← Becker’s Healthcare Podcast
Becker’s Healthcare Podcast
AI Care Coordination and the Future of Patient Engagement with Shailu Verma
00:15:41
Loading summary
Transcript16 lines
- [00:00]
A
This is Scott Becker with the Becker's Healthcare podcast. I'm thrilled today to be joined by Shaylo Verma. Shaylo is the Chief Executive Officer of Milla Health. He's been a founder a couple different times. He's been very successful, extremely bright and intelligent leader. He's going to talk today about what Mila Health does about himself and a lot more. So can I ask you to take a moment and introduce yourself and, and tell us about yourself and about Mila Health.
- [00:30]
B
Thank you, Scott. It's an absolute pleasure to be here. So, starting off, my name is Shelu Verma and I'm the CEO co founder of Mira Health. And at Mila Health we are building an AI care coordination platform that helps healthcare organizations automate all the manual follow up that keeps patients on track with their care. And we do it in a way without having the organizations to replace their tech stack or wait for months, sometimes years in complex integrations. So to tell you a little bit about myself, Scott, I'm an engineer by training, worked at McKinsey for a short while, began my first company in the computer vision space which got acquired by Thomson Reuters and then went on to join Amazon Web Services in their machine learning AI team in AWS in 2016 in the computer vision team. Spent five years there and healthcare has been of deep interest to me. My wife's a physician, I have two daughters, both on physician track. I tried very hard that they go into engineering, but they're both going to be physicians. And when I got a chance to build the machine learning AI team at uhg, I jumped at it. A lot of what you're going to see at Mila is built out of my foundations at AWS and the problems that I saw at uhg. So to tell you a little bit more about Mila, I'll largely focus on the problem. So when we look at large health organizations, health systems spares, there's a complex web of software tools, EMRs, practice management systems, digital front doors, revenue cycle management systems, et cetera. And all of these address the problem in their swim lane. There's of course lots of complex web of integration, but anything that comes outside of that swim lane or anything that is in the middle of all of these things tends to be a manual work. So if I have to connect with you, Scott, to prepare you for your procedure, colonoscopy, procedure, follow up after the procedure, all of this right now is done manually. And this is a large problem. It's a $334 billion problem. And the dream, the vision that we at Mila are building for is a vast, vast, vast majority of these manual work is going to be done by AI agents that are personalized, that are configured for the local settings which each clinic wants or a health system wants across the board. The question is how. And at Mila, we firmly believe that the method to that is to provide a set of tools for each provider that connects up to a larger organization that they can configure, they can integrate into their existing software. They shouldn't have to get new software integrated and it should be able to work in a safe and compliant manner. And that is what we are building Mila for.
- [03:56]
A
Thank you. And tell me the most important thing. The daughters, were you upset with them when they decided to become doctors, not engineers? And what kind of doctors do they want to be?
- [04:07]
B
That's of course not, Scott. I'm super proud of them and I'm very happy that they're taking care of some of the most important things that need to be done in our lives today. So they're both just starting their md. Like the elder one is gotten into school and the younger one has done her MCATs and stuff like that, and she's going to get into the profession and figure out where they want to specialize and what they want to do.
- [04:34]
A
Fantastic. Well, congratulations on that. And talk about what are you most focused on and excited about currently as you build out Mila Health? Where are you most focused and excited? What gets you going in the morning?
- [04:48]
B
Yeah. So you know, when I first started the company in early 2024 and then of course a lot of work I was doing at Mila was coming in from even days before. At aws, I see a couple of trends that are most exciting. So the first trend is that healthcare leaders today do not almost assume, almost want and take it as a very well understood principle that AI agents when automated, when reach out to patients in 40 different languages, 24, 7 and guide a patient through their care plan, that is the way the world is moving towards the future. So we don't have healthcare leaders question that anymore. When I first began, there was a lot of questions around, oh, should I let an AI agent connect with my patient? So that's one trend which is super exciting that people are saying, look, this should happen, let's make it happen. The second trend that I see are healthcare leaders saying, okay, that's great, but I want to make it happen in a way that seamlessly integrates in the software structure that I have. I really don't want to Replace and rip and replace anything and go through some very complex integration. Get me the ROI in weeks, not years. And so those are the two trends that we are seeing which are super exciting and we think we are on the right track with that. But yeah, that's where I see that. That keeps me going and gets me excited to deliver every single day.
- [06:26]
A
And what do you think about health care in the country as a whole? Do you have thoughts on that? What's going right, what's going wrong? How important are some of these types of things like mental health is doing to try and solve some of the gaps and some of the problems? Any overall thoughts on health care in the US and what you see going on?
- [06:44]
B
So, I mean, I think we are honestly, of course, everything has gaps, but we are super privileged to have some of the best health care in the world. I mean, I, you know, I come from India, I grew up in India. My parents went through a very difficult healthcare journey just, you know, the last few years as they got older. And I see what my wife, who's a hospitalist, delivers every day. I of course am privileged to get access to some of that. I know a lot of my friends are too. So I do believe very strongly that we have the best healthcare system in the world. Now, of course it is expensive and so I think the challenge here for us is to keep the quality, in fact even improve the quality, but do that in a way where we can manage our costs down. And a lot of these principles of how we manage our cost down are reasonably well understood and we've tried to embody them in value based care, which is early detection, making sure patients have access to care, making sure that we are doing everything to help them adhere to care and catch problems early. The challenge has been a lot of all what I said to you today, right now is all, every bit of it done manually. It's always a call to me to say, hey Shelu, you need to set up your annual wellness visit or you need to set up, you need to be set up for your, you know, colonoscopy screening. And I think that is an area that we should be able to solve now with artificial intelligence, with agents that are personalized for a patient in the language of their choice, that understands their healthcare journey and guides them through that process and helps them stay on track with of course, guidance from the provider.
- [08:35]
A
And take a second, because I love what you said. I think people don't often give the US system enough credit. It's got great challenges, it's expensive but we're trying to take care of 350 million people. When people compare it to a country with 10 million people, of course, it's very hard to make it work the same as a country with 10 million people. When you compare it to the two largest countries in the world, China and India, and you have that perspective of having been back and forth, you sort of can see, one, it's a lot easier to provide care to 350 million people, and two, we have it pretty good, even if imperfect. Is that a fair assessment?
- [09:09]
B
Absolutely, 100%. In fact, I would say the. The kind of access that we have to care, the quality is just incredible compared to everything I've seen.
- [09:20]
A
Couldn't agree more. So you've been a founder, you worked with Amazon Web Services. You've done all these brilliant things. So take a second. And what advice do you give to other founders and emerging leaders?
- [09:32]
B
I think I would frame this in two parts. One is my first advice would be to very deeply understand the customer problems. And this is something that I learned. I'm very grateful for what I learned at Amazon, which a term that we used called customer obsession and how that translated for me at Mila was I did understand that we needed to automate a lot of these closures of gaps in care, which we could do with AI agents. But if you really understand deeply on how a health system or a peer system would adopt. We also understood that each location has its own custom requirements. Each physician's office has a certain set of way of working which is different from the next one. Even though they may all be part of a much larger organization, there is a level of configuration and customization that is needed, almost like when you bring in a new employee to join your office versus the office next door. The third being, there is a lot of complex web of software. We needed to make it something that embeds itself in the existing stack and not become very complicated to use. So my first message to my fellow founders would be, please understand the problems in not just one layer, but the multiple layers that enable a solution. And consequently the second being design the system to fit those requirements. I see a lot of my fellow founders who would take, for example, a large language model and design a system around it. And I think it's, you know, it's possible to solve that for a demo. It's possible to do a proof of concept. But to really solve the problem at scale, you need to provide systems that are configurable, that are controllable, that are safe, that can be Enterprise integrated into an enterprise stack. And so to comprehensively think of a solution that addresses the core problems that you've understood in spending time with the customers, we lose track of that and we tend to take shortcuts and that hurts us eventually in the long run. That would be my message.
- [11:52]
A
No, and I love that. So tell us, what else would you like to share with listeners today about the journey you're on at Milla Health? What else can you tell us? What else would you like to share with listeners?
- [12:05]
B
Sure. I'd like to tell you a little bit about the areas that Mila is focused on, but I think this is a general trend in the industry as well. I don't want to spend too much time on Mila, but also talk broadly about the industry of how AI agents will enable a lot of this manual work that just dogs us in this space. The first area that I'm focusing on, as I mentioned to you early on, was how do I make these AI agents consistent, safe and configurable. And today some of that is available through large language models, but not all of it. So that's an area that is of focus and an area of innovation, an area that we spend a lot of time on with the engineers. The second being of course we know it's going to happen, it's going to be prevalent, it's going to be embedded across the board. How do you make it super simple so that it works at scale? It's not, you know, a one off solution that takes months off to, to integrate. So there's a lot of obsessive thinking around how do you make that really simple? The third being how well is it working? Are patients, you know, picking up the phone? Do they like to talk to Mila? Is Mila empathetic? Is the AI agent solving the problems for them? Is it safe? Is it bringing a human in the loop? That's you know that when, when, when, when it's required. So these are things where you listen to every single conversation, Scott, and you say could I have done better? Could I change this thing programmatically? And with these three pillars I'm like super excited about where it's not just a journey for Mila, but I see this happening in other industries, even in our industry with some of my fellow colleague entrepreneurs, super excited for the world that is coming to us in the next 6, 9, 12 months. Just to share an example with you, we started working with a large payer and we started with 30,000 member outreach program where we would reach out to members and we would basically make sure that they are set up for their annual wellness visits with a provider with a PCP in the network. And they used to use a call center before that. And the call center had response rates of close to 15 to 18%. We are now reaching response rates of over 50%. And we are now going to start engaging with up to 1.2 million members from the 30,000 members. And when I listen into these calls, Scott, where we are helping an older patient who doesn't speak English, guide them through the journey. Sometimes these conversations go on for 15, 20 minutes. They talk about their kids, they talk about their lives. It is very, very gratifying. And it sort of just makes my day.
- [15:14]
A
It's amazing what you're doing. Again, we're visiting today with Shaylo Verma, CEO, founder of Milla Health. Brilliant, brilliant engineer in person and leader. Just a pleasure to visit with. You've come to us highly, highly recommended by somebody we respect greatly. So it's a great pleasure to visit with you. Thank you for taking the time with us today on the Beckers Healthcare podcast. What a pleasure.
- [15:35]
B
Thank you very much, Scott. It's a pleasure as well.