“Healthcare now is all about empowering the patients to participate in their own care” Claus Jensen, Teladoc Health

00:45

Venky:

In that context, today Teladoc really offers whole person healthcare that straddles primary care, mental health, chronic condition management and many more. And in this digital native firm, pushing the envelope and the boundary is somebody like Claus Jensen, the Chief Innovation Officer and a special guest today.

Now I know Claus for over a decade now. He used to be a Chief Architect at CVS Aetna and then he moved on to Memorial Sloan Kettering as Chief Transformation Officer and now he is in this current role as Chief Innovation Officer at Teladoc. So welcome to Pacesetters, Claus.

01:22

Claus:

Thank you very much. Thanks for having me.

01:26

Venky:

Of course. So Claus, let's start with the basics. I really want to talk about what's happening in the healthcare industry from an overall trend perspective.

From my perspective, if I look at health plans, health plans used to be just benefit administrators but they're now evolving into whole person health. And if I look at providers, providers of course used to be just focused on care management but now they're evolving into value -based care, sharing more risks, etc. And of course, CMS is doing their own things to drive better equity from a health perspective.

Just want to get your take on how you see this trend and where Teladoc is playing in this.

02:12

Claus:

It's a good question. I mean, if you look at it, what you're calling out are sort of national level trends, sort of underlying societal trends in terms of how an industry thinks about itself.

But I think it's triggered by something even more fundamental. I mean, there's a shift in people's perception as to what healthcare should be. You talked about what is our purpose as a company. It's to give people healthcare on their terms.

02:40

Claus:

The definition of what people's terms are is changing. I mean, you're seeing a bleed over from other industries. The whole notion of consumerism is quite real.

And so the one thing I think is driving the change is that people just have different expectations as to what the experience should look like. Nobody has the patience for what's effectively an experience in a bricks -in -the -mortar setting that's not convenient.

03:04

Claus:

That's the one thing. The other thing is healthcare is expensive. We spend a lot of money in the U .S. on healthcare. We're not alone. We've spent a little bit more in the US than anywhere else in the world. But healthcare is expensive. It's one of the biggest expenses that people have is one of the biggest expenses that employers have. It's one of the biggest expenses that governments have. So, to focus on value is quite real as well.

03:26

Claus:

The third one is a little more elusive. The technology, the world's technology did change. If you think about wearables, you think about sensors, you think about old or smartness, you can think about artificial intelligence.

So, the out of the possible went to a step change. If you go back 10 years and you think about what was normal 10 years ago, it went a very different place. So, the third trend I think is happening and still playing out is I'd called it Ambient Healthcare Solutions.

03:55

Claus:

Think about ambience as it's a solution that's just present in the environment around you. If you take those three together, it does drive some pretty fundamental change. And what I think we're seeing is different players in healthcare reacting to those three secular trends.

04:08

Venky:

So, now let's talk about Teladoc specifically and in that, within that context of the trend you spoke about. and you know one of your pet-peeves that I have heard from you is that you know still many people don't have a clear understanding of what Teladoc does they still think it's all about Telehealth.

Tell me in your own words what Teladoc is all about.

04:32

Claus:

I think the one sentence would be; and I'm going quote our CEO, you know Jason said, at one point in the not too far future people just say to tell it because we had to get care. Really what we're all about is eliminating this odd boundary between what the virtual and in-person. The answer is not one or the other. There's going to be things that are clearly superior in a virtual setting I'll give an example.

05:04

Claus:

If you're suffering with mental health challenges getting out of your home is hard, and for those kinds of scenarios having a virtual care setting is actually superior, to going into a you know, a psychologist office.

05:18

Claus:

There are other cases where if I need surgery, yes I have to go to bricks and mortar. If I need to have you know some kind of diagnostic done that requires some kind of lab specimen I have to go to brick and mortar setting.

So we're really out to partner with the rest of the healthcare ecosystem in building a better whole person care model and then that's a little bit different than people think of as it's a place to call a doctor.

05:40

Claus:

I mean the name of the company Telephone a Doc, that was where it started 20 years ago. The company is 20 years old, but in this day and age it's all about giving people care when they need it as they need it in an integrated fashion.

05:56

Venky:

So let's talk about a key constituent in the healthcare ecosystem clause which is; the employer groups right and obviously healthcare costs are continuing to raise every year and to almost a point of being not being sustainable and, you know in a recent survey employers said that they're increasingly looking to virtual care to kind of reduce cost and I think more than half of employers are looking at health plans to incentivize virtual care. To me this is an interesting trend because this is really what Teladoc also is able to play into that trend so, how do you see this unfold from your perspective?

06:41

Claus:

There's a couple different ways of looking at I mean you're right, that there is a push towards cheaper modes of care. What we believe is cheaper does not mean lower quality this is really important we missed a lot in clinical quality we actually have you know very robust clinical quality programs. We run what you would expect in terms of quality councils both from our patient’s perspective and from the clinician’s right. We have a patient safety council so we're doing all the same things that you would do just as a normal healthcare provider.

07:12

Claus:

So I just want to start with cheaper does not mean lower quality, it does mean that you can leverage things and resources more if you think about it just making sure that you can match up providers that have time with patients that need a visit in an optimal fashion. Do that at a national scale that's the problem and we have built actually artificial intelligence technology that helps us understand and predict what that demand is to think that's done to the one part is that yes we need to drive costs down but not at the expense of quality.

07:46

Claus:

The other way of looking at it is to say that there's really two things you're pursuing you're pursuing clinical outcomes and you're pursuing financial outcomes one can't be at the expense of the other right. You can't just pursue clinical outcomes not think about the financials, you can't pursue financial outcomes and have a bad clinical outcome.

The answer lies in the intersection between two, and that brings us to the crux of what modern healthcare is all about - is empowering the patient-health consumer to participate in their own care.

08:13

Claus:

The holy grail in healthcare is to get people to take an interest and take the right health actions to keep them on the right journey.

That actually, by the way, requires a combination of different sciences that historically we have not been able to put together. Think about it. You need clinical science because it's got to work. You need technology science because, as we talked about, it's about reach and driving costs down.

08:34

Claus:

You also need logistics. I gave you the example of matching up the demand for visits with the providers that exist across the country of the world.

You need behavioural science, because behavioural science would allow you to drive a meaningful engagement with the health consumer and help them take the actions they have to and find the data science because there's a lot of data points that you can use to think through what are the best actions and not the least, which ones are going to resonate with the people you're having a conversation with.

09:04

Claus:

On my terms, it doesn't just mean I choose the channel and the time. It also means I make my choices as to what I'm willing to do in my life and what I'm not. That’s a choice we all make.

09:18

Venky:

That's a great segue to the next point I wanted to really bring up, Claus, which is one is, of course, you're a digital native firm. Number two, AI is obviously front and centre of everything nowadays. Now, I have heard that Teladoc publicly stated that you have over 600 proprietary AI models to even drive the care continuum. And even a simple thing like what you just spoke about, which is matching the patient to the right provider. You have some 16 models that you use to drive that to ensure superior outcomes that you mentioned, apart from productivity, cost, etc.

09:59

Venky:

Please unpack this for us. What are we talking about?

What are these models?

How do they actually help drive all the things that we talk about in healthcare, quality, access to care, and of course, cost?

10:11

Claus:

So all the rage right now is about Generative AI. That is important, we can talk about that in a little bit but the more traditional modes of AI which would be predictive, prescriptive and then it's the interactive in terms of chatbot type solutions etc.

But let's just talk about the first two, if you can accurately predict the logistical aspect of providing care, if you can accurately predict what are the messages that may entice people to take the right action, we talked about healthy actions and my responsibility as a health consumer for my own health.

10:51

Claus:

If you can prescribe this is the care journey that is meaningful in the context of the dialogue you add with, between yourself and your clinician, these are all things that are meaningful in terms of operating healthcare at a national scale, so as you mentioned we got, I think the number now is about 90 different models that live in those different dimensions - from logistics to behavioural health to just engagement.

11:19

Claus:

We all get blasted by email all the time. How I do actually know which emails are relevant to send to someone and which are not?

That's a problem that has been plaguing us like forever and AI models help because they're just very good at inferring non-obvious facts from lots of data, when you have 90 million members and that's actually the number.

11:40

Claus:

So we serve that many Americans, you have a lot of data, we have billions of data points that we can throw machine learning algorithms after and it actually does matter in the end.

The objective here is not to replace the existing system, it’s to power up all the good forces in healthcare to deliver a better experience, and sometimes I talk about this notion of solving the village doctor paradox.

12:05

Venky:

What is that? I was going to ask you that. It's a fascinating question.

12:09

Claus:

It's an attempt to put words to the following conflict, I want specialized sophisticated healthcare. Look I worked for two years for one of the premier cancer centres in the world, and I know how good they are. If I ever get cancer, I'm going to go to that kind of institution. I want that.

But I also want all the people that help take care of me act like a single village doctor. Like we had in the late 1800s, I just had to go to my village doctor and he or she would take care of me.

12:41

Claus:

And for a century, we accepted that I couldn't have both.

Because the specialization of healthcare that gave us the academic medical centres also splintered the experience, the medical record and everything else. That's right. What if that's no longer a conflict?

What if you can actually put the pieces back together better?

What if by taking all these five sciences I talked about, clinical science, technology science, logistics, behavioural science and data science, what if we could actually put the pieces back together in a way where it feels like a village doctor experience, but I still have access to modern healthcare?

13:16

Claus:

To me, that's what it means to get care on my terms. Because I don't want to give up the quality or the sophistication, but I do want the experience that the village doctors gave me. That's the paradox.

13:27

Venky:

Fascinating. Fascinating. I really want to dig deeper on this, but with a different lens, mental health. Now, obviously mental health is a major problem in the country.

And if I think about a firm like Teladoc, it's not, and mental health is not just about convenience is important, but more importantly, matching the right therapist to the patient can dramatically have an impact on the outcomes that just clinical or cancer can deliver.

14:00

Venky:

So how are you looking at mental health and overlaying AI and the proprietary models you spoke about that help you match in the context of mental health?

14:14

Claus:

That's a fascinating problem. What you can't do is not give people choice. Think about it. You can't actually disenfranchise the person that's looking for help from making a choice. So going all the way to a model where an AI algorithm decides this is your doctor is not the right solution.

14:39

Claus:

Okay. On the other hand, when you're running a large virtual care practice, you also can't show 200, 300, 400, you know, different clinicians that you might choose.

So what you need instead is the right amount of pattern matching. If you can get people to say a little bit about themselves, a little bit about what matters to them, maybe even know from some of the information that's just contextual, like age, demographics, those kinds of things.

15:09

Claus:

We know something about you. We know what kinds of providers that you would normally might have a preference for because we can see what the results are.

And we'll present you with choice, but it will be a choice that's filtered from the perspective of what we know about you. And you can match that up with the logistical side, which is we also know which providers have time.

15:30

Claus:

Because for a given mental health professional, there aren't so many patients taken. So our ability to track what's the capacity, the idea for what do they have time to take on more?

Do they have availability in the near future? If I'm a new patient, and that's what we're talking about, you want to, if you have someone that works, you just keep scheduling with the same provider. But when you're looking when you're a new patient, there's a timeliness to it. When you're struggling with mental health conditions, you can't wait a month or two or three. I mean, that's just not where you're in the moment, you're willing to make a commitment and schedule an appointment.

16:04

Claus:

So the last component to this is, and I need to be able to schedule an appointment now. Even if the appointment is three days out, four days out, five days out, a week out, the scheduling needs to happen now. So the ability to take that scheduling aspect, the understanding of who you are, and then the ability of, you know, which clinicians actually have the ability to take on new patients.

That's the sweet spot in terms of how you still give people choice, but you give people choice that's tailored to where they're at in their life and in their needs for care.

16:40

Venky:

Wonderful. Now let's switch gears, you know, and you did mention about Gen-AI briefly, and you did make a public announcement recently where you said that you're going to bet big on that with a primary goal of automating all clinical documentation, improving medical data, improving patient convenience, and of course, make sure that providers are liberated to focus on the patient rather than all these, you know, administrative tasks.

17:11

Venky:

So tell us more about it.

What's coming up and how are you orchestrating this whole thing so that the whole experience is dramatically better, and hopefully, you know, it'll feel like a village doctor that you spoke out about.

17:28

Claus:

So it's a fascinating problem, because the notion of having an automated scribe for a clinician is not new.

I mean, if you look through the history of healthcare, you had assistants that were the scribes, right, then you went to electronic medical record systems and, you know, they were intended to help. You have different AI types of solutions.

What makes generative AI a little bit unique in that space is exactly that it's generative. So it doesn't just take bits and pieces of something that other people wrote.

You can actually analyse lots amounts of data and synthesize.

18:06

Claus:

This is the condensed version of what was said. We have had solutions like for years that could take a dictate and turn into texts, feeds to texts.

That problem has been solved. What it doesn't solve is, are you condense a 10 minute, 15 minute conversation to these are the things that matter? So what makes generative AI unique is, you can ask it to create a condensate of the conversation and combined with other types of AI models, it can start providing guidance on which aspects of the conversation are most important.

18:41

Claus:

So to us, that's a step change in what you can do. Just scribing isn't actually objective. The objective is to get to the meaningful synthesis of what was discussed, because that's what you need to keep in the record.

18:54

Venky:

So that's coming up in the future?

18:56

Claus:

That's coming up in the future. There's no question in my mind that that is coming and now we are still early days. We just have to remember that generative AI is a new class of AI models.

Still learning how to do it. It came out of the large language model space, that's a technical term, but that's what it is. Large language model is, take a whole bunch of stuff, of different kinds, audio, video, text, et cetera, bundle it together and let a machine learning algorithm learn from it. We never had the computational power before to make large language model.

It's insanely computationally intensive.

But still learning how to use it.

19:31

Venky:

Okay, okay. Now let me switch gears, Claus. I want to talk about your business expansion.

From an international perspective, you've been focused on it. You had some good success in the UK, Canada, and these are fairly large, complex, nationalized healthcare systems. How are you thinking about it in terms of tailoring your solutions that meets the needs of these kind of, you know, very unique needs of some of these large national healthcare systems?

20:08

Claus:

So there's the obvious, the somewhat obvious, and the not so obvious, mate, there's three parts. The obvious is language, regionalization. There are just different ways that you have to present the experience. So that's the obvious part.

I mean you build technology that can speak multiple languages. We know how to do that.

The somewhat obvious is the payment models are different. So things like claims, billing, you know, how do you tie into the payment system?

You actually have to do work to tie into whatever, you know, national scheme exists for that.

20:44

Claus:

The not so obvious is not every country has the same view on what constitutes standard care. They don't have the same view as to what lab tests should you get for routine exam. They don't have the same view on, you know, what might constitute the, you know, threshold levels of certain test values before this becomes a problem.

21:08

Claus:

So when you're doing real care and when clinical quality matters, you actually have to tie into the backbone, the clinical backbone and make sure that not only do you integrate with, you know, the lab infrastructure, your prescriptions, et cetera, that has to get done, but you also have to be sensitive to the fact that the medical practice is subtly but importantly different from country to country.

21:32

Claus:

So we're very careful in our international expansion to make sure that we roll out services where we know we understand not just the intrinsic clinical quality, but if you also understand how that measures up to the national standard for what good care looks like.

21:48

Venky:

And now do you see regulatory constraints of a unique by the country?

21:53

Claus:

There are. Yeah. You know, every country has its own privacy legislation. Every country has different provisions around how you treat PHI.

And by the way, every country has different regulatory constraints in terms of what type of provider can do what type of service.. In the US we have the notion of registered nurses, APPs and full -fledged physicians and there are rules actually state by state, not even nationally- state by state in terms of what you need to do. Multiply that at a sort of worldwide level and you just get lots of different regulatory constraints in terms of which type of provider.

So that's a logistical problem right there.

22:31

Venky:

All right. Thank you very much Claus. I appreciate the very comprehensive presentation around trends on where you see from a healthcare perspective.

You spoke about AI and how you are leveraging AI to drive better clinical outcomes and you spoke about Gen-AI, how we are trying to use Gen-AI to really automate clinical documentation and then synthesize that information to be available to the provider at the point of care.

22:59

Venky:

Then you spoke about international expansion and the very successful expansion that you're focused on the tricky piece in that in terms of the challenges that you have logistically but then how you're conquering them and driving global expansion.

So good luck on that and thank you for doing this today.

23:14

Claus:

Thank you my pleasure. Thanks for the invitation.

23:16

Venky:

Absolutely.