In this very first episode, I'm joined by Brad Perkins, CEO at Sapiens Data Science, Inc. for a dive into the likely emergence of a new healthcare industry, one built upon datascience (rather than medical practice), human-computer integration, consumer-driven, and for healthy people (as opposed to sick or injured people).
Lee: Hello, and welcome to the inaugural Hyper Wellbeing podcast. On today’s show, our first ever, we have Brad Perkins. His biography is, Brad is a visionary physician and scientist who recently left a position of Chief Medical Officer at Human Longevity Incorporated, found a new startup, Sapiens Data Science. At HLI he was responsible for leading all clinical and therapeutic operations, including collecting and utilizing phenotype data, development of the consumer clinics business, and guiding stem cell therapeutics. Prior to HLI, he was executive vice president for strategy and innovation and chief transformation officer of Vanguard Health Systems. Brad started his career at the Center for Disease Control and Prevention, the CDC, in 1989. After completing his residency training and chief residency in internal medicine, he served as Chief Strategy Officer for six years before leaving the CDC in 2009. Hello and welcome, Brad.
Brad: Thank you very much Lee. I’m delighted to be here. As you know I’m a big fan of your Hyper Wellbeing construct and a big fan of you as well, so thank you.
Lee: Brad, I very much appreciate the kind words and the support you’ve shown for ideas which were very nascent a few years ago. There’s been very few people out there who understood in those early stages what I could see and what I was talking about.
Lee: Let’s go back to the position you just left as the Chief Medical Officer of the Human Longevity Incorporated. What were you doing at HLI, and why did you leave recently?
Brad: I’m a big fan of Craig Venter, for a long time. We interacted several times when I was at CDC. In my early time at the CDC I was very involved in the last revolution in genomics, which occurred in bacteria. And unless you were working in bacteriology at that time, you may have missed it.
Brad: We provided consultation to Craig Venter in his selection of a haemophilus influenzae strain, to serve as the bacteria that he did the first whole genome sequence on back in the ’90s. It was published in 1995.
Brad: And then I worked with Craig again during the anthrax investigation, which I lead after 9/11. And that was the first time that a bacteria had been fully sequenced in the context of an active public health investigation.
Brad: So when I started knocking around after leaving Vanguard Health Systems back in 2012, I came across Craig. And he was just getting ready to start Human Longevity, and I got excited enough about what he was doing that I joined him as the Founding Chief Medical Officer.
Lee: How do you feel the genomics revolution has went? My impression was that it hasn’t been the revolution people were expecting, i.e. a single gene for a single disease.
Brad: Now I think we’ll look back at some point in the future, and I hope it’s not too long, and I think the parallel to bacteriology is a great one. 50 years ago, in bacteriology we grew bacteria on a plate and we poked and prodded them and put them on slides, and see how they stained, and exposed them to sugars to see how they metabolized sugars. We would derive from that set of phenotypic characteristics, the identity of the bacteria and various aspects about pathogenic factors.
Brad: And fast forward to, actually it was a very brief time in the ’90s and early 2000s. We went from bacteriology as it was practiced 50 or 100 years ago, to bacteriology being essentially a quantitative computer science. And I actually hired the first pure computer scientist for the CDC’s Bacteriology Reference Lab right after the 9/11 attacks actually.
Brad: And I think, if you fast forward that to humans of course, we had to wait a while for technology to improve to the point that it started to become affordable. And although it may be slower than people have expected, I actually think it’s quite remarkably fast, what’s happening. When you think about the scale, and everything that needed to go right to get us to where we are today.
Lee: But do you think there was much too emphasis on genetics being the determinant of your health?
Brad: You know my strong point of view is that genomics is critically important, but it’s not going to be a total answer or solution to advancing health and healthcare. Stated another way, it’s gonna be part of everything that we do in health and healthcare in the future, but it’s certainly not gonna be a standalone solution.
Brad: And one of my other strong points of view is that there needs to be a much more deliberate effort at integrating genomics with legacy clinical knowledge, and demonstrating that the addition of genomics actually improves the performance of models and clinical decision making.
Lee: Do you have any opinions I wonder, on epigenetics? Which seem to be the new, I don’t want to call it revolution, but I would call it the latest surge in interest, genomically.
Brad: Yeah it’s however you want to label it, epigenomics is still genomics. And I think we’re learning more about how dynamic the genome is, in terms of its expression. And epigenetics is, I don’t find it a highly useful term to express that realm of knowledge.
Brad: I think it’s critically important for us to get a much better understanding of the correlation between the linear sequence, and phenotype expression and health and healthcare. And of course, there’s lots of things that modify the expression of the genome in that translation from genomics to phenotype. And we clearly need to understand all of them, whether they’re labeled epigenetics or not.
Lee: Well I appreciate the answer. It makes sense. Jumping ahead, you’re now co-founder, director, and CEO at Sapiens Data Science Incorporated. Where and how did you get the idea? I mean, what was the impetus, the flash, the gut conviction?
Brad: Well, I had a tremendous experience working at Human Longevity and working with Craig, and the extraordinarily talented people that he gathered there. I was very involved in building the Health Nucleus, which was a direct medical services platform, that combined whole genomics and other advanced OMICS with advanced imaging. I continue to believe that that is the paradigm for the future of health and healthcare.
Brad: But after four years at Human Longevity I had accomplished what I wanted to be able to do there. The Health Nucleus continues to be a main commercial focus of the company, and I hope they’re able to scale and continue to grow that platform.
Brad: I left back in November, and very excited to have identified two co-founders, Larry Rosenberger who is a important part of Fair Isaac Corporation’s history. He served as a CEO at Fair Isaac, also known as FICO, which has been the long term pioneer in credit scoring.
Brad: And I met him in the context of his service to the Buck Institute which is a longevity research institute in Marin County. He was looking for a partner to begin applying some of the thinking and tools that FICO had used over the years for credit scoring, to health and healthcare.
Brad: And so we, Larry and I partnered with another co-founder, Lou Gerken, a long time venture capitalist in the Bay Area. Also a member of the board of trustees at the Buck Institute.
Brad: All of us got very excited about this being a particular moment in time, where we could start to build a new paradigm, around some of the ideas that FICO has leveraged over the years, and blending in trends that are very omnipresent right now in advancement of genomics, and bio-sensors, as well as consumerism in health and healthcare.
Lee: Could you let people know what FICO is? Because I don’t think the majority of people will know Brad.
Brad: Yeah so FICO is the current acronym for Fair Isaac Corporation, a long time San Francisco Bay based startup that actually has been around since the late ’50s. Was initially led by the founders who gave their name to the company.
Brad: And they decided that, at a time that was pretty early, they decided that data and analytics could be used to help businesses make better decisions. And they began the development of a credit scoring algorithm, and a score that represented the results of that algorithm, that has become the universal standard in credit risk assessment.
Brad: And so they now produce billions of these scores across the globe. And the scores are used in making credit decisions. And there’s now a whole family of these scores that relate to particularly common credit transactions, whether it’s getting a home mortgage, or getting a car loan, or a credit card. These FICO scores, these credit scores are used as a standard in the industry.
Brad: FICO’s now starting to leverage this same notion of gathering data from diverse sources and using high performing algorithms to create scores in other domains. They recently released a driver safety score.
Brad: Sapiens Data Science is founded on the notion that we can apply some of the same principles that have been used in credit scoring, to health scoring, and use that as a way to communicate with customers, individuals, people and families, about their health status.
Lee: Back at the debut Hyper Wellbeing event, where you were kindly a keynote speaker in November 2016, in the hallway you were very kind and you had said to myself that you really liked my ‘The 3 Pillars of Wellness-as-a-Service‘ blog post. Human-computer integration, redefinition of health, and decentralization of healthcare.
Lee: Why did you like it so much and has your view developed between when we spoke in the hallway?
Brad: I think Lee you’ve done a fantastic job of catalyzing things that I’ve been thinking about for a long time, and doing it in a way that is understandable, and compelling.
Brad: I feel like the company that we’re building as Sapiens Data Science is partly catalyzed by this construct. And I would hope that you would see that as you being to see some of our products come to market.
Brad: We’re firm believers on human-computer integration, that the health and the healthcare system needs to move from its current qualitative clinical practice modality, to a firm data science foundation. And there’s an incredible opportunity to do that.
Brad: And that, on redefinition of health, there’s an incredible amount of science that’s available to start that engine. And then move increasing control to individuals and families, for continuous monitoring and protection of their health, improvement and protection of their health.
Lee: But you mention that medical practice today is very qualitative. So what you’re really saying there is a large gap in what could be done, if I put it politely.
Brad: I think there’s two huge gaps. One is the generally poor decision making, which has repeatedly been demonstrated in the current healthcare system. Because we’re constrained by the way our brains operate. So generally poor decisions are made because they’re not based on the level of data that could be available.
Brad: But the other dimension that I’m really concerned and disturbed by is the time gap, between the availability of scientific evidence and the broad scale medical use, or healthcare use of those pieces of evidence.
Brad: And Lee as you know, and as we’ve discussed, it’s been repeatedly measured in a number of developed countries that that gap is about 17 years, with expected variation around that sort of median number.
Brad: But that’s just not gonna work in the place that we’re moving. And unless we have very different platforms to protect and improve people’s health, we’re just gonna get further and further behind of where the science is, and we’ll have more and more missed opportunities for people to protect and improve their health.
Lee: Are you building a cloud based supercomputing service, which is going to only work for the upper echelons of society?
Brad: No, I don’t think so. I certainly hope that’s not the outcome, it’s certainly not the intent. I actually think that we have a tremendous opportunity to do for health and healthcare what mobile phones did for communication, in less developed economic setting where many countries had the opportunity to skip the landlines and go straight to cellphones and increasingly smartphones.
Brad: And I’d like to see many countries have that opportunity with platforms like the one that we’re building at Sapiens Data Science. And I think all the pieces are increasingly available.
Brad: Lee I don’t think we can jump immediately to that solutions. And the challenge is how do you get from here to there? And while you’re building a sustainable business. But I think utilities like the one that I know you imagine are within near term reach.
Lee: Personally I think you’re gonna end up, I don’t mean you personally, but the industry, you’re going to end up with different plans according to your income.
Lee: You could be paying 10 bucks a month to 100, to 10,000 a month if you have the money, and the computing will be run against it. Health is our greatest asset, and I think the more money one has, the more one will be willing to pay a monthly fee to protect and improve one’s health and family health.
Brad: Yeah. And I like your analogy to financial services. As you know, I think the financial services market, the way it has been shaped, is actually a very good analogy for where we’re going with health and healthcare.
Brad: If you take a look at a company like Schwab and you’re interested in getting some help managing your portfolio, they have three broad categories offering. One is machine alone. The other is a little bit of a human and a machine, and the other is all human. And they’re priced according to the cost of delivery of those services. I think increasingly we’ll see health and healthcare offerings along those same dimensions.
Lee: I can only concur there. It’s what I also envision. I don’t think that healthcare today does a great job on prediction. And I definitely don’t think it does a great job on prevention.
Lee: It did great in the 20th century with bacteria and viruses, you know the whole public health thing in fact. It did fantastic. But in the 21st century it’s not meeting the needs of 21st century diseases, i.e. lifestyle chronic diseases.
Lee: And I firmly believe that there’s a new adjacent healthcare, a secondary healthcare beginning to emerge, at least at the seed stage. And it’s one built on the paradigms of prediction and prevention, and optimization.
Lee: And I don’t think the existing healthcare should deal in that space. The existing healthcare should remain for sick people and injured people. What’s your position on that Brad? Do we need a secondary healthcare? Will it be separate?
Brad: I think it’s really and interesting question Lee. I’m not sure how it will shape, or how it will end up being shaped in the marketplace. I think there is a reasonably high likelihood that you’ll be right. That you’ll see “sick care” break off from other health services that are focused on prediction, prevention, and optimization.
Brad: And you know, that’s probably more likely than not, to be honest with you. But what I’m really confident of is that there is now an extraordinary opportunity in prediction, prevention, and optimization. And I don’t see anybody, really executing with the kind of vision in mind that I think would get us there quickly in the marketplace, today.
If you take a look at a company like Schwab and you’re interested in getting some help managing your portfolio, they have three broad categories offering. One is machine alone. The other is a little bit of a human and a machine, and the other is all human. And they’re priced according to the cost of delivery of those services. I think increasingly we’ll see health and healthcare offerings along those same dimensions.Brad Perkins
Lee: Yeah you excite me there Brad. Because for me this was just like pre Android 2006, when I was seeing people putting Linux on mobile phones, and doing their own mobile Linux. And I could see what was coming. Then I saw the Android specs and I was just blown away.
Lee: And once again I saw something as exciting, but this time much more meaningful. For myself I used to be interested in mobile, but now I see mobile as the hub of future health. And as we’re totally aligned, prediction, prevention, and optimization.
Lee: There’s people doing pieces, but it’s all siloed type stuff. And there is no industry umbrella to take it together and to catalyze it. And this is why I started this show.
Brad: Yeah I think you’re on an extremely exciting track and I’m gonna be eagerly awaiting your future guests, and I think this has a really long and interesting runway.
Lee: Yeah I think over the next two, three years you’re going to enjoy a few of the guests. I have a few more lined up.
Lee: Anyway, prediction, prevention, and optimization, do you agree it’s more of a computer and data science than what we would term medicine today?
Brad: Absolute. So medicine today is qualitative clinical practice. And you know, there’s the only word I’m comfortable with among those three is the clinical part. I’m very uncomfortable with qualitative, being applied to the kinds of high risk decisions that physicians are making on behalf of individuals. And certainly practice is a suspicious term as well.
Brad: I do think though, I may have a little bit of difference in opinion with the role of the physician. And I think there’s going to be a prolonged, very constructive period where physicians and other health professionals, and I say that because I’m very interested and committed to behaviorists becoming a bigger part of health and healthcare, are gonna be partnering with machines to optimize outcomes.
Brad: And I think that a partnership between humans and machines is gonna allow humans to move up, Maslow’s hierarchy of health if you will. And begin to connect the things that people need to do in the health realm with larger purposes in their lives.
Brad: And I think it’s gonna be really a long time before we have computers that are gonna be able to play that vital role. And so I’m not counting doctors out of the game anytime soon.
Lee: But it’s a different type of doctor. Because if I was to quote you from the event of mine that you spoke at, you said, if I read it out here, “Medicine has traditionally been a clinical science that’s been supported by data. We are rapidly approaching a time when medicine is about to become a data science, supported by clinicians.” So that means we need an entirely new breed of clinicians?
Brad: I think so. And I still stand by that comment. But I probably I’m highly inclined to not dismiss the role of humans in working with other humans, particularly where behavior change is involved.
Brad: And we’re in an awkward period right now where in many dimensions we live in environments that are pretty toxic to our health. And to overcome that toxicity, you have to be pretty behaviorally resilient, if you will.
medicine today is qualitative clinical practice. And you know, there’s the only word I’m comfortable with among those three is the clinical part. I’m very uncomfortable with qualitative, being applied to the kinds of high risk decisions that physicians are making on behalf of individuals. And certainly practice is a suspicious term as well.Brad Perkins
Brad: And my experience personally, and watching other people is behavior is most likely to change in the context of human relationships, and not necessarily machines alone.
Lee: I definitely agree with that, with B.J. Fogg etc.
Brad: Yeah. So I want both, Lee. And I think it’s gonna be an exciting ride.
Lee: Leroy Hood said that he thinks within a decade, what he calls scientific wellness, a term I actually came across after putting the Hyper Wellbeing thing together, which seemed to be the same paradigm shift, it just put more emphasis on people, and was missing a couple of elements.
Lee: He said that he thinks the scientific wellness as he had termed it, rather than Hyper Wellbeing, will have a greater market cap one day. And he was speaking like within a decade, than today’s healthcare i.e. sick care. Do you think that the wellness industry, the scientific wellness, that data-driven wellness industry is likely to have a higher market capitalization than healthcare today?
Brad: I would agree with Lee [Leeroy]. Big fan of Lee’s and he’s one of my heroes. And I think he’s on the right track. And the things I’ve heard him talking about over the years are also very consistent with the construct that you’ve been building at Hyper Wellbeing.
Lee: Yeah and the difference, I come from a much more machine, device, software viewpoint. Because when Lee first raised this, I think this was 2003, looking back, you know there wasn’t AI and it’s not been his industry, whereas my background has been computer science.
Lee: And I more saw the rise of the machines. And also I see the cost of healthcare going up, and I see that healthcare isn’t aligned with the individual’s health. And I actually don’t see that changing anytime soon in healthcare. So consumer devices and software are obviously, a conduit. And also it drives cost down because people don’t scale.
Lee: In my viewpoint in a decade or two, you’re looking at an end outcome of what I would term ‘the machine knows best’, rather than ‘the doctor knows best’, i.e. trust the algorithm with your life choices. And this ties into what you said about behavior change to live longer and healthier.
Lee: But do you think ultimately we’re going far less top-down, far more decentralized, far more consumer-driven, and far more trusting of the algorithm and the machines?
Brad: Absolutely. But, my caution is that I’m trying to think of ways to get leverage to move there as fast as possible and I think to move there as fast as possible you’ve gotta bring early adopting elements of the existing health enterprise into the construct.
Brad: And one of our lessons early in Sapiens Data Science is that to get to scale with the platform that we’re trying to build, people are telling us they want their doctor involved.
Brad: And so we are building a platform that I think honors your desired direction at Hyper Wellbeing Lee. But I’d like to wire early adopting physicians into the process, believing that there are some small, admittedly a small proportion of them that would like to come on this journey as well.
Brad: And it’s not gonna be all of them, but some of them will want to come along, and that if we take that approach we’ll end up getting to where I think people are best served faster than taking different approaches.
Lee: Well I hope that you do manage to get a fair percentage Brad. At the moment GPs or primary care physicians, from my experience they’re actually even lacking in what I would term basic statistical knowledge.
Lee: So they often confound say relative and absolute statistics, even to their patients. Which is kinda diabolical. I.e. a statin for given patient may give them four days extra life, yet have a whole host of side effects.
Lee: So from my experience, the average physician, primary care physician is lacking in stats. And also things like all cause mortality. I.e. take a statin, and yes it might reduce your risk of cardiovascular disease, but it then increases something else. They don’t look at all cause mortality.
Lee: I find that they’re overwhelmed, have a short time with patients, and it’s really a data problem, and they’re not doing very well with it.
Brad: I would agree with that. But ironically, I think the way to transition primary care physicians, or GPs to a firmer data science based foundation is to work with consumers. And have consumers bring their physicians along, those that are willing along on this journey.
Brad: So many people that are doing what we’re doing in Sapiens Data Science right now have the engine pointed at the healthcare system. And I ultimately think that’s gonna be less effective in the long term than building an engine like the one you’d like to see in the world, and pointing it at consumers. So that’s what we’re doing at Sapiens Data Science.
Lee: Well that sounds very strategic. And it sounds, a guesstimation that it’s more likely to work. And obviously you’re touching the marketplace and I guess you’re already getting all the signals, and potentially some traction, I don’t know how much you’re able to reveal, but I assume you’re getting what we’ll term solid marketplace signals to make that decision.
Brad: Yeah I hope that’s true. But we’re certainly reacting to consumer centered design work that we’re doing and interpreting those signals in regard to product market fit.
Brad: And the dynamics are real, for a venture backed startup in terms of getting traction and getting early revenue. And that capital is essential to being able to build out the long term vision of that platform.
Lee: Well I appreciate you sharing those signals, which takes me onto something else you had said at my event. You’d said, “It’s unlikely that this revolution is going to be led by the incumbents in healthcare. And one of my beliefs right now is that one giant industry which is incredibly well aligned is life insurance.”
Lee: “And life insurance has an opportunity I think to birth this industry,” what I’ve termed Hyper Wellbeing. “And I think that the future of health and healthcare is actually going to be more like wealth management and asset management than what healthcare looks like today.”
Lee: Can you elaborate any on that Brad?
Brad: Lee thank you. I’m glad I said that, I still stand by that. I’ve deepened my experience with the life insurance industry, and the reinsurance industry. And you know it’s not sure, I’m just not convinced that the incumbents in that space are going to make the transition.
Brad: I am very interested in watching a company Health IQ, that’s getting some traction around brokering deals among health conscious consumers, and life insurance companies, to get reduction on the life insurance premiums.
Brad: But it seems like a pretty small step. And I believe, it’s very frustrating to me, I believe the life insurance industry could actually own this industry that you describe so eloquently around prediction, prevention, and optimization.
Brad: They even have a large part of the right workforce in place. They certainly have the plugin to the financial services industry. And the value proposition of life insurance that extends life, seems really compelling, particularly to Millennials potentially, where life insurance is not a growing business sector.
Brad: So still believe it Lee, what I said. But I’m still looking for the signal in the marketplace.
Lee: I don’t know if you’re able to say anything here, and again I don’t want to feel I’m pushing you into my words. But healthcare today is not aligned to the individual’s health. I mean it’s hard not to laugh at that. I mean if health is our greatest asset, but healthcare is not aligned to our health, and I don’t see any way to fix the current system.
Brad: Well that’s right. And I used to start out my meetings with senior executives in the life insurance industry and make that point that I would rather have, as an individual and for my family, I’d much rather have life insurance involved in my health and healthcare than I would my healthcare company. I mean my healthcare insurance company.
Brad: And they’re always sort of startled by that. But I’m at least very confident that life insurance is highly aligned with extending my productive life. And I say productive life because, not only do they want me to live long, but they want me to continue to make my payments. So in order to do that I have to live long and productively. I think it’s a great alignment, but so far it’s been squandered.
Lee: What about health insurance? Because it seems quite happy just to take 15% margin, and keep throwing invoices around and taking 15%. I don’t see how health insurance begins to align under that model of taking 15%.
Brad: Well first of all, you’re gonna touch on an area that I find really frustrating. I mean credit card transaction fees are in the range of 2 to 3%. That’s actually how the health insurance industry should function for what it does right now, because as you know, at least in the United States, administrative services only, is sort of the model.
Brad: And what health insurance companies are doing is claims processing. So to get 15% for that is really frustrating. I don’t see how they can get aligned, unless they’re actually at financial risk, and they’re generally not. So I don’t see a lot of hope with the alignment.
Lee: There is a lot of excitement about precision medicine and I understand that. But in my opinion I’m more excited about what I would term ‘precision lifestyle’. You know a lifestyle is in alignment with say your epigenetics, with your microbiome Etc.
Lee: Because by having a precision lifestyle, as something computed, you’re far less likely to need medicine in the first place.
Brad: I would agree with you. It’s just that in terms of noise in the marketplace, it’s hard for everybody to sort of understand the distinction that you’re making that I think is extremely valid.
Brad: But I think that will be probably a generational phenomenon, but I think it will come.
Lee: Yeah it seems to be that people have been schooled that health is predetermined at birth. That seems to be one major issue in cultural consciousness.
Brad: Yeah I think that may be true. And the whole, it’s actually a pretty distinctly Western idea that you can control your destiny. But I think it hasn’t, although it may be accepted in Western cultures, from sort of an economic or social perspective, I don’t think people really get that that’s gonna cross over into the health realm.
Brad: And you are gonna have more and more opportunity, and more and more responsibility to monitor, protect, and improve your health.
Lee: Yeah. And we’re not talking like a letter in the mail, or a quick chat with your physician. We’re talking about realtime, kinda on your wrist type life guidance.
Lee: “Make this choice, or else you’re shaving seven days off your life,” type thing.
Brad: Yeah. And you and I have talked about our mutual experiences using continuous glucose monitors, and that’s been pretty mind blowing for me to tighten up those feedback loops on what you’re eating and …
Lee: Yeah it’s amazing behavior change. We know Apple’s coming out with a glucose device and a few others, which will make it mass adoption. And they talk about using it for sickness, i.e. diabetes. But once you’re continually measuring glucose, and I believe that glucose is a reasonable proxy for insulin – the main driver behind most chronic disease that we have today. Insulin spikes and high insulin levels.
Lee: And I think that once you can see it on your wrist, I know from my own experience that I change my eating patterns. I go for a walk after eating because, “Hey it’s a bit too high,” after eating.
Lee: Or I have larger gaps, or I eat less. But once I see that number every day, and I know what the values mean in terms of longevity, I start as you said, closing those feedback loops and I start adjusting behavior.
Lee: The issue I think we have today is people are blind to what’s happening in their blood sugars. Like if you’re out at night, you wouldn’t be adding the Coca Cola to the alcoholic drink if you could see what it’s doing, or drinking that Red Bull.
Brad: Yeah. So I’m a big believer that we need to broaden the adoption of these technologies to certainly people with pre-diabetes, and then get out there beyond that.
Brad: And I put these monitors on my adult children, and they had the same response that I did. Is like, “Why doesn’t everybody have this kind of information all the time?”
Lee: Absolute Brad, and I’m wondering myself. Mainly glucose plus a few other biomarkers are gonna become available. And I think we need them all the time. And it’s part of that path of human computer integration.
Lee: Realtime biomarkers.
Brad: Yeah. We’re facing this challenge in Sapiens Data Science right now is, getting passive data on what people eat and drink is extremely challenging.
Brad: And I’ve been thinking a lot about, “Well how much is continuous glucose monitoring, a pretty darn good proxy for what you’re eating and drinking? Do we really need to innovate aggressively in the food area? Or can we use proxies like continuous glucose monitoring? Maybe the microbiome, maybe global metabolomics to actually know what’s going on with eating.”
Lee: Yeah. And it takes me in to ask a question on the side here. Do you also see a rise in direct to consumer lab testing? Because although we’ll have devices that gather more data, I personally see ever rising direct-to-consumer lab testing.
Lee: And that takes in a whole bunch of new data, into the cloud potentially to guide your life. So do you see a rise of DTC?
Brad: I do. I do on the lab testing and of course, I think the entire regulatory environment is shifting in ways that are quite remarkable, particularly given our level of dysfunction in some other dimensions.
Brad: But you know the US government has been consistently driving for several important areas. One is individual control of data, which is a bid deal. Health and healthcare data. Individual control over access to laboratory testing, which I think is overdue and really fantastic.
Brad: And then you know with the 21st century cures act that was passed right before President Obama left office, I’m really encouraged around the support it provides for clinical decision support, and also in particular for patient decision support.
Brad: Guidelines that have recently been articulated for Software as a Medical Device (SaMD) I think are extremely exciting and supportive steps for the industry that you and I would like to see develop.
Lee: Well appreciated for that insight. I can’t remember if I ever said to you, I think I did, and will only briefly mention it here. I didn’t feel good, I had no healthcare because I hadn’t completed paperwork. It would have been free if I had, I was in a Second World country.
Lee: And I was able to walk into a lab, and order any test the doctor could and get it back the same day electronically. So I walked in, said I didn’t feel good. They took a bunch of tests, said my glucose was high. I didn’t really know what glucose was, I knew it something to do with diabetics.
Lee: And then I said, “Hey am I able to change this?”
Lee: He said, “Your food changes it.”
Lee: I thought, “Will I sign up for healthcare or will I play with the data?” So I thought, “I’ll play with the data.” So I just kept eating different things, going back and doing the test again every day. And eventually I reversed my type 2 diabetes.
Lee: I say eventually, it only took me two months, and I felt better than ever. I lost all the weight, sharper mind, slept better, simply had the macro nutrient ratios wrong in my diet. I had high carb, healthy carb I might add, vegan, and then I went high fat, and low carb and suddenly I was the best shape of my life, and the best mental performance of my life.
Lee: And I just saw it as a data problem, and solved it within two months, never thought anything of it. Then two years later, that’s when I began to discover, “Hey healthcare wouldn’t do that.” I haven’t been to the doctor in 17 years.
Lee: So I really don’t know what a doctor would do. I have many doctor friends, and when I speak to them I don’t feel it would be of use to me unless I had a bacterial infection or something, or an injury, and then I would be greatly appreciative.
Lee: I had access to my body data, and what I started doing was walking and after I solved the diabetes I started thinking, “What else can I improve?”
Lee: So I started walking into the lab, ordering tests [inaudible 00:46:29], and studying blood chemistries. And I started, for example I tested magnesium, magnesium RBC I might add, ’cause the standard test and it doesn’t tell you anything [inaudible 00:46:39], only that you can keep homeostasis of magnesium.
Lee: So then when I noticed I had low magnesium I took magnesium, and wow I had a sense of wellness, and then the same thing with vitamin D3. I felt like I hadn’t felt in 15 years. I had more energy, could get off the sofa quicker, whereas before I always had lag and I just fell into this and I’ve not stopped looking at body data and understanding blood chemistries etc.
Lee: And so that brings me on the question that in many countries you can just walk in and get it pretty cheaply and emailed to you, it’s your right. Nobody asks you a question. I find it very unusual in the US, it can be hard to obtain your own body data.
Brad: No I’m total agreement with you. And I think that’s gonna evolve quickly. I love the fact that, less developed economies may innovate more quickly in this space than more developed economies and-
Lee: Yeah it’s fantastic. But I ended up seeing markers later on which can indicate liver cancer, etc. And I suddenly thought, “Gosh this is quite alarming, they just email you this, like the raw data.” But I liked the freedom and the liberty and the personal choice involved.
Lee: Anyway, to move things along, at that event just a final quote from you. You had said, “People are dying unnecessarily as a result of the current medical model. And I think we can fix, or begin to fix that in the near term.”
Lee: So let’s just be clear, people are dying unnecessarily because of today’s model.
Brad: Yes, I think they are, in large numbers. And it’s something really disturbing to me. One of our areas of focus at Sapiens Data Science is to build a set of next generation risk models of for diseases that are associated with premature mortality.
Brad: And I talked about these data at your first Hyper Wellbeing conference. But this particular framing of a problem is not easily at hand for most people. But for men in the United States in 2016, between the ages of 50 and 74, the cumulative mortality rate is a third. And for women it’s a quarter.
Brad: So just to be clear, these are people that reach 50 years of age, but don’t get to 75. And so every one of those deaths is a tragedy for the families and communities involved. My father was one of those deaths, was sudden cardiac death, when he was around 60.
Brad: We can do much better now. And the thing, the opportunity that is near at hand is to profile people for the 27 or so leading chronic age related chronic diseases that are associated with those premature deaths. And it’s really these 27 conditions that are responsible for 90% of that premature mortality.
Brad: And we can build models relatively easily right now that allow a personalization of risk, provides a basis for accelerated screening and early detection of these conditions, and more aggressive prevention.
Brad: So there’s a lot of people out hand waving around longevity right now. For me this is the place to start. And when I see a third of people dying in that age group, a quarter of women, that’s a big fat target that I really want to go after at scale. And that’s one of the things that we’re doing at Sapiens Data Science.
Lee: Would you describe yourself as a predictive analytics company? I.e. you’re skipping the sexy label of AI?
Brad: Yes, definitely. So one of my core hypotheses, and I’m not gonna talk in too much detail ’cause this is areas that we’re following, intellectual property, is that the world has fundamentally changed in terms of data availability for prediction.
Brad: And we’re trying to take advantage of that change. And I think it’s a completely overlooked opportunity by the current healthcare system to change and remarkably improve the lives of [crosstalk 00:51:32]-
Lee: I kind of want to jump into many topics here, I’ll try to restrain myself. By the way, you mentioned Health IQ earlier. They’re on the quest list later on.
Brad: Good. Delighted to hear that. They’re on my list to talk about potential partnerships.
Lee: Small world. I see you recently put up a hello world website for Sapiens DS. So I take it you can’t reveal, many details at the moment. When can we start to expect to see consumer product?
Brad: We plan to be in validation studies with a number of ecosystem partners, at a scale of 1 to 5,000 people, to really look at some of our early products, which will be focused on mortality risk reduction.
Brad: And it will be looking at two dimensions. One is consumer adoption, and the other is measurable health impact in terms of mortality risk reduction.
Brad: So late this calendar year, early calendar year, I hope that we’re able to share results of those validation studies. And then we would look forward to scaling towards the middle of next calendar year.
Lee: I look forward to that for obvious reasons. I see that Jim Mellon is … I laugh because I’m super excited about what you’re doing. I see that Jim Mellon is an advisor. And he coauthored a book I read this year called Juvenescence Investing in the Age of Longevity.
Lee: And I liked the title, because I had concluded early this year that the biggest investment opportunity over the next decade is actually longevity and biological aging.
Lee: Would you like to elaborate any on why you’ve got him as an advisor? And also why I see so many connections with the Buck Institute?
Brad: Yeah absolutely. Jim Mellon is an advisor, but he also importantly is an investor. You know was part of our first round of capitalization of the company, which we were really delighted to have him because of the level of commitment he’s got to the space right now.
Brad: And he brings a lot of dynamic energy, and credibility to the space, from an investment perspective. And so I would urge people to take a look at his book, and to take a look at the field of longevity more closely as an extraordinary opportunity [crosstalk 00:54:14]-
Lee: Have you noticed that on the biological aging side, people have only been looking at say pharmacological interventions? But they haven’t been looking at intervening in terms of biological aging using what I would call digital therapeutics.
Lee: I.e. you can guide lifestyle such as to slow or reverse some biological aging. You don’t need a drug to achieve that. I don’t know if you’ve thought of that?
Brad: That’s an unintended consequence of the way that the marketplace is set up, and life sciences companies are structured to identify new therapeutics in the form of drugs. And so there’s tremendous progress being made there.
Brad: But I totally agree with you that that misses a huge opportunity that is closer to hand. And it’s fairly dramatic in Alzheimer’s there’s fairly good information that a variety of lifestyle factors are responsible for substantial proportions of Alzheimer’s.
Brad: But the almost, the exclusive focus for most of the industry is on finding a drug, when I think we could make a lot of progress with lifestyle [crosstalk 00:55:32]-
Lee: Yeah like feed people fats, like healthy fats and lower their processed carbohydrates.
Lee: And also in Alzheimer’s, it’s also one of these like diabetes, you can predict it up to a couple of decades away.
Brad: Absolutely. And that’s one of the things that we were working on at Human Longevity that I think has tremendous potential, and really demonstrates the potential of combining genomics with advanced imaging of the brain.
Lee: Okay so Sapiens DS, as I’ll call it if that’s okay with you, it’s about measuring health status. I remember for the event I did, at the end of 2006, I realized that health quantification is essential for this new industry being called Hyper Wellbeing.
Lee: And so I invited Quealth and dacadoo to come and speak. And they also assign a health score, on mobile. And dacadoo do insurance partnerships with it. But were you planning to do something superior and further reaching?
Brad: Well we want to do that at the speed of science. And that’s the commitment that we’re gonna make to our customers, is to close that 17 year gap between the emergence of evidence and broad scale clinical use.
Brad: And so it’s very different. We also have a big commitment to cutting edge use of genomics and other advanced biologic measurement and devices. So we’re using the score as a mechanism to get to a fundamentally new place.
Brad: I see some of the other companies in the space using a score to be the value proposition in a consumer facing business, which is fine. But we’re doing something [crosstalk 00:57:24]-
Lee: Do you think you will score different bodily systems, instead of just one score shall we say?
Brad: I think, really what we’re doing is starting with scores, and we’ll have a family of scores that are built around general health, in a variety of dimensions and then specific diseases.
Brad: Where I really want to get to is I want a machine utility that sits quietly in the background of people’s lives and prioritizes what steps that they can take to protect and improve their health at any period of time they want to devote bandwidth to doing that.
Brad: And you know we don’t have anything close to that right now. But I think the pieces are out there to put that together. We can’t jump there immediately, but that’s where we’re headed.
Brad: And I think I like the Amazon Prime price point of $119 a year right now as the kind of price point that you could have that health curation support, running in the background of your lives at very large scale. And have it getting consistently better over time.
Lee: Yeah because today people are Googling up health information, and looking at the likes of WebMD. And that’s kinda 1990s compared to what you’re talking about.
Where I really want to get to is I want a machine utility that sits quietly in the background of people’s lives and prioritizes what steps that they can take to protect and improve their health at any period of time they want to devote bandwidth to doing that.Brad Perkins
Brad: Yeah. And I mean all the pieces are out there to build the platform that I just described. It’s just a question of how do you get from here to there from a business perspective? And how do you build the talent to build it and the credibility to get it adopted?
Lee: Well it certainly seems that you … I can only agree that you’ve identified what I’ll term a green space of exceptional value. And I’m chuckling to myself here that the space seems so open. I’m still kinda surprised at the lack of people that I’m aware of entering into that space so clearly.
Brad: One of the things that we’re doing in our platform is it’s not just about a score. We actually have a way to communicate with people that includes a score and then what we call signals, which are the dis-aggregated components of those scores that they can actually work on to improve or maintain.
Brad: And then we have a services linkage. So our paradigm is built around scores, signals, and services. And Health Nucleus, and Arivale are good examples of the kinds of services that we would like to link people to on our platform.
Brad: And same thing with the Livongo‘s and the OMADA‘s, and probably moving forward the Virta‘s of the world. Pointing people to services like that when they could benefit from those particular interventions. Again it has a very long runway.
Brad: And you know most of the most valuable companies in the world right now have that sort of platform model. We want to be really good at the data science element of this and running the platform at the speed of science. We’re not gonna actually deliver the kinds of interventions that Arivale or the Health Nucleus or Livongo are actually delivering. But we want to facilitate the connection of those services.
Lee: Okay. Thanks for that. At my event you also had said, with HLI you were pushing limitations of cloud computing. You had spoke about HLI using 24 petabytes of data on Amazon, and running queries I don’t think they were capable of handling in the time sensitive manner you needed them.
Lee: Is Sapiens DS also a company that’s pushing limitations of biology and computing?
Brad: Not right now. We’re gonna stay, you know Human Longevity was really operating at the edge of what was possible in a discovery mode. We’re going to, at Sapiens, and I was very stimulated by that, I think that’s critically important. At Sapiens Data Science we’re built around turning validated science into action, on behalf of people.
Brad: So we’re gonna be a little bit further down the stream, and using technologies that are validated, and have reached the level of affordability that allow them to be scaled. So a little further down the stream than where Human Longevity was, or is.
Lee: Does systems biology fit in anywhere Brad?
Brad: Absolutely. So we just hired a new chief science officer, whose training is in bioinformatics and systems biology. This is, it’s gonna be the name of the game going forward, and trying to surf the edge of what’s validated.
Brad: And also we think we have an opportunity with Sapiens to do some crowd sourcing as well. To make it easier for people to build algorithms and access publicly available data sets, to validate those algorithms, and stay at the edge of what’s possible in systems biology.
Brad: Right now the supply chain for high quality clinical algorithms is in the medical literature sort of broken. And people throw those up from academic and other organizations, but they often don’t go anyplace.
Brad: So we think we can one, help to reconcile and improve that supply chain, by getting those out to consumers. And then two, we think we can, sense consumer needs for various algorithms and then crowd source those. And do it in an open environment so that the algorithms can be accessed and improved upon in realtime.
Lee: Brad I see we’re running over time here. So I would love to just jump in, just for fun at the end of our call. Do you have any personal health recommendation? Any routine? Anything you do in the morning? Any little bit of religious behavior you have for your health?
Lee: It can be data-driven, it can be superstitious, it can be a supplement, a food, exercise, a data point you measure, a device you wear? Just do you have one or two health recommendations that you yourself follow?
Brad: No, I would strongly recommend that people get a hold of one of the continuous glucose monitoring devices. I think it’s right now the most compelling example of what the future is gonna look like in tight feedback loops for critically important physiologic data.
Brad: I’m a big fan of the Apple Watch. And we built our first prototype over the top of HealthKit, Apple’s HealthKit. And so I’ve been doing a lot of experimenting with, as a prototype for the company with downloading my formal medical data and of course capturing my informal health data, exercise and such from Apple Watch and pushing that technology construct to the limit.
Brad: So I would urge others to experiment and fool around with that and let me know how it goes.
Lee: Yeah. And the continuous glucose monitors, I concur with you. On the continuous glucose monitor side, that’s been also my first biggest win. And what I discovered with that is that these ideas of, “Hey it’s your volume of carbohydrates which exactly match to your glucose is just not true.”
Lee: Yes it’s approximate, well it could be half. But the other half is very individualized. It seems, according to for example the company [Day Two 01:06:14], a lot of the individualization of glucose response is your gut bacteria, your actual microbiome.
Lee: So for example in my case, if I eat fish, it raises my glucose, and I’ve checked corresponding insulin levels, more than what candy does. So you know, but I’m disappointed at the market today only pushing Dexcom and Abbott and so on, only pushing CGMs for sick people.
Lee: But I’m sure that just as App Store came along for Android and iPhone, I’m absolutely certain that continuous glucose, plus a few other measurements are gonna be the foundation data points of what we’ll call new apps, and new algorithms that compete with each other.
Lee: Decentralized peer-to-peer style of competition to make people well, keep well and be more optimized.
Brad: Absolutely. Absolutely.
Lee: So Brad, I’m very conscientious I’ve overran our allotted time and I highly appreciate your time, it’s been very insightful. Greatly appreciate it as ever.
Lee: I’ll be watching where Sapiens DS is going, and hopefully others listening will. I hope you’ll have something public as planned, and then I hope you’ll be back on the show.
Brad: Thanks very much Lee. And congratulations on the podcast, and all of your efforts with Hyper Wellbeing, and I look forward to your future guests and, keep the force my friend.
Lee: Thank you very much Brad.