Cardio Care Now, Ep. 3

Wearable Technology Data in Health

Seth Martin, MD, MHS

In this podcast series, moderator Seth S. Martin, MD, MHS, aims to explore hot topics in cardiovascular medicine. Gain insight on topics, such as health equity, health technology, genetics of cardiovascular disease, and breakthrough treatments for hyperlipidemia.

​​​​​​In this episode, Dr Martin speaks with Dr Kapil Parakh, MD, PhD, who is the Medical Lead at Fitbit, about the use of wearable technology data to improve health, including the use of sleep data, physical activity data, and the detection of atrial fibrillation.

Kapil Parakh, MD, PhD

Dr Kapil Parakh, MD, PhD, is a cardiologist at VA Medical Center, Medical Lead at Fitbit (Google), and an adjunct professor at Yale University and Georgetown University (Greater Baltimore, MD Area).

Seth Martin, MD

Seth Martin, MD, MHS, is a preventive cardiologist and an associate professor of medicine in the Division of Cardiology at Johns Hopkins University School of Medicine (Baltimore, MD). Dr Martin is the director of the Advanced Lipid Disorders Program and Digital Health Lab at the Ciccarone Center for the Prevention of Cardiovascular Disease and the center director and the principal investigator of the mTECH Center, part of the AHA Health Tech and Innovation Network. He is also the co-founder of Corrie Health, Inc. 



Hello and welcome to Cardio Care Now, a special podcast series led by Dr. Seth Martin. Dr. Martin is a cardiologist and an associate professor at Johns Hopkins University School of Medicine in Baltimore, Maryland. The view of the speakers are their own and do not reflect the views of their respective institutions or Consultant360.

Seth Martin, MD, MHS: Welcome back to the Cardio Care Now podcast. Really appreciate everyone tuning back in. I'm excited for the conversation today with Dr. Kapil Parakh, who is someone I've known for a number of years, dating back to when he was my attending in the Hopkins CCU. And he's just had a fascinating career since then, since his time as an academic cardiologist and heart failure specialist at Hopkins. Kapil's gone on to work as a White House fellow in Washington, and then to wear different hats at Google and has tremendous insights to share around the role of technology in the worlds of medical care and cardiology, and has specifically been leading efforts with respect to wearables, which is going to be the focus of our discussion today. We really want to understand the science around wearables, what can be done today and what's coming down the line in terms of the role of wearables in clinical practice. Kapil, I'm thrilled to have you on the podcast and looking forward to our discussion today.

Kapil Parakh, MD, PhD: Thank you, Seth, and really appreciate that generous introduction. You've had such an amazing career since our time at Hopkins and really glad to be here together.

Dr Martin: Let's kick things off. I think first of all, it would be great just for the audience to learn a bit more about yourself and your journey. I alluded to some of it, but you've had this exciting career path and I'm just curious for the audience to learn more about what really drives you and what's led you to take the career path that you have as you think about the impact that you'll have on patients and the world of medicine.

Dr Parakh: Thank you. I've been incredibly lucky. I did medical school in Zambia, came to the US for further training at Hopkins where we met and I explored a number of different paths, but I think the common theme in my journey has been one around trying to increase my impact. Growing up in Zambia and doing medical school there, I saw a lot of challenges, many of which revolved around public health and people, essentially, for lack of resources, dying from diseases that were entirely preventable.

And I explored a number of paths to see how I could expand my impact. Initially, I did a Masters of Public Health thinking that would be my path, and then pursued research thinking that would be it. And then program building in academic medicine, policy as a White House fellow and then finally I found a home at Google where working through technology and reaching, in some cases, billions of people, has been a wonderful opportunity. In short, you could say I could have benefited from some career guidance early on, but I explored a number of different paths and fortunately have found one that seems to be a great fit.

Dr Martin: For sure, and it's been great to collaborate with you, get to learn from you over the years. I think, let's start diving into our topic at hand, wearables. We have a lot of clinicians in the audience, many of whom will be familiar with wearables, but I think it is helpful just to level set from the beginning just to define what are we talking about with wearables, the basic ideas around how they can be applied in medical and cardiovascular care, and then we can dig further under the surface from there.

Dr Parakh: Thanks, Seth. That's a great question. Wearables cover a broad range of devices and that includes things like smartwatches, which are probably the most common, but trackers as well as rings, even shoes and jackets, so on. The common thread is that these all have sensors that can detect things like movement, heart rate or sleep, and provide that to the user often through a smartphone app.

Dr Martin: There's a lot that can be measured and a lot of potential here. Certain areas have taken the leading position in the race, so to speak, in a world of cardiology. Atrial fibrillation has been one of those. Should we dig into AFib as an example up front around what's being measured there? What does it mean? What can be done with the data?

Dr Parakh: Yeah, absolutely. That's a great one to start with. Many wearables can detect heart rate. In a couple of cases, for example, Fitbit is a range of devices that do this, and the Apple Watch is a couple of those that do that as well. They look at the regularity of heart rates. This is a photo plethysmography-based signal. If you've seen any traces in inpatients where they look at pulse oximetry, it's a similar technology, shine a light through the skin, get it back, and you look at the changes in the light characteristics and you can detect a pulse. And what the algorithms do is they take periods of inactivity, when somebody is not moving and that reduces noise and false positives. And in that period of inactivity, they see how regular the pulse is coming through photoplethysmography. And what that tells you is if it is irregularly irregular, as we all know with atrial fibrillation, then the user can get an alert to say that you might have an irregular heart rhythm and go speak to a doctor about it.

Some of the key things here is that because it's done in periods of inactivity, for example, Fitbit has a multi-day battery life and is often used in sleep. That's when many of these episodes are detected. And it tells you in those periods what's happening. The other thing is that the algorithm tends to take 30 minutes' worth of data and look for recurrent periods of irregularity in this. It's not just a one-off thing that it just happens. Finally, users don't get notified in the moment. It tends to be an after-the-fact notification because of the way things are processed. It's not like you find out as soon as you go into atrial fibrillation, but rather sometime after the fact, usually a few hours.

The other piece that's really important about this is that the FDA clearance that both Fitbit and Apple have gone are around detection of atrial fibrillation, but not a diagnosis. The diagnosis still has to be confirmed with a medical grade device such as a patch, like a Zio Patch or any of the other patches that are out there, a Holter monitor or event recorder, something else that is medical grade that can confirm this diagnosis of atrial fibrillation. That's one example of how wearables can detect irregular heart rhythm and eventually lead to a diagnosis of atrial fibrillation.

The other feature, which is interesting and exciting is that many wearables, Fitbit, Apple, Withings, Samsung can capture an electrocardiogram or an ECG. Now this is a single lead ECG that's usually qualitatively similar to Lead 1, the electrical impulse across the left and right arms. And many of these companies have similar algorithms that can tell a user whether that ECG trace is sinus rhythm or atrial fibrillation or inconclusive. And there are certain reasons why these can be inconclusive and sometimes it's poor signal quality. Sometimes the heart rate is really too fast to tell the regularity, et cetera.

But these are two different ways in which you can facilitate the detection of atrial fibrillation. And again, even with a single EECG, that has to be confirmed with a medical-grade device. It's not a diagnosis, it's more of a detection, somebody who, for example, might be more of a symptomatic person decide to check a single lead ECG and they figure it out. But this is all really exciting and has happened over the last four or five years and we're still figuring out exactly how to deploy these and in which populations they can be most impactful.

Dr Martin: Thanks Kapil. Thanks for that really clear description. I really like how you highlighted the difference between detection and diagnosis. And at a high level, it speaks to the general concept that often technology is not going to totally replace clinicians or the medical system, but there's this collaboration or synergy between the two entities. And the technology can help augment what's happening in clinical care, help identify more people, and they can plug into the traditional medical system, receive a formal diagnosis, and then care can flow from there.

Dr Parakh: Absolutely. It's complimentary. Absolutely.

Dr Martin: Maybe let's dive some more into day-to-day clinical care where we are today with the integration of wearable data, where we're going. In addition to AFib, be interested to hear where you think we are today in terms of what can be used from consumer wearables and how that can complement clinical care.

Dr Parakh: Absolutely. I love this complimentary relationship that you've described. I think that's exactly the right framing to think of it. We see patients every three, six months, sometimes yearly. And there are long periods basically between our visits and we often don't know what's happening to our patients in between visits. And what wearables can help do is fill in those gaps and I think they help fill them in a couple of different important ways. One is just lifestyle changes. We know whether it's diabetes, hypertension, coronary disease, even atrial fibrillation, that the first line recommendation is around lifestyle changes. Whether it's changes in physical activity, I think generally recommended across the board to increase physical activity, changes in diet, whether it's cutting down sugar if you're a diabetic or salt if you're hypertensive and so on, so forth. There's a number of lifestyle changes that we recommend and what wearables can help do is to turn this advice into action and make it easier for clinicians to help patients along that journey.

I was actually talking to physical medicine expert who's an exercise physician. She's worked with all the way from Olympic athletes to helping write some of the guidelines. And in what she described as the lack of physical activities associated with 6% of excess mortality in the world. And when you compare to smoking, which is about 8%, that's pretty close. And just like we talk to our patients about smoking and we can counsel them, I think we can counsel them similarly about physical activity. And what these tools do is make it easy to talk about. It's an actionable thing. You can ask what their metrics were before your visit or set goals for the period after the visit. And I think it is just a way to bridge the gap between visits and help people adhere to the recommendations.

Dr Martin: You get that insight into what's been happening in the home and the community between visits. And physical activity has been one of those pieces and that has been focused on with wearables, partly because it's measurable, but it has such broad clinical and public health importance from primary prevention of cardiovascular disease all the way to heart failure, including heart failure with preserved ejection fraction. There was just a Jack article that came out showing really the importance of physical activity in HFpEF. I wonder if you're seeing one of your own patients in clinic and going through their wearable data, including the physical activity data, what do you focus on? What do you trust? What do you maybe discard a little bit more and say, "You know what, I'm not sure about the accuracy of that." How do you think about these data when you're there in the clinic with your patients?

Dr Parakh: That's a great question and I love how you bring it down to the practical nitty gritty of it. I think that the key thing to remember is a lot of this data, with the exception of the atrial fibrillation data, is wellness data or consumer-grade data. While it's useful and important, it's not medical great data. As a cardiologist for example, I might see somebody's heart rate trends on their Fitbit or the wearable device. I can't actually use that to titrate a beta blocker dose because it's not medical-grade heart rate. Doesn't mean it's useless, but it's not intended for diagnosing or treating or managing conditions. It's much more intended around lifestyle changes. Then what do I actually look at with that disclaimer? Physical activity, let's just double-click on that since you already mentioned the importance of this.

There are two main guidelines. If you look at the scientific report that the government put out on physical activity, which is about 770 pages long, recommended if you have insomnia, but I could summarize it in just a few words. What they recommend is to reduce sedentary time, basically move more. And the metric to look at there is honestly steps. Fitbit popularized 10,000 steps, but really it's just more of a directional thing. You want people to getting more steps. For some folks if their baseline is 3000, 10,000 is too far away, try and reach for four or 5,000. You just want to them gradually increasing over time. But really important just to get them moving. And then the second part of the physical activity guidelines is they recommend 150 minutes per week of moderate to vigorous physical activity. And you could get either 75 minutes a week of vigorous, 150 minutes of moderate or some combination thereof.

Now that's a mouthful to say, let alone decipher. Devices like Fitbit make it easy because it's 150 active zone minutes per week or more. And the way users get awarded active zone minutes is based on their heart rate and their heart rate reserve, which is based on their resting heart rate. If you are in a certain zone, you get one active zone minute for every minute of moderate physical activity. And if you're in a second more vigorous zone, you get two active zone minutes for every minute of vigorous activity. And you don't have to do the math of was it 75 minutes of vigorous or 150 minutes of moderate or some combination. If you get 150 active zone minutes or more, that's great, you achieved the recommendation. And the recommendation is that's least 150. There's benefits up to 300. The relationship between moderate-vigorous physical activity and health benefits of various types in the literature tends to be curvy linear.

What that means is when you go from very sedentary, like let's say 10 or 20 active zone minutes per week to even moderately more like 30, 40, 50 active zone minutes per week, that increase is associated with a greater inflection in your health benefits than if you went from 200 active zone minutes a week to 250 active zone minutes a week. Really the greatest benefit is at the lower end of the spectrum, not that you won't get benefits further on, but for all those folks that we think are rather sedentary and hard to get moving, it's important to remember that these are the folks that have the greatest health benefits from physical activity. I'll stop there on physical activity real quick and then transition over to a couple of other areas. Aside from physical activity, many of these devices look at sleep. For example, we partner with the American Academy of Sleep Medicine and they have two main recommendations that you can look at a Fitbit or any other device data to look at.

There are two main recommendations around sleep duration, which is a sufficient amount of sleep every week, ideally every day. And then consistency, which is you want to have a consistent level of sleep every day over the course of the week and not sleeping less on weekdays and more on weekends. But rather than consistency because of this clock and sleep is actually becoming more and more recognized in cardiovascular disease. The American Heart Association has a checklist, Life Simple Eight. It used to be Life Simple Sevens, but they added sleep duration to it to make it Life Simple Eight. And it's a checklist to help prevent cardiovascular disease, both primary and secondary prevention, that sort of thing. Sleep is another area that I tend to look at in addition to physical activity. I'll stop there because I've said a whole bunch of stuff, but I'll let you react to that.

Dr Martin: Thank you. That's fantastic. I love, first of all with physical activity, how you broke things down there and emphasized, for example, in those that have a lower level there is that even greater potential to have an impact that we need to make sure we don't give somebody with a step count of a thousand per day, an initial goal of 10,000 per day. We need to set realistic goals and gradually work up from there. And I love that you brought in sleep, especially given that's such a timely hot topic with the AHA just releasing their essential eight that brings in sleep data. That's something where I personally feel like in my clinic it's been less prominent. I think the physical activity data has more naturally been something patients bring in, but the sleep data, there's room for growth. And I imagine that may be the case in other clinicians' clinics. Thank you for bringing that in.

Dr Parakh: Absolutely. And I think it doesn't take a lot to dig into these things, but once you started to develop that basic understanding of both the physical activity and the sleep stuff, much like we talked to our patients about smoking, I have a few standard things, "Do you smoke? How much?" And then start going into, "Would a patch benefit you? Gum, et cetera." Once you develop that quick repertoire of a few hit list things that you want to touch base on, you can efficiently incorporate this into a visit. We all know we're pressed for time and it becomes challenging. The other thing I wanted to talk a little bit about is of building these into clinical programs. Aside from what an individual clinician can do, I know you've done some great work in Query Health around incorporating technology into care pathways and post-discharge.

We collaborate with a cardiologist in Ireland who's actually over the course of the pandemic used Fitbit to turn her cardiac rehab program into a virtual program. And essentially what they've done is they give patients who are signing up for cardiac rehab Fitbit devices, and they also have an app and a dashboard. And essentially in the app, patients can report their symptoms and whether they're taking medication so on. And Fitbit keeps track of their steps and other metrics as we've discussed. And nurses will give them a call every week, every whatever the schedule is for cardiac rehab. And they'll discuss not just in the abstract, but very specifically, like, "Hey, how's your activity levels? Here are your step counts, here's your active zone minutes, how's that coming along?" And then they'll go through symptoms and medication adherence and so on and so forth.

And what they've found is that not only are patients more likely to complete this kind of cardiac rehab because the virtual components, the logistics of it are much simpler. You don't have to drive and so on. But that patients are more satisfied with it. They feel monitored because of the device. They feel like they're connected to the hospital more. And the care teams more, remember this is an anxious period, people just had a heart attack. They like this idea of being connected.

And as a result, they're more likely to take their medicines, they have better control of their cholesterol, their blood pressure, and so on and so forth, and are much less likely to go back into the hospital. And I know Dr Connolly's working on publishing this and these results, and she's presented some work of the European Society of Cardiology, and we're also looking into scaling this up across NHS. But you can see how, from this example, if you build the wearable into a greater program, it's not left just an individual clinician to look at that data, build it into that visit, while that's important too, you can build a care pathway that maybe involves a nurse or other parts of the care team who can help manage this data and action on it and then really make the most of it so that patients get the benefit from it, but without over overwhelming the doctor or the clinical care team.

Dr Martin: That's fantastic. I'm really glad you brought up that example of program building and highlighting what Dr Connolly's done in Ireland. I'm definitely excited and inspired by the efforts there. And there are parallel efforts in the US here through our American Heart Association health tech network to scale up cardiac rehab access using technology. I think this actually may be a good chance to pivot to conversations around equity and the digital divide because in addition to augmenting existing cardiac rehab programs for people that can come into the hospital, then to be able to augment the interactions with those participants at home or to shift some of their cardiac rehab sessions to the home there's this relating concept of health equity because so few people, the minority of people that qualify for cardiac rehab participating in cardiac rehab, whether you're in the United States or Ireland or elsewhere. And then there are these long waiting lists to get into cardiac rehab, which really speaks to the need to be more efficient and to scale the ability to deliver cardiac rehab to more interested people.

And I wonder if we could have some conversation around health equity, how you're thinking about that because there's the concern that the technologies could be accessible only to those that have the resources to personally purchase technologies. And certainly, that's a real concern that needs to be addressed. But then, on the other hand, there's the potential to reach more people, to meet them where they are with the technology and actually to improve equity. And I think it's a pretty big topic and certainly, in the years ahead I think it's going to become increasingly part of the conversations around technology as it relates to medical care. I wanted to learn from you on this topic what your latest thinking is around the digital divide and health equity.

Dr Parakh: It's wonderful. We have a whole team focused on health equity, and I actually really like how you frame that. Let's start with how you started, which is cardiac rehab. It takes a certain level of privilege to drop everything and come to a facility three times a week for several hours to exercise and be part of this rehabilitation program so you can get back on your feet. And many times, either our patients or their caregivers who transport them, et cetera, have jobs they cannot get away from, transportation challenges and so on and so forth.

And the folks that do turn up tend to be relatively more privileged than those who are not able to make it. Really being able to deliver programs like cardiac rehab in the home through technology that's provided by either an insurance company or in the UK by the NHS is one way to level the playing field a little bit. We have decades of research that show the benefits of cardiac rehab, but it's only reaching a fraction of the people that are eligible for it. This way you can increase access and make it more equitable. I think there's a huge potential of technology. On the other an angle, I do agree that there is concern that technology might only be available to those that can afford it.

Speaking of Fitbit, for example, there's a whole host of devices. Many of our features are available on every single device, starting from relatively low-cost activity trackers all the way out to the fancy smart watches that we make. But in addition, we work closely with partners, insurance companies in the US, folks like NHS in the UK and all around the world to try and provide devices at either discounted or free to patients as part of clinical programs and other health and wellness programs so that we can reach the people who need it the most. The last thing is, like you said, as we move to the future, it's really important to build research and explore these areas further. The Fitbit Health Equity Research Initiative awarded grants to a number of folks who are looking at everything from postpartum care for rural black women to sleep in transgender youth to diabetes progression in Latino adults using the tools that Fitbit has to see if they can address some of the challenges around health equity.

Dr Martin: Thanks, Kapil, it's really helpful the way you're thinking about it. And you make a good point around coverage through insurance companies and the collaboration between the technology companies, the manufacturers, and insurance companies. We had a recent experience in Maryland where we were able to successfully collaborate between our team at Hopkins and HA and other stakeholders to pass legislation at a state level to provide access to blood pressure monitors for the Medicaid population. And I think it's not too unreasonable to think that years down the line, as the evidence builds, as you say, there could be legislation that actually covers wearables for the Medicaid population, and that becomes more of a commonly covered benefit for insurance across the board.

Dr Parakh: First of all, congratulations, that's a fantastic initiative. I think in terms of the evidence base, I think people don't recognize the level of rich evidence that currently exists. Just for context, there are about 1500 published studies that reference Fitbit so far. There's an independent meta-analysis that was done last year that looked at physical activity programs, whether they included Fitbit or not. And this meta-analysis looked at 37 randomized control trials and found that trials that in included a Fitbit as part of the intervention versus those that did not tend to have more movement, more steps in a day, more activity, as well as greater weight loss. And these are surrogate measures and eventually translate into bigger health benefits, but that's a pretty rich evidence space to begin with. And I think it's one of those things where folks don't quite realize that the extent to which these devices have already been studied and the amount of evidence that exists right now.

Dr Martin: That's a great point. That's a great point. It's been such a fast-moving field and we know that there's such a lag in evidence to guideline adoption to clinical practice implementation that can take over a decade that there is that when the field of technology and evidence is moving so fast, there could be that big lag, which I think as we get towards the close of our conversation, that may be a good segue to talk about. I hope that our conversation today is helpful to clinicians having a better understanding of where we are, but if they're interested in learning more, I wonder what other high-yield resources you may want to point people to, any call to actions that you have that would be helpful to our listeners.

Dr Parakh: I think that's exactly right. There is this gap between evidence and action, and we've built up the evidence base. There are FDA clearances, et cetera, a certain level of rigor that's gone into these devices. And yet many clinicians on the front lines don't know how to use them or what to do. It's something I'm personally interested in working on. If you search provider and Fitbit, you'll find a provider-facing page on Fitbit that we help launch and collaboration with the American Heart Association and the American Academy of Sleep Medicine that lists out some of the features and what the guidelines recommend and how you connect with the different features to the different guidelines. But really there's a big need for education, for CME and just other venues around wearables. People come to us in clinic with data from wearables and from other online research et cetera, and it's a sign that they're engaged in their health and that they want to do more to help themselves.

And it's an opportunity for clinicians to take that enthusiasm and channel it in an appropriate way. Somebody with hypertension may need more movement or more mental health support in terms of mindfulness or deep breathing relaxation to bring down their blood pressure. But if a clinician isn't aware of all of this, they may just ignore that data and it's a lost opportunity to build up the patient relationship and acknowledge a patient's work in what they're doing and guide them in the right way. But even more, it's a lost opportunity to help improve their health in a meaningful way. I do think there's a ton of opportunity around building more resources to help clinicians make the most of wearables.

Dr Martin: Thanks, Kapil. I've really enjoyed this conversation. I'd like to thank you so much for your leadership in this area, for helping teach me and everyone that's tuning in, and clearly we're going to need as many clinicians to be engaged in shaping the way that wearables enter into clinical practice. And we've already come a long way in a short amount of time, but there are huge opportunities ahead from fitness to sleep to AFib and so much more beyond that. But those have been some leading areas. I would be remiss if I didn't pull in a very concise quote you had on LinkedIn recently around fitness. You've looked at a lot of the evidence here and you said, "It all comes down to move more, huff and puff sometimes." I just love that.

Dr Parakh: My six-word condensation of 700 pages.

Dr Martin: And wearables can help measure that objectively between clinical visits. And I really am looking forward to hearing more feedback from listeners to continuing this conversation, the future. I really wanted to thank you, Kapil, for your time in joining us today on the Cardio Care Now podcast.

Dr Parakh: Thank you, Seth. It's been wonderful and it's an exciting time and the future's even more bright, so lots of work to be done, no question. But Fitbit, for example, just launched this continuous EDA sensor. It's a passive sensor that looks at the sweatiness of the skin. It's a measure of stress. And my thesis, when I was doing my PhD in epidemiology, was around depression and heart disease. And the only way I could understand what was happening with patients and subjects was either you had to ask them a questionnaire or you had to interview them by psychologist or a psychiatrist. And now we have a passive way to look at somebody's stress levels and mental health, which is super exciting. And I think it'll be really interesting how this gets incorporated into clinical care and what use cases come out of this. I'm sure we'll have more conversations in the future and it's a really exciting time in this field. Thanks, Seth.

Dr Martin: For sure. I'm looking forward to circling back on that, that's super exciting. Thanks, Kapil.

Dr Parakh: Thank you.

For more cardiology content, visit our website