Speaker: Rick Strobridge, CRF Health
I’m going to bring up Rick Strobridge, a serial entrepreneur, three companies that he’s set up, and he’s now with us at CRF Health. So we’re delighted to have him. I think he’s been mucking around with video technology for most of his career, but was instrumental in getting it into the operating theatres. Rick now is our VP of Healthcare. I say that deliberately, because if I say the name of the company that you used to represent I get a dollar fine. But anybody who mentions that name can give me a dollar for the beer money later perhaps. What was the name of the company, can somebody tell me? Tricky tricky one. Okay, so Rick’s going to talk about best practices of using devices within clinical trials. So Rick.
Thanks Alex, and I think it’s a part of your diabolical plan to put the California guy with the plus-nine-hour time difference last on the schedule. So we’re going to do the best we can to get through it. But I was asked to talk a little bit about using connected devices in clinical trials. My disclosure for this talk is that I am not a clinical research person. I have been supporting clinical research for about ten years in using connected devices. I’m a technologist, so my background is in information technology and integrated messaging, voice, video, and data communications. So I’m going to talk a little bit about my experiences from that standpoint and then maybe get some feedback from you. And so I can stay awake, I’m just going to take this out, so I can talk to the slides.
So I thought I’d start with a couple of quotes from Margaret Mead. And I was thinking about the kind of group that I would be addressing here, and I think this quote really says a lot about you as a group. You are thoughtful. You’re committed, I’ve seen that from the questions and the presentations today. And you guys are going to be the ones that change the world. And with the next thing, always remember that you’re absolutely unique, just like everyone else. I think another perfect one for clinical research.
So 46 billion internet of things devices by 2021. The data is not going to be an issue. The data, getting to the right data, I think is going to be the right issue. And so with all these connected devices, it’s even rise to a think called the medical selfie. So everyone has a medical selfie. Whether you know it or not, you’re out there, from even if it’s only from a social media standpoint or a GPS standpoint, everyone’s being tracked from their phones today down to lots of additional medical data if you allow it.
Anyone know what this is? Anyone watch Star Trek? Heard of Star Trek? This is the Star Trek Tricorder. The Tricorder XPRIZE, if you know the XPRIZE foundation, just awarded their prize for the ability—it was a two year, two and a half year program. It was a $10 million prize. They just announced the winner of that. The goal was to be able to diagnose 13 states of health in a human, 12 indications, 12 diseases, and the last, the 13th was the absence of disease. I talked to one of the judges that had given this award just a couple of weekends ago, got together with him, and he said he wished he had three more years to get there, because they really did the best they could, the winning team actually got $2 and a half million to continue their research. But they only had about eight or nine of those indications plus the absence of disease.
So this is where I started in clinical research. We set out to build the first Bluetooth-connected blood glucose meter. So this is a finger stick meter, it’s been used in many many trials. This meter has been used in about 75 countries since 2007 when we got it going. And we now have a couple of additional models. But what I learned from that was a lot about best practices. In addition, we’re starting to collect a lot of additional data, so I’m sort of the connected devices guy but it’s also connected platforms, not just connected devices. So we’re grabbing not only biometric data, we’ve got a kit out here that we can demo for you. But also we’re starting to collect genomic data, clinical data, claims data, behavioural health data, and also socioeconomic data. As you can see from the cartoon, “Honey go and talk to him, he just found out he’s a placebo.”
So Mark Twain: “It’s not what you don’t know, it’s what you know that ain’t so.” And so again, sorting out from that data good from bad is really what we want to try and do with what we do. And we’re going to do that with five best practices and these are the five. Reliability of the collection of that data, the usability of the device itself, its fitness for purpose, the efficiency of collecting that data, that goes back to the cost-benefit analysis that we talked about in one of the earlier sessions. And then the regulatory and government aspects of that surrounding regulatory clearances, data privacy, data security, intrusion detection, that kind of thing.
So the question really becomes to cloud or not to cloud. And what I mean by that is we can connect directly to these wireless devices and then send it directly into the database, into the EDC, or we can actually collect it from a cloud-based source, such as an electronic health records system. And so some of the implications of that are really going to become known as we go through some of these best practices.
So I went out and did a little survey of what’s going on out in the industry, and there are a lot of different indications that are being looked at. Here’s from Apple, this is a Parkinson’s disease. We have a single EDCG that we can collect ECGs remotely. There’s a clinical trial going on with this particular device, $99 is what this device costs. You stick it on the back of your cell phone, you can collect your ECG anytime you want. One of our employees actually had an aFib episode at work. We were on the way to the hospital, when we got to the hospital he had already taken his device out, taken his ECG, we walked into the emergency room, he gave the ECG to the ER doc, and immediately diagnosed aFib. So that’s the kind of power of these kinds of devices. I was talking to someone earlier about a 12-lead ECG, so if you really want to go 12-lead, this is another at home useable, it’s a chest strap device, it has leads around the chest strap. You put one additional lead under each arm, and one on your waistband, and it gives you Einthoven’s triangle and you end up with the 12-lead ECG that can be transmitted from home. So in the case of being able to collect this data, instead of driving two hours back to the site to do a five-minute ECG, very very powerful from the standpoint of clinical trials.
So the big guys have gotten involved, right. So Google’s health division is called Verily. Verily has announced a number of different things from both a health platform standpoint and a population health tools standpoint that are out there. They’ve also talked about some different devices. Here’s another Apple-based one for autism. Back to Verily, they’ve looked at a miniaturized CGM device as well as a contact lens CGM device. We've got the clinical research person—he’s not there, where is he—clinical research person for Dexcom who is working with Verily and Google on this project is here. So if you have any questions about remote collection of continuous glucose monitoring data we’ve done a lot of glucose monitoring trials as well. And a smart contact lens program, well we’ll see how that goes, and we maybe end up with those sensors collecting more than just glucose by the time that comes to market.
This is one for epilepsy and seizures using the Apple Watch. And then lots of others. Asthma, COPD, diabetes, concussion, and that kind of thing. This is I think the sleeper from a data collection standpoint. This Amazon Echo was the biggest selling device last Christmas in the US, I don't’ know how it’s doing here or whether it’s even been released. But I think interactive voice activation and response could be a powerful technology for this group. As well as Google Home, so you’ve got Amazon versus Google, not only are they competing with each other in self-driving cars, delivery via drone and that kind of thing, they’re also competing with each other in healthcare. And I think that the idea of being able to use these things to easily and naturally talk to patients and get responses from patients could really be a big change, especially given the amounts of devices that will be out there. I mean literally millions of these devices have already been purchased and installed in the US.
So reliability. First of the best practices, wireless versus wired, Bluetooth and the other kinds of wireless technology, Bluetooth low energy seems to be the winner right there. Battery life, battery access, price, and performance, you know, but this is the kind of thing that can happen if the device is not reliable and cannot gain regulatory clearance. So Scanadu was very promising in terms of multi-mode biometric data collection device, it collected body temperature, blood pressure, pulse oximetry. The problem was it wasn’t done accurately, and so the FDA told them that they couldn’t do that anymore.
One of the things that we’ve put together as CRF Healthcare, Alex, is a kit that does—and this has been deployed to hundreds of patients, so it’s not a huge population right now—but hundreds of patients that’ll collect ECG, blood pressure activity, pulse ox, spirometry, asthma inhaler, sensor data, as well as weight and some other things, along with doing video conferencing and surveys to the home. Our whole goal is to provide you all, if you want to use it, with the technology that will allow you to collect this data when it makes sense within your protocols without making the patient come back to the site.
Another project that we’ve worked on is a pulmonary hypertension device, it’s a pulmonary pressure sensor. This is actually an implantable, so we’re not just talking about connected devices and wearables, now we’re talking about actually implantable devices. So this device on the right is an implanted sensor, it goes in the pulmonary artery, when the reader device is put up to the chest, it reads the pulmonary blood pressure 20 times a second and then reports it back to the platform. And this device is a fast follower to a company called CardioMEMS, if you’re familiar with them, but in a much more compact kind of package.
Usability, next best practice. Ease of use, size, and weight, wearable, discreet, comfortable, continuous versus episodic, it goes on and on. But it really comes down to, does the person that you’re giving this device to have the capability of using it and using it on a regular basis, and using it to get you good results. Someone earlier today mentioned this new Verily watch, which is one of the things that I didn’t put on the specs for this is, do people want to wear it. You know, how many people in the room have an Apple watch? Not very many, okay. Fewer than I thought. This is a watch that will do activity. It really is a platform, so it will connect to these other devices that we’ve been talking about. It does gait analysis and lots of other things and was just announced by Verily, so you can see when these kinds of devices are readily available in a retail space, the kind of data that they’re going to be capable of generating.
Again, back to the continuous glucose monitoring, the Dexcom G4 device is a wireless device, in terms of transmitting the glucose data. Their next G5 and then we’ve also had a briefing on their G6 device. Their G5 which is on the market now, connects to an iPhone, soon an Android device, to be able to collect that data just naturally. So I’ve got my Android device in my pocket, if I had one of these G5 devices on I would be collecting that data automatically without doing anything myself, back to the platform.
So some of the things that we’re seeing in what we call at CRF the healthcare market, which is the non-clinical research part that I run of the company is these things called patient engagement indices. We’ve heard a lot today about patient engagement and how do you get them to do the right things. How do you configure these systems. And that really goes back to the usability of the device as well as finding the patient voice in clinical research, and I’m preaching to the choir in this setting.
So the third best practice, fit for purpose. It’s one thing to have a device and to have a device that’s useable. But it’s another thing for that device to be fit for purpose for whatever the protocol is that you’re deploying. And that has to do with accuracy, validation, the cyber security, if that’s necessary, and the connectivity. Is it the correct data sensor and are the technology errors and omissions, if any, going to really be an effective deterrent in terms of your trial. Obviously risk of injury. Since I became part of CRF, which was about nine months ago, the acquisition of the company happened in August, this has been my number one request is for actigraphy devices. So we went out and we did a survey of about 70 different actigraphy devices, everything from the Fitbits on the bottom and the whole line of Fitbits up to the higher end devices. So we’re talking about devices that cost $59 or less. And this is the market leader, is Fitbit, and you can get these kinds of devices much lower cost. You can easily deploy them if you have a secondary endpoint that’s looking at activity—step count, that kind of thing. And we’ve used these in pulmonary hypertension trials—up to $3000 Actiwatch or even more expensive polysomnography-type device and those kinds of devices. So we’ve done that whole survey, and we’ve done a matrix and looked at the price, the battery life, the wearability. Somebody this morning talked about putting several of these devices on a patient. It can be very detrimental and also very expensive from a trial standpoint to use these devices and to keep them operational, charged, that kind of thing. So fit for purpose.
Propeller health has an inhaler sensor that probably most of you have heard about. We’ve got some additional sensors that we can talk to you about. Connected spirometry devices are becoming more and more common. You almost can’t open your newsfeeds in the morning on these kinds of devices without seeing some new kind of device that’s out there that has a Bluetooth chip in it. I talked to my OEM manufacturers. A lot of the new devices today are coming out of South Korea, Japan, Israel, places like that. I had my Korean manufacturer in week before last in San Diego, and he said, we’re not even doing a device that doesn’t have connectivity. So pretty much everything that you can see that will be coming out in terms of biometric data collection from simple things like Fitbits and glucometers up through very sophisticated microassay-type systems are all going to have that kind of connectivity. And then in having that connectivity, with the kinds of usability and fitness for purpose that we’re talking about, they can also have a use at home. So keep that in mind.
So these are a couple of other spirometry devices that we’re working on. The two on the left are Cohero devices, the one on the right was just cleared by the FDA, it’s by a company called Monitored Therapeutics and it’s a GoSpiro device. And then connected medication sensors, I’m getting a lot of questions about these kinds of devices. There are many different ones on the market. This particular one that we have integrated with our platform is an 8-chamber medication sensor that’s cloud connected, so I can from here dispense medications in San Diego over an internet connection, and I can dispense medications from each of those eight chambers into one single chamber, give it to the patient and look at their compliance, because the device also has a camera on it, and so it can actually see them taking the medication.
Fourth best practice, efficiency. Does the device fit within the workflow. We work with a lot of clinicians, as you do. They don’t want to do something that is not out of their natural workflow. And so what we try to do is fit that device as we’re looking at different protocols and different healthcare related remote patient monitoring implementations, we try to look at the workflow, and really in the US and I think in many of the other countries, the center of the clinical workflow is the electronic health record. And so we’re looking at how we can get data in and out of the electronic health record, how can we get the data that we’re connecting on these devices into the electronic health record so they don’t have to log into multiple systems.
Accessibility, standard APIs and Continua, and an ecosystem that allows for a single API. A couple of examples are coming up on that, and I’ll show you, and I think probably many of you have heard of them. And then a standard data set. So if you’ve got a thousand connected devices and 46 billion internet of things devices out there, you obviously don’t want to write software interface to each one of those, you want to write a standard software interface that will apply to all of those different devices, and there are some very powerful standard interfaces out there, the first of which is a thing called ??Personal Connected Health Alliance. This used to be Continua. Continua is now a part of the HIMSS organization. They’re working very hard. This white paper was just released at the end of April. Personal Connected Health Alliance is the fundamentals of data exchange for these kinds of devices. So if your technology team is not looking at the ability to have standard interchange of devices and standard connectivity to these devices without a lot of additional work, they probably should be. So some of the things that Continua are doing, without getting into a lot of detail on this, is they’re really trying to say, okay we’ve got 20, or even many more, connected data devices that are out there. Say glucometers for instance. We want to look at a standard interface, so if we write that interface on our tablet or on our smartwatch or on our iPad, if we write that interface, we want to be able to reliably connect to all of those kinds of devices, all of those same devices. And so we’re having a standard connected data set, as well as being able to pull together medical devices—so regulated medical devices as well as wellness devices and the data that’s coming off these wellness devices, which can be everything from steps, sleep patterns, those kinds of things—in one standard interface.
So one of the cloud-based systems that has a single interface is Validic. Validic relies on the fact—in most cases, not all cases—but relies on the fact that this data resides on the cloud already. And then it gives you a standard interface to go out and grab that data. And they have hundreds of devices. The other ecosystem that’s out there is by Qualcomm, 2net ecosystem. They started with a plug-in-the-wall type system, with an ecosystem device that connects via Bluetooth to that plug-in-the-wall device in the patient’s home, transmits the data and then into a single database.
You can think about now, why would Qualcomm be doing this, right. Most every one of our cell phones in our pockets right now has some form of Qualcomm chip set in it, so their new Snapdragon chip, which goes into every Samsung phone, is going to ultimately have all of this built in. So all of us and all of our constituents and patients and participants are all going to be walking around with a device that automatically has connectivity, so there really will be no issue. As long as they have the devices that collect the biometric readings, they’re good.
Medical grade integration solutions, ReDoc is an interesting one in terms of ecosystem in that they have written interfaces to dozens and dozens of electronic health record systems. So you write one interface to ReDocs, you can get data from all the different electronic health record systems. So similar, but with electronic health records and not devices, to Validic and Qualcomm. And then the HIEs, the health information exchanges in San Diego—this is ours, the San Diego Health Connect—connects millions of patients in the San Diego area by connecting to their providers’ electronic health records systems. So all this data that’s out there from HbA1C to ECG to medical imaging, all of that data is now going to be much easier to access.
So number five, and then we’ll wrap up. Country clearances from a regulatory standpoint, it’s one thing to say jeez this is a really cool device, like the Google watch or whatever. But I can’t use it in Japan, I can’t use it in Russia, I can’t use it in Estonia, so looking at the regulatory clearances of these different devices, and because the wireless part is a new device, they may or may not have the regulatory clearances that they need for your trials. Radio licenses, now we’re not just talking about a medical device, we’re talking about a wireless medical device. Do they have the FCC in the US and other international radio licenses to be able to deploy, even if they have regulatory clearances. Labelling, import/export, all these things that you folks deal with on a daily basis in terms of clinical trial access if they don’t have regulatory, IRB documents and those kinds of things, along with taxes and duties and import information.
So five best practices, I just am going to give you one example. This is the new EU medical directive, medical device regulation. Literally it just was issued a few days ago on May 5. And it is going to dramatically change how medical devices are regulated in the EU. Both medical devices and in vitro diagnostic devices. It goes into effect May of 2020 for the medical device reg, and 2022 for the in vitro diagnostic regulations. But it is going to dramatically change all of these devices and all of the new wireless devices that are coming out.
So in closing, I like this—does anybody know what Waze is? Waze, is that here in Europe? Yeah, Waze is a crowdsourced traffic kind of app. This is by the guy that founded Waze. “Fall in love with the problem, don’t fall in love with the solution.” Because the solution will be different tomorrow in many cases with this technology than it is today. So fall in love with the problem, think about the problem, and what got you here won’t get you there. So just keep that in mind as well.
So these are the best practices—reliability, usability, fitness for purpose, efficiency, and regulatory and government practices. And I’ll leave you with a couple of other ones. this sort of goes with the ones in the first. “The reasonable man adapts to the world, the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on all of us unreasonable people.”
And then this—anyone recognize this formula? No? This is the change formula. So D is dissatisfaction, times the vision of an alternative, times taking the first step, is greater than resistance to change. When that happens, change happens.
A fascinating presentation. Basically if it moves, he can track it. So the bit that I was most interested about was the the concept around remote dispensing. I’m not sure whether that’s terrifying or really really exciting, but nevertheless, really fascinating. So any questions for Rick at all? He does like Star Trek by the way. So any thoughts or questions, or have we thrown the San Diego guy to the wolves?
If I could just ask one. I think that’s really useful to see just how much is out there, and it’s super-exciting, and I suppose for some of us folks that are on the more clinical side, we’re very conservative and we like looking at the exciting stuff but we don’t necessarily know how to approach it. What’s your recommendation for a good first step for clinical teams who are really interested in the possibility of what some of this technology can do, and give us some amazing insight into the patient experience and give the patient more of a voice, but don’t necessarily know how to take those first steps. Any advice for kind of a nice gentle first step that could be taken.
I think a good first step is just use your imagination. When you’re designing protocols, and when you’re thinking about what kind of data you want to collect from these participants, use your imagination, because as you’ve seen, chances are someone is already collecting that data, and there are devices that are out there already collecting that data. And so that would be my thing. And you know, talk to some folks that have done it.
Thank you very much, Rick.
So thankfully we come to the end, I hear you whisper. But just a very very quick roundup. We’ve had some really fascinating presentations today. Paul from PatientsLikeMe talked about the Ice Bucket Challenge. I think the thing that struck me most about his presentation with the condition ALS was the lack of prescribe medicines for the people that are suffering there, and the real interest in the patients in that area to try things off label. And some of the examples he gave us I think were very eye opening. Paul here talked about the patient burden, what that really represents, and how it can be defined. And that was again interesting as a concept because quite simply if you wear a pair of glasses that could be considered as a patient burden. We then asked Michelle to talk about—and I think my key takeaway from your presentation, Michelle, was the impact of change, when you’re going through making the modifications at the design stage, how those have a significant impact as you go towards the end of the trial. So that was really interesting to me. Then Sarah and Nicola came up and talked about the use of eCOA within rare diseases. And I think the thing that most interested me about that presentation was how, almost to the point of personalization to the individual contributor or the individual participant, and how different types of people on those different types of trials require different visualizations of that. So the children versus the adult, that was very—I thought was useful for my own knowledge. Anders, thank you very much, making an incredibly complicated topic very simple, RBM why do we use it, why is it relevant in eCOA, risk-based monitoring, what’s the value of it in terms of how it can be applied to eCOA. I think the thing that I learned most about that was really identification of the risks and what those risks represent when you try to analyze the data at the endpoint. Rebecca kindly came up and talked to us about the relevance of data within the cognitive debriefing process. My key takeaway from that was really the relevance of that data and how you can use that data to drive efficiencies, and I think that’s something that I hadn’t seen before, so thank you very much for that. Rick, well what is there to say about Rick. He likes Star Trek, he tells us about all this funky technology, scares me slightly about if it can be connected he can collect the data, but I think that’s probably a good thing in our industry to associate a value to that and how that value can actually help us start to think about different ways in which we run clinical trials. And then last but not least, I almost forgot Mika—how can I forget Mika—talked to us about the value of eConsent and the process of eConsent in the clinical trial, and really started to challenge some of the paradigms around the paper process versus the e-process during the consent of the patient.
So quick round-up. To those of you who presented, thank you very very much. A really really really helpful day, and I’d just like to ask everybody in the room just to put your hands together one more time for those that presented.
And for those of you that are interested, we are going to be going to the—I think it’s called the Volkhause—is that how I pronounce it, anybody who is local? The Volkhause Basel, it’s about five minutes walk away from here, and I remember the words that were said to us this morning by Paul from PatientsLikeMe, a glass of red wine has medicinal qualities. We will be offering a glass of red wine and a little bit more and I believe dinner as well, should you like to join us. We’re going to be there from 5:30 onwards, so hopefully we can see and enjoy a glass of wine later.
Everybody, thank you very much for your attention and your questions. Thank you.
[END AT 31:50]