Digital health is all the rage in Silicon Valley, especially as consumer technology’s biggest players are getting more interested in tracking health and fitness. But for companies like iRhythm, the health side of the equation comes first.
Housed in tech-savvy San Francisco, iRhythm embodies what it means to be a hard-core medical firm living in a geek paradise. The company shares a building with Zynga, maker of popular Facebook and mobile games such as Farmville, where the common spaces are filled with foosball tables, arcade games and hammocks. iRhythm’s focus isn’t nearly as whimsical as its surroundings suggest, but the environment does help bring in prospective talent and keeps energy high.
"There is a great pool of healthcare and technology talent in San Francisco and it’s also a great place to work, which helps in recruiting," CEO Kevin King told MassDevice.com during an interview at the company’s headquarters. "I worked down in Santa Clara for many years, and I actually think that San Francisco is probably a tad bit more entrepreneurial. I think the people here are bold in their ideas. I think people here are willing to take risks. This is consistent with the values that we have at iRhythm."
Founded in 2006, the company boasts an uber-modern wearable heart monitor backed by sophisticated algorithms and deep mines of clinical data designed to detect and diagnose abnormal heart rhythms.
The palm-sized Zio patch can be worn for up to 2 weeks, surviving showers, workouts and just about any conditions outside of full immersion in water in order to capture data about a patient’s heart beat. But iRhythm isn’t betting on the device; the real special sauce is in the analysis and the ability to collect data on every heart beat for extended periods of time. The Zio patch is a big boost over standard Holter monitors, but iRhythm’s giving them away for free. The real play is in the reports, and the company’s got big plans for the future.
"The Zio patch itself is a data collection device. It’s been designed to be elegantly simple from a patient compliance standpoint. It’s applied to your chest, and then you just forget about it. That’s been a big deal. It’s taken a long time for us to master the ability to do that," King told us. "Where we’re taking the business now is really along the lines of using, not information at an individual patient level, but using aggregate patient information. You hear this term, ‘Big Data.’ This is all about taking the aggregate beat-to-beat information that we’ve collected and combining that with other personal health information to help physicians and health plans make more informed decisions."
In an in-depth interview with MassDevice.com, King told us all about how iRhythm was born and about its big plans for Big Data:
MassDevice.com: What’s it like being a healthcare company in a hub of video-game and tech startups in San Francisco?
Kevin King: There’s a lot of healthcare in San Francisco, and I think working inside of this building is kind of a microcosm of the technology focus and innovation you see in a larger part of San Francisco. Our business is founded on both deep science and deep data. I think you find that that’s what our people are about. That’s what we get excited about. We get excited about how that science and that data can help transform how patients are diagnosed and treated. Overall, it’s a pretty cool place to be.
MassDevice.com: Tell me about your background. How did you get into medical technology?
King: I’ve been involved in healthcare for about 30 years. I’ve been both a public and private company CEO for the last 12 years. Prior to that, I worked for 2 large companies. One was Hewlett-Packard, and the other was GE Healthcare (NYSE:GE), all in the healthcare space.
I was doing some work at Stanford University in their translational medicine program called SPARK, and I got introduced to iRhythm through a colleague there. I was really, really moved by the vision that we have for improving diagnosis and treatment of patients with arrhythmias, and also by the possibility of really changing healthcare in a big way through the use of data – beyond individualized patient treatments and diagnosis. I sort of got swept off my feet at that point.
MassDevice.com: Was there anything in particular that really drew you to iRhythm?
King: Yeah. I think the foundation of what we have here – being both clinically proven as well as analytically proven – is quite unique. There are lot of promising healthcare technologies, but over the past several years we’ve been really, really able to prove this out in scientific literature that the Zio service is affecting how physicians make medical decisions relative to patients with arrhythmias or symptoms of arrhythmias. The Zio patch itself is a product or an offering that patients like to wear, if you will. This prompts strong compliance, which enables us to capture the vast amounts of heart beat data that our proprietary algorithms then curate to provide doctors with meaningful clinical information regarding a patient’s arrhythmia status.
Most of our competing technologies have some drawback associated with them. For example, they’re generally cumbersome to use. We’ve taken that challenge away, so we have very high patient satisfaction scores overall.
MassDevice.com: You joined iRhythm about 2 years ago. What do you think are some of the most significant changes in the company since then?
King: I think there are 3. One is the establishment of that clinical evidence that I was describing. We now have multiple peer-reviewed publications that reinforce that the Zio patch is the new standard of care for how patients with arrhythmia-like symptoms are diagnosed and how they’re treated. This sort of change in medical management is a really, really big deal.
I think the 2nd big contribution is that we have made the technology available to patients through increased payer coverage for the Zio service for diagnosing patients with suspected arrhythmias, further establishing it as the new gold standard. We have pretty close to 200 million covered lives in the U.S. now. That’s a big, big gain, and that’s been done largely through that clinical evidence as well as very, very strong physician advocates and users of our products as well as demonstrated high demand.
The Zio Service has been used with about 250,000 patients by physicians at over 800 institutions since it was commercially introduced in 2011, and those are very big numbers overall.
MassDevice.com: How was the company itself born? Where did the technology come from?
King: The founder of the company and technology was an electrophysiologist at Stanford University who felt there was an unmet need in physicians’ ability to effectively rule in or rule out cardiac arrhythmias in patients suspected of having the condition. He saw that traditional approaches to cardiac ambulatory monitoring were limited in their ability to capture arrhythmias and were difficult for patients to comply with, which further limited their diagnostic yield. This unmet need can be broken into 3 categories. One is that most of the currently available technology is cumbersome.
Another one is that current technology generally lacks completeness of data. In other words, because they do not monitor continuously or for a long enough period of time, arrhythmias – which are often infrequent or without symptoms – can be missed.
The 3rd is that some of the current technologies are relatively expensive, particularly when used in combination with 1 another. So what frequently happens is that current approaches have a poor diagnostic yield, which produces false negative results, and then you have to do another test, which can become quite costly.
Any combination of those 3, he felt was a very strong unmet need, and hence the formation of iRhythm. The Zio patch itself is all about removing that cumbersome barrier, and then it’s really related to the service. It’s a fairly cost effective service that’s now proven to be a better standard than what’s out there. It does it in a way that it collects information continuously during a patient’s wear time.
A patch is applied on a patient’s chest, and it is worn continuously for up to 14 days. You eat, you sleep, you work, you exercise, you shower. The only thing you can’t do with it is swim. During that wear time, we are collecting every single heartbeat of information, which averages to be about 2 million heartbeats during that 2-week period. Then we analyze that data.
It just continuously collects information and the patient doesn’t have to do anything with it. They don’t have to charge batteries. They don’t have to change it. Simultaneous with that, we are collecting patient information, so we’re collecting information about patient symptoms, and we’re correlating those symptoms with the ECG patterns, so when did you have a symptom, what were you doing when you had the symptom, how long did it last, data that helps to inform the interpretation of the algorithm. It helps us to use it to help physicians make medical decisions.
MassDevice.com: How are hospitals charged for this?
King: We don’t sell the devices, we sell the report. It’s a data business, so that the patch is consigned to the provider. Provider writes a prescription, and when the report is generated, that’s essentially when we get paid for it.
Unlike companies that sell the medical devices, watches or Holter monitors or things like that, and then you’d provide the service, there is no capital equipment. There aren’t any software licenses, maintenance licenses. There’s no labor associated with doing this. Everything is done behind the scenes by iRhythm as a service. The physician merely just prescribes it and then gets a report generated for them, a cloud-generated report.
We generally sell to large health systems, and so we have contracts to provide the service. The physician would get a report through our secure site. We tell them about the number of triggered events, where patients feel symptoms and they press a button on the patch, which helps correlate symptoms to heart rhythm.
Then we tell physicians everything about their patients’ types of arrhythmias that they’ve had during the period of time for each and every day – what were the events, where were they, what were they doing – and generate a draft report called, "Findings" or "Preliminary findings" that we list, and then the physician writes in their own final interpretation for the report.
By way of comparison, the legacy gold standard standard has been Holter monitor. It’s a device and it records information for about a day, and then that information is viewed on a screen by a technician scrolling through who sort of cuts and pastes essentially pieces of paper, if you will, into a report.
You end up having 2 million heartbeats to analyze, we use algorithms and tools, software tools to analyze the data and produce the report. Then, we have a clinical team that oversees the final interpretation of that. We do this in a fairly rapid way, and we do this measured in minutes versus measured in hours.
We’ve worked with what you would call "safety net hospitals," where patients don’t have health insurance or have difficulty accessing care. In some cases, the time to get a Holter monitor is measured in weeks or months, so if you show up in the emergency department with symptoms and there isn’t a Holter monitor available because they don’t have capital dollars to buy them, they send the patient off and have the patient come back. Arrhythmias by themselves, if left untreated, could be very dangerous and could lead to sudden death, or they could lead to stroke, and so the time to diagnosis is very important.
With our model where our customers are not buying the equipment, we’re consigning it to them, so there’s no out-of-pocket expense for them. There isn’t that concern about capital availability, so in those safety net hospitals, people have plenty of accessibility to long-term continuous monitoring, the Zio service.
MassDevice.com: You currently have trained and certified technicians looking over these reports. Will there always be human eyes overseeing this analysis or can software take that over 1 day?
King: I think there will always be human eyes to some extent, yes.
MassDevice.com: How does this technology help hospitals cut costs?
King: Most of the savings are in medical decision making, according to our studies. There was a study done by the Beth Israel Deaconess Hospital in Boston that looked at a group of patients that had what’s called paroxysmal AF. This is atrial fibrillation that occurs on an irregular pattern, and patients wore both the Zio patch and the Holter monitor simultaneously. Holter monitor was worn for 24 hours. The Zio patch was worn for an average of 10.8 days, and we detected more arrhythmias than the Holter monitor. We also changed the type of arrhythmias that were seen relative to the Holter monitor, so there was a marked improvement in diagnosing the type of arrhythmia.
It turned out that about 30% of the time, a physician changed their medical decision on how they would treat the patient relative to the Zio patch versus a Holter monitor. That’s really a big deal in terms of clinical management, clinical performance.
MassDevice.com: Tell me a little bit about the technology itself. What’s the special sauce that makes this work?
King: I think the special sauce that makes the Zio service work really is that curation capability on the back end. The Zio patch itself is a data collection device. It’s been designed to be elegantly simple from a patient compliance standpoint. It’s applied to your chest, and then you just forget about it. That’s been a big deal. It’s taken a long time for us to master the ability to do that. We’ve done hundreds of thousands of these, and we also have patient satisfaction scores that demonstrate how well we do. Average patient satisfaction scores on a scale of 1 to 5, show us somewhere between 4.5 to 4.8.
That by itself on the patch is the secret sauce, but the back-end, when you have all of this data and you now need to curate it, analyze it, and you have to do it in a way where physicians have a high degree of confidence in what you’re doing, that’s our secret sauce; the algorithms and the tools that we use for curating the data is more significant than anything that we do.
When physicians do have questions about a report, and we arrive at an answer, about 99.5% of the time physicians have agreed with what the Zio service has provided. That high degree of substantiation of our algorithms and reporting and clinical reporting is what has built physician confidence in our service, and that’s really been helping us to drive growth.
MassDevice.com: What’s iRhythm working on next?
King: Where we’re taking the business now is really along the lines of using, not information at an individual patient level, but using aggregate patient information. You hear this term, "Big Data." This is all about taking the aggregate beat-to-beat information that we’ve collected and combining that with other personal health information to help physicians and health plans make more informed decisions.
It’s for providers and it’s for health plans. Those are the 2 target audiences, so the people who care for patients and the people who pay for the care of patients. The information that we have was previously unobtainable, and we’re doing it at a very, very granular level. Today on the data side we do reporting that allows a particular entity to compare themselves to a regional average, a national average or a local average. Hospital "A" can compare their patients’ satisfaction scores, their diagnostic yields, their prescriber patterns and their patient indications against those comparative benchmarks.
In the future, what we’re doing is looking for ways to measure things that either consume resources – such as emergency department visits, hospital readmissions for example – or things that put patients at a level of risk that could potentially be adverse to them.
So what are those of types of things? Atrial fibrillation is a well-known rhythm disorder, but up until now, we haven’t had the ability to look at this very, very fine granular level. For example, a single 45-minute episode of atrial fibrillation is very different than, say, 45 1-minute episodes of atrial fibrillation, and how patients should be treated should be different.
Our ability to glean that information from our database, compare hundreds of thousands of patients that have atrial fibrillation and say which ones have high AF burden that is of a long duration versus high AF burden of multiple short duration ones, and how they might be treated, is a pretty significant approach.
Similarly, some patients that get atrial fibrillation ablations, EP procedures, are asymptomatic, but in follow-up when they wear a Zio patch, we find atrial fibrillation. The real question is, what do you do? We think the patient’s been cured by the ablation procedure, because they’re asymptomatic, but we still see atrial fibrillation. The ability to follow these patients over a long period of time and figure out which treatments are best for them is another example of how the Zio data will be sort of a Big Data-type application.
MassDevice.com: What’s the next milestone in developing this Big Data pipeline?
King: Right now, on an aggregate level for institutions, we’re providing these reports that I’d mentioned to you earlier about patient satisfaction and diagnostic yield. That’s the information that we collect today. We’re looking to collect additional sources of information about the patient, and we think that mobile tools and things where patients or physicians can provide information to us about the patient will enhance that. That’s the kind of thing that we’re working on for the future.
MassDevice.com: Mobile tools that are patient-facing?
King: Patient-facing tools, yes. Right now, the patients provide information to us in a written booklet when they have their symptoms – what were those symptoms, when did they occur, et cetera. Making these more of an interactive mobile tool will help us to obtain that information in a digital way and use that as part of our decision-making. We are also looking at integrating our data with electronic medical health records at large institutions to better inform patient care. Customers want us to do that, and we want to do that for the obvious reasons. We see great opportunities in digital health.
MassDevice.com: What’s the endgame for iRhythm?
King: Our business is growing so rapidly that we are just focused right now on growth. I’m really not worried about an exit of some sort. It will be whatever it is, but right now we’re not worried about it. Our investors aren’t worried about it. I think probably we have multiple options, if you will, for the company in the future, but right now it’s all about delivering value and delivering growth.
MassDevice.com: I saw that you guys had some recent fundraising? How has your fundraising experience been?
King: We raised $17 million in an oversubscribed round led by Novo A/S, so we had more interest than we had need for cash. We brought in Novo A/S as a new investor. They are a very, very active investor in the life science field and a global investor, so a fund that extends beyond the borders of the U.S.
I think the reason why it was good and oversubscribed is the validation that we present to people about what we’re doing. We’ve got hundreds of thousands of patients using our service, proven technology, a high degree of coverage by health plans, multiple peer-review publications all reaffirming what we’re doing and a high degree of patient satisfaction.
Those things are real and they’re tangible. They’re not sort of fuzzy things that might occur in the future. We’ve actually done it, and we continue to do it every day.
MassDevice.com: Do you think there’s anything unique about being in digital health in terms of the kinds of funders you have, the kinds of experiences you have?
King: Well for us, I think the definition of digital health is not only that the information that we collect is digital information, meaning that it can be aggregated, analyzed, sorted, distributed and things of that nature. I think what we are is an information services company – you might say we’re a digital healthcare information services company. For us, our service is having an impact on the way patients are managed relative to their physicians’ medical decisions. That’s a really, really big deal.
I think in the past or in some other cases, medical device companies have sold their medical device to an end user who was then responsible for providing that service, and I’m not so sure there are enough dollars left in the healthcare system for that to continue. With the Zio service, there are no out-of-pocket expenses for physician and institution users. There’s no capital equipment. There’s no software. There’s no maintenance. We are literally providing a service that doesn’t have a lot of expenses associated with it for users. I think that really is transformational when you think about how to diagnose tens of millions of patients.
The number of patients with arrhythmia symptoms in the U.S. is estimated to be about 25 million, and there are 8 million people diagnosed with arrhythmias, and it’s estimated that about 10 million more people will enter into this arrhythmia pool over the next 10 years as we all age. I think our service model is what really is a big differentiator for us in this digital healthcare environment that we’re discussing.
We are very different from most other companies in the digital health space. A lot of other digital healthcare approaches are starting from a different endpoint. We’re starting from science and data and things that are proven to make medical decisions. Another approach is to start more with consumers and data enthusiasts who want measurements. There’s nothing wrong with that, but it’s a different approach, and I think for us, I’m happy where we are. I think we’re starting with a stronger foundation than just collecting information and then wondering what will you do with it.
MassDevice.com: Does it make any difference in your outlook now that companies like Apple, Google or Samsung are getting more involved with digital health?
King: I think it’s good news. What it’s showing is that there’s an appetite or there’s a mindset for actually changing how healthcare is delivered. That’s a good thing. When you get big companies like Apple or Google or whomever interested in healthcare, that could only mean good things.
MassDevice.com: Could you see the Zio information being listed in Apple’s HealthKit app?
King: Sure, absolutely. Or have Apple’s HealthKit data in Zio.
MassDevice.com: What are some of the big challenges you’re going to have to overcome going forward?
King: The healthcare environment continues to evolve and change, so staying abreast of what healthcare plans and healthcare providers need and want and how they create value is a challenge. You have to pay really, really close attention to customers, both payers and providers, and that is an area where we believe we excel. Additionally, we’ve got a really terrific employee base here at iRhythm and we haven’t had challenges recruiting top talent, but one can imagine that as we continue to grow and expand that we need to continue to focus on recruiting and hiring the best people.
We’ve got a great group of people and long tenure, too, with a lot of our employees. It’s worked out quite well.
MassDevice.com: Are they largely traditional medical technology people, or are you pulling just from the San Francisco tech programming world?
King: It’s interesting. It’s actually a blend. It really is. We’ve got some people that have long history in the ambulatory cardiac monitoring fields. We have just as many people who have come from a variety of data businesses, services businesses, healthcare businesses, financial services businesses, and we have them at all different points along their career. It really is a good representation of a very diverse workforce in many ways. I think that’s helped us to stay nimble and to think differently.
We’re not all of the same background, and so we’re able to challenge one another based upon the experiences that we have.
MassDevice.com: How does the medical device tax affect iRhythm?
King: Well look, for any young company in this space the medical device tax takes operating dollars away and forces the company to pay the government when it could be hiring employees and directing that money towards further innovation. I understand what the medical device tax is intended to do and how it’s meant to fund the cost of care and so forth. It’s just a tough thing to do. In a nutshell, it forces companies to do less than what they normally could do. We were commercial in the 4th quarter of 2011, so we started paying the medical device tax last year.
MassDevice.com: What else is on the horizon for iRhythm?
King: Well, we talked a little bit about the future for iRhythm when we talked about it in the context of data. What we didn’t talk about was additional clinical indications or settings for the use of the Zio service. There are certainly the post-ablation procedure patients, and we’re also increasingly looking at patients in the emergency department and in neurology, as a means of diagnosis of cryptogenic stroke; we’re looking at patients that have what’s called "the silent AF" – people who are at high risk of having atrial fibrillation are great candidates for being evaluated with the Zio. The safety net hospitals that I was describing is another use.
The theme there really is an expansion of the indication of use for our product across multiple care settings, then ultimately to primary care. Primary care physicians are probably a pretty good gatekeeper to expensive therapies and procedures and so forth, and ultimately, that’s where I think the product will get its broadest use, because a lot of patients are seen by primary care physicians but they don’t have access to technology, and we actually can bring the Zio service to the primary care physicians.
MassDevice.com: Does this mean you’ll need a bunch of new FDA clearances?
King: No, we have very broad indication now. We’re talking about building the clinical evidence to allow physicians to feel confident that these are tools that can address populations. At the Heart Rhythm Society meeting here in San Francisco, about a month ago, we presented papers on these subjects, safety net hospitals, silent AF, emergency department, neurology. These were abstracts that were presented that will eventually turn into peer review publications. They’re leading indicators of where the technology will be deployed inside of health systems.
MassDevice.com: It sounds like the patch itself is not going to evolve much as you move forward.
King: The Zio Patch is a really good product. It’s a really good product. We’ve invested a lot of money in it. It will evolve over time, but most of the value we have is really in the back end of the data and in the analytics of the information.
This interview was for clarity and continuity.