (inaudible) going back in the community, going back into health care, getting back at the health care system. I think health care workers that were also vaccinated and also I think, ready to make up for lost income. So I think that's part of it.

So certainly our hope is that we kind of eventually over time, revert back to what normal was prior. There is some phase shift in how patients and providers are behaving. Obviously, we'll try to impact that, but we also feel like we could absorb that cost in 2022 and beyond. So hopefully, that helps give more context.

Your next question comes from Lisa Gill from JPMorgan.

Mike, can you talk about the percentage of the assessments that are done at this point for 2022? You talked about revenue being better in 2022. That would be my first question. And then just as a follow-up, is there any geographic area to call out when you think about some of these higher costs?

Yes. On the first question, obviously, we aim to have our kind of annual wellness visits for 100% of our patients. Usually, we complete it in kind of the mid- to high 80s is what actually gets done based on our patients. Right now, we're, I think in kind of the mid- to high 60s is where we're at on percentage of our patients who (inaudible) annual wellness visits complete. So we feel like we're trending kind of towards those upper 80s, and we'll obviously work hard to push that into the 90s percent complete.

The data points we have as far as the disease we're capturing in our annual wellness, it's a large portion of our patients at this point. So we feel like it's pretty accurate data at this point. Sorry, what was your second question? I apologize.

Yes, the second question just is, is there any geographic area that you'd call out from any of this utilization?

No, that's actually one of the, I would say, trends that has given us that probably increased confidence to continue to grow. We're really seeing this across the board. We're seeing this in new geographies and old geographies. We're seeing this in older centers, older care teams, newer centers. And actually, the reason why that gives us comfort is we don't think this is some problem of the quality is going down as we scale or we're not actually finding the same success in new markets, et cetera. I think it's just -- (inaudible) the external factors that we need to manage through versus some difference in Oak Street performance as we grow.

Your next question comes from Justin Lake from Wolfe Research.

Wanted to talk about the second half guidance and specifically, Tim, maybe you can help us understand in terms of the actual dollars, even the (inaudible) for the back half of the year. Maybe you can go through the 3 buckets and let us know what you're assuming in terms of dollars above typical trend. And again, like I said, MLR would be helpful as well in terms of the target for the back half, if you can share that.

Yes, so Justin, thanks for the question. So with respect to trends and what we're assuming in the second half of the year, what we experienced in the first half, let's say, Q1 and Q2 had different composition. So Q1, as Mike mentioned, more COVID related as COVID lines were relatively high in January and February and started to ebb as our vaccination efforts took hold and the vaccination rate increased. And therefore, about the time we got the end of Q2, COVID expenses were relatively de minimis. Obviously, that's all pre Delta, but I think we were seeing a lot of the benefits of that vaccination work.

What happened in Q2, I'd say was more of this non-acute utilization coming back online. And I'm sure there were a number of drivers, one of which is likely (inaudible) our patients were vaccinated, they felt more emboldened or were more likely to go, seek care because they were more confident about their ability to do so in a safe manner.

As we look to the rest of the year and think about what we factor into guidance from a medical claims expense perspective, we haven't taken the peak of COVID plus the peak of non-acute utilization and combined the 2 of them. We essentially said, look, Q1 utilization over the first half of the year looked like this, again, with a bit of a different mix between Q1 and Q2. And that -- those levels in aggregate represent higher levels than we've experienced. And we're going to -- given the fact that we don't have any data today to suggest otherwise, we will continue that high level of utilization through for the remainder of the year. That works out to MLRs, plus or minus 80%, Justin.

Got it. So right around 80% for the back half of the year.

Yes. I think obviously, from our business, there's some seasonality, so MLR will increase each quarter as new patients come in at a lower -- excuse me, a higher MLR than tenured patients. So Q3 might be a little lower, Q4 might be a bit higher, but you get a sense.

Got it. And then Mike, I just wanted to follow up on the new patients and make sure I understand it correctly in terms of risk score change. What you're saying is if we just say the average patient came in typically at a 1.0 and now they're utilizing 50% more services, you're saying that the risk score of this patient that you're seeing right now, even though you're only getting paid for 1 point of risk score, you actually are submitting enough claims to CMS, where you would get paid for a 1.5. Your revenue would also go up 50%, and therefore, the profitability would return to normal even at this level of utilization. That's the right way to think about it?

Yes, that means, obviously, those numbers are directional. But yes, that is the right way to think about it.

Your next question comes from Elizabeth Anderson from Evercore.

Maybe turning towards the new center guidance increase. Can you talk about over the longer term, like are there certain gating factors that in terms of growth or profitability metrics that you use when sort of deciding like what the right pacing of center openings is?

Yes, I think there's really 2 things we look at when we're determining the pace of center growth. One is how are all of the kind of internal components performing that make a center successful. One is our pipelines and channels are having trouble filling roles. Are we still bringing in phenomenal people to fill all of our roles? How we are doing from a leadership standpoint? Kind of all different pieces that go into the pipeline of actually opening a center payer contracting credential kind of all those nuts and bolts. And do we feel like we're seeing any strain on the system. There's probably the 2 biggest ones that I look at are kind of on the provider hiring, the clinical leadership and kind of the P&L leadership. So we feel really good across those dimensions as these are obviously key drivers and must-haves to be successful.

So one part of it, how do we feel about all those components. And right now, we're obviously finishing up with 2021 expansion and started to turn to 2022 expansion. And so you kind of think about how do I build the infrastructure to perform exceptionally on all those dimensions kind of in advance of having to perform on them, right? And that's why we always think about titration up, right, we're not going to open up 150 tenders next year because we may find some of those components weren't able to kind of succeed at that much higher level. So we always kind of raise the number of centers as long as we're -- we feel like we're able to do so, but doing it in more of a moderate way. So that's kind of one half of it.

The second half of that is more on the financial performance and the leading indicators of the centers. So how are we doing on center cost, direct cost of care, how are we doing on patient growth, how we're doing on kind of the efficacy of our model. The reality for new centers, especially, you don't have a ton of actuarially sound data on 30 (inaudible) on your patients haven't entered risk plans yet. And if they have, it is very new. So we do know from our past history that if we do the right things in the input side of caring for our patients, we're engaging them, we're following our care models to the T, et cetera, and that will drive great results. And so the other thing to really look at is how are the new centers ramping from a patient growth perspective, how are the new centers ramping from a cost? And how are new centers ramping from a kind of execution of the care model perspective. And then as they get more mature, right, then in the year 2, year 3, year 4, then we start looking at more and more of the economic indicators to make sure they're kind of the patients that center confirmation or so needs to be.

So obviously, this year, because of the increased medical costs, that impacts 2021 results of the center. That's where I think looking at the revenue that we're going to receive in 2022, that leading indicators of what gives a lot of comfort that the economics of the centers will remain where we've seen them in the past and hopefully north of there.

Got it. So as we think about 2022, you would expect that sort of number of centers to sort of increase perhaps from where we are in terms of the center guides for 2021, right?

Yes, correct.

Your next question comes from Kevin Fischbeck from Bank of America.

Just want to go back to the cost issue. I guess it's interesting you're seeing these cost dynamics happen across the country. It seems like a lot of the managed care companies are also seeing cost pressures across different books of business. I mean I guess what you're saying makes some sense about vaccinations and higher utilization, but isn't this the type of thing that you would be seeing if trend was actually accelerating that cost would be slowly getting back more faster or there would be high utilization in certain pockets. Why isn't this the first sign of a higher trend that might impact next year's profitability too?

Well, I think there's a -- if we look at the components of the medical costs, obviously, COVID hospitalizations, we hope we don't have a significant headwind of COVID hospitalizations in 2022 for a huge number of reasons, right? And so hopefully, we were able to, as societies navigate through the Delta variant and increased vaccination rates and get to a place where we're not talking about a huge number of hospitalizations from COVID in 2022. The kind of one familiar component around the non-acute care elective care, I think that -- and there's -- in my mind, there's no reason to believe that you're going to see some sea change of how people navigate the health care system and the amount of care they need in the health care system that goes on into perpetuity. So again, I think my belief from what I know today is that, that number will start to normalize over time. Maybe a little bit of (inaudible) but certainly not like we're seeing today. What you're seeing today is just the results of people feeling much more comfortable with part of the health care system. And I think a lot of providers who are kind of ready to get back to what they missed in 2020. So again, I think that those are the factors.

Now I think the second part, what we said what you'll see in 2022. And I would look at it slightly different, what happens if we do see it in 2022, right? So I have reason to believe that these things will start reverting. If they don't revert, I think just because of the increase of the disease burden of our patient population, which I don't think is necessarily the same, I don't even think it doesn't translate that to the rest of managed care. But again, for our kind of pretty narrow demographics of the patients we serve because of the increased disease burden that we're seeing, the revenue offset that level of medical costs persist in 2022. So our hope is it reverts and we actually have expansion of margins if it maintains. We're confident we'll still generate strong and comparable results to what we've seen historically in 2022 and beyond even at this elevated level.

Okay, that's helpful. I guess you're talking a lot about margin normalization next year and rates being better next year because of coding. I guess are you having any conversations with the managed care companies themselves about how they're pricing for next year or how they're treating your payments for next year to potentially adjust to reflect to what you're seeing?

On the second part of the question, I mean our contracts are multiyear contracts at a percentage of premium. And so -- to the extent that revenue per patient goes up, our revenue will go up (inaudible). There's no discussions around kind of adjusting that from either side, nor do we plan to have those.

I think on the bid perspective, we talk to our health plans around how they're bidding, how they're thinking about it. Obviously, we have a large number of health plan partners across a large number of geographies. So there are in the triple digits of kind of health plan, big combinations. So and a lot of times on dynamics become local and health plan dependent, et cetera. But we certainly talk to our health plan to make sure we understand what they're bidding and kind of how the benefits are changing and how that impacts our economics. And so again, we feel comfortable with where they're coming out. Generally, I'll be honest, we mean we like when plans are putting great value proposition out for our patients because that will help get more of our patients under managed care and help keep our patients in managed care. And frankly, making easier for our patients to navigate the system appropriately without undue cost burden. So we generally walk step in what the plans are doing.

Your next question comes from Lance Wilkes from Bernstein.

Two related questions on the customers and the new patients that are coming in. The first one is on the characteristics of those patients? I'm just trying to understand kind of their prior providers, prior plan experiences that that's any different than it was maybe with prior year tranches of new patients. The other question is really related to retention strategies, what your outlook is for retention and what your normal churn is on those patients as well?

Yes, on the first question. As far as percentage of patients who are on a managed care plan, percentage patients who were on Medicare, engagement with providers previously, I think what we're bringing in is generally similar because the model is generally similar. And Medicare Advantage penetration grows across the country and if we go to markets with higher Medicare Advantage penetration, the mix of people who are coming in on Medicare Advantage does go up a little bit.

I would say all else being equal, we are seeing less engagement in the health care system previously from our patients than we've seen in the past. A lot of our patients haven't been to the doctor in years or they use the emergency room as their primary care provider. A lot of them started calling health centers, where yes, they (inaudible) they can go to, but it's not a longitudinal relationship for doctors and panels is almost kind of a it's a different level of care. And so I don't think there's a huge difference from where we're getting patients.

The one kind of hypothesis I have which I can't prove is I think a lot of our patients previously 2019 and before came to us in a lot of really kind of fun energetic community events, which I think attract people who are kind of interested about in a position to go to those type of things. We did lots of that in 2020 for obvious reasons. But we met patients in a lot of other ways. But I do wonder if part of the reason we're seeing kind of a higher disease burden of incoming patients is you're meeting less of the people who are coming to a Zumba class, and you're meeting more people who you are meeting at different places that are kind of referrals from health -- case managers, those types of things. So again, I don't have any quantification but I do wonder if that part of it, which if that's the case, we get more (inaudible).

Then on retention strategies, just what are you normally running for churn with these kind of patients? And do you have any particular insights into this population as far as NPS scores or customer satisfaction or things like that, that might indicate that they're more or less likely to be a normal sort of retention?

Yes, we haven't seen any differences with regards to retention or NPS basis. It still remains very strong on all those dimensions as we've always seen. So again, I think -- the thing I'll make sure at any point on the new patients, while it's certainly a headwind in medical cost this year, having that higher lease burden and associated medical costs coming in, we're confident we'll keep the patients into 2022 and then beyond, right? And we're also confident that once we get into that year and we're able to kind of get revenue to work should be given their disease version that from an economic perspective, there'll still be profitable patients, et cetera. And frankly, we'll work really hard to (inaudible) medical costs burden that they come in with because they've been careful. That's what our model does. And so again, I can say unexpected, it's diverse so much from what we've seen historically, but it's not something we're worried about and can't handle.

(Operator Instructions) Your next question comes from David Larsen from BTIG.

Can you talk a bit about your direct contracting program? How many lives do you have in that? And how is there sort of -- how are the patient economics there with respect to both revenue collected per patient and also medical costs?

David, it's Tim. Thanks for the question. Direct contracting -- the direct contracting value you mentioned is largely consistent with what we -- is entirely consistent with what we discussed in our prior call, which is we had and guidance about 6,500 patients enrolled in Q1, expected to add about 2,000 to 3,000 patients per quarter. And we're right on top of that for our performance right now. So we're in that range from an economic perspective as patients are performing as we expected. So -- the issues that we talked about here, we expect new patients is largely related to Medicare advantage patients, not direct-contracting patients. We have not broken out this financial specifics for our direct-contracting patients relative to our MA patients.

Okay, great. And then just 1 more quick follow-up. With regards to your legacy patients, did you see a significant like 50% increase in utilization? Or was it more limited to simply the new patients? And what I'm getting at is, are we just seeing sort of a catch-up in deferred and delayed care, which we kind of expected and this should sort of dissipate over the next quarter or 2 or is there something else? So the difference between your legacy patients and the new patients? And are you being conservative really in your guide is another sort of way of asking this.

Well, I think about the new patients -- I mean to say is, we don't know what -- who the new patient is that's going to join us in a month. And so our projection in our guide is that the incoming disease burden we're seeing and the associated economics that this burden will be similar for the remainder of the year on new patients. And then obviously, when we get into 2022, we'll get the benefit of the appropriate risk score for those patients.

On existing patients, again, I think the 2 trends we're seeing, as Tim discussed are: one, kind of in the very early part of the year, higher-than-expected COVID admissions, and in the latter part of the year, kind of a bounce back of elective in non-acute care. And then obviously, the kind of the mix of that is this kind of happened in different parts of the first half of the year. The mix of that created elevated medical costs. And again, we're assuming that level of elevated medical costs will continue through the second half of the year. We think that is appropriate just given some of the unknowns around how the pandemic will progress, payer provider behavior. I'm really just -- it's a pretty unprecedented circumstances to navigate. Certainly, our hope is that some of the nonelective care -- sorry, non-acute care, the elective care starts to go down because people have kind of gone back to the doctor, they've gotten the test they're going to get, and that kind of winds itself down. We hope that happens. We're certainly not baking in from a guidance perspective..

Your next question comes from Richard Close from Canaccord.

I think you guys sort of answered this with respect to increasing the number of new centers and you talked about feeling comfortable hiring. I'm just curious, as you think about going forward or second half of '21 and '22, given COVID has really strained health systems and whatnot, are you guys finding it easier to attract people to the primary care side of health care in terms of nurses and doctors and whatnot, just maybe being burned out from health systems and whatnot?

I wouldn't say easier. I think that thank God, (inaudible) from health systems and what we've seen is that they've heard about health care and yes, we can try to convince in the difference between outpatient primary care and inpatient health systems. But I have not heard from our team a huge inflow of people that don't want to be foreigners to a hospital at Oak Street. So I hear the theory. I don't have any data to prove it, but I haven't heard anecdotally that happening.

When we do hiring in Oak Street Health, I think the big thing we really focus on is why Oak Street is an amazing place to work. And obviously, I'm by it, so I think Oak Street is an amazing place to work. And the surveys of our team members certainly support that. And so we talk about the job environment being really tough right now, it is very tough. I remember 2014, when you put a job opening out there, you have 100 applicants within an hour. That's certainly not the case anymore. And so one of the things we've really focused on is, hey, we have an amazing team. Our team really believes in our mission, believes in Oak Street. How do we leverage that team to be the team that drives referrals and drives new hires, right? And that has really, I think, been a huge part of why we've been successful in hiring so far in the year, is referrals from our team and especially on the provider side, right, very satisfied doctors who love the mission and love the support and feel great about kind of being able to actually have the resource to keep their patients healthy and do what they want to do, like they'll tell friends of residency and friends of medical school and former colleagues about it, and that's a great way to grow. And that kind of goes all the way down to our associates and receptionists.

So yes, we have central recruiting, they do a great job. But while that might have been enough to meet all our hiring needs 3, 4, 5 years ago, that's not the case now. And so again, I think that's an opportunity for us to really be even more successful and further differentiate ourselves because we have invested a lot in all the time and the culture in making Oak Street a phenomenal place to work. And I think you're right, places that haven't -- don't have that foundation, will probably have a harder time retaining their employees and attracting new ones. But we feel really good about our position there because we build folks through it the right way.

Your next question comes from Josh Raskin from Nephron Research.

So 2 questions really. First, just process on third-party medical expense notification, kind of when do you find out that your members or your patients have sought care outside of your 4 walls? And then the second is, is it potentially just the new member issue? Is it potentially just a short-term issue of adverse selection, right? I know we see that on the insurance side, but I could imagine a scenario where these patients haven't had access to care. As you mentioned, they've used the ED and community centers and then they see this Oak Street opportunity to get particularly high-quality care in these great, big clean centers. And is it just simply maybe in the short term, you're attracting those members that have been out of the system most recently?

Yes, to your second question, I think that's exactly right, right? I think there's a -- there are more people who are not getting the preventative and primary care they needed who weren't engaged in the health care system in 2020. And so there's already access problems in a lot of neighborhoods we serve. I think those are very exasperated during the pandemic. I think you are seeing the issue there. And I think we are out there in the communities. Some are new communities we've never been in, some are communities we've been here for a while. We are out there every day trying to meet people and trying to get them engaged and out there, and what we're telling them is hi, come to Oak Street Health, we take great care of you. Here's our vehicle center. Our doctors will spend time with you, we'll listen to you, all the reasons why patients while being part of Oak Street have something to communicate. So yes, we are bringing those people in. And so to the extent that more people have more issues and untreated issues coming into Oak Street yes, that does impact us, right? And so certainly, I hope that, that starts to normalize now that there are vaccines now that the provider community outside of Oak Street is back to kind of back up and running, right?

The other thing I think I touched upon earlier question, but I do think -- I'm hopeful that if we get kind of more of the fun events and those types of things, you kind of can attract some of the patients who are going to be attending again Zumba classes and nutrition health eating classes and things like that, that I think just will, by definition, kind of attract a more active, healthier mix of patients.

Final point on the new patients is, again, even if we don't, even if this becomes the new normal, we're very comfortable operating in this new normal. These are not patients who will be -- I think from insurance perspective, you get kind of "bad risk" and that something that you don't want -- it's not that we don't want these patients. We really do want these patients. These patients are exactly why we exist, to take people that are not being cared for and have unnecessary hospitalizations and costs, take great care of them, and that's a higher amount of work. That's a great thing about risk adjustment is you don't have to worry about kind of population poaching, right? You can just take the patients who need your care and the risk set up for that (inaudible) lags by a year. So again that long answer to the new question -- new patient question. I'll turn over to Tim to have more kind of the visibility of that (inaudible)

Yes, Josh. On the data flow for new patients and what their claims might look like, I'd say in limited instances, we do get claims (inaudible) patients, more typically speaking, the claims that we get once that patient comes patient of Oak Street. So we're not always privy to that patient's historical medical expense. There are -- particularly if that patient came over to us already on a Medicare Advantage plan. The patient came from traditional Medicare. Obviously, there's information available through Blue Button or Blue Book, probably get the term around there, where we can access the claims through CMS. But by and large, that data flow begins predominantly when that patient joins Oak Street. So for the new patients we're talking about, as Mike mentioned, there's relatively limited information. That being said, because of the way Medicare Advantage is structured, those things work themselves out over time, albeit with a lag around how risk work works.

Yes, I was thinking more of just someone goes to the hospital for a COVID inpatient stay. How long does it take for you guys to find that out? And what's the process?

Sorry, apologies. Yes, it varies based upon setting of care, but typically speaking, if a patient is going to the hospital, usually one of the first things a hospital does is contact the insurer to make sure that patients cover and manage the extent of the benefit because hospitals obviously want to get -- have certainly getting paid for the visit. So as part of that process, we typically notify that, that patient, a patient is in the hospital, and that one alerts us from a utilization management (inaudible) but we're ensuring that we take care of these patients, but typically speaking when it comes to hospitalizations, we find that out nearly in real time. When it comes to referrals to specialists, that's that can vary. Obviously, we know when we make the referrals because that's all entered through our system. The question ultimately there is what's the flow through. So for every referral you make, how many actual visits are received and that number can move around over time. But that we have better insight into leading indicators because we're the ones actually driving that volume.

There are no further questions at this time. I would now like to turn the call over back to Mr. Mike Pykosz.

Yes. Thank you, operator. Thank you, everyone, for all the questions and engagement. We appreciate it.

I'd probably leave you with just the -- certainly, the core belief in our team that we can make a huge difference in our patients' care, and that will translate to really strong results from Oak Street across all dimensions, really in 2022 and beyond, right? And I think our goal to build a transformative (inaudible) and based on all the data we have, we feel great about the kind of continual evolution of the unit economics and the continual evolution of the business. And so we remain really engaged and excited for the second half of the year and then beyond.

Thank you.

This concludes today's conference call. Thank you all for joining. You may now disconnect.