The Trust-First Model Transforming Medicaid Care w/ Dr. Toyin Ajayi & Mike Roaldi, Cityblock

Welcome to Health care Rounds, where we give you smart conversations with the people moving Health care forward. I'm John Marchica, CEO of Darwin Research Group. As a reminder, if you have a question or comment for me, please send it to Health care rounds at darwinresearch.com. And if you're not receiving our weekly newsletter, you can also subscribe at darwinresearch.com. Today, I'm speaking with two leaders from Cityblock Health: Dr. Toyin Ajayi, the company's co-founder and CEO, and Mike Roaldi, Cityblock's president. Together, they run a company that has raised nearly $900 million to d- to deliver primary care, behavioral health, and social services to more than 100,000 Medicaid and dual-eligible members. The exact moment federal Medicaid policy is undergoing some major changes. So to both of you, um, Toyin and, and Mike, welcome to "Health care Rounds." Thank you. Thank you. Before we get into the questions, why don't we just start off by each of you giving a brief background and maybe your own view of the elevator pitch of what Cityblock Health does, and then we'll get into it. So let's start with you, Mike. Great. I'm Mike Roaldi. I, um, I am an attorney by background. Spent several years practicing law, um, but then had the pleasure of working in state government. Uh, and from there, got very interested in Medicaid policy and joined UnitedHealth care for about nine years in a, a bunch of different roles, initially in a policy role, but then, uh, uh, various P&L management and operations roles, uh, across initially a few different markets and then nationally. Um, I joined Cityblock about two years ago. I initially led our growth functions, but now I lead our, uh, market P&Ls and growth functions. And, uh, I think the short pitch for Cityblock for me is that we are a, a, a company that was specifically designed to provide primary care for the people that need it the most and are most difficult to deliver it to. And so that-- the fact that we were designed that way, as opposed to, uh, what we have seen across the ecosystem, which is trying to bolt on a lot of these capabilities to existing infrastructure, that design really enables us to deliver it in a more effective way than, uh, than those that, uh, try to, try to flex an existing system into that. Um, I'm Toyin Ajayi. Um, uh, I'm a primary care doc, family doc by training. Definitely did not plan to be an entrepreneur or a, a Health care company CEO. Um, started my career seeing patients actually in safety net hospitals and community health centers, um, training residents, um, uh, on how to care better for people with the more complex needs. Um, and then very quickly finding myself in different roles in administration, really trying to figure out how to get to the bottom of why our Health care system is so misaligned, um, uh, in terms of the incentives to produce better outcomes versus the incentives to create profit, um, or margin. Um, uh, my first sort of non, um, clinical role was, was as a medical director for a health plan focused on people who are dually eligible for Medicare and Medicaid. Um, I was then became the chief medical officer of that, of that health plan, really scaling a clinical model that ties to, um, uh, network relationships, but also care management, in-home care provision, engagement, quality. Um, and then about nine years ago, moved to New York to, uh, to co-found, um, Cityblock and, and that's been the journey that I've been on. I think, uh, very much echoing sort of Mike's, uh, framing of what we do. We started focused on the Medicaid population, um, specifically picking off the hardest to reach, hardest to care for, folks with the most complex needs, because the belief I have is that you don't see interventions trickle down particularly well. Um, but we could potentially, if we could crack how to find, engage, deliver care in trusted relationships for people who have the most to lose and the most to gain, and a lot of reasons to not want to engage with the Health care system, we could maybe use that as a scalable model for others. Um, and that's been our journey. We started focused on the, the most complex subpopulations in Medicaid. We now serve dually eligible folks. We serve folks with Medicare Advantage. We're in rural areas, we're in urban areas. Um, really, uh, proving that it's possible actually to, to, to align incentives around outcomes, um, and, and in so doing, improve the patient experience, improve the quality of care they receive, improve the cost of care, um, and, and overall address, I think, what is the most pressing problem for our Health care system today, potentially our society today in the United States, which is Health care affordability. So I'm curious, anytime I get an entrepreneur on the show, particularly a physician, and you c- you, you, you gave us that background, I'm wondering what the, uh, what the spark was or the impetus to go in the direction that you did, because you saw, you know, ear- years earlier, you saw that this may be some of the issues, but what was the spark? What caused you to go out on your own?'Cause that can be kind of scary. Yeah. I think I didn't know enough to be scared. Um, maybe there might be a bit more risk of, um, risk o-off. Um, you know, I think what I've learned over the years is that everybody I get to work with at Cityblock is in some way, has been radicalized in some way by the traditional Health care system. So most of us who go into Health care do so because we genuinely want to help people. We want people to be healthier. Many of us, if not all of us, have some personal lived experience that drives us to wanna improve Health care and health outcomes. Um, and it is true, however, that in most places in the Health care system, if you're working for a traditional organization, so for a traditional Health care provider, even it, even if it's a safety net hospital or a community health center, if you're working for a tr- traditional payer, in many ways the work you're doing is the same work that perpetuates the status quo, that has delivered the outcomes that cr- created the dissatisfaction in the first place. Because our incentives are not fully aligned to, to, um, to really deliver quality and access and experience and dignity and reduce cost. Um, and so what happened for me is what happens to many doctors who, who, um, you know, start off really wanting to help people and, and save their lives and make them better, and you realize you're part of a revolving door, um, that actually doesn't equip you to, to produce better outcomes. It equips you to patch people up and send them back out, and patch them up and send them back out, and do the same thing over and over and over again without any hope of getting upstream of the problem. And then, you know, for many, that turns into burnout and it turns into fatigue and disengagement. Um, and for others, I think, um, uh, it creates this sort of like weird glitch in your brain that tells you that somehow you can make things better, um, and, and you end up on some other journey trying to figure out what that other path is, and that was me. Well, thank you for that. Um- For listeners who haven't spent time with the model, and I want to make sure that I'm getting the model right, so this question may be off or it may be it may be right on the money. But I was thinking about it sort of like in the moment, what happens when one of your care teams shows up at a member's door for the first time? And that could be virtual or, or, but, but I'm thinking really physically. What happens? Can you, can you walk me through that experience? First, our goal in that interaction, uh, unlike, unlike some organizations that I think when they actually are able to reach a member, uh, they wanna check through the boxes that are most important to them. You know, in a payer context, that might be completing an assessment or closing a quality gap immediately. Uh, what we, uh, what we believe is that first interaction should engender trust because that's going to allow for multiple interactions in the future instead of just maximizing that one interaction. So that sounds a little squishy, but, um, to make it a little bit more practical, uh, oftentimes we try to identify something that that member needs and values that we can provide to them. So for instance, if the member's struggling to find food, if they don't have a particular piece of durable medical equipment that they need, if we fulfill that need in that first interaction, um, then we can, uh, engender a trust that's gonna lead to multiple interactions when something more critical is at stake, like they're considering whether or not to go to the hospital or they need a specific screening. So, uh, so the goal of that first interaction is to engender trust. The method of it i- is really phased. Uh, first, they should have some familiarity with City Block before an actual conversation. We, you know, we refer to that as priming, so oftentimes we'll send a mailer. Um, uh, we, we like to put Toyin's picture on the mailer because it t- tends to get a higher response rate, which she, she hates, as you can tell. Um- Uh, but, but, you know, they, they have something that lets them know that Cityblock is a Health care-related organization that might be able to help them before they get that initial call, and we find that reduces scam reflex. Then oftentimes we'll follow that up with a series of text messages specifically tailored and messaged based on what we know about that member, and then we do the personal outreach. Uh, you know, depending on the member's profile, it might be telephone or, or in person. Um, but that's when we really try to identify that, that thing that matters enough to the member that they will trust us in the future. The one thing I would add to that is, is even before we get to them, I mean, I think if you think about the way that Health care traditionally operates, it's essentially a first come, first serve basis, and what you get depends on where you show up, right? So if you show up to the emergency room, um, there's some triage, but by and large you will get seen, um, and you will get seen by the person who is at the place where you went. If you went to your primary care doctor's office and you made your appointment three weeks in advance and you show up that day, you'll get seen by your primary care doctor. Um, what that is not is actually, like, any- in any way optimized for what you need, nor is it responsive to the people who didn't come in that day but who actually may have higher needs than the person who did show up. And so part of what we've done using technology and data is, is take a point of view about who needs us in any given moment and what type of care they need or what type of outreach and engagement they need from us. So it could be that they need a nurse. It could-- because they have a medication issue. It could be they need a psychiatrist because we think they have a serious mental illness and are, are likely to get hospitalized for it if we can't support them with their meds. It could be that they need a community health partner because there are issues around social needs and, um, and around connectivity to social resources. Could be a combination of all three, as an example. But we have a point of view for every member who we're accountable to, regardless of whether or not they picked up the phone that day to say, "I need help," about whether or not they do need help. And we're tailoring our outreach such that it is not on a first come, first serve basis, but it's actually tied to who needs what, and we're optimizing resource allocation and outreach based on, on what people need. This is really what I mean by we're designed for serving this type of membership. If you think of a traditional primary care practice, uh, they are hoping that somebody walks through that door. In some cases they're doing-- I think there's an increasing recognition of a need for proactive outreach, but our entire infrastructure is built on going to find people, and not just proactively going to find people, but doing it in a strategic way that's most likely to actually get them engaged and responsive. That makes sense But I wanna come back, Mike, to the, to your point, and actually I'd like to hear from both of you, to your point about trust. Because it's, it's come up-- First of all, it's one of our core values at Darwin Research Group. We can argue, well, is it an outcome or… Well, it is one of our values. It's something that we look, look for and, and not just in building with each other, but in the products that, that go out there. So trust is huge, and we have a serious trust problem. It's come through, uh, in multiple recent episodes here on "Health care Rounds." And just a little bit of background reading that I, that, uh, that I did before this, I, I, I got that sense that it's really, really important. So can you, can you talk about how you quantify that, how you think about trust, 'cause it was w- literally the first thing that, that you said, what you're trying to do when you show up at the door is establish that. Can you expand on that and have-- we'll talk a little bit about, about that, about trust. Yeah. I, I like to draw this analogy. I once read about the early days of online commerce and, you know, eBay, uh, had trouble getting, getting traction initially because people didn't know if there was any way to hold people accountable if they ordered something and transferred payment, if that thing would ever come. And through online reviews, there was a mechanism established whereby people could understand that this retailer was someone they could trust to send what they needed, and that unlocked an incredible, uh, amount of activity in commerce. And it's just, I think that's just a good way or thought experiment to think about how valuable trust can be to an interaction. So take that to the medical context. What is it that trust enables? If somebody trusts us, they're gonna share information that is critical to their health that they otherwise would not share. So if I don't trust my, the, my care provider, am I gonna share with them sensitive information that is critical to understanding what disease burden they might, I might have? Am I gonna share with them if I h- if I'm still smoking, if I'm, if I'm, um, a diabetic and not adhering to the, um, dietary guidelines that they've provided to me? The answer's no to those things. And then am I gonna do things that they've instructed me to do for my better health, to better my health if I don't trust that they're, they have my best interests at heart, if I suspect that they're asking me to do that thing because there's some benefit to them? So I think it shows up in all sorts of ways, but primarily the information sharing between the patient and the doctor or the care manager, whatever personnel they're interacting with, and then the willingness to actually, uh, do the things that are being recommended because they believe you have their best interests at heart. But, uh- Yeah. And we, I mean, we believe it's so important. Um, uh, and I don't think you can sort of achieve or perform on something you don't measure, and so we actually built our own metric to cover this. Um, and so we, we built and validated a metric. We surveyed folks. We got feedback Um, and, and validated a tool that basically measures trust and satisfaction, um, both in the institution that is Cityblock and also in the individual with whom the person is interacting, the member's interacting. And we measure that metric regularly. It's part of our core performance slate. Um, we found it to be very strongly correlated to engagement and to therefore to outcomes. Um, and so we know, I mean, the sort of the theory of change is that if you can unlock trust, you can unlock engagement, therefore you can ask people to take different actions on behalf of their own health and wellbeing than they would otherwise. Ask them to take the med, ask them to, you know, call us and not 911, ask them to stop smoking or to work to stop smoking. But you need that trust engagement, um, in order to, to engender behavior change. Um, and so we believe it so, so fundamentally that we've actually, you know, inculcated it as part of our core metrics. We measure it, we iterate on it, and we use it, um, to drive performance. Well, it's interesting that you use a change model there, right? Because you want them to change behavior. What you want them to, you know, not be eating sweets all day if they're, you know, let's say diabetic or not be smoking. You're asking people to change, and that's one of the things that, uh, you know, s- uh, I will say Health care in general struggles with You know, like, like, um, I've used this example before, but like accountable care organizations. And physicians say, "Well, okay, the accountability, everything is here with us, but what about the patient accountability?" Right? And use that word too. And so your framing on this is you're asking people to change, but in order to do that, that trust is a requirement. So how do you, how do you operationalize that? I mean, you've gave-- you know, you got the index and you got-- or these, these two metrics, but, but how do you operationalize that? Is there sort of a, a checklist of things that you do when you go in for a visit? Or, you know, I, I'm just, I'm really curious 'cause this, this trust piece is really fascinating to me. Totally. I mean, I think I would start with just reminding us that, like, there are so many ways to make money in Health care that don't actually involve behavior change or Health care outcomes being better, right? And so, and so we picked, we picked a space in which, you know, you actually have to, you have to change utilization patterns in order to create margin, in order to create outcomes. Um, and it's not arbitrage, right? Like, you need, you need the patient to be engaged to do that. Um, how do you do that, I think requires a shared understanding, like full understanding of what's going on for the person, and then the ability of someone on behalf of the team to propose a hypothesis about what might be different. So start with a member who, you know, we think is at high risk for hospitalization because they're having a hard time remembering to take their medicines for their schizophrenia. Um, and they also have really bad diabetes, and they're also struggling with housing insecurity. Um, we based on the information we know about them, in addition to the fact that we're gathering data from them, they're sharing with us in the context of a trusted relationship, "Oh, I'm not so sure where I'm gonna sleep tonight. I'm really worried about getting, you know, healthy, nutritious meals. I might just have to go to Wendy's this evening and the next evening and the next evening after that." Um, we use that data combined with our sort of broader models that have been trained over, you know, many, many years now of serving people to say, "What do we think is the right, most important intervention for this person?" And then we have a conversation with them and try to build alignment around that. Um, what we find is that, to Mike's point, if you have done that on a foundation of, of trust, where we've solved a problem for them that may not be the most important problem, but is the one that's pressing to them, they're more likely to be receptive to us saying, "Hey, you know what? We have this opportunity for you if you'd like. We can send someone to your home to give you an injection of your antipsychotic meds. You don't have to remember to take it every day. It's one less thing for you to deal with. And because of that, you might feel clearer in your mindset tomorrow, so that if we decide to go together to the housing office, we might be able to help advocate for you to get housing. Are you willing to go on that journey with us?" And the person is likely to say yes because they trust us. They think we have their best interest at heart. We've showed up for them repeatedly. Maybe we've solved a problem for them that was pr- pressing, and we take the step from there. And it really is, it's that specific. Um, and when you look at the model we've built and that Mike and his team implements and executes, it's about those, like, very important operational tactical interventions that are backed by data that we're constantly iterating on, testing and iterating on to ensure that we know and we have a point of view about what the next best action is for a person, and that we have the right person in front of them to communicate that, gain alignment, and then actually execute it, and then come back and do the next thing. Yeah. And maybe a bit more tactically, at the early stages, as I mentioned, we do what we can to reduce scan pr- reflex, at least create openness if not trust initially. I, I think to your question about, you know, do we insist that they check all these boxes in inter- interaction, I think we're actually moving the opposite direction so that the interaction can be more organic. So we moved away from a lot of the scripting in, for instance, in our outreach specialists. Uh, when a member schedules a visit at our clinic, uh, the physicians tend to have a lot more time than they otherwise would have. They have the time to walk somebody through, you know-- J- just to give you an example, I was at, in a clinic, um, not long ago, and physicians spent an enormous amount of time helping somebody, uh, with how to use a continuous glucose monitor. Uh, something that would be unlikely to ever happen in a fee-for-service primary care clinic because the, the doctor just would not have time to do that type of thing. So that sort of unstructured time, uh, where the, whether it's the outreach specialist, nurse care manager, community health partner, a physician can have the, the time to do what's necessary to give the person the information they need, um, I think that really engenders trust, uh, mo- more so than, um, you know, any scripting ever could. Well, I'm c- curious, and I'm not asking for you to give away all the secret sauce, but I'm curious how, how you actually operate. Let's use your example, uh, Toyin. Operationalizing the example where somebody has housing insecurity, food insecurity, and a medication adherence issue. Those are, you know, very different things, and so how do you address each one of those components with a physical person or through technology? Like, how do you solve that problem?'Cause literally this is, you're talk- we're talking about some of the biggest challenges in Health care, right? So how do you, how do you go about solving that? I think, um, you know, a few core components. One, it's, it's an interdisciplinary team, so it's not one human. Um, there may be one person in the home or in front of the member at the time, but if you give them the data that they have-- they need and the tools that they have, they can draw on a broader group, right? And so what we do is, first, we put the right person in front of that member. Um, in that scenario, um, we might put a behavioral health community health partner in front of them because they are especially trained. They have lived experience of mental health challenges. They're also trained to support with social needs. They have at their fingertips a directory of all of the social resources for this member. They have suggestions about which ones they ought to connect the person to for housing, for food, et cetera. And they have real-time ability to warm handoff to a physician who's a psychiatrist or to an EMT or a paramedic who can facilitate the medication management. Um, and so we really say, "Look, like, there's no wrong door," and ideally, we're gonna put the right person with all of the tools and resources using technology to ensure that we can meet their needs. And then we prioritize because the rate-limiting step is gonna be the sort of cognitive bandwidth of the person we're serving, their time, their energy, their ability to engage. Um, and we work with them and with our tools to prioritize which is the most important thing to address first, which means that in that visit, we might say,"Actually, the most important thing is to get this person sleeping indoors and warm and fed, and we'll deal with the medication another day. We won't forget about it, but we're gonna get them indoors, warm, and fed today, and then we're gonna work on medication the next day, and then we're gonna work on filling the next problem the next day after that." Um, and, and recognizing you can't do all the things at once, but if you have access to all the tools and you have a point of view that's grounded in data about what the most important and impactful interventions are, you can follow that over time. And then, you know, building in really robust, um, controls and operating systems 'cause off-the-shelf stuff in Health care doesn't work. Like off-the-shelf tools in Health care are basically all geared towards, on the provider side, billing and coding, seeing the person in front of you and getting reimbursed maximally for it. They're not geared towards intervening, following up, closing the loop, measuring outcomes, reaching out proactively. And so we had to build a whole set of custom tools and, um, and management controls to ensure that we actually could keep track of what was happening, um, and we could map that back to our expectations for what makes for good care for this population. Yeah, I think I'd add, um, I recall when I was on the payer side talking to a, um, a, a health system, or I'm so- I'm sorry, a, an FQHC that had built out some, uh, care management capabilities, and we were seeing that they were doing a much better job with both, um, primary care visit scheduling and completion than our care management team was able to do. And we asked them why, and they said,"Well, we can just immediately schedule somebody for one of our physicians right now. You have to go find somebody in your network and hope that, you know, the provider data is correct and hope they have availability and they take Medicaid, et cetera, et cetera." So you think about that across the spectrum of needs that Toyin just mentioned. Where else can somebody interact with somebody who's h- able to help them navigate what their biggest needs are and then immediately get them over to somebody who can help with those things? Uh, it just doesn't exist anywhere else. And so, um, so I think it's a, it's, it's very unique and enables us to create an experience of the Health care and social support system that, that these people really need in order to improve their outcomes. And you layer on top of that the discipline of, of an operator to continually improve, to look at the data to see what's working and what's not, and you can see how that can snowball into some really po- really amazing outcomes. Yeah. Well, what's rolling around in my brain right now is the, is sort of a, number one, how human-centric this feels, how, you know, people-centric. And, and yet that's sort of, I don't know if I'd say against, but going against the grain of a trend of we're gonna solve all of this with technology and AI, and we're gonna put robots in the homes and, you know, home monitor this and… Right? I mean, that's, that's been the argument I know in the, in the home care and the home health world. It's how do we-- We don't have enough caregivers, and so how do we apply technology to be able to address the needs? This is a different audience, a different population. I understand that, but it feels very people-centric. Um, and so that's a comment, I guess, but maybe the question is H- what role does technology play then? Um, is that more in, you know, Toyin, as you were saying with, um, the, you know, or the metrics or tracking using data to see how things are going? It feels very almost technology blind to the person that you're working with. That's what it's feeling like to me, so. Yeah, it's both. It's both. I think, um, uh, for sure, I mean, this is a deeply human model, right? Like, and, um, and by, by design, it is very, very sort of high touch and high trust for sure. Um, the high tech part is really important though, because to your earlier point, in every part of the ecosystem, but particularly in primary care and behavioral health, there just aren't enough humans to do this Particularly with the amount of like administrative waste and friction that we ask most people, most physicians, and Health care providers to wade through. And so the tech can be really important, and we've deployed it in such ways, um, in enabling people to spend more quality time doing the things that only people can do. So that's one piece of it for sure. Um, and that in, in some ways is sort of, is exactly what you described, which is sort of it's tech blind to the patient or to the member, right? Like all they see is that the person in front of them seems miraculously equipped to be there when they need them and has all these tools at their fingertips and is able to be fully present and make eye contact 'cause they're not typing away on the EHR. And when they say they're gonna do a thing, it gets done because we've got tools in the background that are loop closure, like, um, ensuring loop closure and all that. That's one part of it. The other part that has been, I think, surprising, but pleasantly surprising to me, has been that, you know, we have surveyed our members, and we've done a ton of AB testing with them and asked them about their willingness to engage with, say, voice agents for things that are not clinical. Scheduling appointments, um, uh, um, getting information about referrals, um, uh, deciding who they should talk to for a need that they may have, um, fielding their in-inbound call at two o'clock in the morning because that's when they got round to letting someone know that their electric wheelchair is broken. Um, those are things that we would otherwise have people doing. Um, they're not particularly high-skilled, um, interventions per se. Um, and we've been really pleasantly surprised that our members are saying to us, "Have at it. We're, we're-- I'm totally down to talk to an agent, especially if that agent is likely to be able to get me to the right answer more quickly than it would a human." And so, and so we are actually doing both. We're doing member-facing work, patient-facing work that is directly deploying technology, and we are also enabling our clinicians and care team members to be that much more effective and more present also using technology. And I think the future is really exciting to me. Like, I do think that there's some future world that is a combination of robots in the homes and, and people at the bedside, right? And, and I think figuring out that combination about what is uniquely and, and, and critically only ingrained within two people colliding with each other versus where we can use technology to ensure that we have more to give and we can actually be more accessible and more abundant in the resources that we deploy. I think both of those things are gonna be true. And then I'll just, I'll just throw in one more thing that we have to solve for, um, which is a, you know, this is the existential problem for this moment in our country is like net-net, we have an affordability problem. And so we've got everything we do has got to be geared towards explicitly trying to reduce the unit cost of delivering Health care because we can-- if we, we, if our only approach to saving money, um, is reducing the number of people who can access care services, we're gonna end up sicker and disenchanted and disenfranchised and much more vulnerable as a nation, right? So we've gotta figure this out, and technology has to be a deflationary force, um, in terms of unit costs throughout the system. Well, that I think relates to my next question, which is around scalability and, 'cause it feels very localized to me. You know, you've gotta have the connections in whatever community that you're, you're operating in. It can't just be solved by technology. So as you've grown over time Have you struggled with scalability issues, or how does that, if at all, manifest itself? For every dollar of operational costs we add, we have to find medical cost savings to offset that based on improved outcomes and, um, you know, reducing acute utilization. And so in order to make this work, we can't just, um, be, um, undisciplined about the types of, of, um, interventions that we're willing to invest in or deploy. And so the short answer to your question is we have to know what's working and what's not. That's the, the most important thing. And so we, I think, have been very, very disciplined in the data behind the performance evaluation so that we will only continue to invest in things where we're seeing outcomes, uh, that can con- can justify their continued existence. And so, um, that plus, y-you know, having, uh, having versatility in, in our personnel so that they can do each of these types of interventions, um, I think those are the key things. And, and so, you know, what does that look like? That means regular performance evaluation, uh, meetings, uh, um, studies to understand exactly what the outcome of the intervention that we had projected was and what the reality was on top of that. The other thing I'd add is that I think insurers do this quite well. Um, I think they, uh, have the data and the actuarial expertise to understand, uh, what interventions are most likely to drive outcomes. And so we really lean into partnering with our payer partners to understand and be very honest about the things that we're doing that are working and those that aren't, so that we are on the same page with respect to what has the biggest yield. You know, th-there might be an intervention that we feel really strongly about that, um, is-- has the potential to have a, a very positive member impact that we find, uh, we agree with the payer, did not work out the way that we anticipated. And so we, we just have to make hard decisions about those things and not let, not let our, uh, preconceived projections be an impediment to making a good decision going forward. Well, that's-- Mike, this is another follow-up for you Do you see-- It's interesting the way you describe the, the insurance companies and how you're working with them and obviously Medicaid and all of that. But now that you've been sitting in this seat for a while and having all that history, I don't know, do you view Health care differently? view how you solve Health care problems differently? Yes. Okay, next question. I anticipated the answer would be yes, and then the, the follow up of all of this, of course, well, how? Yeah. Um, I think that the, the kind of difference between the, the data and the, the real-life experience at the point of care is probably the, the best way that I can describe it. Um, you know, at a, at a national payer, you have, uh, inexhaustible access to data, and that can often lead to a sort of hubris about what you, what works and what doesn't. Um, and I think, I think that that often obscures what's a-actually happening at the point of care. So you may, you may see a, a study that says, uh, this particular, um, type of intervention isn't producing the outcome that you anticipate, um, and feel pretty good about that and, and decide to move forward with, with, um, going a different direction. Um, now being on this side of things, I can see where, uh, where that might obscure what's actually going on. So for instance, if I'm sitting with a, with, uh, one of our doctors, one of our physicians, one of our nurse care managers, and I can see that that interaction with a diabetic where they took the time to, um, teach somebody how to use a continuous glucose monitor and then, you know, follow up to see how it went with that member, you know, that's, that's very difficult to discern the, the, um, the impact of that in the data in the way that a payer might try to evaluate and, and oftentimes gets washed out in the noise of, oh, the cohort's changing and the, the, um, unit cost has changed. But I have so much more certainty now in seeing the actual point of care in the way that these things impact, uh, impact members from a member experience, from an outcome perspective, but also from a utilization perspective, that you just can't see if you're sitting in an office and sitting with your actuaries and looking at the data. So Toyin, does that f- does it ever feel like maybe you have a higher burden of proof? You've mentioned models a couple of times. You've mentioned data quite a few times, like that, so that it doesn't feel so touchy-feely, or we're just, you know, out there fighting the fight, that you-- that when you're dealing with payers, large payers, Medicaid, so on and so forth, you need to be able to justify the work that you do through doing studies or having outcomes data. Like what role- What role does that play in? Do you feel a sense of, you know, that you have to prove these interactions in a, in, in any way. Totally. And, but I think that we-- I welcome that. I mean, that's what we signed up for, right? Like I, you know, I started my career as a, as a doctor in a community health center and as a hospitalist. I would go to work, and I would see my 25 patients that day, and whoever showed up, showed up, and I would bill accurately for the work I did, and, and I would document in my chart, and I was successful. Um, I did my job. Now, I have no idea whether or not the care I rendered actually translated into better outcomes over time. I have no idea whether they drove Health care costs up avoidably. Um, I have no idea whether the patient picked up the prescription I wrote and actually took the med and found that their hemoglobin A1C got better. I have absolutely no idea whether the work I did translated into improving our ability to fund a Health care system that's gonna actually be there for us when we need it in the future. Um, but, but the metrics that were in front of me said if I did those three things that I described, see the patients, document, and bill appropriately, my job was done. Um, we deliberately opted out of that system, and we complicated it on purpose because we're trying to build a delivery system that proves to not just ourselves and our investors and our partners, but to everyone that it is actually possible to deliver care that translates into better outcomes, better quality, and a lower cost. And so we had to create a much more complicated way of measuring things. It's not enough to say, you know, "How many widgets did you, you know, bill today? How many patients did you see today? How many charts did you write? How many RVUs did you perform?" Like, that's not enough 'cause that actually will just continue to drive the outcomes that we're seeing, which are unsatisfactory to all of us. Um, and so yes, we not only had to sort of build a new measurement system, we then had to measure against that measurement system and demonstrate that it is actually delivering what we believe it's delivering, and be willing to discard, um, leading indicators and lagging indicators if they don't ladder up to the overall outcome. Because unlike, "Did you see the patient?" Yes, no."Did you bill?" Yes, no."Did you write the chart, the note?" Yes, no. The question of like, did you make this, this group of humans overall healthier than they would've been had you not cared for them, and did you do so at a cost that is lower than we would otherwise have paid, is not a binary question that you can answer at the start of every encounter on a yes, no, right? It's complicated. Um, and so I'm not, you know, I'm not sort of belaboring or bemoaning the fact that we have a very high bar that we're held to. Um, it's interesting, right? Like there's all these conversations about does value-based care work? Does it not work? We're not having the conversation about like fee-for-service does not work. The status quo does not work. It will bankrupt all of us. Um, and, and it will undermine the foundations of the social safety net that we've tried to build. Um, so like let's talk about that first, and then let's try to figure out how to make things better, um, albeit- With embracing the complexity that comes along with that. I'm, I'm curious though, the way that you described, not that you didn't enjoy your job as a hospitalist or working in a, you know, FH, um, federally qualified-- I never get these acronyms right, federally qualified health center. Do you find, and not just you, but that your providers and the people in the organization have a higher level of job satisfaction because they're able to see sort of the end points and everything in between? That it's not just a matter of… And it sounds like a loaded question, but I'm genuinely interested if, if you're seeing from doctors and from nurses and care managers and other people that you inter- that are part of your organization, if they have higher well-being themselves? Yeah, I, I think the answer is yes and no. I think it's, it's, it's actually not as straightforward as you would think. I think for some people like myself, and frankly a lot of the providers who thrive in our model, who are willing to take on the complexity, um, and to acknowledge that there's-- it's a much harder road to say, "Did I, did I do well today at my job?" Um, if, if you're being measured against a set of long-term metrics across a population that are inclusive of things that you don't have full control over, where you have to rely on a team to do things in order to achieve the outcomes. For some people, that is liberating and grok-able because they're able to grapple with the complexity and walk them their way back from, "Is this population of people better because I did what I did?" all the way through to, "I did a good job today because I made sure that woman did not leave my office without understanding how to use her continuous glucose monitor." Those people who can understand that and deconstruct the sort of theory of change do really well. They love being part of a team. They love being able to say, "I, I went to work today to spend the time that was necessary to do the things needed to ensure that this population of people under my care are better." They love it. Other folks are like, "God, it was so much easier when you just put me in a room, gave me a white coat, put me in front of a computer, and I took care of everybody I saw, and then I billed and I documented and I went home." Like, that's a, that's-- don't make this more complicated. I don't wanna grapple with the existential question. I don't wanna have to hand off to a community health worker and then have a conversation with them about which continuous glucose monitor I sent and whether or not it's gonna, you know, c-correlate well with the other instructions I gave. I don't wanna deal with that nonsense. And so it's not for everyone. Um, I think it's deliberately for people who are, um, radicalized by the status quo and came to us that way, came to us with an understanding that they can articulate about why the status quo was not working in achieving the goals that they set out for themselves in their career and for the people whom they serve. Those people relish this. Um, and I think that it's also incumbent on us and on Mike and his leadership team in making it easier for people to understand how the work they do on a daily basis, if not tied to RVUs or not tied to encounters or not tied to billing, actually does tie to outcomes. So they can see that path much more clearly and understand, you know, by making the choice they made about spending time educating this patient on their medications, they're actually doing exactly what we want them to do and, and how. Yeah. It's just, it just, it j- do people have, you know, a, a larger sense of self, of worth, of joy in work, you know, that is part of the quintuple aim, right? Um, are they, are they, are they getting more satisfaction knowing that-- Because I understand the frustration for some people. I check these boxes, and I don't know what happened down the, down the line, but I did my job in the moment, and for some people, that might be perfect. But so just really kind of getting at that satisfaction piece of the providers. If I may, I don't know if I can answer it for the providers, just myself being, you know, ex-attorney, ex-insurance executive. Um, but I noticed, you know, when I was working for an insurer, providers didn't really like when I tried to speak for them. So- But I can, I can tell you from an operator's perspective, um, as somebody that, you know, couldn't pass biochem in college and therefore, you know, couldn't go to med school, uh, but still wanted to be, uh, part of an industry where you were making an impact. You know, w- I always really wanted to feel like what we were doing was enabling better care, better member experience, less friction. And, you know, I think the operations side is, is very important to that in a way that it goes unseen. Mm-hmm. And so, so information getting to the physician at the right time, are the, uh, care teams optimized so they feel like they can actually get to all the members that they need to get to? Um, you know, are their schedules optimized? Uh, is the data flowing in the right way so that they're tasked with the right things in the right order in the system? They don't have to click all over their screen. You know, these mundane things that really improve, uh, improve the, uh, experience of giving care. It kind of goes back to that technology question. Are you-- is the technology enabling a better, more valuable human interaction instead of sort of overcoming it, um, or getting in the way of it? And so I can personally say as an operator in my team, that nexus is much tighter than it has been in any other position I've been in in my career, where I feel like if, if we innovate operationally, if we innovate our systems, if we get information to the right place, if we're smart about scheduling care teams, if we're responsive to what the client is looking for, that is gonna have a direct impact on the care providers and therefore the operators on our teams, the, the analysts, the traditional people you don't necessarily think of as, as enabling care who- but actually are, I find that they are much, much more satisfied and much, um, there's much less burnout than in, in previous, uh, roles I've had in my career. I've got two more questions we can wrap up, depending on how, how long the answer is to these questions. One is, has anyone in a more, we'll say a more traditional s- patient panel, patient population. Is anyone looking at taking some of the lessons that you've learned, that you've documented, and applying it more broadly?'Cause it's, you know, you're really identifying a very specific population that you're working with. Are there lessons-- maybe that's a better question to ask. Are there lessons that, you know, the chief medical officer for some, you know, IDN out in, you know, California can, can take from what you're learning and apply to a larger population? Yeah, I think so. I think to some extent, yes. Um, and you know, certainly a lot of the principles we describe, you know, are not, it's not rocket science by any means. I think what is complex to execute against is the operational and technological interventions that are necessary to enable that care. Um, and that's what we've spent the last sort of decade, um, working through. And so, you know, it's too easy an opportunity to, to not take up. But like, you know, to this chief medical officer of the IDN in California, um, I think the lesson is you should partner with us to help you serve your people who really need it in a way that will deliver better outcomes, especially because they have that integration and that alignment between the cost side on the, on the payer side and, and the provider side. And, and we've, you know, we've, we certainly launched partnerships like that where we've actually specifically supported those types of organizations. But I think it's very hard to, to sort of, you know, have a foot in, in two canoes, right? And so you're either a fee-for-service chassis, um, you have a system that requires revenue that comes in the form of patient, um, uh, encounters and interventions, um, or you are a sort of value-based, outcomes-based provider that is actually in many ways trying to keep people out of those same self-institutions where they don't need to be, um, when they don't need to be there, and figure out a way to monetize that instead. It's hard to do both. Um, and that's where partnerships can be very helpful. I, I would add, I think that, um, I think that sometimes in Health care we, um, we abandon the principles of, of just h-hustle and perseverance and execution. Like there's some, uh, magic formula, which obviously we've learned a lot. We've talked a lot about a lot of these things that we've learned that are unique to City Block, are unique to the, to the, um, extrapolations from the data that we're, we have access to. But also we, we are very persistent, deliberate operators. And so what I would say to that, uh, person that's interested in implementing these types of programs are don't abandon the principles of, of PDSA cycles and, and, um, you know, the type of reporting necessary to hold people accountable to drive performance. I think a lot of those principles are still relevant in this challenging field, um, because, uh, they enable, uh, the consistent application of best practices that you learn. And so, uh, so I know you're pr-probably looking for something on the technology side or a groundbreaking new intervention, but my advice to them would be be a good disciplined operator in the way that, in the way that, uh- You know, that drives results in all fields. Well, I intentionally stayed away from the money questions, from the, um, you know, getting too far into the weeds. But I mentioned this in the introduction, and I, I, I'm just curious, you're-- how you're responding to, um, upcoming enforcement of cuts in Medicaid and just changes in Medicaid policy at the federal level that will implement or that will, excuse me, affect things at, in states differently. I'm sure you've spent a lot of time thinking about this, but just sort of your general sense on where you are today and what you're looking at in 2027. Yeah. I think first just historical context is useful, right? Remember that when we came out of the pandemic, um, there had been a period during which there was a moratorium on redeterminations for Medicaid. And, um, and Medicaid enrollment peaked about 90 million people. Um, post-pandemic, um, states conse-consequen- um, in, in sequence, sequentially started sort of re-attesting people for Medicaid. And net-net, um, probably on the order of 15 to 20 million people ultimately no longer qualify for Medicaid. We've been through an exercise of redeterminations for Medicaid on a national scale. Um, and there were a lot of lessons learned. We were right in the middle of it then, um, as we intend to be now, serving a population, helping people figure out, you know, um, whether they're eligible, how to prove that they're eligible if they are, how to make sure that we protect, um, you know, high-risk populations, pregnant women, people with disabilities, folks like that, so they don't accidentally and inadvertently lose their coverage. Um, but this is an entitlement program, and so some mechanisms by which we determine, you know, who's entitled in one given point in time are necessary. And I think there's a broader question here about, um, uh, you know, what happens to those folks who lose their Medicaid, and do we have a strong enough, um, set of infrastructures, um, uh, to catch them and ensure that they have coverage? Those are broader macro policy questions. But, um, but on the ground, um, you know, as we serve Medicaid populations and have done across multiple states for a long time, we're constantly working with our states, with our health plan partners, with our members, with our providers to figure out Who, who's got coverage? Should they-- Are they on the right plan for them? Um, how do we ensure that we support them in documentation? How do we ensure we support people in retaining their coverage? And so, you know, I'm not, I'm not worried about the mechanics of this per se. I think this administration has really tried to, to push us to be more tech forward in the way that we do these things. Um, I think that, um, that states have a lot of leeway here in terms of the guidance that they're using, and so we're working closely with our states. I'm concerned about the broader problem, which is that this came out of a budgetary decision, which is driven by the fact that we recognize as a nation that our Health care costs are outpacing, um, GDP growth and wage growth, and that for most states, Health care is the number one item on the state budget, um, and that we are very, very quickly running out of resources to continue to keep pace with trend, um, because we have not done a sufficient job as providers, as payers, as innovators, as leaders of managing the cost curve in Health care while improving access. And so it's a broader question that we're like in the thick of trying to solve for the, for the ecosystem and hopefully prove that there are new models to deploy that I think is the solution to this, right? Um, and then we've got an operational issue today to support the members whom we serve and to help figure out what happens to people who lose their coverage, because that will be critical for them as well. Yeah, I'd-- I think value-based care is u- is more relevant than ever in this current environment. I mean, especially, you know, the scrutiny on fraud, waste, and abuse, for instance. Um The, the, there's no incentive to commit fraud in value-based care. Mm-hmm. Um, th- there's every incentive to keep members enrolled, which may include helping them through a work requirement or a community, community engagement requirement. Um, and so I think value-based care is more relevant than ever. I mean, that sort of skirts around the near term issue that I think you're alluding to, which is if your panels are smaller because so many people get disenrolled, how does that affect your financial projections? Short answer to that for us is, truthfully, because our cohorts tend to be more acute, they're-- a higher percentage of them are gonna be exempt from those, those work requirements and community engagement requirements. So it's not gonna have as, as big of a, an impact on us as Cityblock. Mm-hmm. But I, to Toyin's point, I'm, I am really, uh, interested and, uh, feel very bullish about value, value-based care because it is so clearly, uh, a remedy for the, the underlying problems that these policy changes are trying to address. Mm-hmm. And so I think we'll see, you know, continued emphasis and investment there. Well, um, Mike Raelde and Dr. Toyin Ajayi, excuse me, um, with, um, with Cityblock Health, I want to thank you for, uh, taking time to, to sp- this has been absolutely fascinating. I just wanted to give you a, a chance if you have any closing thoughts before I wrap up, um, love to hear them. Well, you've been very indulgent. You've let us yammer away a lot, so, um, so I'm, I'm just really grateful to be here and thanks for the conversation. Likewise. I enjoyed it, and, uh, look forward to hearing the episode. Thank you. Well, uh, thanks again, and if you enjoyed this episode, please forward it to a friend and rate, review, subscribe on your favorite podcast platform. And just as a reminder, if you have a question or comment for me, please send it to Health care rounds@darwinresearch.com. I'm John Marchica, and I'll see you next round.