Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#10 - Jonas Albert - Digital Therapeutics and the Digital Transformation
181 Plays11 months ago
Ever heard of digital therapeutics? Most people haven't. I talk to Jonas Albert.
Jonas is a market access and strategy specialist for digital health markets, with a big focus on digital therapeutics. He is currently working as managing consultant for fbeta, Germany's leading consultancy in the area of product conception, evidence generations, general market access and price negotiations for reimbursable digital therapeutics.
We cover many topics starting with what digital therapeutics are and where the innovation lies. How do different countries approach the process and what are business opportunities in this area.
Enjoy
Recommended
Transcript
Introduction and Guest Overview
00:00:00
Speaker
Hey, what's up, everyone? This is Ajmar Zavari, and welcome to another podcast episode. Our guest today is Jonas Albutz. In this conversation, we talk about digital therapeutics and the digital health transformation that is currently taking place. Enjoy.
Digital Health Landscape
00:00:30
Speaker
Hey everyone, and welcome to another podcast episode. If you're new here, my name's Ajmal. I'm a neuroscientist and entrepreneur. On this podcast, we explore the links between science, technology, business, and the impact they have on all of us. Today, we talk to Jonas Albaats. Jonas is a market access and strategy specialist for digital health markets with a big focus on digital therapeutics.
00:00:55
Speaker
He's currently working as managing consultant for F-beta, Germany's leading consultancy in the area of product conception, evidence generation, general market access, and price negotiations for reimbursable digital therapeutics. Jonas has a background of six years in academic research on a European level, investigating digital health business models in European healthcare systems, as well as diversified in the healthcare system, including working in the management of a German university clinic
00:01:24
Speaker
and as a paramedic for the German Red Cross. The list goes on and on. All right, enough background. Let's get into it, shall we?
00:01:35
Speaker
All right, welcome to another episode. It's a beautiful day in Berlin. The sun is shining and we're in beautiful Schoenberg. I was invited to the F beta offices today talking to Jonas. Hello from the beautiful F beta headquarters in the heart of Berlin. It is beautiful. It's very a bit empty today, but with this weather, I can understand. Definitely.
00:01:58
Speaker
Now you and I, we first met while I was working at Lindera. I was heading the clinical team doing, well, basically what a clinical team does, entering this new world of digital therapeutics. It was something that I was always quite interested in. I thought, why isn't this not, why is this not happening in the industry?
00:02:21
Speaker
And I was very surprised that things were happening in Germany. Everyone was, I think. But to set the stage so that all listeners can understand what we're talking about, can you put in your own words, what are digital therapeutics?
Understanding Digital Therapeutics
00:02:40
Speaker
Yeah, that's a good question because I think nomenclature in the digital health sector is a little bit underdeveloped and sometimes people talk about things with certain labels without offering a specific definition and that makes it sometimes difficult to follow or the different expectations or also things are put in specific boxes just because of specific terms.
00:03:05
Speaker
I think quite important to talk about this, digital therapeutics in general are interventions of digital nature. So a therapeutic intervention that is mainly reliant on some sort of digital technology.
00:03:25
Speaker
Of course, how narrow you want to define what is this mainly relying on or what is a digital function, that can definitely be argued about. But in most cases, and because this previous definition was quite abstract, in most cases the
00:03:43
Speaker
The thing that we see in practice when we talk about digital therapeutics, and also what, by the way, the majority of regulators think about when they talk about digital therapeutics, is that you have some sort of software that gives you as a patient certain instructions. And if you follow these instructions, you do better.
00:04:01
Speaker
By doing better, I mean you get healthier or you get sick slower or you stabilize your current state of your health. It can also be used in prevention sometimes, although then we have a bit of a nomenclature problem because
00:04:16
Speaker
a therapy by definition is not prevention, right? Because you have to be sick in order to get the therapy. And that's an interesting point because a lot of people mix that up. But at the moment from a regulatory but also from a nomenclature perspective in the scientific communities talking about digital therapeutics, we really talk about we have somebody who is sick or is on the verge of getting sick. And we help this person to either stabilize their state of health or to improve their health or a relevant health outcome.
00:04:43
Speaker
If I would have to classify digital therapeutics in the overall sphere of digital health, I think it's quite important to differentiate this from something that's telemedicine. Telemedicine being the direct or indirect interaction with a healthcare provider, preferably a physical one. Maybe in the future it might be AI-based healthcare providers, but for now it's still physical ones. Who knows? Let's see, who knows.
00:05:08
Speaker
But also other things like in general calling something digital health or e-health, that's also a word that has been thrown on quite a lot. It's getting a little bit less frequently used by now because I think in terms of e-health, we mainly think about
00:05:25
Speaker
digital infrastructure used in health. Like, for example, electronic health records, that would be something that I associate with eHealth. But for now, I think digital health has been the umbrella term. And then two of the most prominent children of this umbrella term is digital therapeutics, standalone solutions that impact patients directly by the patients interacting with them.
00:05:46
Speaker
And on the other hand, telemedicine, the digitally supported or technologically supported remote communication or interfacing between patients and healthcare providers. Right. Which when you really push it to the limits, what it can be, I mean, robot surgery is a form of telemedicine when you think about it. Yeah. Although then it's a different, it's an interesting question. Where does the digital therapeutic start if the robot is supporting the surgeon? So I think there's an argument to be made there, but that's more of an academic discussion.
Regulatory and Reimbursement Framework in Germany
00:06:16
Speaker
Yeah, I think so too. I think it's definitely an academic one if we drill down to that level. Exactly, completely. I already have to apologize in advance. We might be jumping around a lot. Sometimes that can happen. I don't want this to be just a bullet by bullet question and answers conversation. Well, I can't call that conversation. But you mentioned that it's an intervention based on digital aspects. And you also mentioned
00:06:46
Speaker
a human involvement and that can change things. How does that go? Are then digital therapeutics exclusively standalone software or what happens? I'm just now giving an example. Let's say you have an app that helps you with taking your medication.
00:07:08
Speaker
So there's medication involved, but does the app still count as a digital therapeutic? That's a super interesting question. And in order to answer this question, we have to go to the core of the topic. Why do we even care what's a digital therapeutic and what is not? And the answer is reimbursement.
00:07:24
Speaker
We need, we have, we always have to ask this question. I mean, of course there, if we would be just purely academics, we could maybe delineate a definition or maybe might speak some questionnaires, but for the moment, for us in the digital health space in Europe, especially in front of the recent legislation in Germany, it's quite important. What is the regulator thinking what digital therapeutics are if they want to have digital therapeutics reimbursed? And that's why we have to adhere. And in that regard,
00:07:51
Speaker
The German definition would be, for example, it has to be the standalone kind of efficacy. It has to have this standalone effect. And there can't be medication interfering with it. And also there can't be a human cannot interfere with it. Now, the reason for this is obviously not because we don't want people to get healthier, but it has something to do with the reimbursement of services a human is performing in the healthcare system in Germany and
00:08:20
Speaker
things, medical devices, whether that's physical or medical software are performing. And they want to separate this or they have to separate this. And that's why they needed this narrow definition of, okay, a digital therapeutic has to create its value by itself. The example you mentioned with the pharmaceutical intervention kind of being combined with a digital therapeutic. I think that's the one exception that could work.
00:08:45
Speaker
Because if you would test this digital therapeutic, you would see how good is an individual taking their medication without the digital therapeutic in place and how good are they taking it with the digital therapeutic in place. And of course, we are thinking always on one pill a day or something where it's quite trivial, but if we have
00:09:03
Speaker
I don't know, demand-based or where the patient can choose the specific dosage or something like this, then it gets more complex and then you also have better effects when you take the medication right amount at the right time, right circumstances, eating before, after, left, right, I don't know, taking the necessary precautions. And then that would be then a clear effect where I can say, hey, that's the effect of the digital therapeutic and I can isolate it. And then most regulators I think would agree with that in that regard.
00:09:33
Speaker
And that's, I think, something that a lot of people don't understand when they look at regulation for digital therapeutics. It's oftentimes not about barring entrance for combined or integrated solutions. I mean, everybody likes when humans are involved. It usually has something to do with isolating the effectiveness of the digital therapeutic so that you do not have this. And that's what the German regulators are always afraid of.
00:09:57
Speaker
backdoor reimbursement of human services who should not be covered under the reimbursement scheme for digital therapeutics. Yeah. And if we take it maybe a little more academic again, I guess as soon as the human is involved, you can make the argument the software is purely supporting infrastructure to deliver the service rather than a standalone.
00:10:21
Speaker
Which then would be telemedicine again. Which then exactly. Then we're back to telemedicine. Exactly. And that's always the difficulty. Like when it's super interesting to see how physicians, doctors who practice in clinics or somewhere else, how the way they are talking about digital interventions, they always revert back to the point, I want more information, which is great. I mean, that's their job. And I'm really happy that the majority of physicians is so
00:10:45
Speaker
keen on having more information about their patients and also thinking about how can I improve the care that I deliver. However, that's not really reimbursed. Yeah. And I mean, it's also not the doctor's job to think about how things are reimbursed. I mean, they're in charge of taking care of patients, right? Allegedly. Allegedly most of the time, I think. Dubio por eo, right? Right, right. But in general,
00:11:12
Speaker
It's not this what digital therapeutics are said to be. They are interventions that can have, only in the direct interaction with a patient, that can have or that have a good effect on their health. The nice thing is you can have
00:11:31
Speaker
types of therapies, where you would usually would say, oh, it would be impossible to have this high frequency of interaction with a patient. But if the software in your pocket is interacting with you and you can take out your phone anytime you want during the day, you can have a higher frequency of interaction than with any healthcare professional ever. Because the only thing that could come close would be a private nurse or something, something which is impossible to pay for in a healthcare system.
00:11:56
Speaker
And I think that's where the strength lies. So it's not only a limitation to stand alone, but it's also an enabler for certain technologies. But it also pre-selects the use cases where you can have the most impact, of course. And I think finding these sweet spots is quite important at the moment.
00:12:09
Speaker
Plus it's also, or I should say, it can also be more tailored to the person themselves, which is sad. I mean, you can't do that with, you can only do that with pharmaceuticals to an extent. Yeah, to an extent, exactly. But even therapy, I mean, if healthcare professionals, and I'm not talking only about doctors, I'm talking about physiotherapists and psychotherapists.
00:12:30
Speaker
And everyone, they have such limited time currently to engage with their patients. And of course, personalization is limited. It would be impossible to do anything else. And that's something also, obviously, it's also quite important for me to think about what are the positive aspects of this and what are the defined strengths, because oftentimes we like talking about limitations or problems with certain things, but usually two sides to any aspect of this.
00:12:59
Speaker
Yeah. Then let's talk about another aspect of digital, which if we think about software, you mentioned the smartphone. We all know, I guess, apps. Everyone knows apps. That's why we have now such a massive storage capacity on the phone to accommodate 50 million apps because there's an app for everything.
00:13:24
Speaker
I mean, when it comes to apps or I mean, there was a dot com boom, like begin 2000s, then there was the app boom, I guess. Do you think I mean, it's it's it has been a craze. People call it maybe even a hype cycle. Do you think it's going to be different with digital therapeutics?
00:13:47
Speaker
Yeah, I think so, definitely. Because it's not necessarily a consumer tech product like all these other big hype cycles have been about. And I mean, technically, if you want to talk in hype cycles, the hype cycles for digital therapeutics in the US is already over because peer therapeutics, the biggest player with the biggest investment in the game,
00:14:09
Speaker
went bankrupt and they went out of business and then also investment in digital therapeutics has heavily stagnated. I mean, now it's coming back, but also there.
00:14:20
Speaker
The one thing that is protecting this from a bubble or from a hype cycle, and that's really specifically European development, is the fact that there is reimbursement here. And these are not consumer-based business models or business models that are mainly based on catering towards insurances in the US, which are also acting like private businesses, not even remotely as much regulated as they are in Europe. And the thing is what we're currently seeing is a much more
00:14:49
Speaker
sustainably growing market. It's not like super high, like high end or high stakes, hockey stick growth. It's more like, okay, we're getting a, like substantially, we're getting a sustainably more and more players every year entering the market. But we're also seeing that prescription rates are stable as opposed to subscription rates, which you would have in B2C environments. Yeah.
00:15:14
Speaker
This stability and the steady growth might not be the sexiest thing for any type of investor who wants to make a super quick growth in two to three years. But the midterm, five, six, seven years growth for companies who have dedicated business models to reimbursement as the main revenue source, I think that's definitely the future. And they're also much more stable in terms of economic situation.
00:15:38
Speaker
Out of pocket spending, you can cut, especially for consumer apps, but the German health insurance, they will pay. If this is validated, they have to pay. If doctors think that this is a good thing, they will prescribe it. And that's always what I always say. Do you have hype cycles in pharma?
00:15:58
Speaker
for some specific products, like for example, for the GLP-1 inhibitors, the weight loss injections right now. But in general, you would not necessarily talk about hype cycles in a pharma. It's an established industry with a high margin, with extremely high upfront development costs. That's clear.
00:16:15
Speaker
And I think the digital health, especially the reimbursement based part of the digital health scene at the moment, the digital therapeutics, they are a market that's much closer in its behavior to the pharma market than to the tech markets. And we always compare them with the tech markets.
00:16:33
Speaker
It's always out-of-pocket payment for apps, for weightless, for what? That would be crazy when you think about it. Just imagine somebody has like for, I don't know, like Bledeker theaters or something like this. That would be ridiculous to treat them like an out-of-pocket consumer product. For sure. Yeah. For
DIGA and DIPA Pathways
00:16:49
Speaker
sure. Oh, I didn't know that about this, the biggest player in the United States. So that's quite interesting. I mean, we'll come back to...
00:16:57
Speaker
the differences between Europe and the United States, and also tech development, innovation, et cetera. We'll come back to that, so no worries. I'm very curious what you're going to say, but let's drill down a bit more on the reimbursement part. I mean, the DIFOBE, the Digitallefazogungesetz, was passed in Germany.
00:17:20
Speaker
And it entails two specific aspects when it comes to the listings of digital therapeutics. One is the DIGA, the Digitale Gesundheit zan wendung and the DIPA, the Digitale Pflägeran wendung. So for simplicity, let's just start with the DIGAs. Can you tell us what exactly are they? I mean, except for already digital therapeutics, which we know, but
00:17:44
Speaker
How are they supposed to work? What's the goal? What was the plan? And have there been adjustments along the way? Most likely there were. And I mean, there were a little bit. I mean, just as with any good development, there should be, let's see how this goes and let's fix what's not working.
00:18:04
Speaker
Well, first, of course, the big thing, Germany has been first in something in digital. I think the first time ever, we did it, we had a legislation. We were the first country where there was a transparent and very easily explainable track to obtain reimbursement for a digital intervention.
00:18:26
Speaker
as a dedicated reimbursement track. So not like trying to go over existing tracks for medical devices that happened before. So DIGA is, that was already the answer to the first part of the question, DIGA is a reimbursement pathway.
00:18:40
Speaker
for digital therapeutics in order to be collectively reimbursed by statutory health insurance in Germany. So for anybody of you listening who are not German, Germany has a statutory health insurance system. You have mandatory health insurance and I mean you can opt out to private insurance
00:18:59
Speaker
if you earn above a certain amount of money, although that's only 10% of the population who's opting for that. So if 90% of all Germans are mandatorily statutory or publicly, however you want to call it, health insured. And any of these
00:19:15
Speaker
publicly insured patients can or has the right, that's how it's written in law, has the right to use DIGA if they are listed on the, and it's very German as an official list for the DIGA. All the cliches are true. Germans like the organization and the lists and the regulation. So obviously what did they do or did the ministry of health do? They said, hey, we have these apps and they are really, really the capability to make people healthier.
00:19:40
Speaker
But not in a preventative right away, in a treatment type of way. So true digital therapeutics as established in the beginning. And they said, hey, we will make a central register. That's that list. And this register will be supervised by the Bay Farm. The Bay Farm is the federal Institute for Pharmaceuticals and Medical Devices. Who ever can say this, we'll get the listing. No, I'm joking, of course.
00:20:05
Speaker
That would be great. That would be fantastic. That would be fantastic. That would really kickstart things. And yeah, it's actually quite interesting. Why is the Bay Farm doing this? The Bay Farm is a federal institute. It's directly connected to the Ministry of Health in Germany.
00:20:21
Speaker
And that's kind of a new thing because usually decisions whether or not something was reimbursed was done by the so-called self-managed bodies of the healthcare sector in Germany because usually we have a separation between the state and the actors in the healthcare system.
00:20:37
Speaker
And that means that the Federal Ministry of Health and the Parliament, of course, they make the rules. They make the social code books where all the rules are written down. But they do not have direct control. Like, for example, the NHS would have in England over all the stakeholders in the health care system. So we have these self-governing bodies. The one is the Federal Organization of Statutory Health Insurance, GIGAFAUSPITZENTHABAND. And the other one, the other, or the most important other one, is the KABEFAU KASENERZLIGI BUNDERSFAR EINIGONG. That's the Federal Association of the Statutory Insurance.
00:21:07
Speaker
physicians, the doctors basically. So doctors and insurers, and they have a gremium together with a couple of independent experts, and usually they were the be-all and all of what is getting reimbursed and whatnot. Of course, you can go in much more detail, which I'm not going to do. It's very interesting. Germany has a quite peculiar healthcare system. And they had at least the government at that time when they came up with the DIGA legislation
00:21:31
Speaker
they kind of had the feeling like the existing decision-making bodies were not really willing and also with their processes that they had not really able to incorporate standalone digital health interventions into the reimbursement catalogs. And that's why they devised this new process and gave it to the Bay Farm, the Federal Institute, which was already tasked with greenlighting pharmaceuticals and medical devices, but more from a surveillance standpoint, basically, compliance surveillance, that sort of stuff.
00:21:59
Speaker
But they just said, hey, give it to them and, and that's why you will hear this often together with Degas, make it a fast track. And that's the actual, I mean, the second revolutionary stuff. The first revolutionary stuff was dedicated reimbursement for digital health under certain circumstances, central register, as soon as you're in the register, you
00:22:17
Speaker
Um, you, you are eligible for reimbursement can be prescribed basically. Sorry. I forgot to mention this earlier, the digital therapeutics that DIGA, um, uh, they are intended and they are currently being prescribed as if there were a pharmaceutical intervention. They even have the pharmaceuticals and trial number pizza. And so they have a pharmaceutical register number, same like drugs, which is funny actually, when you think about it. Yeah. And yeah, they made, should I go into the fast track or do you want to talk about the prerequisites first?
00:22:47
Speaker
Yeah, let's go to fast track later because I think that's quite interesting. It is quite unique as well. Yeah, but before we go into that, we also have the D-pass. True. What is up with those? Yeah, that's a very good question. I mean, we have the framework that's there, equally revolutionary. So just as an explanation, D-pass. Digitale pfläger an wendung, so it's digital nursing care applications.
00:23:11
Speaker
But they were, I mean, technically the idea behind the DPA was, hey, DIGA worked well or is working well, or we really want to do that and we really want to double down on digital care. So how about we do something for outpatient nursing care? So you are in a home care situation and you can use an app to do that.
00:23:30
Speaker
It was basically the same idea with DIGA, but let's use a setting and let's lose a purpose. So instead of improving health, it was about improving nursing care, relevant or long-term care, relevant outcomes basically. But otherwise they're completely the same thing. Some of the requirements are different, but I think we'll go towards that later.
00:23:53
Speaker
Unfortunately, they left out a crucial part, which is the fast track. There's no fast track for Deepa and that's the reason why there are no Deepa yet, unfortunately.
00:24:03
Speaker
That would have been definitely the next question of mine, because the DIGA listings, there are quite some already. Yeah, 56 successfully listed. And I mean, six have already been delisted. So overall 60 something who have been in the DIGA directory at some point, but the, yeah.
00:24:23
Speaker
And are the 56, I mean, just based on fast track procedure, are they already permanently, the 56 or? I think I would have to look up the exact number. I think like roughly half of them is permanently listed. Oh, wow. Yeah. Wow. It's pretty good. And also more than anybody expected in the beginning, people were like, ah, this is going to blow over. This is not here to stay.
00:24:45
Speaker
They won't make it. The requirements are too high. I think we'll talk about the requirements later as well. Yeah. But yeah, I'm going to dig out here just a, but really the fast track and maybe that's a, is that a good segue to talk about the fast track? Yeah, sure. We can talk about the fast track. Because it's this.
00:25:01
Speaker
big, huge difference between Deepa and Diga. And the reason we have zero Diga and we have over 50 Diga. And I mean, there are other reasons for this as well, like long-term curbing of that sexy of an area to invest in, which I don't understand, by the way. It's one of the biggest societal problems we have in the near future. Everyone should invest in this. For the Diga they made, because the requirements for the Diga say that you have to prove your effects in a very similar way
Clinical Trials and MDR Regulations
00:25:29
Speaker
as if you were a pharmaceutical intervention. So you need clinical evidence, you need a clinical trial. Usually, well, in almost all cases, usually it's not strong enough. Like you need an RCT, randomized control trial, the control group that's getting care as usual, your intervention group is getting care as usual, plus DGA on top. And then if you can statistically have, if you can have statistically significant effects that are also so strong that they are clinically relevant,
00:25:54
Speaker
Then you will be listed as a DIGA and you can do this the classical way. You make a big powered randomized control trial, couple of hundred patients, enroll them, do the trial, you compile the information afterwards and all the results. You go to the Bay Farm and you write your application and they check if your app is having all the quality.
00:26:14
Speaker
All the quality requirements, if it's safe enough, there's a huge list of requirements for DIGA. And then they check your evidence and then they will decide, A, is this good enough or not? And as you can probably imagine, if you are a venture capital finance startup, you will most likely not be able to raise enough money
00:26:34
Speaker
to finance a huge powered RCT right from the get-go, right? And funny enough, the politics, they knew that. So they said, how can we accelerate this, but obviously without putting any patients in danger. And that's how the fast track was born. So your product still has to be safe and secure, and it has to be medically device certified according to MDR, back then MDD, now MDR.
00:26:57
Speaker
And it has to be, yes, to fulfill security criteria and all that stuff. And it also still needs clinical evidence in order to get reimbursement, but it can have preliminary clinical evidence. So that means an unpowered randomized control trial, 60, 70, 80 participants maybe with strong statistical tendencies, but it doesn't need exhaustive proof. And if you have this preliminary evidence and a robust enough product that checks all the boxes, then you can apply for something that's called preliminary listing. That's the fast track.
00:27:27
Speaker
So the Bay Farm will review your application with your preliminary study, which you can usually easily conduct in half a year, six months. And then they will decide, will you be preliminary listed or not? Together with your preliminary data, you also hand in a thing that's called an evaluation concept. So that is a trial protocol with a couple of more bells and whistles. And this trial protocol then details
00:27:50
Speaker
how you will spend your time in this so-called preliminary listing phase, because you have one year time after you start your preliminary listing, 12 months, to conduct a fully powered RCT, to finally get the evidence to go to final listing, like the one I explained before. But during this time, you will already be fully reimbursed. So you can usually co-finance at least or sometimes completely finance your big RCT with the reimbursement money that you can get in the preliminary listing phase.
00:28:19
Speaker
And that's really the groundbreaking stuff. You can de-risk your trials. You can do smaller studies. You can redo them easier. I mean, a study with 70 or 80 patients, if you have to redo it because you maybe make some mistakes, maybe your product wasn't ready yet, this is not killing your company. Having to redo a trial with 300 participants in a couple of physical centers, which cost like a million euros or 750,000 euros, that could kill your company. And that's the big difference in Deepa,
00:28:47
Speaker
doesn't have this. And so only the companies are actually considering becoming DPUs who have the resources to conduct a fully powered trial. Just to follow up, I mean, you mentioned the preliminary evidence.
00:29:01
Speaker
Is it also possible to argue in front of the Bay Farm and say, look, this is my product. It is built on the latest and greatest scientific insights. And there have been plenty of studies that support that this approach makes sense and already effects were found in this and this scenario. Sure, it's not done in the real world yet, but does something like that also count as a preliminary evidence to then say, look,
00:29:30
Speaker
to really rehammer it home, bring it home, we then do the fast track and we follow this and this and this. You need study evidence like there's no way of doing this, just purely literature based. I mean, it's mandatory to hand in this literature analysis in parallel because the combination of this is extremely plausible what we're doing. Lower back pain can be treated with physiotherapy and video instructions for lower back pain are actually helping people to conduct the exercise in the correct way. That's the minimum requirement, but you need clinical data. You need clinical evidence.
00:30:00
Speaker
preferably in Germany. But also before the fast track starts? Yes, for the pre-study. So in order to apply for a preliminary listing, you need a pre-study with an unpowered RCT that has strong statistical tendencies. So that would usually be like you double the alpha, it's not 0.05, but it's 0.1.
00:30:23
Speaker
Or you have the vast majority of individuals in your intervention group are improving beyond clinically relevant cutoff points and the majority in your control group do not. I mean, it's still descriptive, right? And there's lots of criticism. Oh, you don't have any evidence. I mean, it's usually quite already quite conclusive, but not exhaustive.
00:30:43
Speaker
Yeah, I understand. Now you mentioned the, this sounds like a segue, but I'll get back to this. Just give me a second. So you mentioned the MDD, the previous medical device directive changed to the European Union MDR, the medical device regulation rules, I mean, except for in vitro diagnostics.
00:31:06
Speaker
It specifies four classes. I mean, that's within Europe, within the MDR. We have class one, low risk, like wheelchairs and crutches, class two A, medium risk, like a stethoscope, hearing aid, ultrasound, things like this. We have class two B, medium risk, ventilators, intensive care monitoring equipment. And we have class three, which is the highest risk, things like a hip joint replacement. Now, if we take that into account,
00:31:36
Speaker
Which class do degas actually belong to if we talk about now fast track? Because if we talk about a fast track procedure, my gut feeling would say, if a medical device is class three, yeah, no, you need the full thing. There is no co-finance.
00:32:01
Speaker
But that's actually quite an interesting topic. So originally, DIGAs were locked into Class 1 or Class 2A. So that was the original setup where the regulator said, hey, can't be maximum Class 2A. The majority of DIGA are Class 1 devices, even according to MDR, although it's a little bit funny. Some of the German, like the regional authorities that are responsible for certifying Class 1.
00:32:25
Speaker
Some of them are really okay with digital interventions being Class 1, according to MDR, but there are some regional ones, like the Lagezo and Berlin, for example, who really, really have the opinion that Class 2A is the absolute minimum for any type of medical software, which is interesting that you have different interpretations of the same European
00:32:47
Speaker
regulation in one country, which is, it's weird because it also gives you a regional advantage depending on where you are and which is responsible. And I mean, the majority of DIGA are quite low risk for now. However.
00:33:00
Speaker
since the last, the Diggie Gear, the guitar gazettes, the latest regulation, you are also now allowed to have DIGA in the RISC-class 2B, which is amazing, if you ask me. It's absolutely fantastic, because now you have options to have direct interaction with medication, for example. Or just in general, it's like, yeah.
00:33:18
Speaker
You can help populations of people who are more seriously sick than previously, although obviously that's a bit of a difficult statement. Of course, like somebody with a psychological condition is also seriously sick, although these populations are much more accessible for digital interventions because it's much closer to the psychotherapy that helps them the most.
00:33:40
Speaker
And yeah, but class 2B, and really great that you ask this, because for class 2B, the regulator said, we will not have a fast track procedure for class 2B diggers, which I find terrible. It's pretentious and I'm going to tell you why right now. It's completely the wrong approach. I disagree thoroughly. If something is class 2B according to MDR, it's safe to use and it can be sold. It can be deployed in Europe. There can be studies with it, everything. And why?
00:34:08
Speaker
Why should we make this arbitrary separation between 2A and 2B? The prices will not be significantly different, most likely. But the application is really valuable, and we saw in the past that the fast track for such a young industry with such high-risk projects, because there's not a lot of best practices, they are highly reliant on expert knowledge, because 50-something is not a big sample size, right?
00:34:32
Speaker
And now you have a higher risk class with higher market excess cost anyway because you need the 2B, you need a clinical assessment already for your MDR certification and then you can't have a fast track. That's just punishing people who take higher risk for their businesses. And it's not higher risk for the patients because the amount of hoops that a manufacturer has to jump through to become class 2B, that's the highest quality standards in the world for any type of medical devices.
00:34:56
Speaker
And you can sell them out of pocket. Patients can pay out of pocket for these things if they are applied by a doctor. They can be used in hospitals to be devices, stuff like this. But the fast track then is miraculously excluded just because we have one year of reimbursement for them. So I really don't agree. The risk is not a good line of argumentation for this, not at all. Sorry, I have a very strong opinion. No, no, it's fine. Let me just double check if I understood.
00:35:22
Speaker
I mean, to get the DIGA listed, assuming it's 3B. Sorry, 2B, yes, not 3B. There's no 3B. Like space, organ transplant, gene modification. Maybe beaming in a hip joint replacement. Yeah, exactly. We're not there yet. That would be great. Yeah, we're not there yet.
00:35:46
Speaker
But if it's, so 2B, you are classified as 2B under MDR rules, which is, and you're completely right. I mean, I worked on this as well. Yeah, right. You know how difficult it is. Yes, to get this stamp is, it's work, like a lot of work. I mean, it's also, I mean, it's about safety and patience. If we want them to be safe, that's clear. But everybody who says that is, that there is
00:36:12
Speaker
margin for error there that it's not enough clearly has not done this. Yeah, I mean, it's and it's a lot. And it needs to be it will be checked by third parties by governmental institution. Exactly. Exactly. And that's why it surprises me a little bit because if you get that stamp of approval already, which is not something you can just you can't just go to a ticket counter and just get it. That's that would be nice. But
00:36:43
Speaker
That's from a manufacturer's perspective, from a patient perspective. I want my ventilator to be properly checked. But this seems like a double whammy.
00:36:52
Speaker
Yeah, it is. Like a double whammy as in, hey, you got this ticket, which is great. Yeah, and it means you quality, safety. Yes. All the things. Yeah. But for some reason, it doesn't count if you start to stand in this other line. Yeah, exactly. But that's the big problem. And also some of the regulators not really understanding this correctly. Like even some of the politicians in this area don't understand this correctly. Safety has nothing to do with effect in the end for these things. Like the MDR2B device.
00:37:22
Speaker
is safe to use. Yeah.
00:37:24
Speaker
obviously within the realms of like, yeah, depending how you define safe, like safe enough, it can be used for what it's used, it does what it's supposed to do, and that's fine. And the only question the DIGA fast track is asking is, is this good enough in order to, yeah, to result in reimbursement mandate, a general collective reimbursement? That's the question you ask. And the worst thing that can happen is that you reimburse this thing for one year, then you find out, oh, it's not good enough.
00:37:52
Speaker
And now, company, please give me all the money, all the reimbursement money back. That's the worst case that can happen. And the worst case in the worst case is that this company is then filing for insolvency and then the German healthcare system loses an amount of money that is minuscule compared to the general spendings or other other wastes of money. And don't get me started on home on homeopathy.
00:38:14
Speaker
As long as we pay for interventions who clearly have zero value, then nobody is allowed to complain about because that's in the cell. So far, I think 2D car manufacturers went insolvent because they couldn't pay the money back. One who had a lower negotiated price and the other one because it just was a very tragic story of M sense and why migraine patients have a strong placebo effect.
00:38:39
Speaker
That's a different story and maybe best shared in different contexts as well. I mean, it's very interesting from a clinical research perspective. But yeah, that's the worst case that you pay back. I think it was in one case it was like a million euros or something or even lower. So on the grand scheme of things, not a lot. I mean, if that's the price you pay for having the most innovative way for reimbursing digital health innovations, then I mean, the system should pay it. Like the return on invest or the return on risk is
00:39:06
Speaker
definitely positive in that regard. Right. I mean, the downside is also really, because you had mentioned risk, the downside, so for the opposite direction, right? I mean, there can be a massive upside. Yeah. And the downside is not mirrored in the upside. Yeah. So that's why I, yeah, this is strange.
00:39:27
Speaker
I didn't know this. I thought they would also allow the fast track for that. No, no, it's explicitly excluded. And I don't know which lobby is responsible for this. We can probably take educated guesses on that. Maybe some payers have an interest in making this more difficult because it's also a thing like the DGA pricing is done based on effects, right? Yeah. And if you have a 2B device, which is now really allowed to support
00:39:51
Speaker
the physicians and the patients in a way that you can really, yeah, you can, you can give the patient much, much more granular, more risky recommendations overall, because it's more check than you can go much deeper in the therapy. You can be much more involved. And by you, I mean the app, the digital intervention,
Shifts in Healthcare Industry
00:40:08
Speaker
right? Yeah. And that can have stronger effects and that means higher prices. Yeah. It's quite interesting. That is quite interesting. But does that mean, so we, we talked about the, the four classes under MDR.
00:40:21
Speaker
And you said most DIGAs are Class 1 or Class 2A. Am I automatically a medical device when I try to get listed as DIGA? Well, you have to be, otherwise you can't get listed. I can't get listed. Yeah, exactly. It's a mandatory requirement. It's a mandatory requirement. So I have to jump through the hoops of getting
00:40:42
Speaker
my QMS, the quality management system. Hey, let's start talking about these great ISO certificates. 27001. Yes. Data security and privacy, ISO 13485 quality management system. This is the Holy Trinity. That's the Holy Trinity finishing with ISO 14155 for
00:41:05
Speaker
getting clinical data. Yeah, of course. To prove your effect. It's a good GCP. Yes, that's a good old GCP you have to follow, obviously, anyways. Which you should do anyways. Which you should do anyway. I mean, if you don't, I mean, your evidence is not counted as evidence. Yeah. I mean, some benefactors, I don't think follow this fully, but then you're always open for attack. For sure. Somebody from the outside can say you didn't follow good clinical practice.
00:41:33
Speaker
And that means your results should be, yeah, under most scrutiny or, I mean, in some severe cases, even not be accepted at all. That's true. Yeah. Yeah. Just so for people that don't know, good clinical practice involves, you know, you have to be transparent to your participants. There cannot be any fudging with you calling people constantly in the intervention group and not in the control group. Preferably, you should have two arms, control and intervention. If you have subgroups, maybe even more,
00:42:02
Speaker
But that you should talk to the Bay Farm about because they can depend on case-to-case. Bay Farm always wants RCTs. That's the short version. It says in the legal documentation, even in the guideline document, the LightFard, and it says that you can do intra-individual comparisons or uncontrolled studies.
00:42:19
Speaker
Do not do this. Always have a control group. Please. I mean, it just makes it easier. And if I understood you correctly, the stronger your evidence, which based on the design, the RCT is the gold standard. Do you then also have more leveraging power in the pricing negotiation? Definitely. I mean, of course, effect strength is the anchor for your pricing. They compare it with other treatments which are already reimbursed and then they
00:42:46
Speaker
They kind of form an anchor price. Like for example, some of the depression decas have an anchor price. Like they are as valuable as X amount of group therapy, for example. And I mean, some of them are significantly more valuable than several sessions. I mean, if you look at these studies of results, it's ridiculous. It's crazy. It's amazing. Like you have really super strong effects like Cohen's D almost of one, like 0.9 and really like for some depression patients and stuff like this.
00:43:15
Speaker
But I mean, on the other hand, it's also who would have thought that if I have cognitive behavior therapy validated in my pocket, accessible at all times, I mean, it's kind of intuitive that that would work well, or that maybe patients with, I don't know, like anxiety disorders, they might respond well, not having to go to a therapist, but having in their home, like the therapy step by step on an inanimate object that they use daily anyways without fearing it.
00:43:42
Speaker
There are some things that just lend themselves really well to digitization in the forms of therapy. And that's also the thing that people think like, oh, like the personal interaction is always superior. No, it's not. It's not. Sometimes it is, but in more cases than like the average person in the public might think it's really not superior. It's sometimes like, yeah.
00:44:01
Speaker
Yeah. It's sometimes the individual conversation is inferior to having something available to you at any time of the day or night or in any circumstance. Yeah. And that brings us back to, I guess, the tailor-made intervention, which this makes it much easier. And when we now, I mean, you mentioned psychotherapy.
00:44:26
Speaker
The stigma is still there, no matter what people try to say. The stigma is still there. And even if you try to, and this is not, I'm not trying to dunk on the therapist. I'm just saying the system is so overloaded with people that actually need the help. They cannot get a spot, a therapy spot.
00:44:53
Speaker
I mean, last time a friend of mine, he really needed one. They said, look, you should call this emergency number and then you get rated higher. You get into a higher priority. So you get access faster. So he did. And, um, the earliest spot was six to seven months. I mean, that's, that's.
00:45:15
Speaker
That's a lot. It's a lot. They're just not enough space. It's not enough cousins. It's also one thing that the public health insurers kind of control, like how many psychotherapists per individuals. And just because society became more aware of mental health, a lot more people use it. Of course, that's an own discussion, whether that's desirable or not in its own, of course. But in general, I think one of the nicest indications is that a lot of general practitioners and psychotherapists are prescribing DIGA in the psychological field.
00:45:45
Speaker
also during waiting time. So they know. Oh, that's good. And yeah, that's great. And it helps patients. And it also takes strain away from the system. And if you get patients that are on the threshold just over it, if you get them under the threshold that they have something that they can manage themselves with without them directly needing psychotherapy, that's something good. Even the psychotherapist, because the very American argumentation would hear, oh, they would never do this. They would take away their clients.
00:46:12
Speaker
They're happy when the waiting lists are shorter or at least appropriately sized, because they also don't want to feel like there's such an overwhelming shortage of especially psychotherapists.
00:46:23
Speaker
that, I mean, most of them are extremely happy to see this, which is pretty, I don't know, it's great. It sounds unintuitive in the beginning, but it's really nice. And that's one of these rare triple or quadruple wins. So society wins because we have more people getting the appropriate need of care. The individual wins because, I mean, maybe some people enjoy or think like psychotherapy is the good long-term thing for them, but the majority of people just want to get to a state where they don't need therapy or can help themselves or feel empowered again.
00:46:53
Speaker
feel better exactly feel better exactly just they're doing better all over and and also the psychotherapist win because they have like a low amount of pressure and also health insurance is when because you have people with a more stable mental health overall and that's
00:47:09
Speaker
Yeah. And I don't think that's communicated often enough that we have these triple or quadruple wins or whatever could even include the, like the economy in this equation as well, because you have a low amount of sick days. And if you look at this overall, it's a net positive for society. And I think that's the thing which is really nice about digital therapeutics in general. Then it has this, well, at least, I mean, we don't have definite numbers, but from what we can observe right now, it is a high likelihood that it has this net positive effect overall.
00:47:36
Speaker
Yeah. And this brings me back to, you mentioned the effect size, so Cohen's D, normally between zero and one, not normally, where, you know,
00:47:51
Speaker
0.5 so already that's medium effect. If you like already great, which is already great I mean don't get me wrong, which is I mean if you get that in pharma, this is like amazing It's wow a small effect 0.25 and I went through all the diga listings before before talking nice Yeah, I went through all of them just to check what they did when it came to their evidence and
00:48:14
Speaker
And man, I was surprised to find effect sizes that go from 1.5 even up to two. And that is insane. It's really insane. That is insane. And that basically means you have to do very little
00:48:30
Speaker
To get a lot out. Yeah, exactly. That's basically what it's 90 days interventions. Yeah. Yeah. That have a profound effect on health relevant outcomes. And by the way, that's another thing that's Degas specific. Technically in the law, it says you can have any outcome. You can have ultra structural improvements at some point of criticism, but in the ruling practice of the beef farm.
00:48:51
Speaker
Every digger has to have a medical outcome. And it has to be, at the very, very least, disease-specific quality of life. But in the vast majority of cases, that's just a secondary factor. They have one specific endpoint that's disease-specific. Can be pain, can be a depression scale, can be blood pressure, can be weight, can be all the different things depending on the indication. And they're all having really good effects in that regard.
00:49:19
Speaker
And just to get everyone on board, that basically means if you compare between two groups and the effect is very, very small, you can still get a significant result if you have a lot of people in those two groups. So mathematically speaking, you get an effect, but let's say your pain from a scale from 1 to 10 just goes down by 0.5 points on that scale. You have it significant.
00:49:47
Speaker
like as a p-value statistical, but it is not medically relevant. Or like blood pressure, right? Or like blood pressure. You reduce your blood pressure by two millimeters, Hg. And then, I mean, that's great and can be significantly significant, but it's not clinically
00:50:06
Speaker
relevant, you do not have any meaningful changes regarding your health if you just reduce it by two. I think with blood pressure, it starts with 10 or something. But that's these cutoff points and that in combination with general indicators for effect strength. By the way, that's also one of the most important things that you should discuss with a BFI beforehand. Yes.
00:50:27
Speaker
Now, do you agree, dear Bay Farm, that this is the clinically relevant improvement that I need for the majority of my patients and all of them? Yeah, exactly. MCIDs, minimal clinically important difference. Yes. And I only urge everyone who's listening.
00:50:42
Speaker
just see the Bay Farm as your partner in this, rather than the, I don't know what you call it, the bouncer in front of the club that you have to try to convince to get in. That's not really what they do. Otherwise, you get a heute nicht. Yes.
00:51:03
Speaker
It's a reference for all the Berliners listening. Yeah, for sure. For sure. Now, I'm pretty sure I know the answer to this already, but I'm going to ask you anyway. So we talked about degas, we talked about depas. From a manufacturer perspective, which one should I go for? I mean, the shortest or simple answer is where you have the better effects.
00:51:31
Speaker
But of course, this answer leads to a lot more different things. But of course, you have a go-no-go decision for DEPA, which is, can I finance the study rate? And if I don't have any chance of financing a fully-powered RCT in the long-term care or, well, sorry, not in the home care sector and the nursing care sector.
00:51:52
Speaker
Then I can just completely forget about DPA anyways. But if I have the money, if I really have the choice, then I should really see, do I have better effects in the dimensions of individuals of need of care, like in the home care dimension, like improving their mobility or their cognitive capacity, everyday activities and so on.
00:52:13
Speaker
Or do I have a really good individual clinical medical endpoint for a disease, for a population that I can define by a disease that they're having, by a condition that they're having via an ICD-10 code. That's how the bay fund is classifying them. Right. Yeah. But what's funny, and that's, sorry, that's one business case thing. Deepa, people say, you have a 50 euro per month maximum. And that's not a lot. That doesn't sound like a lot, right? Because Degas pricing, the preliminary pricing is flexible. It's usually around 400 to 500 euros for a 90-day prescription.
00:52:43
Speaker
And then after price negotiation for Degas, it's like usually between 200 and 250 euros for a 90-day prescription, still quite lucrative. And so that's more money per month compared to Depa. But, and that's a big but, Depa is continuous. Like you use it, you don't have a single, you don't have a prescription, you just get it, you use it out of your own volition after six months.
00:53:07
Speaker
The long-term care insurance is making a spot check. I still using this so you're fine and then done. The only way stop using the Depa is when you either go into a nursing care home because then you're not
00:53:22
Speaker
no longer in the setting, or if your condition deteriorates so much that you're no longer capable of interacting with it or using it. If your degree of need of care is going higher, there are these different care degrees. And so we did a very quick early calculation on this. And the average user for a DPAR that has care degree one or two as a target group, people who can still reasonably do stuff while at home, should be around two and a half to three years. That's the usage potential per person.
00:53:50
Speaker
2.5 times 12 times, let's be conservative, not 50, but 30 euros. And then you have a user lifetime value that's significantly higher than for DIGA because for DIGA, the average re-prescription rate is currently around 20%. Of course, there are DIGA's where it's significantly higher. There are other DIGA's where it's significantly lower, but then you discount your, in order to calculate your user lifetime value for reimbursement, it's like the DIGA, average DIGA price times 1.2.
00:54:20
Speaker
you
00:54:22
Speaker
And, but that's still lower than the expected revenue from a deeper, but again, then can you mobilize them? It's a different population, not that used to digital things. So there are a lot of different factors playing into the decision-making, but that's the economic reasoning. And, but the other thing, and that's the more important thing is like, where do I have the stronger effects? Where's, where is it more likely that, because with, with regards to the technical requirements for DGA and for deeper,
00:54:54
Speaker
If you throw enough time and money in it, you will succeed with the technical stuff. I mean, there are best practice for this. There are service providers for this who are very experienced in the whole thing and who can do this really well and very reliable. I mean, it's not easy, but it's definitely not super intellectually challenging for you as an entrepreneur or for you as an investor or whoever, because you can't do it. You can't pay people to do this extremely reliably, like the CE marking. I mean, that people would be better or worse, but it will be done.
00:55:21
Speaker
and all the other technical requirements, safety, security, privacy, all that stuff, robustness. But where you can really fail is the evidence generation. That's where the most risk is associated with. Right. Oh, okay. Cool.
00:55:38
Speaker
So as a manufacturer, I need to definitely take the money into account, the upfront investment. How can I bridge it? You mentioned the fast track. That's a potential, I guess, I guess a loan structure you could even think about. Yeah. With if success, then all fine. Yeah, exactly. If not. If not, goodbye. Give me my money back. Yeah, yeah.
00:56:03
Speaker
Without interest, I hope. I don't know. I don't. I actually don't know. That's a question. No, I don't think that's interesting. Okay. Then no interest. But are there some other pitfalls that I as a manufacturer should take into account? I mean, no, not splitting.
00:56:19
Speaker
between DIGA and DIPA, we can stick to DIGA if you'd like. What about time? What about third parties that I need to need to work with that I should take into account? Yeah. So that's asking quite an interesting question because we had this big development on the market. And so I'll try to answer your question in the context of now.
00:56:42
Speaker
But in the beginning, it was really weird. You had all these heritage digital health companies that never really found good business models, but they developed their products, not with Diga in mind, because Diga wasn't a thing, right? For example, look at the Selfope platform. It was a platform in the beginning, right? We did different things. And then they carved out single pieces where they did very successfully cognitive behavioral therapy.
00:57:06
Speaker
And then they deployed it as Degas and they got the necessary clinical evidence, which was great. But you never really had products who were developed with Degas in mind because it's such, I mean, the first Degas were listed in 2020. So it's like, I think it was in April. So almost exactly four years ago. And.
00:57:23
Speaker
So what you have to take into account now when you do this from scratch is also, you can ask the early diga manufacturers, but do not necessarily follow the way on how they did it all the way through, because it's no longer representative of how the bee farm acts and how their products are developed. And if they have multiple digas, like Selfope is a great example because they have both, like they have these heritage products, but they also have newer products that were then more streamlined with diga in mind.
00:57:51
Speaker
And so in that regard, quite a good example. But then in addition, how it's done right now is we also have some projects going on now where investors go directly with the startup. They want to invest in, they go directly to a consortium of service providers.
00:58:10
Speaker
and have them de-risk the whole process from the beginning on. In the beginning, for example, we do the market access consulting in this area. A company came to us that had a product already and maybe started a study and, hey, how should we do this? Can we pivot or can we adjust this?
00:58:29
Speaker
But now it's going more towards investors come to us and say, Hey, we have this idea for this Degas. We have this, this is the core venture three, four people who are really capable and good at doing, but we already have all the support structures around it. And they raise with your service offers as part of the ask.
00:58:49
Speaker
Okay. And that's really, really, really, really great because you have so little, like you have so much less opportunities for failure because there's less miscommunication. You can communicate timelines. You can share best practices. You can inform investors how much it really costs. You have transparency. You can get multiple offers from different service providers.
00:59:11
Speaker
And that also completely changes up the timelines because you can make, you can conduct certain activities in parallel. And that's really nice. Like you start with the product, the product setup, the product development. If you maybe even have an external developer, there are a couple of very capable ones out there and they developed you guys in a very reasonable amount of time. Like we're talking like half a year.
00:59:35
Speaker
which is extremely fast and it will not be completely done by then. But the basic setup will be half a year or like nine months basically. And this time you can already prepare your CE marking, all the documentation, you can set up the quality management system while the thing is still in development. Or like for ISO 27001, that's an ISO certificate for your company. It's not product specific, you as a company has it. And all these things that usually would have to do piecemeal afterwards, you can do in parallel. And while all this is happening,
01:00:04
Speaker
The product is, I mean, the concepting obviously is the most important part has to be done, but the content is developed and implemented and all that stuff.
01:00:12
Speaker
What are we doing? We're going to the Bay Farm. We're going there with a product concept. The product doesn't have to be ready. We're going there with a study synopsis. The study hasn't been conducted or anything. So in this purely hypothetical state, we can already get initial validation from the regulator. And that's super valuable, because then you can also still adapt the product afterwards, because it hasn't been developed. Or you can adapt your clinical strategy and your timelines, if neither, right? Right, right. And this parallel,
01:00:37
Speaker
technical stuff in parallel, regulatory stuff in parallel, evidence stuff in parallel. All this goes up and then it's quite realistic that you have your product at a point after one year where you can start the pre-study and then you have basically from project start to preliminary listing
01:00:59
Speaker
In an ideal world, it rarely happens, but in an IOG, I hope nobody quotes me on this, but in an ideal world, you can make this in 18 months.
01:01:08
Speaker
Yeah, that's lightning fast. I mean, usually things go wrong. You have to pivot. Sure. The Bay Farm is completely blowing up your evidence strategy because they changed their mind on some points. Yeah. Oh, I remember those meetings. Jesus Christ. Okay. But I mean, also to be fair to the Bay Farm, they also learned and they got better and they evolve and they are really, and they're also really trying. Like it's not, maybe sometimes manufacturers get the feeling that they're just, oh, the Bay Farm is just looking for a reason to say no.
01:01:34
Speaker
They're not they became really like they're developed into enablers. But of course, it's a federal Institute. They have their own their own burdens to bear but It's yeah, they're really they're really trying and they're also improving a lot and it becomes more and more planable but still sometimes it's just like
01:01:52
Speaker
Where's this coming from? Okay, we have to comply. So yeah, that's the thing. But yeah, 18 months, that's the super lightning fast approach. But then it becomes a timeline that's also, I mean, for tech investors, that's too long. And just to be fair, 18 months until you can get potentially preliminary listing. Yeah, yeah. Okay.
01:02:13
Speaker
Hey, of course, everything else would be crazy. Yeah, yeah, yeah. Just to be sure that not that an investor is listening is like, oh, okay. We're not yet. We can't. Not yet. Maybe with a beam in your hip transplant technology now. No, but that's the new age, like the third wave of degas basically now already and where it's all pretty planned or where you have a big pharmaceutical company in the background and they just say, hey,
01:02:41
Speaker
We want to give money and we want to give know-how and we want to also control the digital asset. We don't want to partner with somebody and stuff like that, but we really want to be there from the start. But we don't want our own internal slightly cumbersome processes to make the project slow. So it's kind of semi-outsourced slash. Yeah. Sounds like, you know.
01:03:04
Speaker
This, I don't know if you heard of this, this sort of like a skunk work aspect. Let's have a separate sort of team doing something, but they have access to all the resources the big company actually has. Exactly. And that's really where it's going at the moment. Because also big pharma, I mean, they're slow and it's fine that they're slow. They're all riding on their big slow boats made of money. It's fine. It's pharma.
01:03:29
Speaker
I mean, they have to, right? Yeah, exactly. They're coming around and they have, I mean, they may, they learned their lesson with like the whole the beyond the pill stuff, which didn't work, but also because they didn't believe that the digital aspects of care could be their own standalone thing.
01:03:50
Speaker
And that's, for example, something that we will see with GLP1 inhibitors. I mean, for now, they're not being reimbursed super widely yet. I mean, US is slowly starting. But in Europe, it will take, I think, a couple more years until they are being widely reimbursed. But with them, it will be that the whole beyond the pill stuff is no longer valid, but the DTX plus pill. And to kind of be in the treatment pathway with multiple assets and digital and pharmaceutical assets, and that will create
01:04:19
Speaker
Funny enough, again, the double win, like it will create more revenue for you, but it will also create more value for the patient because everything, especially when it's reimbursed in Europe has to prove patient value. And that's one of the things that we will be really seeing is like that you have an early stage digital intervention where you can get, you can service all patients.
01:04:40
Speaker
And then if you have non-responders to conservative therapy, that's based on, I don't know, maybe your existing early stage medication is, you can optimize it, doesn't work, or I don't know, physical therapy or like a nutrition or whatever, like anything behavior that you can do. And if, but if they're non-responders to this conservative therapy, and then they can just directly move on towards the medication therapy, because you also know you will not get recourse.
01:05:06
Speaker
You know that they are an appropriate candidate for the medical intervention. Sorry, for the less conservative, like the more invasive intervention. Yeah. I think obesity there is really interesting because now we have the full thing. You have like prevention, you have conservative therapy, which by the way, the only real conservative therapy that's used at the moment in Germany reimbursed documented that combines movement and nutritional therapy is DIGA. There's nothing else really. Yeah, I know there is. And that's also why they will be the gatekeepers, I think for the GLP1 inhibitors.
01:05:35
Speaker
in that regard. You mentioned the pharmaceutical companies. You also said there were some other medical device manufacturers that saw the dig up possibility and then squeezed their product into the mold, let's say. But do you think then, I mean, you see the trends.
01:05:59
Speaker
Where do you think, who do you think is gonna have an easier time to jump on this wave? Is it gonna be the medical device manufacturers or is it gonna be pharma? Yeah, so I think from a pure monetary perspective, pharma is better equipped and also they are more, I have the feeling, but maybe, I don't know, I'm not the biggest expert on like pharma sales and medical device sales, but from what I currently see is that pharma is better equipped to
01:06:27
Speaker
to do the sales aspect of DIGA well and really get it to prescribers because it's not like P2C sales, the whole Instagram advertisement stuff. Hey, get your code from your insurance, which is also fine, but it's not working well. You have to get doctors to prescribe your DIGA. You are like a pharmaceutical product.
01:06:46
Speaker
pharmaceutical companies are historically better than this than Medtech, but also because the market works a bit differently there, right? And the Medtech market seems to be, from a sales perspective, more classical B2B. But on the other hand, from the preparedness to get a medical device in reimbursement of the market in contact studies, obviously medical device manufacturers are much better at this because they have been doing this for longer. A, they know the Bay Farm better, but also they are more comfortable with doing studies with hardware involved, right?
01:07:15
Speaker
and their GCP constructs are not as cumbersome as for pharma companies. It's funny sometimes because you have a big pharma company and they're treating the DIGA study as a phase three pharma trial with all the bells and whistles. It's a medical device of a low risk, you don't have to jump through the same hoops that you would do with heart medication. You shouldn't.
01:07:39
Speaker
But because they have internal rules, they kind of have to and then it gets weird and slow and expensive and everybody's like,
01:07:46
Speaker
I mean, there are things missing here in this equation, so it would be unfair to ask for the same levels. I mean, there's no pharmacokinetics, there is no pharmacodynamics for this. Of course, there's risk involved with the DIGA giving you wrong instructions or something like this, but that's what the MDR is for, to name the risk.
01:08:15
Speaker
Then I'm wondering where we can go. I guess I just let my curiosity take us. When it comes to pharma and the, let's say the pot of gold they're sitting on, most of it
01:08:30
Speaker
is made in the United States. Correct. I'm not trying to say that Europe is an unattractive market, but it's very different. Plus in the United States, I guess pharma really figured out how to sell their stuff. Yes. For better or for worse. For better or for worse.
01:08:47
Speaker
And, but when, when we now look at the United States, you also mentioned at the beginning peer, which, which kind of went down. Unfortunately, when we look at innovation, digital stuff, digital innovation, you know, software development in Europe, we often look at the United States.
01:09:05
Speaker
They are, you know, Silicon Valley, they keep pushing and pushing and pushing, you know, move fast and break things. Yeah, exactly. I would say just break things and doesn't matter how fast you move. Maybe that's the new model.
01:09:23
Speaker
What's happening over there? Because with all these hurdles that you mentioned that are happening in Europe with the rules and regulations, did some companies that may have started in Europe said, you know what? Fuck this. I'm out. I go somewhere where I have easy access to VC money. And I also have easy access to a sales pipeline to make my money back.
01:09:48
Speaker
I think the access to VC money is really a topic because that's really difficult at the moment in Europe. I mean, the coffers are full, right? They should. Bursting. Yeah. I mean, everybody, all the funds are full. People have invested their money in the last couple of years and they put it into these funds and they're not investing, or they're just getting their own portfolio companies giving them more money. But I think getting venture capital in the US is much easier.
01:10:13
Speaker
But however, a couple of companies, when Kaya expanded to the US, it really depends on how the European business will go. Famously, that's not a giga company, but General Digital of ADA fled to Europe.
01:10:28
Speaker
Yeah. Like flat, like really flat? I think they, I don't know. I think they, they don't have any active, they don't have any activities in Germany, France or so like that. Yeah. Also very strange, like such a company with such a European DNA, but yeah, fair enough.
01:10:44
Speaker
And, but I don't know, I mean, I'm not super familiar with the current number, so I don't know how, I mean, a couple of companies went to the US and I think a couple of them, sure they get some traction, they get some insurance contracts probably, but to get paid for actual clinical value is also not that common in the US. I mean, you have these contracts with the employer-based health insurances.
01:11:08
Speaker
And I mean, they're interested in reducing sick days and having people still perform well, also people just feeling nice. And it's not necessarily all about keeping them super healthy or I mean, that the US health system is so inherently flawed. And if you can play on these flaws, you can make a lot of money.
01:11:30
Speaker
But in health insurance, because it's employer-based, if you're really facing the risk of unemployment and being unemployed or getting fired, that's something where the health insurance not necessarily has a benefit from, because if you're seriously sick and then you get fired from your job, you also lose your insurance coverage.
01:11:47
Speaker
Yes. I mean, of course, the Affordable Care Act and I mean, the part that hasn't been reversed by Trump. Yeah, different topic, of course. Also, I'm the biggest expert in this one. But yeah, so it's a bit more complex than that. And there's also Medicare and Medicaid, of course. But in general, for business models in the US, I think, yeah, sure, they're easier.
01:12:02
Speaker
But I also see a lot of like, yeah, companies claim that they're active in the U.S. and then they're active in two and a half valleys somewhere in California. I mean, it's not a single market. That's, I think, what a lot of people, they always say, oh, the U.S., it's one market and I have one market existence. Absolutely not true.
01:12:17
Speaker
You have a lot of regional contracting, you have regional differences, not only in culture, but also in the organizations that are active there. And of course, there are these huge national entities like Kaiser Permanente and whatever, of course, everybody knows them. But yeah, I mean,
01:12:34
Speaker
It's still super fragmented. And just because you have an office in Boston, we have successfully managed your market access in the US. And I think that's where Europe really has an advantage. And also we see a lot of US based digital therapeutics company looking to Germany at the moment. Yeah. It's like.
01:12:50
Speaker
We knew that our companies went over there because its market is juice and ripe and nice and tasty and lots of reimbursement money. But now we see it for the first time, like it's the other way around as well. And they're looking for sustainable income, like sustainable business models that are crisis stable, that will happen with contracts that they cannot be canceled within a year or something like this, where you just have, yeah.
01:13:13
Speaker
constant revenue with OK upfront investment. And I mean, compared to the with the amount that is invested in venture capital in the U.S., market access in Germany is not expensive. I mean, for a new diga manufacturer, it's like one million euros, maybe two if it's super expensive or if you have to, I don't know, do several trials or something. But that is not a lot. No, that's not a lot. It's really fine, actually.
01:13:40
Speaker
But it's often done, I mean, the strategy that people would say that most of the time companies use now is massive market penetration. What bleeds your business, builds your business. And as soon as you capture it most, squeeze and squeeze and squeeze.
01:13:57
Speaker
but that doesn't necessarily work well for healthcare. Yeah, and the regulated market doesn't work. Yeah, I mean, in general. And the regulated market is terrible because then the regulator will just say, no, why won't that you squeeze? That's not what I want. That's not the best for the population, for the society. That's not in my KPIs to have someone in a position like this. I mean, with pharma patterns in the beginning, sure, it can be like this. But yeah, again, pharma is maybe not the best example because they make the most turnover in the US, right?
01:14:27
Speaker
Right. But still. On the other hand, still, Europe is not a bad market. No. And if you can make some money there, why wouldn't you? And sure, they do make most of their money in the United States, but drug manufacturing also takes place in Europe. Exactly. From American companies. Yeah, exactly. I mean, AstraZeneca does it, Pfizer does it. Johnson and Johnson. Johnson and Johnson, JSK does it. Or Janssen, I think, in both. Yeah, I don't know how to get it. Janssen. Oh, yeah, that's true.
01:14:55
Speaker
And the manufacturing takes your place as well. And if that was so shared, they wouldn't do it here, right? So there must be something. There must be something. Yeah, exactly. And I think, yeah. And for digital therapeutics, it's just these reliable market access pathways that are in Germany and now in France as well. And Austria starting in the third quarter of next year. And Switzerland has kind of like a Degas-like pathway. So we have Central Europe, Dach.
01:15:25
Speaker
Yeah, like the german-speaking countries and we have france for now and I mean that's critical mass I mean that's far beyond like that's like over 150 million people and yeah, it's ridiculous when you think about it, right? Yeah, that's massive and um And that's also the thing and because it's really collective reimbursement It's such a nice time because it means really you have access to the whole population if you're on the list, right? And that's something you will never have in the us
01:15:51
Speaker
I mean, of course, for certain drugs, yeah, sure, but like it's not this. You have zero or almost zero out of pocket payment in the majority of the European country. You have copayments for drugs, but for the digital therapeutics, for example, in Germany and in France, you don't pay anything out of pocket. There's no copayment. There's nothing preventing you from using this. And in both countries, you have 90% or anything in France, over 90% of the people in this public insurance, public coverage scheme.
01:16:15
Speaker
And when you're on the central list in the Pecan or LPPR list in France or on the DIGA directory list in Germany, like the whole population is open to you. You just have to find the prescribers, but like it's the whole country market exists done. Now it's sales time. And in the US, it's never like this. It's like, I have a hundred thousand different insurance groups. I negotiate with the single contracts and get it to the people and all that stuff. Yes, it's very lucrative.
01:16:41
Speaker
But if you think like a pharma company, if you think like like like an actor in the healthcare field who has a long term vision of long term sustainable growth and also market penetration is not having one product. That's such a stupid tech thing. My product called WhatsApp.
01:16:57
Speaker
and I'm scaling WhatsApp, and my company's called WhatsApp, and our logo has WhatsApp written on it, and I'm going to sell WhatsApp to the big company who really likes WhatsApp, and everybody has to use WhatsApp. That's not how healthcare works. Buyer is not aspirin. Right. And also Novo will not stay semaglutid. No. Right? I mean, they have their flagship products that go well, but what is pharma good in? Developing and bringing to market access and selling drugs.
01:17:26
Speaker
And Degas manufacturer, and that's what we're seeing. We'll see a new genesis like of a new healthcare area here. Degas manufacturers also will have to become good at this. They're not good at the one thing. It's not the WhatsApp approach. I have this one app that will find this one niche messaging and everybody will use it and then I'll sell it to somebody else because they will, yeah, do God knows what they're doing with it, whatever Facebook is doing with WhatsApp.
01:17:49
Speaker
It's about you being good at developing, bringing to market access and, or bringing to market in general, and selling digital interventions. And that is a multi-product approach. It's a portfolio approach. It's not a single product thing.
01:18:04
Speaker
And that's the biggest pitfall we have. And if you look at it from this perspective, then having this in Europe with this huge amount of coverage across wide areas of the population, it's much nicer, much more appealing because when you're in, you're in with this one product and you don't have to renegotiate with every single entity in the whole country all the time. And also you can use your sales force in a much more homogenous way.
01:18:28
Speaker
And yeah, you have your power prescribers. You have your doctors. And I think diga manufacturers who are active in specific disciplines, like, for example, Carnos Health, Men's Health is their main, I mean, now they also have an OAB diga for women. So more urology, let's say, in the widest sense. And they are leveraging their sales network. And they're doing a really, really great job at it.
Innovations in Digital Health
01:18:53
Speaker
It's one of, can I, one Carnos anecdote? Yeah, I hope it's okay. Sorry, I just,
01:18:58
Speaker
Imagine you have an app for male individuals with erectile dysfunction and they have access to, what is it called, PED5? Yeah, PED5 inhibitors, blue pill, Viagra, all that stuff. Yeah. So it's really easy. So you take the pill and everything down there is working. But then you develop an app and this app tells people with erectile dysfunction that they should go running and that they should stop eating unhealthy and they should stop smoking and they should lead a healthier life.
01:19:27
Speaker
And you would really think, who would use this? And they're doing it. If the urologist tells you, you do not need the pill, the blue pill. You can basically change your life, change your habits, change your behavior, be more healthy, and you will have less problems with your erection.
01:19:49
Speaker
And it works. Also big effect size. Exactly. Huge effect size. No pun intended. And sorry, apologies. If it fits, it fits. Exactly. And that's amazing when you think about this, because on the one hand, it shows that people like conservative treatments and agency, but on the other hand, it also shows that
01:20:15
Speaker
If your doctor recommends you something, you do not necessarily have to directly believe that this is the best shit ever. You also trust your doctor. And that's something that's so underestimated. And that's why I said initially, like with the bladder catheterization, like these devices, when you would look at them, like something really unpleasant. But if you would look at them like a tech VC looks at Degas and then they would always say,
01:20:38
Speaker
Oh yeah, you have to show some market traction. If nobody buys this product out of pocket, why should I bring it in front of the regulator? Why should I give you 1 million for the market access? Show that people pay for this out of pocket. It's medicine. People don't need to pay out of pocket, at least in functioning health assistance, such as in Central Europe. But you treat it differently. It has a positive effect. And medical professionals can assess, OK, this has been proven that this is good for my patient. And then they tell to the patient, please,
01:21:06
Speaker
Do this, although it might be even at times unpleasant or inconvenient, even painful, right? And then the patient will do it because they trust the healthcare provider and they trust that they check the evidence and that they know that this is good for you and then you do it. Although there might be a shortcut or something else or something else, but if the healthcare provider tells it to you and they made their decision based on good evidence,
01:21:29
Speaker
then it's again a really good situation, not only for the individual, but also for the system as a whole. And I think Kranos is a really good example that you can do this with a combination of good evidence, well, obviously good intervention, but also a good sales force that creates lots of trust with doctors. Yeah. Wow. That was an interesting example. Just to complete the cycle of getting listed.
01:21:53
Speaker
You mentioned already some, some amounts of the reimbursement structure. Now let's say I got my evidence. I got all the ISOs, got the QMS, got the ISO 27001. I did everything according to GCP. And now I'm about to get listed. Who do I have to talk to about saying this is how much money I want?
01:22:17
Speaker
Yeah, that's a good question. No, it's the Geckarwauspitzerbahn. So, well, two things first. So the first year...
01:22:24
Speaker
of your listing, whether you are preliminary listed or whether you are final listed. First 12 months, you have your manufacturer's price. You can have a price of your own choosing. You just tell them. What? Are you serious? Yeah, of course. That's what they always criticize. That's why a lot of diggers have initially quite high prices for the first 500 euros. I was about to say I saw some of those. Yeah, like a thousand euros or something.
01:22:48
Speaker
I thought, how did they negotiate this? They didn't. They did not. So it's a preliminary listing price for the first 12 months. And for example, if you have the fast track, you are during preliminary listed. And then in the first year, yeah, you have a high price. And then you get, and then after you finish your final listing, you start price negotiations together with the Gigafel Spitzenfahbahn, that's the umbrella organization, like the Federal Association of Statutory Health Insurance. So they are your partners.
01:23:17
Speaker
as if you would negotiate a price for a new pharmaceutical intervention. Same procedure with them.
01:23:23
Speaker
Really, like absolutely the same procedure. There's a new, I mean, there's a different framework agreement with the responsible lobby groups from the industry side, which governs something slightly differently, but the overall process is quite comparable. Of course, if you have people who only know how to negotiate on the pharmaceutical side, they might not know everything around DIGA. So that's always a little bit iffy in the beginning. It was a little bit difficult sometimes, but now, yeah, things got grooved in with the price negotiations. But yeah, to come back, so preliminary price,
01:23:53
Speaker
first year, 12 months, whatever you want, which is weird, right? I mean, technically, whatever you want seems why don't people take 10,000 euros or something like this? Well, there are two reasons for this. One,
01:24:10
Speaker
health insurances have to greenlight your DGA if you come as a patient there. And then I think there would be lawsuits quite quickly if you would go that high. But the more important point is the vast majority of DGA's are prescribed. And if doctors see, I mean, they see the price in their system. If they see 10,000 US, they will not prescribe because doctors also have
01:24:34
Speaker
Yeah, they have the responsibility of working in an economic manner. Of course, patient well-being is the highest priority. But if they can, they should do it in an economic way. And I don't actually know any doctor who would prescribe an app for 10,000 euros, maybe a few. But yeah, keep it reasonable. Also, there are maximum prices according to specific DIGA groups. So if you have enough DIGA in a specific indication,
01:25:01
Speaker
Then a maximum price is formed and that's then basically adjusted in a small downward spiral. So it's getting less and less over time, the more come in, but only slightly so. So that's kind of in line to do that. Still a preliminary listing price, a conservative preliminary listing price would be 400 euros per prescription.
01:25:23
Speaker
By the way, that's also your quick and dirty calculation for the first year. If you manage 5,000 prescriptions for a conservative preliminary price of 400 euros, then you have a yearly turnover of 2 million euros in the first year. Hypothetically, you do not include your running cost in this. You would theoretically have your market access cost of around 1 million euros covered plus another million euros which you could theoretically put into a sales team, for example. That's your quick math.
01:25:52
Speaker
These are easy rules of thumb. Yeah, exactly. That's much easier than pharma for sure. Yeah, yeah, exactly. Much easier than pharma. And of course, it's not that easy. There are smaller indications. Are your doctors, are they primed for Degas already? How big? Yeah, lots of unknowns, but this rule of thumb, 5K prescriptions in the first year is good. But I mean, you can also be like Endo Health had like a team of, what was it, like 17 or 18 people. So really small team. And I think they made like
01:26:16
Speaker
16,000 prescriptions in the first year? Yeah, completely crazy. On the other hand, also, would have thought endometriosis, underserved condition, patients are suffering massively. They are mostly young and they are really well organized in online support groups. Who would have thought that a completely reimbursed digital therapeutic that addresses this crazy great need among a very, very suffer like high
01:26:41
Speaker
burden of pain and of discomfort patient group, who would have thought that this would work? Well, of course it works well, because it addresses a need from patients quite specifically. Yeah. And just for me, so the 400 euro reimbursement, that's for three months use. 12.
01:26:58
Speaker
It's for a 90-day use. And that's a bit unintuitive also for DIGA. There is no ruling on the prescription or the prescription timeframe for DIGA, but everybody seems to kind of work with 90 days at the moment because it's a quarter of the year. That's the billing period in which health insurers think. So it was kind of easy to do this, although technically the pricing for DIGA is expressed on a daily rate.
01:27:23
Speaker
Okay. So when you negotiate the price, you get a daily rate, but usually we talk about the price per prescription and that's usually 90 days, but there are some Degas who are lifelong. It's a one-time payment. And, but theoretically you could make a four month Degas or a six or nine or 12 or whatever. Hmm.
01:27:41
Speaker
So there is choice and I hope that some manufacturers will do this. But yeah, it impacts on how we generate the evidence and three months seems to be the sweet spot for a lot of behavioral interventions where you can see good results after a quarter of a year. I was about to say 12 weeks between 10 to 12 weeks. Can make a good behavioral change program for lots of things. Exactly. Whether it's exercise or nutrition.
01:28:04
Speaker
And the follow-up periods, I mean, the ones, everybody can read it up online on the Bay Farm website on the listings. It's also in English, it's not all in German, where you can see that, you know, the intervention is done for three months. And then, of course, as a primary endpoint, often they use the, after the 12 weeks, so the three months.
01:28:26
Speaker
point, but many of them did follow up periods of up to six or 12 months in total. And the effects are still there, which is giant. I mean, that's, that's quite something. I mean, if a farm farmer doesn't have that necessarily, I mean, depends on the condition. Yeah, it depends on the condition, but it's much, much less often the case.
01:28:52
Speaker
Yeah, exactly. And yeah, that's nice. And of course, but like we're just talking about the preliminary pricing, right? And so it's, yeah, it's, uh, also was not without criticism in the past, definitely not. So people said, Hey, you are not really fully validated and you're getting so much money, but the amount of manufacturers that don't make it is really, that's the, it's the minority. Like the vast majority is making it.
01:29:15
Speaker
And you have strong preliminary effects. And then afterwards, of course, the final price negotiation starts. So you have acquired your final listing, like the official letter from the Bay Farm. You are accepted indefinitely into the DIGA directory.
01:29:31
Speaker
And now the price negotiations with the Gigafau Spitzenfabant can start. And then usually you end up somewhere between 200 and 250 euros, which is a bit sad because it seems like we are caged in from the current pricing decision because
01:29:47
Speaker
In most cases, you do not finish your negotiations with the Giga for Spitzen for band with the statutory health insurance umbrella organization. You go to an arbitration board. That's the status quo and the arbitration board. I mean, I don't know for certain, but sure. It looks from the outside like as if there is a politically inspired
01:30:07
Speaker
artificial border around DIGA prices not lower than 150 and not higher than 250, which is a bit sad. On the other hand, I think with a risk class 2B and when we are getting into indications such as, I don't know, oncology or something like this, we really have different endpoints, maybe even survival at some point could be.
01:30:27
Speaker
I think then we will have to see a breakout from this mold because I am absolutely certain that there are digital interventions, digital therapeutics, DIGA, that have effects that are worth significantly more than 200 euros a quarter, significantly more. But of course,
01:30:48
Speaker
Anchor pricing, what's the reference treatment? What is the arbitration board saying? It is a hurdle to overcome. But then on the other hand, still significantly more lucrative than anything anyone would pay out of pocket for an app.
01:31:02
Speaker
Yeah, for sure. Yeah, but they're definitely not. When I get 250 euros for now, I was like, no. That's three months of use, even if you have one of the more expensive ones. It's like 30 euros per month, which I would kind of, I mean, that's already pushing it, right? Yeah. And then you would have 90 euros basically for three months. That's not even half what most digger manufacturers make it. Yeah. Now I have, I think, well, I have plenty more questions for you, but I think these two are
01:31:33
Speaker
I would say quite important to address.
01:31:38
Speaker
AI, hype.
AI and the Future of Digital Therapeutics
01:31:43
Speaker
I mean, I worked in this field for a while, for a long time actually. And man, I can't hear it anymore. It's pissing me off. AI is done more with PowerPoint than with anything else. Yeah, for sure. But, I mean, irrespective of me being annoyed, or I guess all of us being a little annoyed that actually knows something about this, how is AI?
01:32:08
Speaker
taken advantage of at the moment or can be taken advantage of. I mean, digital therapeutics basically lent themselves to this. This is one of the best use cases of all times for like, especially, I mean, machine learning, of course, is in its own way, like picture recognition stuff, also cool and fine. But the real power is really with the newer development in LLMs, large language models, because
01:32:33
Speaker
Okay, then we need to make a little detour because that's one of my favorite points, but it takes a little bit of explaining. Cool, do it. We've got time. So as we said in the very beginning, right, DIGA and especially the German regulators, the Bay Farm is quite keen on this, DIGA have to work by themselves. So if you put a human coach in combination with a DIGA, the Bay Farm will say, uh-uh.
01:32:57
Speaker
We don't know, has the app done all the work or was the coach responsible for the clinical effect? We cannot separate this now anymore. So yeah, sorry. We assume that the app did nothing. The coach did everything. So sorry. So back to telemedicine. Exactly. Yeah. And then, but then what we see in studies where we have this set up is if you have a human coach, for example, for nutrition counseling, then you create this kind of this little bit, well, dependency is such a strong word, but
01:33:25
Speaker
The patient's getting used to it, and they also feel responsible not to let their coaches down in a way. And they feel taken, they feel cared for, and it has a positive effect. You have a commitment, you have an invest. Exactly. It's an invest. And also this person is asking you how you did. And so long story short, coaching increases adherence and thereby trial outcomes. And adherence and not only mean like, okay, I'm taking this or I'm doing this now, but like people are more engaged and more involved and they're really more interested in doing things correctly as well.
01:33:55
Speaker
However, coaching is not allowed in your evidence generation studies for market access and also for good reason, right? I mean, backdoor financing, or nutrition counseling would be weird for a digital intervention. I completely agree with the regulator that in most cases, not in all of them, but in most of them.
01:34:14
Speaker
What we are seeing now, and we have, I mean, we at Fbeta, we also have a unit that deals mainly with AI and its implementation. We train on large language models and we implement them in specific cases, different environments, super interesting stuff. And what we are seeing in some early stage investigations that we did was that
01:34:37
Speaker
you can come close to the effectiveness of a human coach with a good LLM. So this whole, I have an entity here that cares for me that asks how I've been. I mean, people know that this is an AI and machine, a chatbot. They know that's about, but it doesn't really matter that much. I mean, it's not, I mean, we're super in the beginning. I don't have any super reliable numbers. This is just from first impressions and also from fast results that we're seeing, but
01:35:06
Speaker
it seems to come quite close to the effect of this human. And it's funny, I mean, and again, really early stage, but if you give the chatbot a face and a voice, it works even better. And then people really identify with their supporter there. And a lot of people really like it as well. And so where we see fantastic potential is you have your weight loss decal, for example, and you build in this chatbot and you can scale this
01:35:34
Speaker
I mean, even if coaching was allowed, it would be horribly cumbersome for scaling, right? The chatbot is infinitely scalable. So you're like two birds, one stone. Even if you want to do coaching later, the chatbot is doing this so nicely. And you increase your clinical outcomes. You can negotiate a better price and you increase scalability. So I mean, DTX is just a question of time.
01:35:53
Speaker
And whoever is willing to go to a notified body and fight with them about what's this AI doing, what's it doing with the risk classification, that's more the hurdle to jump over. But that's also the intellectually interesting part of it. For sure.
01:36:08
Speaker
I have an obese patient and my app tells me, I don't drink the bottle of olive oil. It's completely fine. Then they might have diarrhea for a couple of days, but they will not be in mortal danger. But if you tell a diabetic, eight cinnamon buns, go ahead, buddy, just eat them all. That is quite dangerous, right? And that could be potentially deadly or at least extremely damaging to that individual's health. And we need to make a differentiation here.
01:36:33
Speaker
But then we're also in, yeah, I don't want to go too deep into this, but that's one of the topics that we're currently working with. And what I personally find super interesting is how do we hard code limitations into AIs or how do we test them and how do we test their limits and how do we test when they break, when they give wrong information and how can we explain to regulators in a transparent way, no, my AI will not make a bad decision. And there are very interesting methods to do that for transparency.
01:36:58
Speaker
partial explainability of AI or extreme case testing, stuff like this. And that's amazing. It's one of my favorite topics at the moment because it has so much potential because as soon as we're able to talk transparently with a regulator about the limitation and the risks in our AI, then we can start deploying it.
01:37:14
Speaker
Yeah. And I see huge potential benefits there. Huge. I mean, better outcomes, better scaling, better accessibility as well. And the nice thing about also localization of apps, like currently you translate all the texts and the videos and stuff like this. If your app is mainly chatbot based, it's done. It's done. You just push a button and then it's in French. And that's awesome.
01:37:38
Speaker
I mean, if you think about it, it's actually insane. From an investor perspective, that is insane. Yeah. But again, you can't be afraid of having to fight with a couple of regulators. You have to. But it's also fine because the return is so high. Let's say discuss. Maybe fight is wrong.
01:37:57
Speaker
I think in the beginning it will be a friendly, you know, like a friendly match between fences. For sure. It's very polite and everybody's going back and forth. And like sometimes you tip yourself with, oh, sorry, that was a mistake. I can't let you through. Oh, yeah, right. But here, maybe your opinion was a little bit too extreme, huh? Yeah, exactly. Exactly. And I think a friendly fencing match, I think, is a good metaphor overall. Yes. Nobody's getting hurt. Everybody's very polite. The rules are strict. But if somebody scores a hit, it's being recognized by a loud bap. Yeah, for sure. It's like a boxing match with a ref.
01:38:27
Speaker
Exactly. With a referee and with friendly sparring. Friendly sparring, yes. Yes, for sure. Now, I remember, I mean, I was involved in the audits of when products did involve AI aspects. And I was shocked about the questions that I was asked from the people auditing the system.
01:38:51
Speaker
I mean, I remember one time they, the person, I mean, of course, for, for these audits, you have to prepare a lot of stuff, right? It's not like these people don't just come in and ask you random questions. They, you have to file massive levels of documentation so they can see the holes and then come to you and then basically poke around to see if those holes make sense or if they, uh, if they actually found something or not, or if there was a misunderstanding, right?
01:39:20
Speaker
And I remember it was about the AI part and they asked specifically in the quality management system for how do we validate or basically make sure the efficacy of the system.
01:39:37
Speaker
And if it's a deterministic system, yeah, one plus one equals two. Okay, yeah, fine. But it's deterministic. But it's deterministic. So then, for example, one aspect they they kept on talking about was, well, how did you calibrate the system? What were like the measurement instruments? And I looked at him and I
01:39:57
Speaker
What are you? What? What measurement instruments do you talk about? Like a ruler? 5,000 randomized prompts. You mean like a ruler? Yes, exactly. A ruler. Did you use a ruler? No. This is like this. And then we checked it like that. And then you have cross validation metrics and all of this. And this is the accuracy. And he just looked at me and said,
01:40:24
Speaker
And where's the ruler? Say, what the hell? Did you just listen to a word I said? But they're also not prepared for this, but there's a classic way of like, yeah, we have this with, yeah, with everything. There's some new technology and then regulators have to kind of learn on the fly how to deal with that. And that's quite
01:40:42
Speaker
It's challenging. I mean, in all fairness, right? In all fairness, yeah, sure. I don't want to bash on them too much, even though I can't be annoying at some point. Exactly. Awareness of them being taxing sometimes, I think, is fair. But then also enough, regulator empathy. Seems like a nice course. Yeah. That's the app besides the mindfulness app for like, be nice on the workplace. For sure. It's regulator empathy. Yeah, yeah, yeah. 15 meditations on the beef farm. Oh, yeah.
01:41:11
Speaker
And just to be fair, I mean, for the ones listening that don't know that yet, basically what they asked for was, one of them literally said, assume this entire place burns down, all your data is gone and everything, and all you have left is this document describing how you did it.
01:41:33
Speaker
Jesus. It should be explained in here in a way that you can start from scratch and roughly get to the same output. And I thought, OK, you know, that's fair. Yeah, that's fair. That's fair. And after they phrased it this way, it was very easy to to provide them with all the details. Yeah.
01:41:51
Speaker
But I think, yeah, we will get to a point also where we may might also have to save generic training data sets somewhere or something. We say like, Hey, this is the core set that we trained it with and, or maybe have some.
01:42:05
Speaker
It's going to happen with the AI act anyways. I mean, it's been passed. It's going to come. They talk about higher risk applications, which all medical stuff is automatically categorized there. The pre-categorized stuff. And then you have to do it anyways.
01:42:23
Speaker
Then you have to do it anyway. I just hope that they will have a rational discussion on the whole thing. But I'm in a couple of discussion groups on AI in healthcare. It's always the same thing. Nobody talks about specific applications, but there's always like two to three guys or girls, and they go like, hey, but what if the AI makes a decision on life or death? And it's like nobody said that they want to deploy this there. Nobody said,
01:42:45
Speaker
the AI is going to do triage in the emergency room. Like where is this coming from? Was there somebody accepting the idea of people randomly deploying? I mean, who thinks about this? SkyNet. Yeah, but everybody always thinks about the Terminator.
01:43:00
Speaker
That's another of my favorite anecdotes. I once presented it in front of a group of doctors on nursing care innovations. But this was more on the robotic side. And it was like, hey, there was this wearable. And it measures galvanic skin response. And maybe you can see if somebody's dehydrated or not with this wearable. And then a little pepper robot comes into your room and knocks. And they say, hey, we would like to be too drunk enough. Please drink this bottle of water. And then one of the doctors stood up.
01:43:28
Speaker
a doctor from Bavaria, and then she set up and said, yeah. And then the robot goes in the room and forces the old lady to drink the bottle of water. It's like, wait, wait, where is this? Have you seen a pepper robot? Like, what is this? Like, how am I going to get this into a nursing home? Like, this is the drink forza 9000. And then it's just like, I don't know. People always seem, I mean, it's nice to try to find cases that break the system or extreme cases to kind of define your playing field, Union.
01:43:58
Speaker
But if your first reaction with every technological innovation is to always think about the Terminator, it's just weird. And I hope that with AI, that it will be a bit more differentiated. I really hope so.
Resources and Guidance for Digital Therapeutics
01:44:10
Speaker
But for DIGA, I think it's a huge chance overall. I mean, it has so many advantages. It would be complicated, so there will be first mover advantages, definitely. Yeah, for sure. For sure.
01:44:22
Speaker
but also just the scalability or just to have if you have a really good app with a really, really good mix between interactive content and an LLM that's kind of masterminding the whole thing. Well, technically it will always be multiple LLMs, right? You know, it is. And if you have this as a really good basis, and then for any type of condition of disease, of sickness, where you have a roughly similar
01:44:50
Speaker
roughly similar mechanism for the treatment. Anything has to do with your own behavior where you can impact it. You can scale this across different types of conditions very easily in a short amount of time. And then the only thing that's left to do is that you have to become good at making multiple studies and then at selling it. And that's not only scalability on a product level, but also on a pipeline level. LLMs will revolutionize it.
01:45:13
Speaker
So please, if you're listening, if you know someone who's willing to invest, be the first, please. It's definitely a good case. Investors are not seeing this at the moment because they see digital therapeutics and they think peer therapeutic and it's bad. But in Europe.
01:45:28
Speaker
I know it's not scaling as super fast as some other cases, but from a sustainability perspective and a crisis stability perspective, and also the opportunity to have this much impact on the healthcare market and this powerful position in the market, and I'm not talking in an altruistic way here, talking completely
01:45:47
Speaker
money, capitalists, revenue. I mean, this is a good position to be in. It's a lucrative position to be in. I mean, the bonus is here due to the fact that it is a reimbursement-based market. You cannot be economically successful if you do not make patients better. But that's great, right? I mean, that's like the clean conscience way to earn a lot of money and also improve society. I mean, talking about purpose.
01:46:13
Speaker
Yeah, right. Making the world a better place. Yeah, but of course, I mean, I am a liberal at heart. So I always say you have to put the right incentives in place and then people will make the world a better place. But currently in Europe, in the healthcare sector, the incentives are there and the time is there and the technical tools are there and the patients are there and the need is there. I think we are in quite a good position to kickstart this. And I am convinced that
01:46:40
Speaker
2024, maybe late 2024, and then 2025, this will be the two years where the investment train is leaving the station. It will start rolling again. I'm convinced. Yeah. Wow. I think that's a great way to wrap it up. I mean, for everyone listening, what would you recommend if they are interested in more things, resources to read up? Given, sure, this is all very new. I'm sure there is no
01:47:09
Speaker
Finnish written road, this is the roadmap to follow book, but which resources would you recommend? Yeah, so I think start with the DIGA Lightfin, the DIGA guidance document, the DIGA guidelines. I'm sure they will be linked under the podcast as well. And for DIGA and DIPA, and I think that's a good way to start, just to say, hey, what is the regulator thinking? Well, of course, feel free to hit us up if Beta was giving out
01:47:37
Speaker
Advice, no strings attached for, for people who are interested in this type of, in this type of topic. And otherwise, yeah, just ask yourself, do I have a potential product that can stand clinical testing that really has added value for a patient? And then that's the starting point. And then see if you are capable and willing to jump through all the hoops as detailed in the, in the, in the guidance document, in the Degas light fade in or deeper light fade in.
01:48:06
Speaker
And yeah, that's a good starting point then, definitely.
01:48:09
Speaker
And if people don't want to read, but actually directly talk to you, they can reach you on, I mean, they can find the FBTA website. I'll put it in the show notes anyways. You're on LinkedIn. Yes. I'm on LinkedIn. LinkedIn is the best. You want us to look for the guy who looks like a hipster emoji, beard and man bun, the whole shebang, easy to identify. And yeah, but also yeah, reach out via email or LinkedIn. I think that's the, these are the two best channels, definitely. Yeah. You don't have to write anything down. It will be in the show notes.
01:48:39
Speaker
Jonas, I can't thank you enough for this great conversation. It was a lot of fun. Thank you for letting me ramble on for such a long time. Highly appreciate it. Thank you for rambling. It's always more interesting to listen to the guest than to me. No, thanks. Thank you for having me. Thanks a lot, man. Highly appreciate it. All right. Then to everyone listening, have a great day.
01:49:04
Speaker
Hey everyone, just one more thing before you go. I hope you enjoyed the show and to stay up to date with future episodes and extra content, you can sign up to the blog and you'll get an email every Friday that provides some fun before you head off for the weekend. Don't worry, it'll be a short email where I share cool things that I have found or what I've been up to. If you want to receive that, just go to ajmal.com. A-D-J-M-A-L dot com. And you can sign up right there. I hope you enjoy.