Bridging the Primary Care Gap: Dr. David Carmouche on Lumeris’s AI-Powered Future

[00:00:01] This podcast is produced by Outcomes Rocket, your healthcare exclusive digital marketing agency. Outcomes Rocket exists to help healthcare organizations like yours to maximize their impact and accelerate growth. Visit outcomesrocket.com or text us at 312-224-9945.

[00:00:35] Hello everyone, welcome to the Beat Podcast recorded live at the Vive event in Nashville, Tennessee. I have the privilege of hosting Dr. David Carmouche today. He is the Chief Clinical Transformation Officer at Lumeriss. He is a visionary leaner in transformational healthcare delivery with a unique blend of provider, payer, retail, and integrated delivery network leadership experience.

[00:01:03] Welcome to the podcast, David. Such a pleasure to have you here. David Carmouche It's great being here. Thanks. David Carmouche Hey, listen, how has the meeting gone for you? We're here on the last day. David Carmouche Fighting the snow in Nashville. Yeah, hoping my plane makes it out of here. It's been a good meeting. We've had a busy meeting around the meeting. We had a big event yesterday evening that kind of took most of my energy. We were at the Country Music Hall of Fame. David Carmouche Oh, nice. David Carmouche Oh, nice. David Carmouche Oh, nice. David Carmouche Oh, nice. David Carmouche Oh, nice.

[00:01:33] David Carmouche David Carmouche

[00:01:45] David Carmouche

[00:02:22] David Carmouche

[00:02:24] David Carmouche To support care teams to engage patients and he never lived long enough to see the technology that we have today. He died, sadly, five years ago. But his legacy was to teach us a lot about how primary care needs to be organized. And in so that the company's history has been one where it's tech heavy, but that tech has largely been in the data aggregation space. So health system data is dirty, fragmented.

[00:02:52] David Carmouche David Carmouche David Carmouche data, consumer data into an environment where it's normalized and able to be used to support value-based care is a good starting point now in the LLM era to launch actions autonomously on behalf of patients because of our ability to understand their health journey, their trajectory,

[00:03:18] and what the next clinical best action would be. And then build agents that can help solve that, keeping humans in the loop where they need to be in the loop, but extending care more proactively to patients and to take care of things that, frankly, we don't need clinicians to think about. So what's unique is that the data capabilities, the data aggregating capabilities of Lumeris allow us

[00:03:41] to launch multiple coordinated agents on behalf of patients' needs, not launch point solutions that exist on their own. So that's the differentiator. No, I think that's fantastic. And also thanks for the background on the company. It's great to hear the name of the new product and the story from Tom. Let me tell you what's really cool. So the voice that we use in our demos is the synthesized

[00:04:08] voice of Tom Doerr. His wife gave us access to that. So last night at the event, his widow and their two children and his sister were all in attendance and they got to hear their loved one speak live synthesized for the first time in five years. And they were really emotional. And you can imagine how cool that was for us. Wow. That's amazing. That's, that's really a great touch. And so primary

[00:04:35] care capacity is a huge challenge and Lumeris has set out to bridge the 2 billion hour gap. How does Tom fit into that effort? And what kind of difference do you hope it'll make? Yeah. So first of all, let's, let's talk about the 2 billion. So we think we've estimated that there's about 500 million hours of primary care delivered in the United States right now. There's a hundred million people who don't have access to primary care at all. It's crazy. But in this country, a hundred million people don't have

[00:05:04] regular primary care. Another 120 million who do have primary care don't have perfect primary care. They don't get all of the services and care that the evidence or guidelines would suggest. So to, to fill that gap, we think that's 2 billion more hours of primary care. If we do that with humans, that's going to come at a cost of about a trillion dollars. It's probably not going to happen. And so for sure to extend primary care to all Americans, we're going to need technology to extend

[00:05:32] the limited workforce that we have. And then it happens a lot of ways. There's no one way. I mean, first there's three ways at least that this helps. One is it takes work that is currently time consuming for humans that doesn't need a human to do a lot of care coordination, a lot of scheduling, a lot of outreach. And then it, it also provides services that should happen, but that aren't happening. A good example of that is a patient who comes out of the hospital

[00:05:59] and needs to get back into the primary care office. So, you know, during that period of time, they're really vulnerable. They may have new medications, they may have new conditions. And over half the time, these patients aren't getting back in to the, to their doctor within a week. And we know that when they get back in and in a week, they're less likely to bounce back into the hospital. Well, Tom can reach out to someone in an automated fashion, say, Hey, I heard you were in the hospital. It looks like you were there for congestive heart failure. Do you understand your

[00:06:26] diagnosis? Can I answer any questions? I see that you received two new medications. Were you able to fill them? Do you have any questions about them? Can I help you schedule an appointment? And all of that just happened with, with a computer, right? So that's example of work that just needs to happen. That doesn't happen today. And then the third is that we can upskill the delivery of primary care. An example would be, so Leanne Rimes last night was interacting with Tom and she was talking about

[00:06:51] her history of cervical cancer scare that she had. She's publicly spoken about that. And, and so we simulated a, an outreach to her that was just to prompt her that it was time to have another pap smear. And so Tom's interacting with her and she gets into the conversation about that as she's approaching perimenopause, she, her mood around her menstrual cycle is, is not great. And she, she says, I'm willing to go do a pap smear, but when I'm there, I want to have a conversation with the doctor about

[00:07:20] the pros and cons of adding hormone replacement therapy. Tom actually says it's a great conversation to have. I'm going to summarize the fact that you want to have that conversation. I'm going to send that to your clinician in advance of your visit, which is next Thursday at three o'clock. In my mind, what happens is that message comes into the doctor. He's going to see it just before he walks into the room that, Hey, Leanne Rimes here for pap smear, but she also wants to talk about hormone replacement therapy with a link to like up to date content that might be the most recent evidence so that the doc,

[00:07:48] if they want to take a look at like, what is the most recent guideline on that could take a look at it before they walk into the room that upgrades the performance of primary care and it ensures, or it makes it less likely that Leanne's going to leave that office without having the crucial conversation. It happens a lot in practice. People come in there. They have an idea of what they want to talk about. They get anxious. They get nervous. They get distracted. They leave and they get home. They're like, God, I forgot to even ask the question I wanted. So that's, those are some of the

[00:08:16] examples that we see with Tom. That's fantastic. Thanks for sharing that. You've had a lot of very interesting leadership roles with companies like Walmart health, Optum, and so an Ochsner health and beyond. What are you seeing as the key trends and technologies that will have the biggest impact on healthcare over the next five years? So it's a great question. If we think about the technology

[00:08:41] that's come into primary care over the last 15 years, I'd say largely you could argue it's not been great. And it's mostly been driven by electronic health records, right? And the work that has been added to physicians' plates has not been super satisfying for the profession. It's not just a primary care problem, but probably clinicians at large haven't loved having to be data entry clerks and having their patients flood them with in-basket messages, et cetera. I think what's happened in the

[00:09:07] last two years, though, with like ambient voice technology and transcription services, AI-based technology, I think clinicians for the first time are seeing that technology can solve problems for them, right? So I think this is a good time for us. I think having a conversation about using AI agents in primary care three years ago might've been more challenging. I think now what we're seeing is clinicians are open to this. I think there was a study out from JAMA or in JAMA recently that

[00:09:35] basically said for the first time, physicians see more positives related to AI than they see negatives. So that means that their mindset is changing to where we're going to accept that. I think this is the time where we're going to see the most impactful technology. I think it's going to be the use of AI to support diagnosis, to support treatment, to support patient engagement education. I think this is the era that technology will be integrated most helpfully into primary care. And I

[00:10:04] think there's a need across all of those organizations that you mentioned that I've spent time at. I think they all struggle with the same challenges. Yeah. Thank you for that. So AI keeps coming up. It's exciting. It's making a difference in healthcare, but people also want to make sure that the technology enhances the provider patient relationship rather than replacing it. How does, and let's get back to Tom, right? Like how does Tom strike that balance? So first of all, Tom's a new, new technology and new technologies have to be delivered carefully.

[00:10:32] So there's a lot of, as we construct the activities that Tom's going to take part in, guardrail setting, where do we allow Tom to start and stop? What happens when Tom runs against the end of his capability? How do we communicate back in a thoughtful way to the practice, to the right person in the practice, not always the physician? How do we test? We use LLM. Actually, we simulate patients to interact with our own AI and we put our own AI conversations through

[00:11:01] tens of thousands of simulated patient encounters. And we were able to assess how it performs. We deploy it to all of our employees. They give us feedback. They're given, you know, free range to kind of take the conversations wherever they want. And then we've got a group of early adopting clinicians where we're able to kind of perfect this and see it in the wild. So I think we're going to learn how to do it effectively. We are being very cautious and very thoughtful about how we design

[00:11:29] the interventions. We're probably not going after the highest risk interventions initially as well. I think there's going to be a trust cycle here, right? Clinicians are going to delegate more and more to Tom as they see Tom do more and more good things for them. And so we eventually in the vegetable world where Tom's probably doing several hundred different services at maturity, but that'll be earned over time as Tom, you know, becomes, we talk about Tom as a care team member

[00:11:56] that has work delegated to it. And I think there's some configuration that allows doctors to choose what they'd want turned on and what they don't want turned on. And so, you know, we'll learn as that all plays out. That's great. Yeah. And I love that idea of thinking of Tom as a care team member. Yeah. Who's always on. I mean, you know, unlike us, he doesn't have office hours. He's there 365, 24 seven. And, you know, for some patients, the challenge is that we try to engage them with

[00:12:24] humans during work hours. Well, I mean, people are busy, right? And so the fact that Tom could sense that your preference is to call between eight and nine at night allows Tom to do his work at eight o'clock at night. And I just think that creates so much more flexibility from a patient's perspective to not be disrupted. Yeah, no, that's great. Super promising. And within the realm of all of the challenges this could solve, are you guys pointing this mainly to value-based care arrangements?

[00:12:53] So that's the pivot for the company. I think, you know, our current business model is that we're a risk-bearing operating partner with tech and services, and we have long-term deals, multi-year deals with health systems. That's a relatively small market. Not every health system wants to enter into a five to 10 year relationship to take risk. So really we're pointing Tom at all aggregators of primary care. Basically anyone who thinks they need primary care as part of their

[00:13:20] business model at scale could be a retailer like a Walmart or a CVS could be an Optum who's bought, you know, lots of practices. It could be large health systems. It could be large medical groups. It could be the VA. I mean, who knows? It's where there's a lot of primary care that's needed and they're all struggling with the same problems of workforce and subsidization and access. And we think we're a solution to solve that. Well, that's super exciting, David. And certainly we will be

[00:13:48] keeping our eyes open for updates on Tom. If you wanted to leave our listeners with a call to action, what would that be? And where can they reach out to you and the Lumeris team to learn more? Yeah. So I'm, I'm available. I'm on LinkedIn. I'm at Lumeris, dcarmouche at lumeris.com. Certainly welcome direct emails. The thing I'd leave them with is like, I think this 2 billion hour problem, whether it's 1.8 billion or it's a lot of hours that we need. Big gap. We need to

[00:14:16] agree that we should get high quality primary care to all Americans. I think, you know, from my job at, one of the jobs I've had over the last few years was at a health plan, the Blue Cross plan in Louisiana, the best of commercial insurance in the state and, you know, 30% lower costs for anyone who had a primary care doctor. And half of our patient members at the time didn't have a primary care doctor. So we want that for everyone. I think it's, it's really part of the answer to

[00:14:42] the problem of healthcare in America. And I would like others who see the world the way we do and who believe primary care enabled with powerful modern technology is kind of the answer to kind of reach out and see how we can collaborate. That's fantastic. Well, thank you so much. Ladies and gentlemen, Dr. David Karmouche with us today. He's the chief clinical transformation officer at Lumeris. Just want to give him and the team a congratulations on their recent innovation

[00:15:08] with Tom and just to thank them for joining us, but also reach out to him, reach out to his team. You'll find all the ways to do that in the show notes. And David, thanks for being with us. I really appreciate you having me on. This podcast is produced by

[00:15:35] Outcomes Rocket, your healthcare exclusive digital marketing agency. Outcomes Rocket exists to help healthcare organizations like yours to maximize their impact and accelerate growth. Visit outcomesrocket.com or text us at 312-224-9945.