Beyond the Buzz: Making Healthcare Innovation Last

Speaker 1Our first presenter here.

Speaker 1Is uh Benjamin Schwartz.

Speaker 1Uh, Benjamin, uh, is a fellowship trained orthopedic surgeon with expertise in Digital Health and Healthcare innovation.

Speaker 1He serves as the chief Innovation officer for Healland, a preseed startup on the Medical advisory Board for Bo Bo Health. How do you pronounce it? Bu Buo Health, a CVC company, and, uh, as an advisor to Quadrant Eye and osteoApp.ai.

Speaker 1Doctor Schwartz has uh mentored over 60 health tech startups through Mass Challenge, MIT Hacking Medicine, and Matter Health. He was named a LinkedIn top voice for frontline Healthcare in 2020. And that's how I met him. I met him on LinkedIn, and um he has been amazing at uh delivering great content.

Speaker 1And very thoughtful advice for our startups. So, please welcome Benjamin Schwartz, Doctor Schwartz.

Speaker 2He's been a great mentor to me, he's been very open and generous with his time, and I greatly appreciate the advice that he's given me over the past couple of years.

Speaker 2We have the slides. So, the title of my talk is Beyond the Buzz, Making Healthcare innovation Last. So, briefly about me, I won't get too much into this, it really did a good job of introducing me. In my day job, I'm a Private practice orthopedic surgeon in the Boston area. I've done some work in Healthcare Technology and Innovation, and I also serve with several professional organizations within Orthopedics.

Speaker 2So, by way of brief introduction, we all know that interest in Healthcare innovation has never been greater, that's why we're here, that's why we attend a conference like this. If we look at this chart, this graph on the right-hand side from Rock Health, that looks at Digital Health funding from 2011 on, if you look all the way to the right-hand side, H12021 is the first half of 2021.

Speaker 2And if you compare that to 2020, which was already a record-breaking year, in the first half of 2021, we beat all of 2020. And if you look below that at the average deal size, you can see the average deal size has increased over time.

Speaker 2The Other aspect that we're seeing as well in healthcare, Technology and Innovation is a continued push by big tech and big retail into healthcare. So there's a lot of interest, there's a lot of money flowing into Healthcare Technology and Innovation.

Speaker 2But I would argue we also have some headwinds, particularly big tech and big retail have shown some struggles. It was recently reported in The Wall Street Journal that Apple's Casper project, which was their attempt to open their own Primary care clinics, ran into some snags and has stalled at this point. Google Health recently went through another reorganization and actually shrank their healthcare team.

Speaker 2There was a very public failure of Haven Health, which we all know was the joint venture of Amazon, Berkshire Hathaway, and JPMorgan that finally gave up the ghost a few months ago. Walmart was very aggressively pushing into health and had plans to roll out a lot of brick and mortar stores, and those plans seem to have been scaled back quite a bit. Walmart seems to be pivoting more towards Telehealth by buying a Telehealth company, and all of these companies have also seen high-level people leave their healthcare efforts.

Speaker 2Now, I put this slide together a few weeks ago before the Delta surge, but at some point, we anticipate emerging from the Pandemic, and what effect is that going to have on Healthcare innovation. The chart on the right-hand side is from Trillion Health, and it looks at Telehealth adoption. No surprise, February 2020, March 2020, there's a big spike in the use of Telehealth as the Pandemic.

Speaker 2Limited access to in-person care. If you look to the far right-hand side, in the spring of this year, as numbers were starting to come down before Delta spiked, we saw a gradual decline in Telehealth. If you look at April and May, that trend continued. Now, we all expect Telehealth usage to remain above Pandemic levels, pre-pandemic levels.

Speaker 2We know that Healthcare innovation was already taking off before the Pandemic. The Pandemic is felt to have accelerated health care Innovation.

Speaker 2The question is, is this sustainable? When we finally get to a post-pandemic world, are we going to see a return to the old way of doing things? We know healthcare is traditionally very resistant to change, and will we see a sliding back and a slowing down of Innovation? Will the appetite for Innovation persist over time, once we get to a post-pandemic world? We're also seeing, as the Digital Health and Healthcare Technology space matures, consolidation, we're seeing Healthcare Technology companies joined together.

Speaker 2Seeing bigger deals. The question is, is this going to squeeze out the little guy? In general, consolidation may decrease Innovation or may stifle Innovation. There'll be a lot of attention, I think we anticipate more mergers and acquisitions of Healthcare Technology companies going forward. SACs came and went and may be coming back again, and there'll be a lot of attention paid to IPOs as Healthcare Technology companies go public and allow us to peer behind the curtain a little bit on how they're doing financially and how much success they're having in the public markets.

Speaker 2So this leaves us with questions. Are we headed for a Healthcare innovation bubble, as we see

Speaker 2The acceleration as we see increasing interest and money flowing is, is this sustainable or will we hit a bubble? Will big tech and big retail eventually burn out of healthcare? We've seen some of the most big, uh, successful and biggest, most resource rich companies in the history of mankind struggle to make inroads in healthcare. Will they eventually decide it's not worth the effort, it's not in their wheelhouse, something more exciting, more attractive comes along, and they turn their attention back away from healthcare.

Speaker 2I think a very important question we have to address is, will anything really change with all of this interest in Healthcare Technology, at some point, we have to prove that we're making a difference, that we're solving the problems in healthcare that we hope Technology will solve. And really the focus of my talk is going to be how do we make sure that Healthcare innovation lasts over time.

Speaker 2So, I'd like to propose 3 principles, 3 keys to sustainable Healthcare innovation, reality, trust, and validation.

Speaker 2So when it comes to reality, this is the Gardner hype cycle, we're all familiar with that. We know it's not really a cycle, and we know it doesn't apply perfectly in all situations, but it's a good framework to work from when we talk about hype, there's a Technology trigger, there's a lot of interest and excitement, so we reached this peak of inflated expectations, the hype doesn't match the reality, so we fall into the trough of disillusionment, we regather ourselves and end up in the slope of enlightenment, and eventually on the plateau of Productivity.

Speaker 2So in traditional Technology, the hype cycle uh really surrounds product launches, whether it's a new cell phone, a new operating system, a new device, there's a lot of hype that goes into that, and that's because there's fierce competition in traditional Technology amongst these companies, and there's a real need to generate consumer excitement about new Technology and traditional tech. And the typical Approach is to release some sort of minimally viable product and iterate, make it better and better over time.

Speaker 2This is a little bit different than what we're seeing in Healthcare Technology in terms of the hype cycle. Despite all this money flowing into Healthcare Technology and Innovation, it's still a challenge to raise funding, it's still a challenge to get your company from pre-seed to Series A to Series B to Series C and

Speaker 2So you have to be able to attract investors. Once you have investors, once you have some success, eventually you have to justify your evaluation, whether you're looking to be acquired, whether you're looking to exit through an IPO, there's a lot of pressure on justifying these big valuations that we're seeing in Healthcare Technology, and I think that leads to the potential to overpromise what you're doing, your product, your service.

Speaker 2And risks underdelivering on that promise. And so, traditional Technology hype cycle is not the same as the hype cycle in Healthcare Technology.

Speaker 2There are plenty of examples of the reality not matching the hype in Healthcare Technology. Theranos is one of the most egregious and well-known examples of the reality not matching the hype. There was a lot of interest around IBM's Watson and the ability to change, particularly Cancer care.

Speaker 2that has been scaled back significantly. Google, through big data, tried to predict flu seasons through its Flu Trends project that didn't really pan out. And we've seen various applications of Artificial intelligence and machine learning. There's been a lot of excitement around it, but that excitement hasn't necessarily matched the reality.

Speaker 2There are dangers of hype. We're all a little bit, we're all a little bit fake it till you make it. I think that's OK, particularly when you're dealing with Innovation, Entrepreneurship, but eventually you have to make it, and if you're just faking it and you never make it, there's a danger there to the hype.

Speaker 2We know healthcare is already resource strapped. We know there's already a lot of waste in healthcare. We really can't afford Healthcare Technology and Innovation to contribute to that misdirected resources or the wasted funds. And over time, if the reality doesn't match the hype, you risk eroding trust of doctors and patients, which we'll see uh in a, in a little bit. It's already questionable how much doctors and patients are believing in the hype of Healthcare Technology.

Speaker 2There's the very real danger of patient harm when it comes to Healthcare Technology and hype. We saw that with Theranos and potentially fabricating lab results. It's different. It's one thing to tell somebody they're holding their cell phone wrong and that's why they're getting bad reception, as opposed to having a medical device where you're delivering care or treatment or diagnosis. You really have to be sure that what you're doing is, is accurate and validated, and we'll talk about that too.

Speaker 2We also risk worsening the problem. I think we're all excited. The reason we attend meetings like this is the potential for Technology to address the problems in healthcare. If the hype and the reality don't match, then we risk worsening the problem and not really fixing the healthcare problems that we all hope Technology will be able to achieve. And ultimately, this can impede progress and limit our ability to apply Technology in healthcare.

Speaker 2So I think we need a reality check. We have to understand that Healthcare innovation takes time. It's something that doesn't occur at the same speed as you might see in traditional Technology. I think it's great to dream big in healthcare and Healthcare Technology, but to understand that small wins are still wins, Healthcare is a huge problem. You can't tackle it all at once, and even small gains, small wins are important within healthcare.

Speaker 2I think, at the end of the day, it boils down to honesty and transparency, and that's what's going to lead to sustainability over time. And I don't wanna harp on the Theranos example, but the hype cannot become the product. At some point, there has to be some reality, at some point, the fake it has to become a make it. You don't want to be that red arrow on the right-hand side in the chart there.

Speaker 2Moving on to the second pillar of sustainable Healthcare innovation, I think that's trust. So, in traditional Technology, there's this understanding, this implicit or explicit understanding that there's a trade-off in using Technology for free. That when we click on that agree on the end user license agreement to use Google Search, Gmail, Facebook,

Speaker 2Facebook, Twitter, LinkedIn, Apple Pay, whatever it is, we're giving up a little bit of our privacy for the convenience to use this Technology, and at some level, we're all comfortable with that. At some level, we're all willing to trade a little bit of our privacy to use these products. But the famous quote is, when something online is free, you're not the customer, you're the product.

Speaker 2When it comes to Healthcare innovation, data is still the goal, but I think we would argue that the data is a little bit different. It's different for Siri to know that I like to order pizza from this restaurant every Tuesday at 5:30, as opposed to Alexa knowing that I order this particular medication every month. So we view our protected health information as patients and physicians.

Speaker 2A little bit different than the information we give up, sort of in our consumer day to day lives. And I think this creates this conundrum. How do we fully realize the potential of Healthcare Technology without giving up too much of our privacy? And if we can't find that sweet spot, then it's going to continue to limit how we can apply Technology to healthcare.

Speaker 2So who do patients trust? This is another graph from Rock Health. They asked the question of who patients trusted with their healthcare information. So if you look at the top of the graph here, it's physicians are number one, not surprising that physicians would be the most trusted. However, and it's beyond the scope of this talk, it's a little bit concerning that if you look at the 2018 and 2019 data, physicians have not maintained that level of trust. We've actually lost some trust with patients.

Speaker 2And that's concerning and something to keep an eye on. If we go further down, Healthcare Technology companies, only about 20-something% of respondents were willing to share their information with Healthcare Technology companies. And when you go even further down at the bottom of the graph, only about 10 to 11% of patients were willing to share with traditional Technology companies. So there are some trust issues when it comes to patient Healthcare Technology companies and traditional Technology companies.

Speaker 2This is a slightly different survey from Accenture in 2020. The question was, trusting these entities to secure the information, and hospitals, pharmacies, physicians, 80-something% did pretty well, but tech companies, only about 45% of respondents trusted tech companies to keep their information secure.

Speaker 2Switching over to doctors who the doctors trust. So this was a survey from Intel where it asked doctors different questions about how much they trust AI or their concerns about AI. 54% of doctors felt that AI was going to be responsible for a fatal error. Doctors, about half of doctors felt that AI was going to be overhyped, and not match the hype. So some serious concerns on the part of physicians, about half of physicians thought that AI was going to lead to a serious complication or problem or lack of trust within healthcare.

Speaker 2A similar survey from Medscape in 2019 asked, as a physician, I am uncomfortable or anxious about using AI powered software, and about 50% of physicians responded, yes, they were concerned about using AI and AI powered software. That was actually up from 41% in 2018. Interestingly, if you look at Europe and Latin America, the physicians there aren't quite as skeptical of AI powered software as we are here in the US.

Speaker 2But all of this creates questions around trust. Who owns the data? Is it the patient? Is it the physician? Is it the hospital, the health system? Is it a third party? Is it the EMR company? How do we regulate and oversee how the, how data is being collected and used? We know there's a fire hose of petabytes, zettabytes of healthcare data.

Speaker 2How do we regulate and oversee the way that's being collected and used? Who can be trusted with the data? As we just saw, there's some real concerns on the part of doctors and patients in terms of trust with healthcare data. Is de-identifying data enough? There's concerns about re-identifying data. Are the tools that we're using to de-identify data enough? Are people who don't come from a medical background but have access to this data, getting the proper training and how to treat it respectfully.

Speaker 2So I think building trust is a balancing act. We have to have better safeguards of protected health information. We have to close the information gap of perception versus reality, because as we just saw, the perception among patients and physicians is that this data is not being used or they don't trust companies to collect the data to give their information to

Speaker 2Certain companies, tech companies, Healthcare Technology companies, and if we can't bridge that gap, then we will never realize the potential of Technology and data to improve healthcare. And again, it comes down to honesty and transparency. How's the data being collected? When's it being collected? Who's it being collected by? What is it being used for? We have to answer these questions to build trust.

Speaker 2The final pillar, I would argue in sustainable Healthcare innovation is validation. So, in traditional healthcare, we're not perfect with this, but at least we try and we have mechanisms in place to validate things that we do. We try to practice evidence-based Medicine. So, evidence-based Medicine is this confluence of clinical judgment of relevant scientific evidence, and patients' values and performance or values and preferences. We try to base what we do on peer-reviewed literature. We aim

Speaker 2For reproducibility, the ability to do things over and over again and get the same results, and generalized ability, the ability to take these tools and apply them broadly across the healthcare system to different patient populations in different settings. We have oversight mechanisms, we have registries where we collect and analyze data. We have professional societies, we have Conferences, we go to national meetings. All these are mechanisms by which we try to validate what we're doing in traditional Medicine.

Speaker 2The question is, in Healthcare Technology, how are we validating health tech tools? And often, we don't hear about the outcomes of Healthcare Technology, what you hear about is funding rounds, Series A, Series B, Series C, big valuations, and we have to be careful about attaching valuations and funding and IPOs to, uh, that type of success to validation of

Speaker 2These tools and products. And what is the Primary goal? We have to define what our Primary goal is in Healthcare Technology and Innovation. Is it to save an employer money? Is it to improve access to care? Is it to improve workflow for physicians? Is it to improve interaction with the healthcare system for patients?

Speaker 2And I still think we haven't fully defined what the Primary goal is with a lot of Healthcare Technology. And then how do we measure that we're achieving these goals? Are there implements in place to allow us to measure the success of Healthcare Technology and its impact on healthcare?

Speaker 2So there's plenty of examples where I think we need to start thinking about validation of Healthcare Technology and Digital Health. Digital Health startups are becoming more aware of this. I think they're realizing that Peer review will distinguish them from their competitors, but there's still a long way to go. A lot of Digital Health apps are not regulated by the FDA. There's not great.

Speaker 2Mechanisms in place to make sure that the information they're providing, the care that they may be delivering is really achieving the goals that they say that it is. There's a lot of work to be done there. You know, Apple is increasingly positioning its watch as a medical device, to their credit, I think they're trying to use more traditional channels, working with academic medical centers.

Speaker 2Submitting what they're doing for Peer review, but it may require taking some steps that a company like Apple may not traditionally be comfortable with, and that is allowing independent evaluators to peer behind the curtain, what do your algorithms look like. Uh, and that's not something that traditional Technology companies are necessarily used to doing. We've seen a lot come out recently about Epic and their EMR AI tools, particularly around diagnosing sepsis in an ICU setting where the algorithm didn't perform very well. There's been stories

Speaker 2About Epic paying hospitals and health systems to use their algorithms. And so we have to be careful that these things are being validated and used the proper way. Google has made a pretty strong push into Artificial intelligence, whether it's predicting kidney injury, whether it's diagnosing skin Cancer, breast Cancer, diabetic retinopathy, and the Technology can be very good, but there's another part of validation as well where you have to validate, again, that these things are generalizable and reproducible. So this headline, uh, was an MIT Technology.

Speaker 2Review a story about Google's diabetic retinopathy algorithm worked very well in the lab, but when they tried to implement it in real life in a clinic in Thailand, there were barriers to be, to, to being able to use that tool, and that's a concern. If something works great in a lab, if it works great in canned data sets or clean data sets, that's one thing, but it has to translate to the real-world setting, and that's another way we have to validate these tools. It's not just that the Technology works, but it can be applied in the real-world setting and out in the wild.

Speaker 2There are challenges to validation. Validation takes time, it can be very expensive. We have a saying in Medicine, and particularly in Surgery, that nothing ruins a good outcome like follow-up. The same thing could be said of Healthcare Technology tools that nothing could ruin a great Healthcare Technology idea like validation if you have to prove that you're achieving the goals that you're saying you're going to achieve. And I think it requires a framework that maybe doesn't exist quite yet to balance the speed at which technological

Speaker 2Innovation takes place with the safety and efficacy that we expect in healthcare. This is a, a chart, a proposed framework of the Nature of Digital Medical Journal of how a validation process might work for a Digital Health tool. So the, the blue bar at the top is developing the tool and then sort of the peach-colored box in the middle is an independent evaluation that before these tools go to market, they're submitted to an independent evaluator to test them, to validate them, and then that process continues post-market.

Speaker 2I don't know how much this is actually happening right now, but it's a good framework to look at, and I think it's something we need to seriously consider as Healthcare innovation Technology takes off, and we're validating the tools that we're using.

Speaker 2So in conclusion, I would argue that sustainable impactful Healthcare innovation depends on embracing reality over hype.

Speaker 2On engendering trust, particularly with patients and physicians.

Speaker 2And seeking validation of the tools that we're using, because at the end of the day, Healthcare innovation is about addressing real-world healthcare problems, and that's what we all are excited about when it comes to Healthcare Technology and Innovation. I think the best way to do that is to validate what we're doing and prove that what we're doing is making a difference in healthcare.

Speaker 2And I thank you, this is my contact information. I'm happy to take questions.

Speaker 1Thank you, Benjamin. Uh, those of you who have questions can walk up to this microphone in the middle.

Speaker 1And uh ask uh Doctor Benjamin questions.

Speaker 1Otherwise, I'll look up our

Speaker 3solution is to the problem of who really owns the data and how do you put more data in the hands of the individual.

Speaker 2We're talking about individual patients. That's a great question. Uh, I think it's a key question that we have to look towards. We know that once the data becomes de-identified, maybe the patient doesn't own it anymore. I think to me, the key is really transparency. I think physicians and patients are going to be more willing maybe than we think to allow their data to be used as long as we're transparent about how it's being used, uh, who's using it, the way it's being collected. So I think

Speaker 2The first step is just being transparent. Hey, your data is being collected. It's being used for this purpose. Maybe it has to be opt in to some degree that the patient feels like they have control over it. The question about monetization, I think is a, a, a, a very important question for ethicists, you know, to determine, is it right for a third party to take the identified data that's patient data and be able to monetize that is a very tricky question.

Speaker 3I guess my question more is in the sense of how do we allow it to be in the hands of the individual who actually, the data is referred, is referenced to, right? And how do we use that data in order to then start really informing people and educating people and helping people to lead a more preventative lifestyle instead of a

Speaker 3You know, reactive and, and restoration type of life.

Speaker 2Yeah, yeah, I think there's efforts underway. We all know that are in healthcare that recently the rule was passed that that patients have access to their healthcare records, tests, uh, results, and, and things like that. And I think.

Speaker 2Uh, patients are becoming more informed healthcare consumers. There are tools out there, whether it's on your cell phone, be able to take your records around with you. Um, so I, I think it's coming. I think it's early on, uh, but I think there are tools and processes in place to allow patients to have better ownership of, of their information.

Speaker 4Ben, Gary Goldman. Um, so how do you, you know, here we are at the precipice of a Digital Health revolution. I mean, things are changing at the speed of light here, especially in the RPM arena. So, you know, with a model like this, how do you, how do you adapt to the rapidly changing ability to collect data, either non-invasively or with new Technology? How are you kind of shifting with that?

Speaker 2Yeah, it's a great question. I think it's a question we have to answer because we know things in traditional healthcare don't move that quickly and we don't necessarily wanna pump the brakes or miss out on opportunities to adopt Healthcare Technology, which is gonna move more quickly. I think it's that sort of constant validation, that constant oversight, that constant checking in of are we achieving our goals? Are the, the outcomes that we're getting, are we validating the outcomes that we're getting? Is our tool achieving what we

Speaker 2Hope it achieves. If it isn't, you know, can we change that on the fly? Are we doing things that are risky to patients? And if so, do we need to back off? So I think it's, it's sort of a constant validation process, a constant evaluation process, which isn't easy, it's labor intensive, but I think you always have to be keeping an eye on, on what your outcomes are, what your tools are producing, is it accurate? Is it working? Is it safe? Uh, so I think it's a constant, uh, validation process.

Speaker 5Uh, great presentation. I got a question to ask in regards to the legal uh implications of, uh, Artificial intelligence tools or anything like that in regards to the level of disclosure of their performance matrix and also legal implications, who's at fault when you use an AI tool.

Speaker 5But the AI too had bad data and it wasn't validated. Who's at, who's at fault here? That's really, and how, you know, right now, a lot of tools are not really registered as a medical device, you know. So we got a lot of unknown Technology being flooding in the market, but pretty soon, I'm sure there's gonna be a crackdown on this.

Speaker 2Yeah, I think it's a great question. I think it's a question we don't have an answer for right now. I think it's part of the reason that we see skepticism on the part of physicians and using these tools, cause I think a lot of physicians feel like, ultimately, at the end of the day, I'm on the hook for making that final decision whether or not AI told me to do it or whether or not I used the tool.

Speaker 2To guide that decision making at the end of the day, if I'm the one making that decision, keeping that, writing an order, deciding, you know, the treatment, that I'm on the hook for that. I think that's an issue that has to be addressed in order to get physicians to feel more comfortable. We saw that in the Intel survey that 50% of the physicians thought that AI would lead to the death of a patient. That's a pretty significant thing and barrier to overcome.

Speaker 2Uh, and I think it's about, again, trust validation, but we'll have to answer the question, uh, who's responsible for that? Because if the physician's responsible, I think there's gonna be a lot of resistance to using algorithms, particularly if they go against what we know from our experience. The algorithm might be right, but if in my experience, I don't trust what it's saying, I'm not, and I feel like I'm gonna be responsible for that, I'm gonna fall back on my experience, and then we may miss out on making healthcare better because of that.

Speaker 1Great. Thank you very much.