Empowering Better Health with Wearables and Other Digital Tools - Dr. Kapil Parakh, Senior Medical Lead at Google





Shiv Gaglani: Hi, I'm Shiv Gaglani. As regular listeners to Raise the Line know, I'm a big proponent of direct-to-consumer healthcare because it helps engage people in their own health. That's one reason I've been really looking forward to today's conversation with Dr. Kapil Parakh, who has helped launch products that reach a billion people and has pioneered partnerships with a range of organizations, including the World Health Organization and the American Heart Association. 


Dr. Parakh is senior medical lead at Google, where he has spent the last nine years leading projects to expand access to health information and help people achieve their fitness goals. Before Google, he served as a White House Fellow and was the Principal Health Advisor to the Secretary of Veterans Affairs. Dr. Parakh also currently sees patients part-time at the VA Medical Center in Washington, D.C., and serves as an adjunct associate professor at Georgetown and adjunct assistant professor at Yale. 


As a clinician scientist, he has over forty publications -- many on psychosocial factors and heart disease -- and his book, Searching for Health, was published by Johns Hopkins Press. 


Before we get into it, it's a very small world. We know many of the same people, including Garth Graham at YouTube Health, who was on the podcast, as well as my dean at Johns Hopkins School of Medicine, Dr. Roy Ziegelstein. When I look back at my email archives, he actually connected Dr. Parakh and me back in 2015 when Osmosis was just getting started. So it's kind of a nice circular moment here. 


Dr. Parakh, thanks for taking the time to be with us on the Raise the Line podcast. 


Dr. Kapil Parakh: Oh, hey, thank you so much for the invite. I'm really glad to be here. And yeah, small world, indeed. It's great to be reconnected.


Shiv: Before we started recording the podcast, you mentioned that Dr. Ziegelstein has had a big impact on your career. We know you're a cardiologist. Do you mind just telling our audience in your own words about what got you interested in a career in medicine and then



Dr. Parakh: Yeah, sure. Happy to. Roy is just fantastic and a really key sort of person in my journey. But I think if I may, I'll take you back sort of to the origin story and how I got here. I grew up in Zambia, a small country in southern Africa, and did medical school there. I saw a lot of people dying of preventable illnesses. You’d treat somebody with cholera or malaria, they'd go back to where they live and get either that or some other disease and come right back, and so this idea of retail medicine being insufficient was an early sort of observation. 


I initially thought I would go on to do global public health based on what I was seeing, and infectious disease. I came to the US for further training, was fortunate to get into Hopkins for a master's in public health, again, in the same international health sort of line, figured I'd need some biostats and epidemiology and stuff like that. So, I did that as well to beef up my sort of technical skills and I met Roy when I was trying to figure out a residency spot. He's like, “Look, all this stuff is great. I just want to make sure you have clinical skills.” So, he set up a rotation for me at Johns Hopkins Bayview Hospital. I did okay at that and so he offered me a spot out of the Match, which was remarkable at the time. I quit everything, so I had to pull myself out of the Match and I accepted, and I said, “Look, I need a visa to get in, and a specific type of visa.” Roy said, sure. 


Then fast forward, like a few months went by, I came to fill in the paperwork at the international office, and they're like, “Oh, you're that guy.” I’m like, “What guy?” “Wait, you don't know?” And I’m like, “Know what?” And they're like, “Well, it turns out Hopkins doesn't give out the kind of visa you wanted and we told them to just give him the other one.” Roy said no. “He's made it very clear that he wants this specific type of visa.” And he took my case all the way to the top. They made a once in a university exception for me, and so I was very well known in that department. Roy never told me. I had no idea until that moment that he had done all of this for me, and of course, I thanked him. 


I ended up in residency in internal medicine at Bayview. He was a program director at the time and sort of had one of these mentoring sessions where I'm like, “Okay, I'm headed towards ID and global public health, and it's going to be amazing.” He said, “Look, all that's great. And I know you grew up there, but have you actually ever worked in global public health?” I'm like, “No.” He's like, “Why don't you do that? Just make sure you know it's for you and that you actually enjoy that work.” I'm like, “Sure, I can do that.” 


I found myself a gig in Zambia to go back and like, I'm going to go home for the summer. That's awesome and so I'm getting paid as a resident. I'm home. This is amazing. Except I hate global public health. Honestly, I think the work was interesting, it was more that I found it hard to justify the ethics of it where I wanted a comfortable life and I knew that I would want a certain salary and a certain way of living. I felt like if my job is to help all these poor people, you just take that money and give it to the village and they might be able to do more with it than I could with that salary. Also, I’m honest enough with myself that I'm not so self-sacrificing like some of the folks who truly live within very few means just to support their community and things like that. 


So, I had that moment, came back and I'm like, “Roy, you were right. I don't think I want to do this.” He asked me “Well, what else do you like?” And then he gave me some really great advice. “Listen, I know you’d like to do something outside of medicine. You want to do this public health stuff. Maybe you'll end up doing research. Maybe you'll end up doing something else. Whatever that is, you do that. But what you do clinically most often doesn't have to be directly related to that.” 


So, as his own example, he started his career as a basic scientist, had publications in Cell and stuff like that -- super amazing stuff -- and he decided to pursue cardiology. Then he moved over to then focus on depression and heart disease, psychosocial factors, things like that. It didn't matter that he was a cardiologist doing that work. He could have been a nephrologist or endocrinologist. That was his point. He said, “When choosing your clinical specialty, choose the kind of patients that excite you. Which rotations do you enjoy? What do you like doing?” And so I said, “I really like cardiology.” 


That was honestly one of the things I loved in medical school and coming to the US, it's even more and Roy's a cardiologist, was one of my mentors, as well as many others who are really fantastic. So he's like, “Well, do that.” So, that is the story of how I ended up in cardiology and the role that Roy played in that journey. 


Shiv: That's incredible. That's a really great story. I'm sure we'll send this podcast to him. I'm sure he's had a very similar impact on many people and you've touched upon something about the importance of mentors here, as well as the importance of truly questioning your assumptions. Because you know, medicine is such a long journey that a lot of people began med school with a very specific idea that they want to be, say, a neurosurgeon or a fill in the blank and then by the third year or fourth year, you sort of very much change course based on the types of patients you're excited about seeing, and also very much so in the mentors they have. So, that's really good advice. 


Tell us more about your journey after becoming a cardiologist, because you've had such an interesting group of companies you worked with and for. 


Dr. Parakh: Yeah, I feel like Roy was so instrumental, but I feel like if I'd gotten potentially more career advice even early in my career, I wouldn't have had such a circuitous path. But I explored a lot of areas. After global public health, I'm like, “Okay, I’ve got to get into cardiology fellowship.” So ,I looked around for projects. Roy had some projects in depression and heart disease, so I wrote some papers on that and there was some heart failure, which I'm interested in. There were some papers around that. I started getting involved in that work just to more do research, but also like just beef up my application. As you just said, medicine's a path and so you have to get to the next rung and there's a series of checkboxes that you have to follow and part of that also included getting my PhD in epidemiology in the time between residency and fellowship. I'd done some of the work during the masters. 


Well, that taught me that I also don't like research. At the end of that experience, I got into fellowship. I got into Hopkins and I went to the main hospital for fellowship and I was like, after I did my thesis, I didn't want to look at anything about depression ever again, or if possible, like not write papers if I could help it. That was probably too strong. I still do like some of that, I just knew it wasn't going to be my primary career. 


So then like, fast forward, I finished cardiology fellowship. I focus in heart failure transplant because within cardiology, that was the stuff. Those are the kinds of patients I like to see – and there's a job opening at Bayview. They need someone to start a heart failure program there. I'm like fresh out of fellowship, but everybody knows me and so they're like, “Do you want this job?” And I'm like, “Actually, I would love it because I think that'd be super interesting and fun.” 


So I've now looked at global public health. Now, I moved to cardiology, not going to do academics. I'm going to try and build this program. It was amazing. It was really an amazing job. I got an opportunity to build a program that focused on reducing heart failure readmissions. This was 2011 to 2013 timeframe. I got to go to a workshop at IDEO and learn about design thinking. I met some folks on LEAN startups. I met Eric Reese and Steve Blank and read all their stuff and talked to folks doing startups. I got involved with a group that was doing Lean Six Sigma and I'm like, “Hey, use my program as a pilot.” And they did. That was out of the Armstrong Institute. Peter Pronovost was great. He gave me some advice. So, like anybody who would listen to me, I'm like, “Help me with my program.” And you know, they were great. Hopkins is such a wonderful place, but also just like the world at large...there were folks from Spain who gave me advice on how to set up a clinic. Like, it was really a ton of fun. 


I did that for about two years and I started hitting a ceiling of there's only so much you can do at Hopkins and then they sort of expect you to be more senior before you can do more meaningful work. So, I started a nonprofit. It was called Health for America. The idea was that we would have a cadre of young students who would learn about design thinking and Lean startup and all this stuff and become innovators. Primarily it's from this idea that medical education is a tunnel and can we expose these people to other ways of thinking and doing before they end up ten years in, out of fellowship, on the other end of that tunnel and all of that creativity and passion and energy has been channeled into becoming a clinician or a clinician scientist or clinician educator or clinician you know, the paths that currently exist. 


Honestly, I think we were too early. We're also on the East Coast in DC where like there wasn't as much of an ecosystem as in the West Coast. We started about the same time as Rock Health and didn't get anywhere near as exciting or as much traction as they did. We did a few cohorts of trainees who went through the program and then ultimately, it got acquired by MedStar Health. I was doing all this work and I met somebody -- this is serendipity now and it's one of those things where you know, you can't plan for these things, but sometimes life takes these moments -- he was like, “Look, all this stuff is great, but you should consider being a White House Fellow.” 


This was at dinner with my wife, who was my co-founder of the nonprofit. And I'm like, “I don't know.” She like elbows me. She's like, “He's interested. Tell him more.” She’s honestly the secret to a lot of my success and in many ways helped me when I migrated and just taught me about the ways of the world and how America works and things like that. So anyway, I got great advice. He was really wonderful. He mentored me and I got in as a White House Fellow, which is amazing. Garth Graham, who you mentioned earlier, actually was a cardiology fellow with me, but served as a White House Fellow years before and so we have that in common as well. 


I served a year as a White House Fellow, which is a phenomenal experience. I served in the VA. It was a turbulent year. There was a government shutdown, which is a much bigger deal. The secretary had to step down because there were a lot of backlogs from wait times and colonoscopy and some folks had passed as a result. I helped onboard the deputy secretary who had never worked in health and is now in charge of this big health system. I was able to do some really meaningful work and I was looking at what's next, because I also knew government wasn't for me. It felt very bureaucratic and slow and political and complicated and I didn't have the patience for it.


Looking back, I probably could have stayed a little bit longer and had more impact, but I had like maybe a hundred different coffee dates that year and everyone was like, “Look, you're great to have coffee with. You're so interesting. I don't know if I can give you a job. I want somebody who's done this for five years and then I'll give them that job. You're not that guy.” So, I met somebody who was at Google. He was like, “I have no job for you, but I'm working on this project. Can you give me some advice?” I did. We talked every few weeks and eventually he's like, “Man, your advice is really helpful. Would you consider joining us?” I'm like, “Sure.” So, he checks and he's like, “Well, we don't really hire doctors. I know how to do this. I can hire you as a temp. These intern positions that we have...we'll just hire you as one of those.”


So, I'd gone from being an assistant professor at Hopkins and director of Heart Failure, to being a White House Fellow, to being a glorified intern. Everyone's like, “You're going the wrong way.” I took the leap. I had to quit Hopkins. They wouldn't even give me an adjunct or anything. It was like, “Look, the rules are if you leave, goodbye. You’re on your own.” So, I closed that door and I was an hourly employee at Google. I had like none of the perks...no paternity leave. I had to work double time for two weeks so they gave me two weeks off to be with my daughter when she was born, and she was born on COBRA insurance because we had no health insurance. 


Shiv: Your wife must have loved this path. 


Dr. Parakh: Oh my God, she's a saint! Not only is she the reason for my success, but she puts up with my craziness and just really in more ways than we have time to discuss, I'm really lucky to have her. In any case, so that was almost nine years ago. I joined search. We launched health knowledge panels, which is like essentially a summary of information. But what we launched was a graph underneath it. That connects points...like cough is connected to asthma is connected to influenza and so when you then type in influenza, you see all the details. But then when you type in cough and fever, you can see related conditions to that query, because Google now understands how those conditions and symptoms and terms are connected. So, we built that infrastructure, then we internationalize it to Brazil to India to a whole bunch of places and built more things on top of that. That reaches billions of people every day now. Really cool work. Then over the years we’ve have done a bunch of things at Google, which has been a really interesting journey. 


I think that's home. I've found my passion around innovation, around digital health, this intersection of consumer and health, as you were saying. It took a while. It took a lot of trying different things out and being like, “Yeah, that's not for me, that's not for me either, not that one.” You know, it would have been nice to figure this out sooner, but I’ve been very lucky to end up where I am, and it has worked out for me. 


Shiv: Yeah, it's an incredible, incredible journey. You've talked about zigzag paths. Do you want to share your advice for our listeners, many of whom probably are questioning, do I want to be a researcher? Do I want to go into tech? Do I want to, you know, practice full time all the time? What do you say to them? 


Dr. Parakh: Those are all great questions. About every few weeks, somebody reaches out and says, “Hey, I need career advice. I'm looking for a job. I'm trying to figure out what I'm doing,” One of the things I live by is that autobiography is not advice. I read that somewhere. I can't claim any benefit, any insight from that, except I liked it. You know, people will say, “Hey, tell me your story, maybe I can get some out of it.” But I’ll say, “Tell me what you're dealing with and I'll give you the parts of my story that might make sense, as well as some advice and things.” I appreciate them giving me the opportunity to give my story and you feel important and cool, but it doesn't really help people as they're trying to navigate things, right? 


So, some of the common pieces of advice that I end up giving are around ‘try different things.’ 

Many times, you sort of have the set path and don't be afraid to fail. Somebody else said -- again, I take no credit for it -- life rewards risk disproportionately. You and I are fortunate to be in a place where you could take a risk with Osmosis, and if it went great, that's great. If it completely went berserk, you can go back to medical school, you get a job at McKinsey, you could do something else. We're so privileged that when we take risks, the reward is massive, and if we fail, we learn something, and we still have a place to land. 


Part of why I say that we're fortunate to do that is the inverse is also true. There's this great piece in The Atlantic about folks who are underprivileged. If you work an hourly job, your car breaks down, then you get fired because you were late to work a couple of times. Now you can't afford to repair your car, now you can't get to work and like, that spirals backwards and that's really terrible and it is what it is. The point being, that knowing this, in medicine we’re all taught to be risk averse and conservative. So, I think appropriately and thoughtfully taking risk -- don't go to Vegas and say, ‘Kapil said, take to take risks’ -- but appropriately taking risks with your career, trying new things...I think that's something worth doing. So, that's one. 


The second is something you touched upon, which we just talked about, which is around like, do I want to be a clinician or not? I think that's a really, really important question. You don't have to be a clinician to have an impact on health, Osmosis being a great example. But if you want to be a clinician, then my advice is always do that first and do that really well. Don't try to double dip like, “Oh, I'm going to be coding on the side and learning how to take care of heart failure.” No... that just doesn't work. Like, really become a good doctor, because if you mess that up, that's somebody's life you're dealing with and it's hard to overestimate how much damage you can do. I've lost friends from medical errors. It's terrible. So, if you're going to be a doctor, then do it right. Focus on that and then do the next thing, right? You might do a little stuff on the side -- maybe take an elective or two, try different things out, make sure you're on the right path -- but do that really well. 


So, that's like the other piece of advice I have. So, when that fork happens, choose that and then you can choose what you're passionate about in terms of your clinical work, what kind of patients you like to see and all that. But if you do decide to be a doctor, be a good doctor. 


Shiv: Yeah, that's great advice. And actually very timely, because as you know, I've come back to med school myself. Since Elsevier bought Osmosis, I've had more time to consider what's next and that's largely why I've come back. 


I want to dive into some of the work at Google as well, just things you've been involved with. We've been very fortunate to get to know many of the clinicians and other people working in health at Google. I mentioned I met up with Michael Howell, chief clinical officer there at this AI med conference. Garth Graham's wonderful. And then the team, I'm sure you worked with some of them...like Claudia Amar on the search panels. It's very, very cool work and the scale of impact is what gets us really excited. We basically grew on YouTube. That's how Osmosis became popular with 2.8 million YouTube subscribers. 


So, one aspect we learned about this year in COVID especially, is this idea of health literacy, patient education, and combating misinformation and disinformation. Do you want to comment a bit about some of that? This is a very big question, but in the age of generative AI, when more content is being produced than we can handle -- I thought there was too much content years ago, let alone now -- how are you thinking about it personally, or whether you can share about Google?


Dr. Parakh: Absolutely. I can't speak on behalf of Google. I have to give you my perspective. I started writing a book five years ago called Searching for Health which was released last year. At that point, we had a much simpler world and the idea was like, you get all this information online, it's very confusing. How do I go talk to my doctor about this? How can we empower people so they can have more meaningful conversations with their doctors? It came from my personal sort of disconnect between seeing patients get very confused with this stuff online, and not being sure how to bring it up with me, and on the other hand, I've like built products that are supposed to simplify that information and make it easier to access, and still there's a gap.


I felt the gap that was that people don't really understand how to take what they find online and insert into a short interview with the doctor. There's such a big information gap between the two. So, we provided a lot of tools and things to make this happen. Ironically, after I wrote my first chapter, I showed it to my wife and she's like, “Look, you're great and everything, but go talk to Anna. You need her help.” Anna Dirksen is my wife’s friend and my eventual co-author. She took all of these dry ideas and we found a whole bunch of anecdotes, brought it to life and put in a bunch of tools, and got over a hundred something references. It takes all the helpful literacy literature and makes it useful. 


I can't claim any credit for the original work. I've got decades of work on this stuff and tried to turn that into useful things that people could use. We published in April of 2021, which is like right in the middle of the pandemic. Now looking back, we’ve gone from this relatively simple problem of like, “Hey, this is a lot of information. How do you take this to your doctor?” Some of it is overwhelming.” But then, none of it was malicious. At that time, it wasn't quite the same level of politicization. There's misinformation and active misinformation, right? Like, deliberately leading people in directions that are unhelpful from various motivations. 


So, it's a space that I've obviously long been interested in and have written about. I agree 100% with you that it's only gonna get more and more challenging. I think we've historically relied upon organizations like the WHO and the CDC and public health bodies to be smart about this stuff and help us make sense of these complicated scientific things. I think the level of investment in health communication is an area that's under-invested in vis-a-vis the need. Most experts are great at talking to each other, but not so great at communicating to the lay public and I think there's now an increased recognition of the need to do that. More and more people are doing that or part of it...connecting the dots between experts and the lay public and my hope is that as much as generative AI could add fuel to the misinformation fire -- because now you can generate tons of content and make it super engaging because the model knows how to do that and very click baity and all of that sort of stuff -- there's an opportunity for experts to also use generative AI in a way that's engaging and useful that they didn't have before. 


In the past, if you asked me about hypertension treatment, I'd be like, “There’s five different drugs and this one does this, but sometimes you want to use this one first.” Three paragraphs later the patients’ eyes glaze over. Whereas now I could say to Bard or ChatGPT -- pick your generative AI -- here's a summary of the guidelines. Turn this into five bullet points. Make it funny. Make it interesting. Add some emojis. Boom. Like, now in literally ten minutes, any scientist who wants to get their message across can translate their science into engaging content and then proof it and be like, “Oh yeah, this makes sense.” 


When we did this for knowledge panels back in the day when we launched, first we had to have a UX writer write this stuff. Then I read it as well as a whole bunch of doctors internally at Google to make sure it's legit. Then we had the Mayo Clinic read that and then we put it out in the world because it was reaching billions of people. Now, you could do that at scale, at speed with generative AI that anybody else has access to. So, I'm hoping that if we're smart and if we can use education platforms like Osmosis and others, we can use generative AI to close that gap rather than let it widen. 


Shiv: Yeah, I couldn't agree more. I mean, this is definitely what motivates us and I mentioned that the YouTube partnership has been incredible for us, not only for generating new content with Garth and his team -- we just started working on some rare disease content for this Year of the Zebra initiative we've been doing -- but also just the fact that there's community driven and technology driven translations and localizations, right? Our main YouTube channel has 2.8 million subscribers, but a community of doctors and other health professionals in Vietnam created Osmosis Vietnamese and have translated a subset of our Vietnamese videos, not just the subtitles, but the text on the screen and the spoken audio. There are 100,000 subscribers to this channel and we've done nothing. We’ve just given the ability to do it.


Now with Gen AI, what I'm excited about -- if I can put a plug for you to take back to your colleagues at Google -- I would love to snap my fingers and/or have Google snap its fingers and have all of our content in any language. I mean, I was born in Namibia, you're from Zambia. You go to these countries and the English proficiency isn't as strong, but this content is important for community healthcare workers or a pregnant mother...whatever it may be. So, I would love for our content, trusted content, to get to as many people as possible using these tools and I think Google is a great, great force for good in that way.


Dr. Parakh: Yeah and I mean, I love what you all are doing with video. That's such a great way to get past literacy barriers, right? You don't have to read a video, you just listen and watch. I do think by nature, I'm an optimist and an enthusiast, but in as much as I have that part of my brain, I also have the like, realist, clinician, researcher part. I can be critical about digital tools and criticize snake oil when I see it. But I do think when you look at some of these technologies, like video, for example, it legitimately has this ability to cross barriers and have a real impact. So, again, technology is a tool and it's up to us. I think all too often medicine is, by the conservative nature of the field, slow to take advantage of these and we're sort of being forced to move quicker and I think if we do it well, that can be a good thing.


Shiv: Yeah. And that's why it’s important to get our audience proficient in what's happening so they can help their future patients. You know, when 23andMe came out with direct-to-consumer genetic tests, patients would show up at an appointment with this report and ask, “Does this mean I'm going to get breast cancer?” They didn't understand it and there weren't enough genetic counselors to go around. So, being able to raise the bar of what patients understand with health literacy, but also scale what health professionals can do in terms of explaining it, is again, one of the things possible through platforms like Google.


Dr. Parakh: Yeah, it's super exciting. It's a great time.


Shiv: So, switching gears to the other part of your work -- which is remote patient monitoring, Fitbit, the work that you've done there -- a decade ago, my summer project when I was starting Osmosis on the side was working with Daniele Rigamonti. Do you remember him from Hopkins? 


Dr. Parakh: I remember the name, yeah. 


Shiv: Yeah. He's a mentor of mine. He's been on this podcast. We worked on several projects. One of them was slapping Fitbits and other commercial grade pedometers on patients with hydrocephalus and using that data to predict whether their hydrocephalus would get worse and they'd see other symptoms. Clearly, we've been talking about this for years. I saw you recently shared Mitesh Patel’s paper in Nature -- another mutual contact of ours -- about hospital readmission rates and predicting that with Fitbit and other devices. 


Dr. Parakh: Yeah. 


Shiv: I'd love to hear more about how the space has evolved in terms of remote patient monitoring and consumer health devices and where do you think it's going in the next couple of years? 


Dr. Parakh: I think I would elevate that a little bit beyond remote monitoring. I think there's a ton of opportunity for consumer devices, wearables, phones in health. Remote monitoring is one aspect of it, but you sort of think of population health writ large, right? To answer your question on remote monitoring, I think where I see consumer data on that front is, I think there's a careful distinction you have to make on appropriate use of data. Remote monitoring of vitals, like blood pressure and heart rate, these are medical grade devices that measure medical parameters that clinicians are very comfortable dealing with. If your blood pressure is high, I know what to do with it. If your heart rate is low, maybe cut down on that beta blocker, right? It's a very straightforward way of managing that. 


Whereas if your step count drops or if your sleep is erratic based on your consumer grade wearable -- whether it's Fitbit or not -- one, that data is consumer grade data, so it's not meant for clinical decision-making. Its accuracy is lower. It's still useful, but it's not the same as a blood pressure monitor, which by the way, has its own problems. Like, blood pressure is not a perfect measure, as you well know. It sort of goes up and down. But we are trained how to use that and understand its limitations and its fidelity and so on and so forth. Most clinicians don't have the training on what to do with this consumer grade data. They don't understand it.


Then two is, in my opinion, I think it's an incredibly powerful signal that's best used as a way to get more information. If a patient’s step count is down, give them a call, see what's

going on. Your sleep's disturbed. Give them a call, see what's going on and it might be that, “Hey, you know what? I had some friends in town. We're drinking a little too much and so I'm kind of hungover. That's what's going on.” “Okay. Don't do that again. Take care and hydrate.” Or the patient could tell you, I'm having chest pain. That's why I'm not walking around as much. Bigger problem. You should come in. Let's talk. So, I do think this consumer grade data and consumer devices -- when it comes to remote monitoring -- are powerful not because the quality of data is phenomenal. I think it's quite actually the opposite, the quantity of data. Most people will wear a Fitbit 24-7, charge it once a week or something because it's got a great battery life, and sort of keep it on just for a long time. Whereas getting a heart failure patient to just stand on a scale and get a blood pressure once a day, we'd have nurses calling them to remind them, and they don't like doing it because they don't like looking at their weight or it's inconvenient or whatever. So, you'd get a couple of discrete, very high-quality data points that were inconsistent. So it was much harder to understand what's happening with them over time and so this

is where the larger volume of data, even though it's poor quality, could be useful. 


The analogy I sometimes give is like smartphones and landlines. You have amazing signal on a landline, you just don't have it where you need it. Whereas you have a lot of poor signal, poor quality on a cell phone, but everyone uses cell phones because it's the convenience of having it all the time. So, understanding what that data is, what the regulations around it are, and how you can use it in an appropriate way, I think is interesting and once we get that done and we get the systems around it, I think we'll see really powerful things and we're starting to do some of that. 


So, one part of it is remote monitoring, but then there's a whole bunch of other things. If you look at population health, there's stuff around behavior change. Fitbit helps people get more active, something that was invisible before sedentary time. Now you have steps, 10,000 steps, and people recognize, oh man, I must be sitting a lot because I'm not taking many steps. And not just a Fitbit, your phone measures steps...it’s sort of everywhere, right? It started with that, but it got popularized and so you could use that to reduce sedentary time -- which by the way is one of the WHO guidelines and is associated with a whole ton of health benefits. So, all the behavior change around physical activity, sleep, stress, et cetera, to promote wellbeing at a population scale...that could have huge impact on disease prevention. 


Then there's a whole separate use of that data around predictions, which is different from remote monitoring. When COVID rolled around, for example, there's a paper in Nature Medicine,

which showed that Fitbit data could be used to predict the onset of symptoms a day before they actually happen, and this will happen for more and more diseases over time. The one that's FDA-cleared is around atrial fibrillation. Fitbit can detect irregular heart rhythms. They did it in 455,000 people, the largest study of its kind, and they showed that we could predict when these irregular heart rhythms would happen and correlate that with atrial fibrillation on a diagnostic patch. So, it's not a diagnosis, it's a detection, but it's a prompt for a doctor to prescribe a diagnostic tool like a patch and then make that diagnosis. We actually help prevent stroke if that's confirmed and they are put on treatment. 


So, there's all these different buckets of work around where you can take these consumer devices and make them useful. There's work being done around recovery. If you've had a heart attack, for example, we recommend cardiac rehab, and that's something that's been shown to reduce mortality and morbidity, readmissions, and all those kinds of things. Really important stuff,

thirty years in the making. Like, this has been a known thing. But cardiac rehab involves you coming in three times a week into a facility, doing some exercise under supervision and so on. It's very hard for people to do, and during COVID, it was impossible to do. So, we've actually partnered with some folks in Ireland and did this program where they got Fitbits and did the cardiac rehab remotely, and it showed some great results. Now, we have a randomized, controlled trial across three NHS trusts that's being done in partnership with the British Heart Foundation Data Science Center and funded by the NHS to study this in a more rigorous way. 


We have a similar program that I helped launch in Manipal, India, where we're doing this for post-surgery as well as for cardiac rehab. The list goes on, but you get the idea that you can take these consumer-grade tools intelligently and appropriately use them in many different clinical and population health settings. But you just have to understand what that data means and how to use it.


Shiv: Those are some great examples and again, I think in this age of big data -- we've been talking about big data for many decades now -- but it feels like the data has just gotten bigger and our ability to process it and draw inferences and turn it into insights and prediction is even bigger. So, we've touched upon a couple of big things and already gone through the advice. I want to be respectful of your time. What else would you like to share with our audience about what's top of mind for you? You're still practicing, you wear multiple hats, you have this book. What else is like keeping you active and excited these days? 


Dr. Parakh: Yeah, I mean, that's funny. But I think for your audience, I'd encourage them to learn more. One of the things I helped the Fitbit do is we put up a page for physicians so that clinicians can learn about our partnerships with the WHO and the American Heart Association and the American Academy of Sleep Medicine because we follow their guidelines and our tools help you get to those guidelines as well. There's a ton of science behind this. There are about 1,700 papers and counting that are published on Fitbit and most people don't understand that level of evidence that exists. So, as you embark in these careers, yourself included, more and more patients will come to you and say, “Hey, my device said I might have atrial fibrillation.” What do you do with that? When I present at conferences, I've asked the question, so what do you do with that information? Do you just ignore it? And some hands will go up and say, “Yeah, we just ignore it.” And then I say, “Or do you prescribe a blood thinner?” And some hands will go up. Then I'll say, “I hate to say this, but both sides are wrong.” 


On the one hand, you don't want to ignore it because it's actually FDA-cleared and useful, and atrial fibrillation is a real disease that can have consequences like stroke. On the other hand, it's not a diagnostic tool, so you can't start prescribing a blood thinner on it. Somebody will be like, “Yeah, but I saw that single lead ECG that came from the device. It looked pretty good.” I'm like, “That's not what it's FDA-cleared for. You still have to validate it with an FDA-cleared medical grade device and use it appropriately.” 


We understand that level of nuance for other tools that we use, so we're capable of this, right? Like somebody comes into me and says, “Oh, doc, I wrote down my blood pressure readings for the last two weeks,” and I take a quick glance at it. I'm like, “Oh, yeah, take your lisinopril ten and make it twenty,” because I know how blood pressure machines work. I know that yes, he's got ten numbers there, but six or eight out of those ten are above range. So, that's meaningful enough for me. How far above range? Enough that I can increase his medicine. How much? Oh yeah, if I increase this medicine X amount, that's how much benefactor I estimate to get. 


If somebody comes back with wearable data, you're like at a blank slate. It takes probably hours of reading to get to that thirty second conversation reviewing those data points and changing the lisinopril dose. If you put in some level of effort -- maybe not as much as for learning blood pressure readings -- into understanding consumer data and those things, and we certainly need to build the tools out for that, you'll be able to have similar conversations with your patients. And why is that important? 


This is true for wearable data, but also true for health information that people seek online. If you think about it, somebody coming into your office with either stuff they found online or some data from the wearable or something like that is telling you, “I care about my health. I'm investing in my health. I spent the time to put on this device, to look this information up online. Here's what I got.” And then if you say, “Look, all that stuff's junk. Let's talk about real medicine. Let's talk about your medications that we need to adjust, because that's on my agenda.” Well, the patient's going to be like, “That doctor doesn't care about me.” But if you take the thirty seconds to engage with it, “Oh, amazing. That's fantastic. Let's take a look. Yeah, you know, this doesn't apply to your condition. I'll explain to you why.” Because you took literally the thirty seconds it took to look at the information they collected -- whether it's the online information, whether it's wearable data, whatever it is -- and validate the work that they did, then you can take that energy and point it in the right direction. 


You can say, “Don't worry about the step counts. What I really need you to be doing is set a reminder for the medication so you don't skip those and maybe check your blood pressure twice a week and you know what, your sleep's kind of off, so why don't you focus on getting to bed every day at a reasonable hour.” Right? You validated them, pointed them in the right direction and now they'll listen to you. You've built that relationship and they'll come back eager, like, “Doc, you told me to do these things. Here's where we're at. What's going on?” So it's an opportunity. And it's like a little bit of a judo move where you take their energy and put it in the right way, as opposed to turn into a confrontation where you are thinking, “I can't bill for this stuff. I know how to make sense of it. I'm just going to do what I'm comfortable doing. What's more important, at least for me.” So, that's my sort of parting thoughts on how we think about these consumer tools because we're just going to see more and more of them over time. 


Shiv: I think that's great advice. I love that. The judo patient education. Maybe that's a course or a video series we'll build. Because I agree. A lot of clinicians I've met are fairly dismissive of patient-generated health data, whereas others maybe put too much stock in it and we all know about the “worried well” and cyberchondriacs. We've heard those terms and it's nuanced. I think you appropriately hit upon nuances. It's good advice. Any last words? Anything else you want to share before we let you go for the day? 


Dr. Parakh: No, this is fantastic. I'm really excited for Osmosis as a platform, because I do think you guys reach a ton of people. So, I'd love to collaborate on wearable education and I think over time, you should have courses in Gen AI and really bring clinicians up to speed on the emerging parts of technology, as well as more traditional education. So, yeah, I’m grateful to be here and excited for where things are going.


Shiv: Absolutely. You're speaking my language. It's why we have you and other leaders on the podcast, to get our students thinking more beyond step one, beyond NCLEX, and thinking about actual practice and delivery of medicine. So, Dr. Parakh, thanks so much for taking the time to be with us today and more importantly, for the work that you've been doing across multiple very interesting roles to raise the line and strengthen our healthcare system. 


Dr. Parakh: Likewise. Thank you for the opportunity and thank you for all the work you've done. Thank you. 


Shiv: And with that, I'm Shiv Gaglani. Thank you to our audience for checking out today's show. Remember to do your part to raise the line and strengthen our healthcare system. We're all in this together. Take care.