Hey everyone, welcome to the Venture Capital Podcast. Today, Peter Harris and I are joined by Max Marchione. Am I saying that correctly?
Yep, that's close enough.
He's the CEO and co-founder of Superpower.com. They provide health and data intelligence through very important blood tests. Peter has taken it, and we have John Dayton - he's an MD, MBA. He's with the Intermountain venture fund, he's with Spring Tide, and a few others. And so today we're just going to kick off and get started. So, welcome, guys. Before we begin, Max and John, do you want to give a brief intro of who you are and what your background is? You go ahead, Max.
Totally. So I'm Max, one of the co-founders of Superpower. We're building a new healthcare system focused on preventing disease and enhancing human capability. And previously I spent a little bit of time starting other small companies, and I've done that for several years. Lots of things - some are alive, some are no longer alive. That's the nature of starting things. I spent a little bit of time on the investor side. I am still a founder of a fund. I invested around a company a week, and then I spent some time on the finance side as well. And oh gosh, what else is interesting? So I don't know. I think that basically around two and a half years ago, I was thinking what kind of problems actually really matter on Earth. And there were really only two things that I thought really mattered: one was human health, and the other was things that happen in the physical world. They still matter post-AGI. And I think in a post-AGI world, health is potentially the most important thing - enhancing and extending human life - and hence here we are.
Awesome. Thanks, Max. John Dayton, do you want to give your background?
Hey everybody. I'm an emergency doctor. I work with Intermountain Health, and on faculty at Stanford, where I'm involved in research that has to do with innovation, particularly bringing acute care innovation into the ER. As John mentioned, I'm an adviser for Intermountain Ventures, and I'm a venture partner with Spring Tide.
All right. Well, today Peter and I wanted to talk about what's the future of health tech - from the founder side, from the VC side. And the first thing is, let's talk about first what's broken in the system. What do you guys see?
So, as a doctor, I'm happy to take that. Our system is based and reimbursed on fixing problems. It is not based on prevention. It's not based on longevity. It's not based on health span. So that's one of the major issues, but also that provides lots of opportunities.
Awesome. John, have you looked at systems like Max's company? What are your thoughts?
Yeah, that was really interesting to look at it. Particularly, you know, you take different business models. You know, we saw these startups in 2021 that were focused on digital health and making things work, and a lot of money was spent on those. Not a lot survived. Except lately, Omada and Hinge have turned the corner, and the reason for that is they were able to say, 'All right, well, how do we fix something in the system? How do we make people get better?' But how do we work with the existing reimbursable models? And so I'm always interested when someone has, you know, something may start out as concierge or may start out as something within the insurance system focused on longevity and health. That's always very interesting to me. I'm always curious, you know, when people say we're going to look at all these biomarkers and these monitors, what specifically they're looking at and what data and research informs those choices. But I always think that's a good way to look at a forward-facing health approach.
Yeah. What do the two of you think are the most important metrics to be looking at right now in 2025?
I think what's really interesting is how do we summarize a lot of different biomarkers into one. The reality of how healthcare has always evolved is: let's have static single biomarker cutoffs that we look at, and we try to predict things based on a single biomarker, and it's quite binary - you have something, you don't; you're in range and you're out of range. I think that makes a lot of sense in a human-led paradigm, where we try to simplify things to a smaller number of data points because it has to be comprehensible for the human brain. I think what we fundamentally get to is the fact that we can ingest many data points and look at them simultaneously, and we can create scores that are the aggregate of many biomarkers. So one example of that is some of these biological age tests. They're not actually telling you your biological age. Instead, they're saying here's 10 or 20 biomarkers regressed against time to death, right? And that as a predictor is typically better than just a single biomarker. So I think that's one thing that is quite interesting - biological age.
But we built a Superpower score, which says: look, the biological age algorithm takes into account 10 biomarkers. Let's take into account 40 biomarkers and work with a scientific team to say how do we weight these different 40 biomarkers into some sort of linear algorithm to give you a score out of 100. And my general belief is that those biomarkers often reflect, or are proxies to, the foundations of health. If you improve the foundations of health, you improve many of the diseases that sit on top of the foundations, and therefore improving your Superpower score correlates strongly with improving overall health. So that's probably my answer.
Peter, what are your thoughts?
So I'm curious, from John's perspective. So traditional medicine has a ton of data, right? Collecting all kinds of stuff - the diagnostic tests, blood tests, you have procedures that have been done, medications that have been done. There's a lot of universities that are doing studies on patients. You've got clinical trials. So there's this huge wealth of data. Why is that wealth of data not being used in a multi-linear regression analysis, where you're pulling in lots of these data sets and data points in the way that Max is talking about? Right - why is it always so point-solution focused, if that makes sense?
That's one of the big problems in healthcare right now is pointillism, right? Everybody wants - well, I'll build this system, and this will help solve this GI issue, and I'll build this system, and this helps with mental health. One of the biggest issues is we need to have data that works together. The big issue is interoperability of data. And to take that one step further - not only do we need to have the EHRs connect to the clinic records, which connect to lab records, but we also need to overlap that with data from remote patient monitoring. And that's the next step: multimodal AI, right? How do we take the text from the charts? How do we add that to the imaging data, like the CNN data? And how do we utilize that? Ultimately, the big issue right now is there's a lack of interoperability - it's a major issue of fragmented data. And so that's why I'm really interested in companies that focus on interoperability, or when Epic says we're going to do something that brings data together or brings that in a way that's usable.
Because, I mean, it's hard, right? What you think about from a patient perspective: if you're watching your blood pressure, you're interested in what your blood pressure is on average. If you start a medicine to address it or a lifestyle change, you want to know if it's coming down. But and so there's a lot of data points that are available to you. Unfortunately, the only data point you get is, well, I ended up in the ER because I had high blood pressure and I had a headache, and my blood pressure was really high there. That's compounded by so many variables - you know, you're nervous about being in the ER, you're uncomfortable. So rather than using all that trended data, you end up with one data point that's not very accurate or representative of multiple things, and then hopefully that trends down while you're in the ER. But that's the issue: you have to have components that can bring things together and inform others, and that's one of the major processes, one of the critical aspects we need - but also very, very challenged: how do you take all these data points and make sense of it? And I think multimodal AI is going to be the way to do that.
Yeah, for sure. Agree with all of that. And then I'd also add that we have a lot of data relative to where we were two years ago or five years ago, and this is an exponential curve, so we actually have no data today compared to what we're going to have in a few years' time. And I think that as a result, many limitations actually come from a lack of data still. And many of the things that we would like to do - for example, trying to predict disease, not just, or trying to predict health not just based on a single biomarker, but based on clusters of different biomarkers in an embedding space - that's not possible because we don't have enough data, and many problems are actually limited by the lack of data. So I think that we're actually still early in collecting all the things we need to move to a model of medicine that is more algorithmically driven rather than human driven.
I'm curious, Max, to build on that: what areas do you feel like we're lacking in? You know, where maybe for the people that are listening that want to build startups to start collecting some of this data, right, to integrate into a platform like yours - where are the holes?
Like everything. So I think that we still lack omic data, meaning biological assays. There are all sorts of modalities of omics - there's the microbiome, there's the genome, there's metabolomic data. 23andMe has 2% of your genome for only 15 million people, with very limited clinical data on top, and they sold for $300 million out of bankruptcy, right? Like, now we can assess the entire genome, not just 2% of it, and we can hopefully do that for more than 15 million people around the world. There are 8 billion people on this planet.
So omic data is one; clinical data is the second, and it's maybe even more important. Clinical data is what has happened to someone over time, and the 'over time' is very important, because we want to be able to map from what we know at the level of biology or the omics to what actually has happened, and that allows us to perform prediction. And then the other thing that's interesting is continuous wearable data, or ambient data. And the reason that's interesting is: today we try to go from biological assay to prediction. In the future, what we might be able to do is go from continuous wearable data to prediction, and that means that we can actually predict things continuously rather than requiring a blood test or some sort of biological sample.
Max, just to add to that - I just got back from Amsterdam, from the Health Europe conference. Really interesting. But one of the best talks was Tom Hale from Oura and another device expert from a company called Metabolite. And they were saying the ultimate goal is we want two-thirds of what is going on to be monitored data that we can put into a predictive model. And then, you know, one-third of that is your physician, but the two-thirds is coaching: what should you eat that day, what exercise should you do, are your levels trending in the wrong direction. You know, where it started out it was like a sleep aid, right? It was interesting, but now they've got glucose monitoring, and now they have temperature, and they went through the first FDA trial to predict when women are most fertile based on their temperature. And what's interesting on top of that is we've seen companies in the value-based care space say, well, let's use some of these wearables to inform this. Let's match that with clinical coaching. Let's match it with a diet. And you know, we have only so much money to take care of so many patients with so many ailments, but if we can prevent these issues, we're paying for a wearable instead of an ICU stay. And that's such a better model.
How do you look at the effectiveness of a wearable like an Oura? Is it accurate enough?
Not accurate. Yeah. So that's - I'm a guy. They really only did their first FDA trial. I got one of the initial ones, and it always said, 'Well, your sleep's bad.' I'm like, 'I'm an ER doctor - of course my sleep's bad. That's not very helpful for me.' But I think if you could move beyond that to something like, you know, you get a continuous glucose monitor, you pair that with A1Cs, and you pair that with your diet coach, and you pair that with your physician and modulating your meds. I think that's where you get the value, because those are endpoints that are clinical, right? You can say my A1C went from, you know, seven to five, or my average glucose went from, you know, the 200s to the 100s.
But not just that - that's the part that's critical in the way healthcare is built. Unfortunately, it's built in a reactive model as well. But now, with using RPM tools, we can say, you know, with this many data reads during the month, this glucose is coming down, it's better, and then the physicians can get billed for that and build a healthcare model around that. So I think you have to take very specific data points, very specific biomarkers where reimbursement is available, and kind of build around value-based care. That seems to make the most sense right now. But ultimately it'd be nice - you get the results that you're judging for, you're reimbursing for, so no surprise that we make a lot of money in the healthcare system with surgeries or ICU visits, but really there's so much more value keeping people out of the hospital or doing remote care or hospital at home. And so it would be nice to see more of the funding shifted to that side. So that's - I'm interested in companies like the one Max is building.
So, John, just to dig in on that a little bit more: why is it? Because, I mean, it does feel obvious, right? Like, why - wearing an Oura ring and wearing a Whoop device, whatever it might be, my primary care doc doesn't have access to that, nor, I would argue, in some cases wants access to it, right? It seems the most obvious thing in the world, and yet it doesn't get done, and the devices that are implemented are frankly pretty terrible on a comparison basis to, like, usability, portability, UX - a lot of these things - compared to an Oura ring that are out there. And I'm curious, why is there that breakdown?
Yeah, I mean, let's take the Oura ring specifically, right? It's only reached FDA clearance for temperature modulation, right? That's the only one that's stood up to clinical trials. So you can say, you know, if you're basing on usability, absolutely some of these wearables are much better than what you use in a hospital. From the data side, though, you really have to value that, but that has to be within the framework of the system that's reimbursing. Well, if you are relying on FDA clearance to get CMS coverage, you really have to have those high-level data points, and so that's tricky. Unfortunately, like the FDA, there's a lot of modernization that's taking place, specifically with software as a medical device and AI, but really it's based on: is this healthy? Are these results replicable? And that's kind of the standard that they're going on. It would be nice to see a little bit more of where's the ergonomic wearability component, but I mean that's - you've got CMS with a certain amount of money for their Medicare and Medicaid patients, and they can say, well, we have this much for heart patients, let's figure out the best way to utilize it. You really have to be stringent on the data that you're using. So I totally get what you're saying. Usability is not valued by the FDA as much as they should. But at the same time, withstanding clinical trials, that's kind of the benchmark.
Yeah. I mean, it's a bit of a catch-22, because if the usability is not there, nobody will use it, and you don't get the outcomes, right? Even if it did go through rigorous FDA approvals. I'm curious, Max - do you have a perspective on this? Like, why has there been a need for you, frankly, to start Superpower, right? Because it feels like you're trying to solve this problem.
Well, our fundamental belief - there's a few. One is that there's a big gap between the best of healthcare and what most people have access to. And some people don't agree with that; I think if you go deep enough, that is probably true. The second is that the only way to close that gap is going to be through technology versus trying to hire and train and pay more doctors - partially because of the limitation of the human brain, partially because of cost, partially because of just supply scarcity. I also think that in an AI paradigm, we don't just match the best of medicine. We don't just close the gap. We actually far exceed it. And look, these AI systems may be a little bit off that today. They're better than the average doctor. I don't think they're necessarily as good as the very best doctors yet. But that's today, right? Again, these curves are exponential, not linear, and we're very bad at predicting the exponential.
And the implication of these things is that the structure of healthcare starts to fundamentally change. And I would contend that we actually have a new consumer front door - a new first place people turn for their health today. People might say it's the primary care doctor. I'd contend it's a Google or ChatGPT. And I think that Google or ChatGPT don't do an awfully good job, because they don't know that much about you. They frankly don't know that much about medicine, and they can't take action, right? They can't prescribe, they can't diagnose, they can't refer out - they can't actually take action. So the implication is, I think we get to a world where we have a new consumer front door, a place people go to first that knows about them, that knows about medicine, that can close the gap between the best of healthcare and what most people have access to. Can do that largely using technology, and can also do that with a fantastic consumer experience, a brand that people love. There's no one walking around wearing UnitedHealth T-shirts or One Medical T-shirts, right? Like we have people sending me text messages and hoodies around the world - Superpower hoodies around the world - like every day. There's no real software that people love. Everything is kind of confusing. You never really have your data in one place. There's a lot of frustrations that customers have as a result of this.
People are like, 'Well, I heard about the gut microbiome on Instagram. Then I went to my doctor, and I was told the gut microbiome is not a real thing and it's pseudoscience.' Or there's the Oura ring problem, which is, oh, I used my Oura ring to track my sleep and I saw some problems with my sleep - my doctor refused to look at the data. So all of these consumer problems as well. And I think if we fundamentally start by saying what is best for our members, or the customer, or the patient - if we start with that, how do we work backwards from that - whereas I think the way healthcare has evolved, primarily because of incentives, is instead starting with payment mechanisms rather than customer experience.
Yeah, makes sense. I'm curious - what do you think is driving the interest in this? So if you look at companies like Lifeforce, Function - you know, somewhat competitive to what you're building - have grown incredibly fast in a relatively short time period. We're seeing tons of different companies and all kinds of diagnostic testing, new therapeutics get funded. I'm curious, what's driving all this now, from your perspective?
I think you have two things. On one end, patients just used to be very passive, and it was a very paternalistic system - the doctor would say do this, and they'd say okay, and that was kind of the end of it. It's much more common for me to see a patient in the ER and say, not only did my heart feel funny, but on my Apple Watch I actually have it - or they'll bring in their Kardia tracing; they'll say this is what my EKG looked like while it looked funny. And Kardia is very - like, you know, that's something that we utilize, we recognize. The big issue with the data is it has to be reliable to both parties and in a way that's shareable. And so when someone can say my heartbeat felt funny and they show me the EKG tracing, I'm like, oh, yeah - I could keep you on a cardiac monitor here for a couple hours in the ER, and I'm probably not going to see this. But now that you have it, it's totally verifiable. We don't need to keep you overnight. We can get you on a monitor. We can get you on medication.
And so I think you have a rise of consumers that are just more knowledgeable and more involved in their health care. And at the same time, while there's very traditional medicine and a very patriarchal kind of mentality, most young doctors use these devices themselves. They're familiar with it themselves. And there's a goal in medicine to kind of upskill. You know, we went through med school, residency, and AI didn't exist. And there's an interesting quote from the AMA, and it says, you know, AI is not going to replace physicians, but physicians that use AI are going to replace the ones who don't. And it's like, we need to make sure we've maximized our ability. Well, I'm going to have nowhere near the ability a large language model does to bring correlations and bring that together, or give a patient a discharge instruction in their native language according to their education level. But I could totally utilize those in my process, in my care, to personalize care. And I think you have kind of a wave of patients, super users, and then physicians wanting to provide better care. That's why they went into it in the first place. And they're realizing, boy, we need to understand how these tools can help us in our roles.
What are your thoughts, Max?
Just like timing, right? Why not? Yeah, it's really interesting. I remember in 2016 I would wear an Oura ring, and everyone would make fun of me for tracking my sleep. And in 2019 I wore a CGM - a continuous glucose monitor - and people thought I was actually psychopathic, putting a microneedle in my arm, right? Then - so this has panned out time and time again. And so many of the things that I've been obsessed with for 10 years that everyone has made fun of, people now are constantly messaging me about and texting about and asking about. I think I do believe to an extent in the technology-shift version of history as well as the key-man or key-person version of history.
I think the key-person version of history here is that Huberman and Bryan Johnson came along and started popularizing health. I think the technology-shift version of history is pro-social media to an extent, which valorizes health and aesthetics, and then also AI to an extent, where 8 to 12% of questions are about health. So there are a few of the things that are catalyzing this push towards preventative medicine and what many are calling longevity medicine.
It's really interesting - if you look at why people follow Bryan Johnson, it's very rarely to know how to live longer. The stuff that he's actually selling and they're following him for is very basal needs: how do you not lose your hair, how to reverse your skin age, how to sleep good and feel good, how to maybe put on a little bit more muscle. They're actually surprisingly basal needs. And this is - people have now realized that health is a tool to help you achieve those base needs. And just as new consumer products come along, when someone sees a new tool, I contend that there's a consumerization moment here. And some people say, 'Oh, we're consumerizing healthcare.' I don't actually think that's what's happening. I think we're just healthizing consumer - which is the way consumer works: someone takes a consumer product that's packaged, a new CPG product, and they find one that they think people are going to like, and they realize you can sell it. Hims realized that with pills. Function realized that with blood tests. I think many of these companies, at least in the short term, are well understood as CPG companies doing CPG marketing, and then the big bet has to be: can they actually turn into health companies long term? Because CPG caps you at a billion to maybe 10 billion market size, whereas if we can build a new health system, that becomes a little bit more interesting.
No, for sure. It definitely feels, especially when you look at an Oura ring, right? They're trying aggressively to make that transition, right? It's like, we used to sell a ring; now we sell all of these services, down to helping you get pregnant, right? - that are more health related. I'm curious what your perspective is on Bryan Johnson. Talk about him for a second, because he is a little bit of a controversial figure out there with all the crazy things that he's done. And I'm curious to get John's...
Perspective on him, from maybe the medical perspective first. So he's actually a really interesting figure. I'm part of a group called Clinicians UBC, and we had him as a guest, and people wanted to talk with him, and suddenly the things... he doesn't stress as much as he used to, but ultimately I like where he's coming from, that age isn't inevitable, that we're not going to get worse, or if we are, there's preventive steps that we can do to make that happen. You know, he's kind of like on the shock value of that side. Someone in that space that I put a little more credence behind is Eric Topol, who just wrote a great book called Super Agers, and that talks about, you know, we have all these problems with healthcare, the fragmented data, all these things, but we could use more immunology, and we could focus more on things like community and meaning that make people's lives last longer. And obviously we're not in a situation where we have a healthcare system that reimburses for meaning and community, but if you're focusing on what helps people live longer, you're focusing on longevity. You know, it's hard to ignore the data that he has associated with that.
So, you know, Brian's kind of like the rock star, and he's interesting to me also because he's like, you know, I think he's a BYU guy, isn't he? He's from Utah originally. Oh, yes. Yeah. Um, so, you know, I'm always interested in folks from Utah and BYU doing interesting things. But at the same time, you know, Eric Topol will write a book every five or six years, and then the next decade goes on to prove that everything that he said was exactly right. He wrote, you know, The Patient Will See You Now to start digital health. He wrote Deep Medicine, right? Before we started using AI or even comprehending how to use it. And now he's writing Super Agers, and so I put a lot of stock in what he says. If he's focusing on utilizing components of immunology and recognizing that and recognizing what keeps people alive, he's got a lot of street cred, I think, on any spectrum you're working with.
Any perspective to add there, Max?
I have a lot of respect for Brian. I reflect a lot of the things John said. I think that Brian thinks very deeply. I think that he's creating a religion with really syllogistic underlying beliefs and principles, and I've come across many wannabe messiahs that don't have strong underlying beliefs and principles. I think he thinks very deeply about the world post-AI, and I think very few people have thought as deeply about that, or have come to conclusions that I think map as accurately to where we're going. And one of the conclusions is that we do not know, right? And the implication of 'we do not know' is: what do we know? We know at this very moment I do not want to die. And that is a really powerful foundational belief. I think that he's learning when it comes to health. Like, I won't say I'm any better, but I think that he's not quite at the frontier yet, right? Like I see a lot of people, a lot of my doctor friends, like, 'Why is Brian doing that? It's so dumb.' And then like two months later, Brian will update his beliefs.
I think he focuses a little bit too much on data. I do believe in the fundamental principle that we should just follow the data, assuming we have all of the right data points, but we do not, right? We do not have data to capture everything yet, and as a result you cannot just follow the data. I do think that the subjective is also important. We cannot be purely objective here, the subjective actually matters. I also think he's rotated too far in the direction of homogeneity, and I would contend that one purpose of life is experiences per unit. Right. And what he has basically said is no experiences: same place every day, same time, same meals, same everything. And is that living? If we draw that to the extreme, like you eat through a test tube and you sit in a chair. So I think he's rotated a little bit too far in that direction. I think the marginal utility of some of these things is quite low. So that's kind of my read. He's rotating back a little bit right now though. He's not having his Whoop perfect sleep days anymore.
Yeah, look, I think there's a strong... if I had more time to do Twitter, I would love to create a strong, like, opposing narrative which is reasoned and comes from a place of productive discussion. Which is: what happens if wine is actually good for you? What happens if not fully obsessing over sleep is actually fine for you, right? And if we look at just epidemiology alone, we can look at blue zones, and they do not have perfect sleep. They drink wine. They eat like... they do a lot. Some of them eat red meat. Like, they're not perfectly optimizing these things, but they have meaning and purpose in life. They're positive. They're not constantly judging. They have a positive mental attitude. Um, so I do think that there's an opportunity for people to rotate slightly back in the direction of what a lot of people would call living, rather than the purely objective follow-the-algorithm.
Yeah, it is interesting. He came and spoke at an event that I was at, and I mean, he made the argument — he's like, 'I've never been happier.' He's like, 'People ask me like, hey, you're doing all of these things, you cut out all of this, you know, all these things that bring a lot of other people happiness, right?' And he was like, 'Yeah, but I just have never felt happier in my entire life.' So, I don't know. It is interesting because I hear what you're saying and resonate with it, but also not sure if what he's saying is actually accurate or not. And don't know if I'm ready to make all the sacrifices to test it and see either, right? You're not really going to use it on yourself. I can relate to his focus and meaning a lot though — like he's found something he's after. He is definitely goal oriented. He's definitely setting metrics for himself, and he's achieving a lot of them. So I think there's something for that. I am curious though to double click on this idea of more of that holistic view on health. It's not just the data. It's not just, you know, getting a perfect night's sleep, but maybe enjoying life. How do you both think about how that fits within the forward-looking paradigm of healthcare? Like, how are you fitting that into Superpower? How are you fitting that into your daily practice as a clinician, John?
That's a tricky thing for me because I'm more likely to, you know, treat a heart attack or fix a dislocated joint, and so it's harder to say, 'Now you need to find your meaning with this exercise with your new shoulder.' But at the same time, I think that has to be recognized. If you're having a life without any purpose, you know, there's no ICD-10 code for that, there's no billable tool for that, but it's meaningless, and you see that. And like the longevity studies, people need food with skinny vision, but the stress might be the thing that put them into the emergency room, right? I mean, they had a weak heart, but then it was, you know, stress at work or family, whatever — the genetic weakness, and then you needed that component of the stress. Now, that's — I think that's fair because what I do is the quintessential version of the reactive healthcare. I'm not in a space for that. But I think, you know, once we get that guy's heart attack diagnosed, we get that clot buster in, we get it fixed, I think, you know, where that component should take over is his cardiac rehab, right? Let's get some exercise in your life. Let's get these other things that are preventive in your life. Rather than, you know, just stenting a vessel when it gets blocked, let's prevent your other vessels from getting blocked.
