How Medicare Advantage winners use data and incentives

Data incentives and care models redefining MA strategy

Vivek Garipalli, John Kao, Julie Smith, and Martin Graf

3 min read

Double Quotes
If you identify the small share of patients driving most costs and deliver the right care to them, you can bend the cost curve, improve outcomes, and create sustainable value across the Medicare Advantage system
John Kao, Chief Executive Officer, Alignment Healthcare

In this episode of the Oliver Wyman Health Podcast, Martin Graf, a partner in Oliver Wyman’s Health and Life Sciences division, chats with John Kao, Alignment Healthcare’s Chief Executive Officer, Julie Smith, former president of Government Business, Anthem and Blue Cross Blue Shield of Michigan, and a senior adviser at Oliver Wyman, and Vivek Garipalli, Clover Health’s co-founder and Chief Executive Officer, to discuss stars, risk adjustment, and more.

“There are different perspectives on how to win in Medicare Advantage. Our perspectives, the collective group of us with differing views, is a good way to start to frame what that next generation of strategy might look like,” Martin says. Below, hear our full podcast conversation, recorded from this year’s Oliver Wyman Health Innovation Summit in Chicago.

Key talking points:

  • A small share of Medicare Advantage members drives the majority of costs, and success depends on identifying and managing that population effectively.
  • Data — how it’s captured, used, and applied in real time — is a core differentiator across approaches.
  • Much of today’s Medicare Advantage success comes from backend levers like risk adjustment, stars, and pricing mechanics. There is growing tension between optimizing those levers and delivering real clinical value to patients.
  • The future of Medicare Advantage will be defined by better outcomes, consumer experience, and more meaningful care delivery rather than financial engineering.

This episode is part of our Oliver Wyman Health podcast series, which includes conversations with leaders pioneering healthcare market transformation. 

Subscribe for more on: Apple Podcasts | Spotify

This episode was first broadcast in December 2019.

Vivek Garipalli

I think that’s what MA is about. Really our whole thing starts with making sure you capture the data, and then making sure you apply the data and do the right thing, taking care of the people who need care the most, and then productizing those care models into products that can serve more and more seniors.

Narrator

By 2030, 20% of the US population, will be over 65. On this episode, three healthcare leaders explain their organizations’ different approaches to the Medicare Advantage advantage, to support the biggest, most lucrative consumer population of the future.

The Oliver Wyman Health Podcast is brought to you by the global management consulting firm Oliver Wyman. For more insights on the business of transforming healthcare, visit our online publication, health.oliverwyman.com.

Martin Graf

Hello and welcome to the Oliver Wyman Podcast. I’m Martin Graf, partner in the Health and Life Sciences practice here at Oliver Wyman. We’re here at the seventh annual Oliver Wyman Health Innovation Summit from our mobile podcast studio at the Chicago Loews Hotel, where Oliver Wyman brings 400 senior executives from 200 companies, 38 states, and 12 healthcare ecosystems together over three days to build for impact and design the healthcare landscape. Joining me today are John Kao, Alignment Health CEO.

John
Hey, Martin, thanks for having me.

Martin
And Vivek Garipalli, Clover Health’s co-founder and CEO. Hey, Martin.

Vivek Garipalli

Thank you.

Martin

And Julie Smith, the past president of Anthem and Blue Cross Blue Shield of Michigan’s Medicare business.

Julie Smith

Looking forward to our session, Martin.

Martin

We’re going to talk about some key takeaways from the Medicare Advantage advantage winning in the next generation of MA. Now, what I’d like to do is talk a little bit about our discussion earlier today, and I think one of the things that was interesting as feedback has been that not all the views that we shared were shared by everyone in the industry. That there are different perspectives on how to win in Medicare Advantage, and that our perspectives, the collective group of us with differing views, is a good way to start to frame what that next generation of strategy might look like, because I think there are different experiences and different drivers of success.

So what I’d like to do is start by talking a little bit about what some of those different operating models in the different business models actually look like. So with that, maybe John, maybe you can talk a little bit about Alignment and the way Alignment has continued to grow and win in the market.

John

Sure, Martin. Thank you very much for having me. The Alignment healthcare model is really predicated on what we believe the Medicare Advantage construct was designed to achieve, which is to provide beneficiaries with really high quality at a very good price point. And in order for that to occur, in our opinion, it starts with what we would refer to as a virtuous cycle where we use our data and our technology to identify individuals that need care that could be categorized as having polychronic conditions and were frail. And that’s kind of the proverbial 20/80, where 20% of the members cost 80% of the spend.

And if you do that and do the right thing and take care of those individuals that need the most care, there’s going to be an opportunity to start bending the cost curve. And if you can provide high quality of care at a very good price, you’re going to be able to grow because by law, the Medicare Advantage model requires you to invest that savings in higher degrees of coverage for beneficiaries. And so, if you got good coverage and good benefits and good quality care delivery and you’ve got good member experience ratings, you’re going to be incentivized to grow.

John

And I think that’s what MA is all about. And so really our whole thing starts with making sure you capture the data and then making sure you apply the data and do the right thing, take care of people that need the care the most, and then productize those care models into products that can serve more and more seniors.

Martin

Yeah. And along those lines, the way Alignment has gone at this, with obviously intervening in the care to a certain extent, has been an interesting difference between, say, network-based health plans. And I know, Julie, you’ve spent most of your career on the network-based side. So, what’s your thoughts on, given what John said, because I think there are a lot of facts that fit that market, right, that 80% of the members do only use 20% of the cost and vice versa. What’s your sense in terms of the network-based plans and how they address that same issue?

Julie

The network-based plans and the larger-scaled incumbents in this space have obviously leveraged their scale in a number of key areas. First, geography, as well as products and their networks. They are very focused on serving consumers with benefits and services that will drive outcomes that will improve the health. In terms of specifically meeting the needs of the 80% and the 20% populations, they use varying degrees of enablement capabilities, partnerships with provider organizations, as well as benefits and services to drive behavior.

Martin

And Vivek, I know your model is kind of unique and different than either the network-based or the Alignment model. What’s your take on how to address some of that? Expensive small component of the population that drives most of the cost.

Vivek

When we think about Medicare Advantage, we got into the space as Clover not at the outset of trying to be an MA plan. Right. We were trying to figure out any other way to build a business without having to be an insurance company. And one of the things we think has been missing in healthcare for a long time is where you can align the business model with … I think everyone, a lot of reason people got into healthcare is the mission of trying to improve health outcomes for consumers.

What we found unique around MA is it’s one of the few constructs that exist in the US where you can build a consumer product, where you can look at a customer over many, many years, seven, eight, nine years, and if you actually drive improved care, better management of that patient, you’re actually going to generate real business value while improving care for that consumer. And that, I think, is something that, because those rules evolved over the last 20 years, it wasn’t as if when Medicare Advantage was started, or launched as Medicare Plus Choice back in the late 90s, that wasn’t the exact construct.

Vivek

So all the large incumbents that are in there today launched a long time ago, and so they built their companies and approaches around the original way of how Medicare Advantage was started. And obviously, as CMS saw how the business models were designed, they realized those business models were designed because of how the rules were set up. And they kept adjusting the rules to try to create the right incentives. And I think if you started a business 20 years ago, you would have purpose-built it for a specific reason. And you now had to kind of evolve your way to what I think is going to need to be in place to be successful.

Going forward, because we only started our business as Clover about five years ago, under the new construct of rules. If we’re trying to build a business 20 years ago around driving improved care management, improving clinical decision-making for physicians, focusing on the most at-risk population in terms of direct care management, we would have been out of business in 2001 if we started that as the model in 1997. Are you solving the most important problem? And for us, the most important problem is always going to be, whether it’s in healthcare or otherwise, which area do you have the highest amount of leverage to drive value for the consumer and the physician?

And so if we break out that 80/20, the 80% of cost driven by 20% of the members, and we look at it as 3% to 5% is driving 20% to 30% of that cost structure, where it is harder for the typical primary care physician to manage that patient, in their current practice setup. But it starts first with developing the direct relationship with primary care. And so the big problem we’ve identified in a lot of the markets we operate in is time has gotten constrained in terms of office visit time. There’s lots of information you can look at whether has the office visit time really decreased or not. What has changed is the amount of interaction time that’s directly related to patient care.

But there’s been a lot of advances as it relates to actual evidence-based guidance on managing not just chronic conditions globally, but when you think about what is a specific evidence-based plan for this particular diabetic who has this comorbidity, who’s on this set of medications. There’s many examples of that across whether it’s one chronic condition or multiple comorbidities. The value we bring, we think, is one, how do we remove primary care physicians from the notion of billing insurers for services or having to worry about what we’ll call acuity.

So we think about like full capitation practices. Our problem with that is we think it can create the wrong incentives. You have physicians that are overly focused on accurate coding, which sometimes can lead to kind of the wrong behaviors. How do we ensure physicians are paid an attractive reimbursement but not having to bill us? So for us, that’s where we introduce what we call our Clover Assistant platform, and that’s our point-of-care software where everything we’ve been compiling over the last five years is presented in a way where it’s actually actionable for the physician, purely around clinical decision-making. And it removes the paranoia from the physician’s perspective as to, am I going to get paid or not? It’s really an actual clinical guidance tool. It’s not as if they’re paid more or less based upon our satisfaction of whether they answered questions or did things that they should or shouldn’t have done.

Really, the value is, if we think about, all the data points that get captured today in healthcare, whether it’s lab results or imaging results or information of EHRs, for us, there’s a ton of metadata that we found valuable where when we surface a piece of information to the physician, what did they do with that? Did they act upon it or not, and why? That feedback back to us is highly informative, not just on helping understand how that physician makes decisions, but what do we need to improve internally in terms of information we present.

And you hear this all the time where physicians aren’t just going to go off of something that’s presented to them. They need to see the clinical evidence. And that’s where we also present that back to them. Then the other side, we’ve got the really complex population where we actually will take direct management of care over them with our own physicians. But when you have alignment with that 90, 95% of the membership and their physicians, you’re going to get a much higher matriculation rate in your own direct clinical programmes.

Martin

If I understand it correctly, you’re really strongly enabling the primary care, right, both with technology, evidence-based guidelines, making it easy on them to actually get paid as well. In addition, helping with respect to, for the chronic population, what’s done with your own intervening models, right? And then, John, how does Alignment do it? I got the model of Clover. What’s your view of how it’s done in Alignment and where it’s perhaps the same or different than the way Clover does it?

John

I would say the use of technology is something that is probably more apt to say we really ingest and clean and use data to help us in clinical decision support. And we rely on our primary care partners to care for that 80% of the population that costs that 20% of the actual claims cost. And it’s not that they can’t take care of everyone, it’s just that the realities of it are difficult for them to see that chronic patient on a reliable basis. And so that’s where we would take our clinical resources that are employed, doctors, nurses, et cetera, figure out how we deploy that with and in concert with our primary care partners. If you’re able to address and repeat the ability to care for that chronic frail membership, it’s going to start bending the cost curve.

And then I think the ability to “productize” that care delivery and that technology in creating value for the end consumer through, I would say, three things. Very specifically, you got to have very good benefits. You got to be number 1 or number 2 in the marketplace. And the price that the consumer has to pay is getting lower and lower. And there are a lot of zero-premium plans and products out there. So the ability to manage that care quality at a lower price point is going to get more and more important.

And then, as I mentioned also, what I think was historically just doctors and hospitals and ancillary services, have these supplemental benefits that now you have Part D. So Part D has been a big differentiator to consumers in their selection. But then transportation, health, fitness, dental, and vision have all been standardized supplemental services. But I think the expansion of the definition of that now includes groceries, home companionship, over-the-counter drugs. I think those are the kinds of things that are going to impact social determinants, and I think those things are going to be drivers that further distinguish one option from another.

Martin

So both models, Clover, Alignment, both seek to drive down the cost of care, right, through better engagement of primary care or clinical evidence-based guidelines, combined with data.

John

While maintaining quality.

Martin

While maintaining quality. So doing that then gives you the financial headroom, to your point, to increase benefits and differentiate. Did I get that right?

John

Yeah, the way I look at it is all of it we just talked about, primary care engagement, specialty networks, hospital networks, claims, authorize it, all of that is on the backend, right? The complexities of our healthcare system, that backend ecosystem has to be glued together properly. But from a consumer perspective, do they care about any of that backend stuff? Not really. They care they have enough access at a really good price point. Unless they want fee-for-service, they’ll stay and they’ll get full access. But for a Medicare Advantage product, either HMO or PPO, they get enough access, enough choice, but really they’re buying on price.

And I think the definition of that value equation is going to begin to shift from what historically has been network divided by cost. The network definition was how they define quality, divided by cost would equal value. And you’re going to start seeing, I think, the definition of value shifting to not only your clinical outcomes, plus your access, plus your experience. So net promoter scores, so you have to have the product accessibility, you have to have good clinical outcomes, and you have to be happy, and you divide that by your price, and your price is going to be the nature of your benefits. So the definition of value is going to be tougher.

Vivek
Broader.

John

It’s going to be broader. It’s going to be more competitive.

Martin

Right. So to that end, Julie, since nationals, for the most part, have some assets in the care delivery space, but I would argue, and I think the data helps support this, that they are spending a lot of energy trying to differentiate on benefits in a variety of ways. What are your thoughts in terms of what John just said around the expansion of the definition of value? Are large national plans or multi-regionals equipped to compete in that space?

Julie

I think that the large network-based plans are competing in that space. They have competed in that space. They were the first to enter the market. We’re seeing more than 50% of members with zero premium plans, both in HMOs and PPOs. That’s being driven by their ability to flex their scale on a couple of the areas that I mentioned in my first comments and bring in supplemental benefits. John talked a little bit about those, bringing digital solutions to the consumers, really beginning to empower the consumers.

The network-based plans, I understand that they are competing with a different type of local entity, both of we’ve heard a little bit about today. With Alignment and Clover are different. You’re locally based and very integrated with the provider community. The network plans have to have a differentiation where they’re reaching the consumer. Obviously, the provider’s always going to play a part in that, and you’ve seen them create partnerships and be local in the approaches, but they have to vary across their geographies because the healthcare still is local and they need to adjust for that in their value proposition.

Martin

If I think of many of the national players, they often, if we look at their financial picture of the world, they’re often winning on the lowest admin cost per member per month. And with that, a lot of strong hand on the provider community, either through delegation or incentive models or trying to, to Vivek’s point earlier around driving risk scores. Maybe the models will be, I think, as we talked about, under pressure in the future. I think that the question for those, I think, larger plans is they get there to a large extent on benefits and low admin. I don’t know that they get there on better medical cost historically. Would you disagree with that?

Julie

I think that the broad network plans have been very focused on stars. They’ve been very responsive to the CMS requirements around stars. I think that you saw early when stars were introduced, a lot of the larger plans evaluated their networks and looked at who were the high performers and engaged with those performers so that their members were served well. But to your point around where do they go and how do they compete long-term, it’s going to require discipline across a lot of different dimensions of the business to maintain the scale value that they have today. That is what they bring in terms of incremental benefits. They’re leveraging that low cost point that you’ve already referenced.

Martin

Right. So Vivek, and as you think about where your company is relative to the larger players in the space, it sounds like you’re, to a large extent, winning on a better medical cost, better care management, better engagement of the provider. Is that a fair statement, or do you think it’s broader than that?

Vivek
Yeah, I think it’s a fair statement. I think a bit of it also is trying to build an organization for what we think is going to be truly valued in five to 10 years versus what’s valued today. If we were trying to build something valuable for today, we wouldn’t have launched in New Jersey. If I was inside of a plant, a large incumbent plant, I would have never gotten inside. Starting an MA plan, I wouldn’t have gotten sign-off for New Jersey being the first market. It was fortunate I didn’t have to go through those hoops. So, I think if we think about …

Let’s go through each of the basic levers of MA math-wise. So, let’s take stars. So, stars, it’s not a big secret among large incumbents, you got to pick the right geography. Stars is not done on a local comparison basis. It’s done on a national comparison basis. So what that means is plans aren’t really driving the value historically in stars, it’s provider groups. Now if you look at very recent data, as, you know, as recently as, whether it’s the most recent star ratings about to come out or last year’s, you’re starting to see large incumbent plans that haven’t been able to move lives around, they’re getting huge drops on star ratings in harder markets because they haven’t been able to ride off of the other markets. And what that’s exposing is …

Martin

That’s one of the arbitrages. Yep. You know, that we mentioned.

Vivek

So is that arbitrage still going to be a business model in five years? I don’t know how much of it is or isn’t, but it’s going to be less important in five years. Even recent MedPAC recommendations were trying or making a recommendation to do stars comparisons locally. Now there are a lot of systems issues CMS has to figure out internally to be able to calculate effectively, and that will take years to get there. But again, we don’t think MA is going away anytime soon, so we’d rather build something that’s valuable in five to 10 years around that.

If we take risk adjustment, for example, companies spent huge amounts of dollars to capture paper charts, pull data out. ROI and efficiency is viewed as how do you get the data most efficiently in time for risk adjustment season. And there’s some cash flow dynamics. You want to submit it by a certain time. We look at it differently from the perspective of, do we really think in five to 10 years, if we put ourselves in the shoes of taxpayers, which we are, is that really how we want clinical data capture, where the primary driver is for revenue, not for clinical value?

If it was for clinical value, it should actually be, well, let’s capture it when the data point was produced at the point of care. So, I would bet that in five to 10 years there’s more value from a taxpayer and CMS perspective if relevant data points are captured by the plan real time and then you can validate how they’re being used as part of enhancing clinical decision making. If it’s not being used for that purpose and it’s just for revenue, how is that different than just a pass-through? And then how is that different than fee-for-service?

So then let’s talk about how it’s different than fee-for-service. Then we’ve got med-ex. Well, around medical expenses, you have utilization management. So what is utilization management in an MA plan? Let’s ensure admits are classified as observations instead of hospital admits. OK. Does the patient even know what the difference is? There’s no difference inside the hospital. It’s literally a billing classification. Or let’s ensure that a hospital is not upcoding relative to a higher severity DRG versus another. Does the patient really understand that difference? No, because it’s not affecting his or her care. It’s a chart making it to the HIM department, and there’s a 3M coding machine there that’s determining how to optimize the DRG. Is there cost savings there? Yeah, in terms of if the MA plan can look at that effectively, but there’s no actually clinical value to that.

If we think about site of service, let’s make sure a diagnostic service is being done in an outpatient facility, non-hospital owned, versus a hospital-owned outpatient imaging facility that would be billed at an HOPD rate. Is there … yes, there’s some modicum of probably clinical value if it’s done in a true outpatient setting than a hospital facility, but there’s not good data on that. Again, it’s a cost savings thing. So a lot of those are true commodity functions that no doubt will be viewed as commodity in five, six years.

So then we look at Part D. Every single plan has a PBM, whether it’s United that owns OptumRx or it’s Aetna now with CVS Caremark or Anthem with, I think it’s Ingenio, or Cigna with Express Scripts. What does a PBM do today? They’re incentivized to have high branded pricing. There’s very minimal margin made by independent pharmacies on branded.

And then on the back end, what’s happening is PBMs are taking DIR fees in the tune of like 10%, 12%. Well, what’s really happening though, that higher pricing is hitting CMS on the catastrophic dollars. So they’re getting reinsurance coverage back from the government and they’re pocketing the DIR fees. Perfectly fine to do that from a legal framework, but those are dollars that otherwise could be given back to consumers. And so, you have a class of trade issue. So, do I really believe in five to 10 years that plans will be permitted to own the PBM stack? I wouldn’t want that if I was in charge of the FTC. So, I think there are a lot of math components in terms of making a plan successful today that I wouldn’t want to build a business around for and I bet that that’s going to be the smartest way to do it for five to 10 years.

Martin

Yeah. And John, you mentioned, what Vivek just described is a lot of the backend machinery that you talked about.

John

I think he’s exactly right. I think the large players that have the share have done a brilliant job of leveraging their strengths. And they have optimized what they do within the rules defined by CMS. And so they’re pushing risk adjustment. OK. We would argue risk adjustment is an extension of care delivery and ensuring that the care plan for an individual that’s got a higher level of acuity is in place. Some agree with that. Others would say that it’s a form of revenue cycle management, which I think is Vivek’s point. You shouldn’t think that way. It’s really a care delivery-centric way of looking at that. Stars, it’s just … it’s quality of clinical outcomes and member experience.

As he mentioned, you’ve got certain XYZ Health Plan of Wisconsin doing business in North Carolina because they’re arbitraging starts. Is that the right spirit of what was intended? No, not really. The whole point is to have good quality outcomes in a local geography. Now, I think that loophole has been closed. But those are two large drivers of revenue, unit pricing, right? And then in terms of the scale, I agree with Julie and you, the unit costs that the big guys get from providers is typically lower than what we would get.

But is that the spirit of Medicare Advantage, to pound providers getting lower than Medicare? Probably not. And then the admin fees, ironically, are a huge advantage for them in the way in which they calculate their bids. So, I think what the big guys are doing is absolutely brilliant, but I think that the business models are going to evolve where the value to the end consumer is going to ultimately win out. And I think that’s directionally what Vivek is saying, is at the end of the day, a lot of these vagaries of the backend and the calculus this are going to sort itself out at the consumer level. There is more opportunity in the system as constructed where there’s more value to be received by that consumer. And Dr. Mark McClellan is on our board, who is the author of the Medicare Modernization Act of 2006. I was talking to him and you compare what MA was in 2006 …

Martin

Yeah, pre-risk adjustment, pre-Part D.

John

… to where it is now. Right. It’s a chassis that is really designed to improve quality and do so with more efficient and innovative care delivery. That’s the whole purpose of it. It’s a higher value. And I think to Vivek’s point, some of the folks have optimized the calculus in a very smart business way, but it doesn’t mean the value is realized by the consumer.

Martin

To that point, Julie, what’s your take on that? Is the value being realized or is there more to go?

Julie

Well, I would say that value is being realized. When you look at what the cost of the health plans are today, the big larger network plans have really reduced that cost for our consumers.

Martin

But if they hadn’t used their scale, we wouldn’t be …

Julie

We wouldn’t be at a zero premium PPO today without that. And I think that when I look at the trends in the larger plans, they’re using their data and their analytics to drive information to the providers. They’re doing a lot of interesting things with natural language processing and AI so that they can continue to enhance some of that, the data that they’re bringing. They’re also focused on value-based care. And then third, consumerism and what can they do to empower the consumer to engage the consumer.

This is a hard market to engage your consumer in. We know that. The consumer doesn’t want to necessarily talk to their health plan. They do want to have a relationship more aligned with their providers. But stars in the structure of CAHPS has actually forced health plans to want to participate in that and need to have that good relationship and an engagement model that’s different. So I think as we look about at innovation moving forward in the large-scale plans, we’re going to see continuation of the integration of services, which is really what they’re intending to do to drive quality.

Martin

Well, one final question in the spirit of the summit and our discussions this week on building for impact across the industry, and this is a little spin on what I had said at the end of our session today. If you had no limitations on resources, money, or talent, and the sky was the limit, what would you fix in Medicare Advantage? What would you do differently?

Vivek

Selfishly, I’m very happy that there are very high barriers to entry. But in terms of what is Medicare Advantage about, it’s about the consumer. So realistically, taking my Clover hat off, dramatically lowering the barrier to entry is really important. If we think about insurance, there’s such great actuarial data that CMS has. The amount of capital required to get into the space prevents a lot of really intelligent people who probably have access to capital, who could build very logical enablement solutions, are blocked out. If we think about a ton of the operational functions, everything from claims processing I was describing, utilization management, not that they’re simple, but they are commodity-like in terms of the value. If we think about Part D and PBMs, it shouldn’t be a way to capture additional margin. A lot of those constructs could be easily rented from CMS. Easily is a loaded word. It would take a long time to build a lot of that effectively and intelligently that can compete against incumbent constructs, but it would bring a whole host of new participants into the space, create a ton of competition. You already have, to some extent, rate parity. That would be the best thing for consumers. John, what do you think?

John

In a nutshell, I’d say we think healthcare in the country is actually pretty good. We think sick care is really bad. And I think that the more care coordination, the more transparency, that we have with both providers and consumers is just going to improve care outcomes for everybody.

Julie

Similar to John, I would really like to see the Medicare program support well care and help preventively with benefits. And I think we’re moving in the right direction with the supplemental benefits and focusing more on public health types of issues. And opening up the opportunities to improve the health of people before they need the insurance.

Martin

I want to thank the three of you for taking the time to talk to us today. It’s been great chatting with everyone here in Chicago at the Oliver Wyman Health Innovation Summit. Thanks so much for joining us.

John

Thanks, Martin. Thank you. Thank you.

Narrator

The Oliver Wyman Health Podcast is brought to you by the global management consulting firm Oliver Wyman. For more insights on the business of transforming healthcare, visit our online publication, health.oliverwyman.com.

This transcript has been edited for clarity.
 

    In this episode of the Oliver Wyman Health Podcast, Martin Graf, a partner in Oliver Wyman’s Health and Life Sciences division, chats with John Kao, Alignment Healthcare’s Chief Executive Officer, Julie Smith, former president of Government Business, Anthem and Blue Cross Blue Shield of Michigan, and a senior adviser at Oliver Wyman, and Vivek Garipalli, Clover Health’s co-founder and Chief Executive Officer, to discuss stars, risk adjustment, and more.

    “There are different perspectives on how to win in Medicare Advantage. Our perspectives, the collective group of us with differing views, is a good way to start to frame what that next generation of strategy might look like,” Martin says. Below, hear our full podcast conversation, recorded from this year’s Oliver Wyman Health Innovation Summit in Chicago.

    Key talking points:

    • A small share of Medicare Advantage members drives the majority of costs, and success depends on identifying and managing that population effectively.
    • Data — how it’s captured, used, and applied in real time — is a core differentiator across approaches.
    • Much of today’s Medicare Advantage success comes from backend levers like risk adjustment, stars, and pricing mechanics. There is growing tension between optimizing those levers and delivering real clinical value to patients.
    • The future of Medicare Advantage will be defined by better outcomes, consumer experience, and more meaningful care delivery rather than financial engineering.

    This episode is part of our Oliver Wyman Health podcast series, which includes conversations with leaders pioneering healthcare market transformation. 

    Subscribe for more on: Apple Podcasts | Spotify

    This episode was first broadcast in December 2019.

    Vivek Garipalli

    I think that’s what MA is about. Really our whole thing starts with making sure you capture the data, and then making sure you apply the data and do the right thing, taking care of the people who need care the most, and then productizing those care models into products that can serve more and more seniors.

    Narrator

    By 2030, 20% of the US population, will be over 65. On this episode, three healthcare leaders explain their organizations’ different approaches to the Medicare Advantage advantage, to support the biggest, most lucrative consumer population of the future.

    The Oliver Wyman Health Podcast is brought to you by the global management consulting firm Oliver Wyman. For more insights on the business of transforming healthcare, visit our online publication, health.oliverwyman.com.

    Martin Graf

    Hello and welcome to the Oliver Wyman Podcast. I’m Martin Graf, partner in the Health and Life Sciences practice here at Oliver Wyman. We’re here at the seventh annual Oliver Wyman Health Innovation Summit from our mobile podcast studio at the Chicago Loews Hotel, where Oliver Wyman brings 400 senior executives from 200 companies, 38 states, and 12 healthcare ecosystems together over three days to build for impact and design the healthcare landscape. Joining me today are John Kao, Alignment Health CEO.

    John
    Hey, Martin, thanks for having me.

    Martin
    And Vivek Garipalli, Clover Health’s co-founder and CEO. Hey, Martin.

    Vivek Garipalli

    Thank you.

    Martin

    And Julie Smith, the past president of Anthem and Blue Cross Blue Shield of Michigan’s Medicare business.

    Julie Smith

    Looking forward to our session, Martin.

    Martin

    We’re going to talk about some key takeaways from the Medicare Advantage advantage winning in the next generation of MA. Now, what I’d like to do is talk a little bit about our discussion earlier today, and I think one of the things that was interesting as feedback has been that not all the views that we shared were shared by everyone in the industry. That there are different perspectives on how to win in Medicare Advantage, and that our perspectives, the collective group of us with differing views, is a good way to start to frame what that next generation of strategy might look like, because I think there are different experiences and different drivers of success.

    So what I’d like to do is start by talking a little bit about what some of those different operating models in the different business models actually look like. So with that, maybe John, maybe you can talk a little bit about Alignment and the way Alignment has continued to grow and win in the market.

    John

    Sure, Martin. Thank you very much for having me. The Alignment healthcare model is really predicated on what we believe the Medicare Advantage construct was designed to achieve, which is to provide beneficiaries with really high quality at a very good price point. And in order for that to occur, in our opinion, it starts with what we would refer to as a virtuous cycle where we use our data and our technology to identify individuals that need care that could be categorized as having polychronic conditions and were frail. And that’s kind of the proverbial 20/80, where 20% of the members cost 80% of the spend.

    And if you do that and do the right thing and take care of those individuals that need the most care, there’s going to be an opportunity to start bending the cost curve. And if you can provide high quality of care at a very good price, you’re going to be able to grow because by law, the Medicare Advantage model requires you to invest that savings in higher degrees of coverage for beneficiaries. And so, if you got good coverage and good benefits and good quality care delivery and you’ve got good member experience ratings, you’re going to be incentivized to grow.

    John

    And I think that’s what MA is all about. And so really our whole thing starts with making sure you capture the data and then making sure you apply the data and do the right thing, take care of people that need the care the most, and then productize those care models into products that can serve more and more seniors.

    Martin

    Yeah. And along those lines, the way Alignment has gone at this, with obviously intervening in the care to a certain extent, has been an interesting difference between, say, network-based health plans. And I know, Julie, you’ve spent most of your career on the network-based side. So, what’s your thoughts on, given what John said, because I think there are a lot of facts that fit that market, right, that 80% of the members do only use 20% of the cost and vice versa. What’s your sense in terms of the network-based plans and how they address that same issue?

    Julie

    The network-based plans and the larger-scaled incumbents in this space have obviously leveraged their scale in a number of key areas. First, geography, as well as products and their networks. They are very focused on serving consumers with benefits and services that will drive outcomes that will improve the health. In terms of specifically meeting the needs of the 80% and the 20% populations, they use varying degrees of enablement capabilities, partnerships with provider organizations, as well as benefits and services to drive behavior.

    Martin

    And Vivek, I know your model is kind of unique and different than either the network-based or the Alignment model. What’s your take on how to address some of that? Expensive small component of the population that drives most of the cost.

    Vivek

    When we think about Medicare Advantage, we got into the space as Clover not at the outset of trying to be an MA plan. Right. We were trying to figure out any other way to build a business without having to be an insurance company. And one of the things we think has been missing in healthcare for a long time is where you can align the business model with … I think everyone, a lot of reason people got into healthcare is the mission of trying to improve health outcomes for consumers.

    What we found unique around MA is it’s one of the few constructs that exist in the US where you can build a consumer product, where you can look at a customer over many, many years, seven, eight, nine years, and if you actually drive improved care, better management of that patient, you’re actually going to generate real business value while improving care for that consumer. And that, I think, is something that, because those rules evolved over the last 20 years, it wasn’t as if when Medicare Advantage was started, or launched as Medicare Plus Choice back in the late 90s, that wasn’t the exact construct.

    Vivek

    So all the large incumbents that are in there today launched a long time ago, and so they built their companies and approaches around the original way of how Medicare Advantage was started. And obviously, as CMS saw how the business models were designed, they realized those business models were designed because of how the rules were set up. And they kept adjusting the rules to try to create the right incentives. And I think if you started a business 20 years ago, you would have purpose-built it for a specific reason. And you now had to kind of evolve your way to what I think is going to need to be in place to be successful.

    Going forward, because we only started our business as Clover about five years ago, under the new construct of rules. If we’re trying to build a business 20 years ago around driving improved care management, improving clinical decision-making for physicians, focusing on the most at-risk population in terms of direct care management, we would have been out of business in 2001 if we started that as the model in 1997. Are you solving the most important problem? And for us, the most important problem is always going to be, whether it’s in healthcare or otherwise, which area do you have the highest amount of leverage to drive value for the consumer and the physician?

    And so if we break out that 80/20, the 80% of cost driven by 20% of the members, and we look at it as 3% to 5% is driving 20% to 30% of that cost structure, where it is harder for the typical primary care physician to manage that patient, in their current practice setup. But it starts first with developing the direct relationship with primary care. And so the big problem we’ve identified in a lot of the markets we operate in is time has gotten constrained in terms of office visit time. There’s lots of information you can look at whether has the office visit time really decreased or not. What has changed is the amount of interaction time that’s directly related to patient care.

    But there’s been a lot of advances as it relates to actual evidence-based guidance on managing not just chronic conditions globally, but when you think about what is a specific evidence-based plan for this particular diabetic who has this comorbidity, who’s on this set of medications. There’s many examples of that across whether it’s one chronic condition or multiple comorbidities. The value we bring, we think, is one, how do we remove primary care physicians from the notion of billing insurers for services or having to worry about what we’ll call acuity.

    So we think about like full capitation practices. Our problem with that is we think it can create the wrong incentives. You have physicians that are overly focused on accurate coding, which sometimes can lead to kind of the wrong behaviors. How do we ensure physicians are paid an attractive reimbursement but not having to bill us? So for us, that’s where we introduce what we call our Clover Assistant platform, and that’s our point-of-care software where everything we’ve been compiling over the last five years is presented in a way where it’s actually actionable for the physician, purely around clinical decision-making. And it removes the paranoia from the physician’s perspective as to, am I going to get paid or not? It’s really an actual clinical guidance tool. It’s not as if they’re paid more or less based upon our satisfaction of whether they answered questions or did things that they should or shouldn’t have done.

    Really, the value is, if we think about, all the data points that get captured today in healthcare, whether it’s lab results or imaging results or information of EHRs, for us, there’s a ton of metadata that we found valuable where when we surface a piece of information to the physician, what did they do with that? Did they act upon it or not, and why? That feedback back to us is highly informative, not just on helping understand how that physician makes decisions, but what do we need to improve internally in terms of information we present.

    And you hear this all the time where physicians aren’t just going to go off of something that’s presented to them. They need to see the clinical evidence. And that’s where we also present that back to them. Then the other side, we’ve got the really complex population where we actually will take direct management of care over them with our own physicians. But when you have alignment with that 90, 95% of the membership and their physicians, you’re going to get a much higher matriculation rate in your own direct clinical programmes.

    Martin

    If I understand it correctly, you’re really strongly enabling the primary care, right, both with technology, evidence-based guidelines, making it easy on them to actually get paid as well. In addition, helping with respect to, for the chronic population, what’s done with your own intervening models, right? And then, John, how does Alignment do it? I got the model of Clover. What’s your view of how it’s done in Alignment and where it’s perhaps the same or different than the way Clover does it?

    John

    I would say the use of technology is something that is probably more apt to say we really ingest and clean and use data to help us in clinical decision support. And we rely on our primary care partners to care for that 80% of the population that costs that 20% of the actual claims cost. And it’s not that they can’t take care of everyone, it’s just that the realities of it are difficult for them to see that chronic patient on a reliable basis. And so that’s where we would take our clinical resources that are employed, doctors, nurses, et cetera, figure out how we deploy that with and in concert with our primary care partners. If you’re able to address and repeat the ability to care for that chronic frail membership, it’s going to start bending the cost curve.

    And then I think the ability to “productize” that care delivery and that technology in creating value for the end consumer through, I would say, three things. Very specifically, you got to have very good benefits. You got to be number 1 or number 2 in the marketplace. And the price that the consumer has to pay is getting lower and lower. And there are a lot of zero-premium plans and products out there. So the ability to manage that care quality at a lower price point is going to get more and more important.

    And then, as I mentioned also, what I think was historically just doctors and hospitals and ancillary services, have these supplemental benefits that now you have Part D. So Part D has been a big differentiator to consumers in their selection. But then transportation, health, fitness, dental, and vision have all been standardized supplemental services. But I think the expansion of the definition of that now includes groceries, home companionship, over-the-counter drugs. I think those are the kinds of things that are going to impact social determinants, and I think those things are going to be drivers that further distinguish one option from another.

    Martin

    So both models, Clover, Alignment, both seek to drive down the cost of care, right, through better engagement of primary care or clinical evidence-based guidelines, combined with data.

    John

    While maintaining quality.

    Martin

    While maintaining quality. So doing that then gives you the financial headroom, to your point, to increase benefits and differentiate. Did I get that right?

    John

    Yeah, the way I look at it is all of it we just talked about, primary care engagement, specialty networks, hospital networks, claims, authorize it, all of that is on the backend, right? The complexities of our healthcare system, that backend ecosystem has to be glued together properly. But from a consumer perspective, do they care about any of that backend stuff? Not really. They care they have enough access at a really good price point. Unless they want fee-for-service, they’ll stay and they’ll get full access. But for a Medicare Advantage product, either HMO or PPO, they get enough access, enough choice, but really they’re buying on price.

    And I think the definition of that value equation is going to begin to shift from what historically has been network divided by cost. The network definition was how they define quality, divided by cost would equal value. And you’re going to start seeing, I think, the definition of value shifting to not only your clinical outcomes, plus your access, plus your experience. So net promoter scores, so you have to have the product accessibility, you have to have good clinical outcomes, and you have to be happy, and you divide that by your price, and your price is going to be the nature of your benefits. So the definition of value is going to be tougher.

    Vivek
    Broader.

    John

    It’s going to be broader. It’s going to be more competitive.

    Martin

    Right. So to that end, Julie, since nationals, for the most part, have some assets in the care delivery space, but I would argue, and I think the data helps support this, that they are spending a lot of energy trying to differentiate on benefits in a variety of ways. What are your thoughts in terms of what John just said around the expansion of the definition of value? Are large national plans or multi-regionals equipped to compete in that space?

    Julie

    I think that the large network-based plans are competing in that space. They have competed in that space. They were the first to enter the market. We’re seeing more than 50% of members with zero premium plans, both in HMOs and PPOs. That’s being driven by their ability to flex their scale on a couple of the areas that I mentioned in my first comments and bring in supplemental benefits. John talked a little bit about those, bringing digital solutions to the consumers, really beginning to empower the consumers.

    The network-based plans, I understand that they are competing with a different type of local entity, both of we’ve heard a little bit about today. With Alignment and Clover are different. You’re locally based and very integrated with the provider community. The network plans have to have a differentiation where they’re reaching the consumer. Obviously, the provider’s always going to play a part in that, and you’ve seen them create partnerships and be local in the approaches, but they have to vary across their geographies because the healthcare still is local and they need to adjust for that in their value proposition.

    Martin

    If I think of many of the national players, they often, if we look at their financial picture of the world, they’re often winning on the lowest admin cost per member per month. And with that, a lot of strong hand on the provider community, either through delegation or incentive models or trying to, to Vivek’s point earlier around driving risk scores. Maybe the models will be, I think, as we talked about, under pressure in the future. I think that the question for those, I think, larger plans is they get there to a large extent on benefits and low admin. I don’t know that they get there on better medical cost historically. Would you disagree with that?

    Julie

    I think that the broad network plans have been very focused on stars. They’ve been very responsive to the CMS requirements around stars. I think that you saw early when stars were introduced, a lot of the larger plans evaluated their networks and looked at who were the high performers and engaged with those performers so that their members were served well. But to your point around where do they go and how do they compete long-term, it’s going to require discipline across a lot of different dimensions of the business to maintain the scale value that they have today. That is what they bring in terms of incremental benefits. They’re leveraging that low cost point that you’ve already referenced.

    Martin

    Right. So Vivek, and as you think about where your company is relative to the larger players in the space, it sounds like you’re, to a large extent, winning on a better medical cost, better care management, better engagement of the provider. Is that a fair statement, or do you think it’s broader than that?

    Vivek
    Yeah, I think it’s a fair statement. I think a bit of it also is trying to build an organization for what we think is going to be truly valued in five to 10 years versus what’s valued today. If we were trying to build something valuable for today, we wouldn’t have launched in New Jersey. If I was inside of a plant, a large incumbent plant, I would have never gotten inside. Starting an MA plan, I wouldn’t have gotten sign-off for New Jersey being the first market. It was fortunate I didn’t have to go through those hoops. So, I think if we think about …

    Let’s go through each of the basic levers of MA math-wise. So, let’s take stars. So, stars, it’s not a big secret among large incumbents, you got to pick the right geography. Stars is not done on a local comparison basis. It’s done on a national comparison basis. So what that means is plans aren’t really driving the value historically in stars, it’s provider groups. Now if you look at very recent data, as, you know, as recently as, whether it’s the most recent star ratings about to come out or last year’s, you’re starting to see large incumbent plans that haven’t been able to move lives around, they’re getting huge drops on star ratings in harder markets because they haven’t been able to ride off of the other markets. And what that’s exposing is …

    Martin

    That’s one of the arbitrages. Yep. You know, that we mentioned.

    Vivek

    So is that arbitrage still going to be a business model in five years? I don’t know how much of it is or isn’t, but it’s going to be less important in five years. Even recent MedPAC recommendations were trying or making a recommendation to do stars comparisons locally. Now there are a lot of systems issues CMS has to figure out internally to be able to calculate effectively, and that will take years to get there. But again, we don’t think MA is going away anytime soon, so we’d rather build something that’s valuable in five to 10 years around that.

    If we take risk adjustment, for example, companies spent huge amounts of dollars to capture paper charts, pull data out. ROI and efficiency is viewed as how do you get the data most efficiently in time for risk adjustment season. And there’s some cash flow dynamics. You want to submit it by a certain time. We look at it differently from the perspective of, do we really think in five to 10 years, if we put ourselves in the shoes of taxpayers, which we are, is that really how we want clinical data capture, where the primary driver is for revenue, not for clinical value?

    If it was for clinical value, it should actually be, well, let’s capture it when the data point was produced at the point of care. So, I would bet that in five to 10 years there’s more value from a taxpayer and CMS perspective if relevant data points are captured by the plan real time and then you can validate how they’re being used as part of enhancing clinical decision making. If it’s not being used for that purpose and it’s just for revenue, how is that different than just a pass-through? And then how is that different than fee-for-service?

    So then let’s talk about how it’s different than fee-for-service. Then we’ve got med-ex. Well, around medical expenses, you have utilization management. So what is utilization management in an MA plan? Let’s ensure admits are classified as observations instead of hospital admits. OK. Does the patient even know what the difference is? There’s no difference inside the hospital. It’s literally a billing classification. Or let’s ensure that a hospital is not upcoding relative to a higher severity DRG versus another. Does the patient really understand that difference? No, because it’s not affecting his or her care. It’s a chart making it to the HIM department, and there’s a 3M coding machine there that’s determining how to optimize the DRG. Is there cost savings there? Yeah, in terms of if the MA plan can look at that effectively, but there’s no actually clinical value to that.

    If we think about site of service, let’s make sure a diagnostic service is being done in an outpatient facility, non-hospital owned, versus a hospital-owned outpatient imaging facility that would be billed at an HOPD rate. Is there … yes, there’s some modicum of probably clinical value if it’s done in a true outpatient setting than a hospital facility, but there’s not good data on that. Again, it’s a cost savings thing. So a lot of those are true commodity functions that no doubt will be viewed as commodity in five, six years.

    So then we look at Part D. Every single plan has a PBM, whether it’s United that owns OptumRx or it’s Aetna now with CVS Caremark or Anthem with, I think it’s Ingenio, or Cigna with Express Scripts. What does a PBM do today? They’re incentivized to have high branded pricing. There’s very minimal margin made by independent pharmacies on branded.

    And then on the back end, what’s happening is PBMs are taking DIR fees in the tune of like 10%, 12%. Well, what’s really happening though, that higher pricing is hitting CMS on the catastrophic dollars. So they’re getting reinsurance coverage back from the government and they’re pocketing the DIR fees. Perfectly fine to do that from a legal framework, but those are dollars that otherwise could be given back to consumers. And so, you have a class of trade issue. So, do I really believe in five to 10 years that plans will be permitted to own the PBM stack? I wouldn’t want that if I was in charge of the FTC. So, I think there are a lot of math components in terms of making a plan successful today that I wouldn’t want to build a business around for and I bet that that’s going to be the smartest way to do it for five to 10 years.

    Martin

    Yeah. And John, you mentioned, what Vivek just described is a lot of the backend machinery that you talked about.

    John

    I think he’s exactly right. I think the large players that have the share have done a brilliant job of leveraging their strengths. And they have optimized what they do within the rules defined by CMS. And so they’re pushing risk adjustment. OK. We would argue risk adjustment is an extension of care delivery and ensuring that the care plan for an individual that’s got a higher level of acuity is in place. Some agree with that. Others would say that it’s a form of revenue cycle management, which I think is Vivek’s point. You shouldn’t think that way. It’s really a care delivery-centric way of looking at that. Stars, it’s just … it’s quality of clinical outcomes and member experience.

    As he mentioned, you’ve got certain XYZ Health Plan of Wisconsin doing business in North Carolina because they’re arbitraging starts. Is that the right spirit of what was intended? No, not really. The whole point is to have good quality outcomes in a local geography. Now, I think that loophole has been closed. But those are two large drivers of revenue, unit pricing, right? And then in terms of the scale, I agree with Julie and you, the unit costs that the big guys get from providers is typically lower than what we would get.

    But is that the spirit of Medicare Advantage, to pound providers getting lower than Medicare? Probably not. And then the admin fees, ironically, are a huge advantage for them in the way in which they calculate their bids. So, I think what the big guys are doing is absolutely brilliant, but I think that the business models are going to evolve where the value to the end consumer is going to ultimately win out. And I think that’s directionally what Vivek is saying, is at the end of the day, a lot of these vagaries of the backend and the calculus this are going to sort itself out at the consumer level. There is more opportunity in the system as constructed where there’s more value to be received by that consumer. And Dr. Mark McClellan is on our board, who is the author of the Medicare Modernization Act of 2006. I was talking to him and you compare what MA was in 2006 …

    Martin

    Yeah, pre-risk adjustment, pre-Part D.

    John

    … to where it is now. Right. It’s a chassis that is really designed to improve quality and do so with more efficient and innovative care delivery. That’s the whole purpose of it. It’s a higher value. And I think to Vivek’s point, some of the folks have optimized the calculus in a very smart business way, but it doesn’t mean the value is realized by the consumer.

    Martin

    To that point, Julie, what’s your take on that? Is the value being realized or is there more to go?

    Julie

    Well, I would say that value is being realized. When you look at what the cost of the health plans are today, the big larger network plans have really reduced that cost for our consumers.

    Martin

    But if they hadn’t used their scale, we wouldn’t be …

    Julie

    We wouldn’t be at a zero premium PPO today without that. And I think that when I look at the trends in the larger plans, they’re using their data and their analytics to drive information to the providers. They’re doing a lot of interesting things with natural language processing and AI so that they can continue to enhance some of that, the data that they’re bringing. They’re also focused on value-based care. And then third, consumerism and what can they do to empower the consumer to engage the consumer.

    This is a hard market to engage your consumer in. We know that. The consumer doesn’t want to necessarily talk to their health plan. They do want to have a relationship more aligned with their providers. But stars in the structure of CAHPS has actually forced health plans to want to participate in that and need to have that good relationship and an engagement model that’s different. So I think as we look about at innovation moving forward in the large-scale plans, we’re going to see continuation of the integration of services, which is really what they’re intending to do to drive quality.

    Martin

    Well, one final question in the spirit of the summit and our discussions this week on building for impact across the industry, and this is a little spin on what I had said at the end of our session today. If you had no limitations on resources, money, or talent, and the sky was the limit, what would you fix in Medicare Advantage? What would you do differently?

    Vivek

    Selfishly, I’m very happy that there are very high barriers to entry. But in terms of what is Medicare Advantage about, it’s about the consumer. So realistically, taking my Clover hat off, dramatically lowering the barrier to entry is really important. If we think about insurance, there’s such great actuarial data that CMS has. The amount of capital required to get into the space prevents a lot of really intelligent people who probably have access to capital, who could build very logical enablement solutions, are blocked out. If we think about a ton of the operational functions, everything from claims processing I was describing, utilization management, not that they’re simple, but they are commodity-like in terms of the value. If we think about Part D and PBMs, it shouldn’t be a way to capture additional margin. A lot of those constructs could be easily rented from CMS. Easily is a loaded word. It would take a long time to build a lot of that effectively and intelligently that can compete against incumbent constructs, but it would bring a whole host of new participants into the space, create a ton of competition. You already have, to some extent, rate parity. That would be the best thing for consumers. John, what do you think?

    John

    In a nutshell, I’d say we think healthcare in the country is actually pretty good. We think sick care is really bad. And I think that the more care coordination, the more transparency, that we have with both providers and consumers is just going to improve care outcomes for everybody.

    Julie

    Similar to John, I would really like to see the Medicare program support well care and help preventively with benefits. And I think we’re moving in the right direction with the supplemental benefits and focusing more on public health types of issues. And opening up the opportunities to improve the health of people before they need the insurance.

    Martin

    I want to thank the three of you for taking the time to talk to us today. It’s been great chatting with everyone here in Chicago at the Oliver Wyman Health Innovation Summit. Thanks so much for joining us.

    John

    Thanks, Martin. Thank you. Thank you.

    Narrator

    The Oliver Wyman Health Podcast is brought to you by the global management consulting firm Oliver Wyman. For more insights on the business of transforming healthcare, visit our online publication, health.oliverwyman.com.

    This transcript has been edited for clarity.
     

    In this episode of the Oliver Wyman Health Podcast, Martin Graf, a partner in Oliver Wyman’s Health and Life Sciences division, chats with John Kao, Alignment Healthcare’s Chief Executive Officer, Julie Smith, former president of Government Business, Anthem and Blue Cross Blue Shield of Michigan, and a senior adviser at Oliver Wyman, and Vivek Garipalli, Clover Health’s co-founder and Chief Executive Officer, to discuss stars, risk adjustment, and more.

    “There are different perspectives on how to win in Medicare Advantage. Our perspectives, the collective group of us with differing views, is a good way to start to frame what that next generation of strategy might look like,” Martin says. Below, hear our full podcast conversation, recorded from this year’s Oliver Wyman Health Innovation Summit in Chicago.

    Key talking points:

    • A small share of Medicare Advantage members drives the majority of costs, and success depends on identifying and managing that population effectively.
    • Data — how it’s captured, used, and applied in real time — is a core differentiator across approaches.
    • Much of today’s Medicare Advantage success comes from backend levers like risk adjustment, stars, and pricing mechanics. There is growing tension between optimizing those levers and delivering real clinical value to patients.
    • The future of Medicare Advantage will be defined by better outcomes, consumer experience, and more meaningful care delivery rather than financial engineering.

    This episode is part of our Oliver Wyman Health podcast series, which includes conversations with leaders pioneering healthcare market transformation. 

    Subscribe for more on: Apple Podcasts | Spotify

    This episode was first broadcast in December 2019.

    Vivek Garipalli

    I think that’s what MA is about. Really our whole thing starts with making sure you capture the data, and then making sure you apply the data and do the right thing, taking care of the people who need care the most, and then productizing those care models into products that can serve more and more seniors.

    Narrator

    By 2030, 20% of the US population, will be over 65. On this episode, three healthcare leaders explain their organizations’ different approaches to the Medicare Advantage advantage, to support the biggest, most lucrative consumer population of the future.

    The Oliver Wyman Health Podcast is brought to you by the global management consulting firm Oliver Wyman. For more insights on the business of transforming healthcare, visit our online publication, health.oliverwyman.com.

    Martin Graf

    Hello and welcome to the Oliver Wyman Podcast. I’m Martin Graf, partner in the Health and Life Sciences practice here at Oliver Wyman. We’re here at the seventh annual Oliver Wyman Health Innovation Summit from our mobile podcast studio at the Chicago Loews Hotel, where Oliver Wyman brings 400 senior executives from 200 companies, 38 states, and 12 healthcare ecosystems together over three days to build for impact and design the healthcare landscape. Joining me today are John Kao, Alignment Health CEO.

    John
    Hey, Martin, thanks for having me.

    Martin
    And Vivek Garipalli, Clover Health’s co-founder and CEO. Hey, Martin.

    Vivek Garipalli

    Thank you.

    Martin

    And Julie Smith, the past president of Anthem and Blue Cross Blue Shield of Michigan’s Medicare business.

    Julie Smith

    Looking forward to our session, Martin.

    Martin

    We’re going to talk about some key takeaways from the Medicare Advantage advantage winning in the next generation of MA. Now, what I’d like to do is talk a little bit about our discussion earlier today, and I think one of the things that was interesting as feedback has been that not all the views that we shared were shared by everyone in the industry. That there are different perspectives on how to win in Medicare Advantage, and that our perspectives, the collective group of us with differing views, is a good way to start to frame what that next generation of strategy might look like, because I think there are different experiences and different drivers of success.

    So what I’d like to do is start by talking a little bit about what some of those different operating models in the different business models actually look like. So with that, maybe John, maybe you can talk a little bit about Alignment and the way Alignment has continued to grow and win in the market.

    John

    Sure, Martin. Thank you very much for having me. The Alignment healthcare model is really predicated on what we believe the Medicare Advantage construct was designed to achieve, which is to provide beneficiaries with really high quality at a very good price point. And in order for that to occur, in our opinion, it starts with what we would refer to as a virtuous cycle where we use our data and our technology to identify individuals that need care that could be categorized as having polychronic conditions and were frail. And that’s kind of the proverbial 20/80, where 20% of the members cost 80% of the spend.

    And if you do that and do the right thing and take care of those individuals that need the most care, there’s going to be an opportunity to start bending the cost curve. And if you can provide high quality of care at a very good price, you’re going to be able to grow because by law, the Medicare Advantage model requires you to invest that savings in higher degrees of coverage for beneficiaries. And so, if you got good coverage and good benefits and good quality care delivery and you’ve got good member experience ratings, you’re going to be incentivized to grow.

    John

    And I think that’s what MA is all about. And so really our whole thing starts with making sure you capture the data and then making sure you apply the data and do the right thing, take care of people that need the care the most, and then productize those care models into products that can serve more and more seniors.

    Martin

    Yeah. And along those lines, the way Alignment has gone at this, with obviously intervening in the care to a certain extent, has been an interesting difference between, say, network-based health plans. And I know, Julie, you’ve spent most of your career on the network-based side. So, what’s your thoughts on, given what John said, because I think there are a lot of facts that fit that market, right, that 80% of the members do only use 20% of the cost and vice versa. What’s your sense in terms of the network-based plans and how they address that same issue?

    Julie

    The network-based plans and the larger-scaled incumbents in this space have obviously leveraged their scale in a number of key areas. First, geography, as well as products and their networks. They are very focused on serving consumers with benefits and services that will drive outcomes that will improve the health. In terms of specifically meeting the needs of the 80% and the 20% populations, they use varying degrees of enablement capabilities, partnerships with provider organizations, as well as benefits and services to drive behavior.

    Martin

    And Vivek, I know your model is kind of unique and different than either the network-based or the Alignment model. What’s your take on how to address some of that? Expensive small component of the population that drives most of the cost.

    Vivek

    When we think about Medicare Advantage, we got into the space as Clover not at the outset of trying to be an MA plan. Right. We were trying to figure out any other way to build a business without having to be an insurance company. And one of the things we think has been missing in healthcare for a long time is where you can align the business model with … I think everyone, a lot of reason people got into healthcare is the mission of trying to improve health outcomes for consumers.

    What we found unique around MA is it’s one of the few constructs that exist in the US where you can build a consumer product, where you can look at a customer over many, many years, seven, eight, nine years, and if you actually drive improved care, better management of that patient, you’re actually going to generate real business value while improving care for that consumer. And that, I think, is something that, because those rules evolved over the last 20 years, it wasn’t as if when Medicare Advantage was started, or launched as Medicare Plus Choice back in the late 90s, that wasn’t the exact construct.

    Vivek

    So all the large incumbents that are in there today launched a long time ago, and so they built their companies and approaches around the original way of how Medicare Advantage was started. And obviously, as CMS saw how the business models were designed, they realized those business models were designed because of how the rules were set up. And they kept adjusting the rules to try to create the right incentives. And I think if you started a business 20 years ago, you would have purpose-built it for a specific reason. And you now had to kind of evolve your way to what I think is going to need to be in place to be successful.

    Going forward, because we only started our business as Clover about five years ago, under the new construct of rules. If we’re trying to build a business 20 years ago around driving improved care management, improving clinical decision-making for physicians, focusing on the most at-risk population in terms of direct care management, we would have been out of business in 2001 if we started that as the model in 1997. Are you solving the most important problem? And for us, the most important problem is always going to be, whether it’s in healthcare or otherwise, which area do you have the highest amount of leverage to drive value for the consumer and the physician?

    And so if we break out that 80/20, the 80% of cost driven by 20% of the members, and we look at it as 3% to 5% is driving 20% to 30% of that cost structure, where it is harder for the typical primary care physician to manage that patient, in their current practice setup. But it starts first with developing the direct relationship with primary care. And so the big problem we’ve identified in a lot of the markets we operate in is time has gotten constrained in terms of office visit time. There’s lots of information you can look at whether has the office visit time really decreased or not. What has changed is the amount of interaction time that’s directly related to patient care.

    But there’s been a lot of advances as it relates to actual evidence-based guidance on managing not just chronic conditions globally, but when you think about what is a specific evidence-based plan for this particular diabetic who has this comorbidity, who’s on this set of medications. There’s many examples of that across whether it’s one chronic condition or multiple comorbidities. The value we bring, we think, is one, how do we remove primary care physicians from the notion of billing insurers for services or having to worry about what we’ll call acuity.

    So we think about like full capitation practices. Our problem with that is we think it can create the wrong incentives. You have physicians that are overly focused on accurate coding, which sometimes can lead to kind of the wrong behaviors. How do we ensure physicians are paid an attractive reimbursement but not having to bill us? So for us, that’s where we introduce what we call our Clover Assistant platform, and that’s our point-of-care software where everything we’ve been compiling over the last five years is presented in a way where it’s actually actionable for the physician, purely around clinical decision-making. And it removes the paranoia from the physician’s perspective as to, am I going to get paid or not? It’s really an actual clinical guidance tool. It’s not as if they’re paid more or less based upon our satisfaction of whether they answered questions or did things that they should or shouldn’t have done.

    Really, the value is, if we think about, all the data points that get captured today in healthcare, whether it’s lab results or imaging results or information of EHRs, for us, there’s a ton of metadata that we found valuable where when we surface a piece of information to the physician, what did they do with that? Did they act upon it or not, and why? That feedback back to us is highly informative, not just on helping understand how that physician makes decisions, but what do we need to improve internally in terms of information we present.

    And you hear this all the time where physicians aren’t just going to go off of something that’s presented to them. They need to see the clinical evidence. And that’s where we also present that back to them. Then the other side, we’ve got the really complex population where we actually will take direct management of care over them with our own physicians. But when you have alignment with that 90, 95% of the membership and their physicians, you’re going to get a much higher matriculation rate in your own direct clinical programmes.

    Martin

    If I understand it correctly, you’re really strongly enabling the primary care, right, both with technology, evidence-based guidelines, making it easy on them to actually get paid as well. In addition, helping with respect to, for the chronic population, what’s done with your own intervening models, right? And then, John, how does Alignment do it? I got the model of Clover. What’s your view of how it’s done in Alignment and where it’s perhaps the same or different than the way Clover does it?

    John

    I would say the use of technology is something that is probably more apt to say we really ingest and clean and use data to help us in clinical decision support. And we rely on our primary care partners to care for that 80% of the population that costs that 20% of the actual claims cost. And it’s not that they can’t take care of everyone, it’s just that the realities of it are difficult for them to see that chronic patient on a reliable basis. And so that’s where we would take our clinical resources that are employed, doctors, nurses, et cetera, figure out how we deploy that with and in concert with our primary care partners. If you’re able to address and repeat the ability to care for that chronic frail membership, it’s going to start bending the cost curve.

    And then I think the ability to “productize” that care delivery and that technology in creating value for the end consumer through, I would say, three things. Very specifically, you got to have very good benefits. You got to be number 1 or number 2 in the marketplace. And the price that the consumer has to pay is getting lower and lower. And there are a lot of zero-premium plans and products out there. So the ability to manage that care quality at a lower price point is going to get more and more important.

    And then, as I mentioned also, what I think was historically just doctors and hospitals and ancillary services, have these supplemental benefits that now you have Part D. So Part D has been a big differentiator to consumers in their selection. But then transportation, health, fitness, dental, and vision have all been standardized supplemental services. But I think the expansion of the definition of that now includes groceries, home companionship, over-the-counter drugs. I think those are the kinds of things that are going to impact social determinants, and I think those things are going to be drivers that further distinguish one option from another.

    Martin

    So both models, Clover, Alignment, both seek to drive down the cost of care, right, through better engagement of primary care or clinical evidence-based guidelines, combined with data.

    John

    While maintaining quality.

    Martin

    While maintaining quality. So doing that then gives you the financial headroom, to your point, to increase benefits and differentiate. Did I get that right?

    John

    Yeah, the way I look at it is all of it we just talked about, primary care engagement, specialty networks, hospital networks, claims, authorize it, all of that is on the backend, right? The complexities of our healthcare system, that backend ecosystem has to be glued together properly. But from a consumer perspective, do they care about any of that backend stuff? Not really. They care they have enough access at a really good price point. Unless they want fee-for-service, they’ll stay and they’ll get full access. But for a Medicare Advantage product, either HMO or PPO, they get enough access, enough choice, but really they’re buying on price.

    And I think the definition of that value equation is going to begin to shift from what historically has been network divided by cost. The network definition was how they define quality, divided by cost would equal value. And you’re going to start seeing, I think, the definition of value shifting to not only your clinical outcomes, plus your access, plus your experience. So net promoter scores, so you have to have the product accessibility, you have to have good clinical outcomes, and you have to be happy, and you divide that by your price, and your price is going to be the nature of your benefits. So the definition of value is going to be tougher.

    Vivek
    Broader.

    John

    It’s going to be broader. It’s going to be more competitive.

    Martin

    Right. So to that end, Julie, since nationals, for the most part, have some assets in the care delivery space, but I would argue, and I think the data helps support this, that they are spending a lot of energy trying to differentiate on benefits in a variety of ways. What are your thoughts in terms of what John just said around the expansion of the definition of value? Are large national plans or multi-regionals equipped to compete in that space?

    Julie

    I think that the large network-based plans are competing in that space. They have competed in that space. They were the first to enter the market. We’re seeing more than 50% of members with zero premium plans, both in HMOs and PPOs. That’s being driven by their ability to flex their scale on a couple of the areas that I mentioned in my first comments and bring in supplemental benefits. John talked a little bit about those, bringing digital solutions to the consumers, really beginning to empower the consumers.

    The network-based plans, I understand that they are competing with a different type of local entity, both of we’ve heard a little bit about today. With Alignment and Clover are different. You’re locally based and very integrated with the provider community. The network plans have to have a differentiation where they’re reaching the consumer. Obviously, the provider’s always going to play a part in that, and you’ve seen them create partnerships and be local in the approaches, but they have to vary across their geographies because the healthcare still is local and they need to adjust for that in their value proposition.

    Martin

    If I think of many of the national players, they often, if we look at their financial picture of the world, they’re often winning on the lowest admin cost per member per month. And with that, a lot of strong hand on the provider community, either through delegation or incentive models or trying to, to Vivek’s point earlier around driving risk scores. Maybe the models will be, I think, as we talked about, under pressure in the future. I think that the question for those, I think, larger plans is they get there to a large extent on benefits and low admin. I don’t know that they get there on better medical cost historically. Would you disagree with that?

    Julie

    I think that the broad network plans have been very focused on stars. They’ve been very responsive to the CMS requirements around stars. I think that you saw early when stars were introduced, a lot of the larger plans evaluated their networks and looked at who were the high performers and engaged with those performers so that their members were served well. But to your point around where do they go and how do they compete long-term, it’s going to require discipline across a lot of different dimensions of the business to maintain the scale value that they have today. That is what they bring in terms of incremental benefits. They’re leveraging that low cost point that you’ve already referenced.

    Martin

    Right. So Vivek, and as you think about where your company is relative to the larger players in the space, it sounds like you’re, to a large extent, winning on a better medical cost, better care management, better engagement of the provider. Is that a fair statement, or do you think it’s broader than that?

    Vivek
    Yeah, I think it’s a fair statement. I think a bit of it also is trying to build an organization for what we think is going to be truly valued in five to 10 years versus what’s valued today. If we were trying to build something valuable for today, we wouldn’t have launched in New Jersey. If I was inside of a plant, a large incumbent plant, I would have never gotten inside. Starting an MA plan, I wouldn’t have gotten sign-off for New Jersey being the first market. It was fortunate I didn’t have to go through those hoops. So, I think if we think about …

    Let’s go through each of the basic levers of MA math-wise. So, let’s take stars. So, stars, it’s not a big secret among large incumbents, you got to pick the right geography. Stars is not done on a local comparison basis. It’s done on a national comparison basis. So what that means is plans aren’t really driving the value historically in stars, it’s provider groups. Now if you look at very recent data, as, you know, as recently as, whether it’s the most recent star ratings about to come out or last year’s, you’re starting to see large incumbent plans that haven’t been able to move lives around, they’re getting huge drops on star ratings in harder markets because they haven’t been able to ride off of the other markets. And what that’s exposing is …

    Martin

    That’s one of the arbitrages. Yep. You know, that we mentioned.

    Vivek

    So is that arbitrage still going to be a business model in five years? I don’t know how much of it is or isn’t, but it’s going to be less important in five years. Even recent MedPAC recommendations were trying or making a recommendation to do stars comparisons locally. Now there are a lot of systems issues CMS has to figure out internally to be able to calculate effectively, and that will take years to get there. But again, we don’t think MA is going away anytime soon, so we’d rather build something that’s valuable in five to 10 years around that.

    If we take risk adjustment, for example, companies spent huge amounts of dollars to capture paper charts, pull data out. ROI and efficiency is viewed as how do you get the data most efficiently in time for risk adjustment season. And there’s some cash flow dynamics. You want to submit it by a certain time. We look at it differently from the perspective of, do we really think in five to 10 years, if we put ourselves in the shoes of taxpayers, which we are, is that really how we want clinical data capture, where the primary driver is for revenue, not for clinical value?

    If it was for clinical value, it should actually be, well, let’s capture it when the data point was produced at the point of care. So, I would bet that in five to 10 years there’s more value from a taxpayer and CMS perspective if relevant data points are captured by the plan real time and then you can validate how they’re being used as part of enhancing clinical decision making. If it’s not being used for that purpose and it’s just for revenue, how is that different than just a pass-through? And then how is that different than fee-for-service?

    So then let’s talk about how it’s different than fee-for-service. Then we’ve got med-ex. Well, around medical expenses, you have utilization management. So what is utilization management in an MA plan? Let’s ensure admits are classified as observations instead of hospital admits. OK. Does the patient even know what the difference is? There’s no difference inside the hospital. It’s literally a billing classification. Or let’s ensure that a hospital is not upcoding relative to a higher severity DRG versus another. Does the patient really understand that difference? No, because it’s not affecting his or her care. It’s a chart making it to the HIM department, and there’s a 3M coding machine there that’s determining how to optimize the DRG. Is there cost savings there? Yeah, in terms of if the MA plan can look at that effectively, but there’s no actually clinical value to that.

    If we think about site of service, let’s make sure a diagnostic service is being done in an outpatient facility, non-hospital owned, versus a hospital-owned outpatient imaging facility that would be billed at an HOPD rate. Is there … yes, there’s some modicum of probably clinical value if it’s done in a true outpatient setting than a hospital facility, but there’s not good data on that. Again, it’s a cost savings thing. So a lot of those are true commodity functions that no doubt will be viewed as commodity in five, six years.

    So then we look at Part D. Every single plan has a PBM, whether it’s United that owns OptumRx or it’s Aetna now with CVS Caremark or Anthem with, I think it’s Ingenio, or Cigna with Express Scripts. What does a PBM do today? They’re incentivized to have high branded pricing. There’s very minimal margin made by independent pharmacies on branded.

    And then on the back end, what’s happening is PBMs are taking DIR fees in the tune of like 10%, 12%. Well, what’s really happening though, that higher pricing is hitting CMS on the catastrophic dollars. So they’re getting reinsurance coverage back from the government and they’re pocketing the DIR fees. Perfectly fine to do that from a legal framework, but those are dollars that otherwise could be given back to consumers. And so, you have a class of trade issue. So, do I really believe in five to 10 years that plans will be permitted to own the PBM stack? I wouldn’t want that if I was in charge of the FTC. So, I think there are a lot of math components in terms of making a plan successful today that I wouldn’t want to build a business around for and I bet that that’s going to be the smartest way to do it for five to 10 years.

    Martin

    Yeah. And John, you mentioned, what Vivek just described is a lot of the backend machinery that you talked about.

    John

    I think he’s exactly right. I think the large players that have the share have done a brilliant job of leveraging their strengths. And they have optimized what they do within the rules defined by CMS. And so they’re pushing risk adjustment. OK. We would argue risk adjustment is an extension of care delivery and ensuring that the care plan for an individual that’s got a higher level of acuity is in place. Some agree with that. Others would say that it’s a form of revenue cycle management, which I think is Vivek’s point. You shouldn’t think that way. It’s really a care delivery-centric way of looking at that. Stars, it’s just … it’s quality of clinical outcomes and member experience.

    As he mentioned, you’ve got certain XYZ Health Plan of Wisconsin doing business in North Carolina because they’re arbitraging starts. Is that the right spirit of what was intended? No, not really. The whole point is to have good quality outcomes in a local geography. Now, I think that loophole has been closed. But those are two large drivers of revenue, unit pricing, right? And then in terms of the scale, I agree with Julie and you, the unit costs that the big guys get from providers is typically lower than what we would get.

    But is that the spirit of Medicare Advantage, to pound providers getting lower than Medicare? Probably not. And then the admin fees, ironically, are a huge advantage for them in the way in which they calculate their bids. So, I think what the big guys are doing is absolutely brilliant, but I think that the business models are going to evolve where the value to the end consumer is going to ultimately win out. And I think that’s directionally what Vivek is saying, is at the end of the day, a lot of these vagaries of the backend and the calculus this are going to sort itself out at the consumer level. There is more opportunity in the system as constructed where there’s more value to be received by that consumer. And Dr. Mark McClellan is on our board, who is the author of the Medicare Modernization Act of 2006. I was talking to him and you compare what MA was in 2006 …

    Martin

    Yeah, pre-risk adjustment, pre-Part D.

    John

    … to where it is now. Right. It’s a chassis that is really designed to improve quality and do so with more efficient and innovative care delivery. That’s the whole purpose of it. It’s a higher value. And I think to Vivek’s point, some of the folks have optimized the calculus in a very smart business way, but it doesn’t mean the value is realized by the consumer.

    Martin

    To that point, Julie, what’s your take on that? Is the value being realized or is there more to go?

    Julie

    Well, I would say that value is being realized. When you look at what the cost of the health plans are today, the big larger network plans have really reduced that cost for our consumers.

    Martin

    But if they hadn’t used their scale, we wouldn’t be …

    Julie

    We wouldn’t be at a zero premium PPO today without that. And I think that when I look at the trends in the larger plans, they’re using their data and their analytics to drive information to the providers. They’re doing a lot of interesting things with natural language processing and AI so that they can continue to enhance some of that, the data that they’re bringing. They’re also focused on value-based care. And then third, consumerism and what can they do to empower the consumer to engage the consumer.

    This is a hard market to engage your consumer in. We know that. The consumer doesn’t want to necessarily talk to their health plan. They do want to have a relationship more aligned with their providers. But stars in the structure of CAHPS has actually forced health plans to want to participate in that and need to have that good relationship and an engagement model that’s different. So I think as we look about at innovation moving forward in the large-scale plans, we’re going to see continuation of the integration of services, which is really what they’re intending to do to drive quality.

    Martin

    Well, one final question in the spirit of the summit and our discussions this week on building for impact across the industry, and this is a little spin on what I had said at the end of our session today. If you had no limitations on resources, money, or talent, and the sky was the limit, what would you fix in Medicare Advantage? What would you do differently?

    Vivek

    Selfishly, I’m very happy that there are very high barriers to entry. But in terms of what is Medicare Advantage about, it’s about the consumer. So realistically, taking my Clover hat off, dramatically lowering the barrier to entry is really important. If we think about insurance, there’s such great actuarial data that CMS has. The amount of capital required to get into the space prevents a lot of really intelligent people who probably have access to capital, who could build very logical enablement solutions, are blocked out. If we think about a ton of the operational functions, everything from claims processing I was describing, utilization management, not that they’re simple, but they are commodity-like in terms of the value. If we think about Part D and PBMs, it shouldn’t be a way to capture additional margin. A lot of those constructs could be easily rented from CMS. Easily is a loaded word. It would take a long time to build a lot of that effectively and intelligently that can compete against incumbent constructs, but it would bring a whole host of new participants into the space, create a ton of competition. You already have, to some extent, rate parity. That would be the best thing for consumers. John, what do you think?

    John

    In a nutshell, I’d say we think healthcare in the country is actually pretty good. We think sick care is really bad. And I think that the more care coordination, the more transparency, that we have with both providers and consumers is just going to improve care outcomes for everybody.

    Julie

    Similar to John, I would really like to see the Medicare program support well care and help preventively with benefits. And I think we’re moving in the right direction with the supplemental benefits and focusing more on public health types of issues. And opening up the opportunities to improve the health of people before they need the insurance.

    Martin

    I want to thank the three of you for taking the time to talk to us today. It’s been great chatting with everyone here in Chicago at the Oliver Wyman Health Innovation Summit. Thanks so much for joining us.

    John

    Thanks, Martin. Thank you. Thank you.

    Narrator

    The Oliver Wyman Health Podcast is brought to you by the global management consulting firm Oliver Wyman. For more insights on the business of transforming healthcare, visit our online publication, health.oliverwyman.com.

    This transcript has been edited for clarity.
     

Authors
  • Vivek Garipalli,
  • John Kao,
  • Julie Smith, and
  • Martin Graf