Traditional medical care is being revolutionized before our eyes. COVID-19 has driven an unprecedented pace of digitization, and remote care and monitoring have proven themselves as critical pillars of health systems the world over.
In this episode, Israel-based partner and life-sciences specialist Gila Tolub explores the biggest trends in digital health today with David Maman, founder, CEO and chief technology officer (CTO) of Israeli health and wellness monitoring solutions company Binah.ai. Together they discuss how the industry might look like in five years’ time, and the unique challenges healthtech companies face, from regulation to AI bias and beyond.
Before introducing our guests, let’s start with a quick warm-up question.
Peleg Dekalo: If you had to give the Nobel Prize to someone within digital health, David, who would that be, and why?
And the Nobel Prize in digital health goes to…
David Maman: For me, my Nobel Prize winner is the person that started with PPG—photoplethysmography—1937, [wrote the] first paper. Back then, most people still didn’t have a working light at home, and he already wrote about PPG. So that’s fascinating.
Peleg Dekalo: Tell us about this article. What is PPG?
David Maman: So PPG, photoplethysmography, means that—I'll start with a very simple explanation—if we take a look [at our skin surface] and monitor the tiny changes that are happening to our skin, [on] any part of the body, by the way, that provides us an indication that we can actually generate the signal that helps us transform it to heart rate, to other types of vital signs, [and] to endless amounts of information that actually exist on our skin surface. Before that, we needed to put some very serious sensors and monitor on our chest and get a lot of information, sometimes from our head, in order to get the most basic vital signs. So PPG, [which] then evolved to RPPG—remote photoplethysmography—actually changed this revolution so that medical devices can be used at home.
Peleg Dekalo: Gila, the bar is high.
Gila Tolub: Right? We're at the cusp of digital health. I think for me there are two things that stand out. One, there is a person that stands out to me, and it’s Joy Buolamwini, who's doing work at MIT [Massachusetts Institute of Technology]. She's a PhD candidate at MIT, and she’s doing so much work on algorithmic justice. She also has a nonprofit organization. Her work has actually prompted IBM, Amazon, Microsoft to think through, to pause what they were doing around facial recognition, because she showed them that the algorithms were actually biased against people of color. And while she’s not working on digital health, I think a lot of the work that she’s doing on raising awareness for the biases that are in AI are going to help with digital health and healthcare, in general. We still have algorithms that are discriminating against certain populations because they’re not based on data or real-world data. The other thing that excites me a lot with digital health innovation is the fact that we have a lot of reverse innovation happening. What we’re seeing with digital health is that the innovation is coming from the areas that have the biggest need. They’re trying to solve an issue, right? Issues of access, issues of unmet needs that don’t interest everybody. And so we have innovation coming out of the Southern Hemisphere for the first time, right? Out of Africa, out of India. Then we realize, “Hey, we have so many areas, even within the US, that have the same issues because they’re remote, because they’re rural, and we don’t have as much access,” and I think we’re going to see much more of that coming up.
David Maman: I think just something to support what Gila is just saying, we live in a Western world. And we’re used to that, but you just take your flight out of Israel for 30 minutes or 60 minutes, [and] you get to countries [where] you have one physician for every 10,000 people, one physician for every 20,000 people. And I want to talk about something really serious. Papua New Guinea: 14.9 million people population. The entire country [has] 463 physicians. That’s one for every 35,000 people. That’s unbelievable. Okay. So Gila is right. Those countries, by the way, understand they didn't have those ten years of research for a new product. And the capabilities of the cloud infrastructure, the AI capabilities, allows them to actually to catch up very fast. And that’s what’s happening. In Latin America, in India, in a lot of countries that are way behind, they will actually start producing very interesting stuff.
Peleg Dekalo: And now introductions. A little bit late. But let’s do it. David, you go first. You’re the guest.
David Maman: My name is David Maman, the husband of Mirit, the father of Alma. I’m living in the center of Israel. I’m currently the founder CEO and CTO of Binah.ai, a very interesting company. A few words about myself. I guess I started pretty early. I graduated with a master’s degree in computer science and applied mathematics at the age of 18. I’ve been always into technology, like hardcore, low-level technology, and not on a specific subject because this is actually the thirteenth start-up that I've co-founded and eight of them were in cybersecurity, which is definitely my comfort zone, but I love everything, from water saving to grid processing to healthcare, of course, to everything. I'm just a real technology junkie.
Peleg Dekalo: Gila, tell us about yourself, please, and about your role in McKinsey today.
Gila Tolub: Sure. So, I’m Gila Tolub. I’m a partner in the Tel Aviv office. I made Aliyah about six years ago. I'm originally from France, as some of you may hear from my accent.
David Maman: No way.
Gila Tolub: But I’ve spent 12 years in the US. I really love to do work that has an impact on others. So, after my MBA at University of Chicago, I Joined McKinsey in this spirit. And when you join McKinsey, they ask you all the time, “What is your passion? What is your passion? And what is your purpose? What do you want to do?” And I said I want to help companies that are doing more than just bringing new products. I want to help companies not just making more money. I want to help them because there is something more than that. And very quickly, I went into healthcare. And I served hospitals, I served pharmacy chains, I served pharma companies, medical device companies, and started to work in vaccines way before it was sexy, when people were asking me, “Why are you doing something that is so niche?” Nobody has ever said something like that since 2020.
Peleg Dekalo: Yes, most definitely.
Gila Tolub: But then Corona [COVID-19] happened. and instead of traveling, going to all my clients all around the world, I was based in Israel. And I said, “This is maybe a gift in disguise because there is so much innovation happening here in Israel.” And so I’m leading the healthcare vertical of our high-tech hub in Israel, where we're trying to see how we can help start-ups accelerate the pace of innovation. And so, whether it is getting it right or whether it is getting it done faster, what can we do so that this innovation gets to market? And I think our purpose is twofold: bringing the best of the world to Israel and bringing the best of Israel innovation to the rest of the world. That’s how I view my role.
Peleg Dekalo: So now, you’ve been exposed to a lot of the innovation and trends that are happening in digital health. Walk us through what are the biggest trends right now that are happening in this realm?
A major healthcare trend is care being pushed out of hospitals
Gila Tolub: Yeah. So I think one big thing, especially after corona [COVID-19], is we realized a lot of the care is going to be pushed out of the hospitals. During corona [COVID-19], nobody wanted to go to the hospital. And you know that it’s the worst place to be. But you know what? It was like this before corona [COVID-19]. Actually, no one should want to be in the hospital. Hospital-acquired infection is a real thing. And you want to be at home as much as possible. And I think because of what happened with COVID-19, we’re seeing more and more of the innovation that was existing on remote monitoring, being talked about, being adopted much faster than before. You have companies, who were around from the early 2000s, and it was so slow—the adoption was so slow. And the doctors were like, “Nobody’s ever going to be able to do remote care. People still want to come to me. People want to talk to me.” And then suddenly, they had that option during corona [COVID-19], and it was great. And they liked it better, and the patients liked it as well. And so now, we’re seeing this huge adoption on remote monitoring, and I don’t think we’re going back. I mean, I think we’re going to get a hybrid kind of care, right, where we’re going to have some of those things remote. Some of the in-person [care] is never going to go away, right? But I think this is going to be a huge shift.
Peleg Dekalo: How do you prepare for that?
Gila Tolub: I think people are now only starting to get together to think about healthcare of the future, which is the theme of this podcast. But it’s really thinking about how do you create a connected environment. So I think technology is a huge enabler. There’s a lot of questions around data privacy. How do you connect different hospitals, different people at home? How do you monitor them? So, a lot of technology enablement. But I don’t think anyone is fully thinking about it yet, to be honest.
Peleg Dekalo: Indeed. Interesting things to come. David, how do you see healthcare in five years from now, and also digital health? What should we expect?
COVID-19 sped up transformation in the healthcare industry
David Maman: So definitely, I think that the healthcare industry is one of the key industries that were left behind in the digital transformation. And no doubt, I completely agree with Gila that COVID-19 has pushed healthcare ten years forward, easily. And I talked with companies that started in the late ‘90s and didn’t move anywhere, not in sales, not in product definition. And suddenly with COVID-19, now they’re new unicorns. And the offering proposition that COVID-19 started requiring, it’s not just about remote care. It's actually about having self-control and a self-availability for healthcare services anywhere. And in the past, we really used to completely separate healthcare and wellness, and everything about wellness was nice if you’re training, if you’re doing some sports, if you’re going to the gym. But healthcare, no, no, no, you have to see the physician for that. Everything now is combining. The wellbeing industry is actually double the size of the healthcare industry today, which is unbelievable. So I think due to the available technology—not only that Binah is doing—a lot of companies are working on making things available everywhere, I think that healthcare is going to be dramatically personalized. And what do I mean by that? Today, when we take a pill, the pill was designed to support the commodity, which means the majority of the population, but there are no personalized healthcare solutions unless you have a lot of money. So I know this physician that started a practice that if you have half a million dollars—I'm not kidding, half a million dollars—and you have cancer, 25 scientists will work with you for six months, and will build you a treatment with over 99 percent success criteria. That's unbelievable, which means cancer is treatable. So personalized health is going to go very, very deep up to cancer treatment and rare diseases treatment, but on the other hand, I think personalized help is going to be at the wrist of your hand. And many of us have those type of smartwatches, but let’s talk practicality. There are countries that don’t have a sewage system, but everyone has a smartphone. And those are the key players in Africa, in Latin America, in rural area in China. And when you have so much information that can be actually generated on a daily basis, actually on an hourly basis. When you have so much data, personalized health will go forward. And not only from a smartphone, but from wearable devices too.
Peleg Dekalo: And with those tech advancements, personalized health will come in a cheaper cost, and will be more accessible to more and more people.
David Maman: Completely.
Gila Tolub: It’s about changing the mindset so that we can move into this precision medicine arena. I think what we’re seeing is a lot of patient empowerment, right?
David Maman: A lot.
Gila Tolub: Just the data that’s out there, the fact that we laugh that people go and Dr. Google before they see their doctor—but the point is, they are just much more involved, right? And they want to learn. It’s actually great that people want to know more about their health. And what we’re seeing in the data is that 80 percent of people actually engage in social media about their health. So whether it is in platforms like PatientsLikeMe, or in Europe or in France, it's Doctissimo—all of those forums where you ask questions about health or you talk to people who have something similar—80 percent of people do that already today.
David Maman: One more thing about this specific—because we’re talking about what's going to be the—where we’re going to be in five years or less. So, what we’ve been able to see in Binah now, we have a lot of major insurance companies that are customers of ours. And when we talk with them, we see the amount of money that they’re spending now on wellness, and they’re actually enabling their customers to have wellness monitoring capabilities. That’s why they bought Binah as well. But why it’s so interesting—and I truly asked an insurance company, “Why are you spending so much money and you’re willing to go with this path to enable your customers?” And they said, “On every dollar that we will spend on wellness, we will save four and a half dollars of claims within the next five years.” And what makes you think that the organization that’s going to make the world a healthier place are actually insurance companies?
Peleg Dekalo: That relates to the integration of wellness and healthcare that you spoke about.
David Maman: That’s one of the integrations that we see that is happening in real-time now.
Peleg Dekalo: Gila, you said that in your capacity, you help incumbent players, healthcare giants, and also insurance players that are in digital health and also start-ups that are in the field of digital health. What is the difference in how we serve them? What are the main challenges that each encounter?
Healthcare giants should be more agile and start-ups should be more focused
Gila Tolub: You really need to think about the culture of the organization and think about how to make it more prone to innovation. And when we look at what is a more innovative company, usually it’s about patient-centricity. So really being clear that it’s not about selling your own product, but it’s about solving problems for the patients. It’s about breaking silos because those big organizations very often have different business units doing different things. But when you’re trying to solve a problem for a patient, you need to all work together to get there. The other thing is, obviously then being more agile. So this idea that start-ups know you have to fail fast. But large organizations don't have the mindset of “Let’s try things quickly, and then let’s fail fast.” And so it’s also something that we work a lot on with large organizations, [which is] especially hard I think in healthcare because you have a lot of people that are PhDs and who are used to testing things for a very long time. The other reason is, I mean, very simple to understand. You have people’s lives in your hand, right? So you do want to be a little bit more cautious than just let’s test and see where it goes. Now on the start-up side, it’s very interesting. It’s the opposite, right. I mean they know they’re moving really fast. They’re trying new things. But sometimes they lack focus. They want to do a lot of different things. They want to go after many different business problems. They found an innovative way of doing something, and they're like, ’Okay, but this could help payers this way. This could help the provider that way. I could even go help pharma companies with their clinical trials.” And so many different things you could do. But then you’re actually not going deep enough into one area.
David Maman: As a young entrepreneur, 15 or 20 years ago, I didn’t know that strategy is, first of all, to know what not to do, because, as an entrepreneur, you always try to keep the doors open. You want to do that and that. “Oh, yes, and I can supplement this specific vector.” No. First of all, know what you don’t do. I agree with you Gila.
Gila Tolub: So that’s what we help start-ups with. We help them think through their options, but also what should they be focusing on to make sure they increase their probability of success in those areas, and it’s not always easy. Sometimes it’s about bringing a little bit of the human side of things, right. It’s not only about the numbers.
Peleg Dekalo: The simplification of it, it sounds [like], for the giants, we are kind of digital experts, and for the start-ups, we’re kind of business and management experts.
Gila Tolub: Yeah, that’s a good way of putting it.
And I think one of the big roles that we play is actually making the link between those. So actually introducing—a lot of time, we have large clients who say, “Hey, I'm looking for the latest innovation in skincare.” And you were able to tell them, “Well, look, there’s those five start-ups that are doing very cool things. You should look into what they’re doing.” And then we can even make introductions if that helps them. Most of the healthcare inefficiencies occur at those intersection points. I think if all the players knew how to work together better, we would have way less inefficiencies, right?
Peleg Dekalo: David, as you said, you have a lot of experience in a lot of different fields within the tech sector as an entrepreneur. Give us a taste on the unique challenges you face with a digital health company as opposed to a cybersecurity company or a web company.
Regulation and compliance: a unique challenge for healthtech entrepreneurs
David Maman: So I think that, first of all, let’s talk about the positive things. If it were ten or 15 years ago, there were really, really niche investment companies that used to invest in healthcare startups. Back then, no one used the definition digital health. It was hardcore health because you needed a lot of time in order to generate value. no one would have ever touched any product without the medical approval, without all the full clinical work. And that’s why those companies, until they started selling a product, it would be ten to 12 years. And usually, investment companies, —when we're talking about tech and technology in general—want to see a much better return, like a few years to have a serious, mature product in the market. And a lot of the new technologies have actually helped to close the gap. There is a lot of data, health data. So, the great thing is that we have unlimited power today, processing power. The cloud supports everything today. There are no limitations to data management, data processing, data analysis and also architectures that can generate value from the data.
But I think that a lot of companies chose during the past, probably ten or 15 years when they started working on healthcare, they said, “No, no, we are wellness. We do not need the medical approval. We do not need your regulation approval.” But this is changing now. Even wellness companies will have to support the FDA requirements, whether it’s about accuracy, whether it’s about compliance, about regulation, about processes, about eQMS [electronic quality management system] systems. So I think that those worlds of wellness and healthcare, we’re just going to have another grade in FDA. That's how I see it happening. So I think that regulation is a huge part of the challenges.
Peleg Dekalo: Yeah, regulation and compliance, definitely unique to healthtech.
David Maman: Second challenge is the data that you can actually use. So let’s separate the type of companies in digital health or in healthcare to three different types. There’s a company that used current available data. And they do sort of a data mining, sort of an AI-based insight. Eventually, they’re analyzing the data, and they can give you very interesting insight. The second [type] of companies are actually utilizing current medical devices data. Basically, they’re taking radiology images, and they are getting the insight. Whether you have a hemorrhage in your brain [or]whether you have a broken knee, they are knowing to analyze medical imaging. Medical imaging is not the only piece of data that companies analyze, of course, but they’re using data. And there’s the third party that actually generate their own, let’s call it, sensor data and each and every sensor data have its own challenges, its own sampling rate, which means you have to generate your own data. And generating your own health data, throughout the process of Helsinki approvals and throughout the process of having enough subjects for each and every data generation, is a very expensive and annoying process. In order to have enough information, you have to get, not just to record people, you have to get the ground truth, which means you need subjects with specific observations and specific health issues so you can generate this kind of health—we just released our blood pressure as part of Binah solution. It was a six-month process just to get access to sick people, very hypertensive or hypotensive, very high blood pressure or very low blood pressure, so we can actually record them with a medical grade—let’s call it ground truth—device, so you can actually have the data.
Peleg Dekalo: But if I understand correctly, it’s not due to privacy reasons it’s just that it’s not out there, the data that you’re after. You need to create it.
David Maman: Exactly. Even if it’s out there, it’s not according to the devices that you’re actually using. We’ve been exposed to dozens of data sets of people connected to PPG devices. The data is worthless for us. We couldn’t even get one insight and actually optimize it to help ourselves in any way. We needed to create everything from scratch.
AI bias in digital health, its effects on health equity, and best practices to minimize it
Peleg Dekalo: So, fortunately, there is an aspiration to advance towards health equity by companies, by people, by systems. And this issue is getting bigger and bigger, especially now when the pandemic hit. And with that being said, plenty of digital health solutions are reliant upon artificial intelligence algorithms that might be biased. For, I think, two main reasons, First of all, the sample that people might use to teach the machine is not a true representation of the population that is being measured. The other reason might be that the mathematical relations inside that algorithm are not accurate and are not a true representation of the objective truth, so to speak. And there is the concern that relying on those biased algorithms will perpetuate health inequity. So, Gila, you mind telling us about that really fascinating issue?
Gila Tolub: Humans are, by definition, biased, right? I mean, we have so much information going through our brain. If we were not biased, we would die out of too much information, right? We wouldn’t know what to do with it. So, we need to have biases to actually help us get some order in our brain, right? It’s meant to protect us and to help us make decisions or else we would all be paralyzed by the amount of information that we’re faced with, and we wouldn’t be able to move forward. But because we have our biases, and it allows us to identify is this a person I should trust, a person I shouldn't trust, we make better decisions. Now, the problem is we are biased. And so we are perpetuating those biases when we build new solutions. The first step is just to realize that it happens. You don’t have to beat yourself up over it. But you just have to make sure that your organization is set up in a way that the culture recognizes that there are biases and then puts in place ways to fight them, whether it’s somebody who’s going to be devil’s advocate at every meeting and remind you, “You might be using this. You might be seeing this because of a bias”, right? Or whether it is making sure to have some external organization review your algorithms, right? Just recognizing that you are, in a sense, biased by definition. And there’s nothing wrong with that, right? I’m way more worried about people that tell me, “Oh, I’m not biased." That actually doesn’t exist, right? You wouldn’t be alive if you were not biased. So let me give you an example. You have some AI in healthcare that are trained based on what doctors have done in the past. You have some algorithms that are trained based on [New England Journal of Medicine] best practices. Which one is less biased? Well, the [New England Journal of Medicine] is the best practices that are published based on their research, right? What the doctors have been doing the last ten years is also full of biases because they’re practicing medicine, hopefully, through the lens of best practices but also based on who they have in front of them. And so, actually, by training algorithms only based on what doctors have been doing, you’re putting in more human biases inside of your algorithms that if you were going straight back to what are the best practices, how should you treat that patient? Not how have you treated that patient before.
David Maman: Gila do you think if it’s even possible to get something that is non-biased? No, I’m not sure.
Gila Tolub: Probably not. But you can work to make sure that the biases are not hurting people.
David Maman: Yeah, probably. But it will never get to a perfect place.
Gila Tolub: It will never get to a perfect place. It will get to a better place.
Peleg Dekalo: Yeah. So basically, to break down the problem into first principles as much as possible and to really check your premises.
Gila Tolub: Exactly. So, for example, we have a few things that we do with companies around patient-centricity or around—we actually have a workshop that we do on inclusion and diversity, right, where we just take a patient-centric approach or look at a product and say, “Hey, where could biases enter this product?” Very open minded. You do it in a workshop with your employees, 30 people, whatever it is, and say, “Hey, what do you guys think?” And then you go one by one, right? It’s an hour-and-a-half workshop. And then you think through where it could get in and what are ways that we could mitigate it. You’re raising awareness. Also, you’re giving a message to all your employees that this is important to you and that this is something that everybody should be on the lookout for. And then can we fix it 100 percent? I don’t know. But even if we can fix it 50 percent, that will be a huge win, right?
Peleg Dekalo: For sure. Minimizing the damage that’s being caused.
Gila Tolub: Can be done. Yeah.
Peleg Dekalo: David, Gila, it was a pleasure.
David Maman: That’s it? Wait.
Peleg Dekalo: It was fascinating.
David Maman: Something AI. Come on. No, I’m kidding. Thank you very much for having me.
Peleg Dekalo: It was a fascinating conversation, really. Thank you so much for joining us.
Gila Tolub: Thank you for having us.
Comments and opinions expressed by interviewees are their own and do not represent or reflect the opinions, policies, or positions of McKinsey & Company or have its endorsement.