- Guru Difference
- Customer Experience
What are the similarities and differences between an in vitro diagnostic (IVD) device machine and a medical device?
In this episode of the Global Medical Device Podcast, Jon Speer talks to Milton Yarberry, Director of Medical Programs at Integrated Computer Solutions (ICS).
Jon and Milton discuss practical pointers for determining and describing the similarities and differences between the two types of devices. Milton also offers some helpful recommendations for IVD manufacturers to consider when it comes to regulatory compliance and their quality system.
“That’s the key difference. There’s no patient. There is a sample.” Milton Yarberry
“It clearly gives you a target population, a way they’re going to use it, a way they can misinterpret it, and all of those should become mitigations in your design.” Milton Yarberry
“You can tell if you hit the right spot if the person wants to punch you.” Milton Yarberry
“It’s hard to translate some of that science, that research, that clinical, that laboratory side of things into something that’s commercialized and into something that’s regulated.” Jon Speer
“It’s a mindset of the manufacturer that I think is the main difference.” Milton Yarberry
Announcer: Welcome to the Global Medical Device Podcast, where today's brightest minds in the medical device industry go to get their most useful and actionable insider knowledge direct from some of the world's leading medical device experts and companies.
Jon Speer: On this episode of the Global Medical Device Podcast, I had a chance to catch up with Milton Yarberry. Milton is the Director of Medical Programs with Integrated Computer Solutions. You can check out Integrated Computer Solutions by visiting ICS.com. Three letters, real easy. We talked a little bit about IVDs and kind of compare and contrast an IVD or in vitro diagnostic with a medical device. How is it a medical device? How is it not? What are the similarities? What are the differences? In my mind, I don't want to say it's always been clear, but it's been clear for a while when I see an IVD, I'm like, "That's it." Sometimes it's hard to describe and the good news is, through the course of talking with Milton, I think we've got some practical tips and pointers and advice for how to determine that, but also if you are an IVD company, what are some things that should be focuses for you with respect to regulatory compliance, quality systems and things of that nature. So enjoy this episode of the Global Medical Device Podcast. Hello, and welcome to the Global Medical Device Podcast. This is your host and founder at Greenlight Guru, Jon Speer. Joining me today is Milton Yarberry. Milton is the Director of Medical Programs at Integrated Computer Solutions. So Milton, welcome.
Milton Yarberry: Thank you. Glad to be here, Jon.
Jon Speer: So you and I were catching up just a little bit ago and you offered a topic that I thought was really intriguing and the topic is IVD, or in vitro diagnostic and medical device. How are they the same? How are they different? And you started to share some of your learnings. I'm like, "Hold on Milton, this would be great content for us to talk about on the Medical Device." So I thought, maybe we can explore that a little bit more in depth. My own experience, I've worked with both. In my mind, it's sort of clear, but then again at the same time, it's pretty fuzzy at the same time. So maybe a good place to start is IVD, what is it? Maybe, what does it not? Maybe that would be a good place for you to just start to dive at and then we can start to peel back layers of this.
Milton Yarberry: Sounds reasonable. Yeah, yeah, sure. And I should even say, as a precursor, I consider myself largely a medical device guy, but recently into IVDs. And so I'm sort of describing this as, from that perspective, from somebody who is coming from one to the other, noting the differences. But yeah, for the people who aren't as familiar or maybe have a fuzzy idea around IVD, IVD stands for in vitro diagnostics. So this is a fancy name for a test machine. And so this is going to be like in the news now, of course there's COVID and the PCR test machines that are the ones that are the good test, that tell you whether or not you have COVID. But more generally, this consists of reagents, instruments, systems that are used for the diagnosis of disease or other conditions, even health as it were. So typically, this would be things like HIV, flu, hepatitis, diabetes, that sort of thing.
Jon Speer: Yeah. I guess the way I think about it, and feel free to critique my lay person explanation, is I always think of an IVD, there's usually some sort of equipment or analyzer of some sort that's involved, but it's not typically invasive from a patient perspective. I usually think of it as usually there's some sort of sample of some sort that's involved, saliva, urine, blood or something along those lines that's taken from the patient. So I guess depending on how you get that sample, some may say that's invasive. But nonetheless, the IVD itself is taking that sample, puts it through some sort of chemistry or some sort of analysis and that sort of thing to determine the results of that particular test.
Milton Yarberry: Exactly. Right, that's the key difference, is there's no patient, there is a sample. And to that end, the scope of the IVD includes the whole ball of wax, collection of that sample as well as preparing it for analysis. And then of course, the actual analysis. So that's maybe one of the key differences here, patient versus no patient. And instead, your proxy for the patient is this thing called a sample. And obviously, whereas in medical devices, you're worried about the safety of the patient, in this case, it's more like protection of the samples and not to let the information get out. But just on the fidelity and accuracy of the test of the sample.
Jon Speer: Right. I know this is over simplifying, and there is risk, and this is one of the things that I think I've heard from time to time, sometimes people are like," Oh, I'm an IVD company. I don't have any risks to patient." Well, you kind of do, because what if you take the sample, it's analyzed and the results that are provided are incorrect.
Milton Yarberry: Exactly.
Jon Speer: The false positive, false negative, what does that mean? There is risk, but I think if you think about it from the pure patient interaction perspective, you could say," Oh, my risk is pretty minimal." But what is your test doing?
Milton Yarberry: Exactly, exactly. And for that reason, interestingly enough, the risk- based categorization that the FDA does on these devices is actually around what is it that you're testing for? So it no longer becomes, how dangerous is the technology, because if the machine breaks down, it becomes more, are you testing for something that is a non life- threatening disease or a life- threatening disease?
Jon Speer: Yeah, just to kind of help people wrap their heads around, to me, a classic IVD example that I think a lot of people can probably visualize the at home blood glucose meters, where you do the finger prick and you put the drop of blood on the test strip, and you put it in the little analyzer and it gives you an idea of what your insulin level is. I think another example is a home pregnancy test, is probably, I guess I don't know this, but I would assume that that's probably considered a type of IVD as well. Granted there's no equipment, it's a pretty simple chemistry type of thing.
Milton Yarberry: I haven't actually looked those up, but I've been working primarily on the ones that are sort of replacing labs or at point of care at a physician's office. But the interesting thing is, of course, in the examples you mentioned, there is a user, there is a patient involved. And IVDs are sort of strictly the domain of, there is no patient involved. Now whether the test is considered a therapy, I think is probably where that gets parsed along.
Jon Speer: Right.
Milton Yarberry: But the ones that I think of typically are, these either big lab machines the size of a small closet or a big closet where it's processing thousands of samples, right on down to these little, if you ever see them in the back office, yeah, little devices about that big. You have a little cartridge or a stick or something you put in there. The difference being, the operator on that device is some form of, I want to think clinician, because a lot of them aren't licensed clinicians, they are lab people. So because of that, there's a whole category of regulation for labs that applies to them.
Jon Speer: Right. And so you get into good laboratory practice, I'm sure. Right? And CLIA probably is involved, right?
Milton Yarberry: Yeah, definitely. Yeah. CLIA's an interesting one because for these point of care machines that are going to go at CVS or your drug store or a doctor's office, they want to be this thing called CLIA waived. And CLIA waived means that basically we've dumbed down the interface, perhaps dumbed down is the wrong term, we've simplified the interface and made it sort of bulletproof so that even somebody who isn't trained on the machine won't accidentally misinterpret the result.
Jon Speer: Almost like binary, it's either black or white or red or green or whatever. It's very obvious as to the outcome of that rather than just spitting out numbers and somebody has to interpret where that goes on a chart and what that means.
Milton Yarberry: Exactly. And yet at the same time, it's both that simple and also being that simple, it's somehow still complex because there's gray in everything, meaning a test might not run to completion, a test may have an invalid result, it may have a result that doesn't make sense. There are internal controls in these systems that may have a marginal result. So all of a sudden, they're still gray.
Jon Speer: Yeah. You and I have had conversations in the past where we delved into things like human factors. This seems like, I don't know how the right way to describe it, but almost like a human factors expert, this is like a kid in a candy store type of opportunity because there's all sorts of challenges there, I would assume.
Milton Yarberry: Exactly. And like the standard 62366, which it takes a risk- based approach for designing for usability, really.
Jon Speer: Right.
Milton Yarberry: So this is a prime candidate for that. If you have a machine, I think they actually cite in the regulation that a CLIA waived machine is supposed to cater it to an eighth grade reading level. So I think they actually specify that, which means that, yeah, it clearly gives you a target population, a way they're going to use it, a way they can misinterpret it and all those should become mitigations in your design.
Jon Speer: Sure. And going back to my example of the blood glucose meter, that information has to be crystal clear because the user population in that case is, well it could be a child, it could be an elderly person, it could be somebody with good eyesight, it could be somebody with poor eyesight. But they have to know what that means because if I do take the wrong action or misinterpret and take the wrong action, that could lead to death. I think this is an example where an IVD product is not without risk. But even within the lab environment, even if it's CLIA waived, there's still risk. It has to be clear to that person what that means.
Milton Yarberry: Right, right. And the even more sort of dangerous part of this is that it's not all in the machine.
Jon Speer: Right.
Milton Yarberry: It's how did you collect that sample? Having gone for a couple of COVID tests lately, how long was that swab up your nose?
Jon Speer: And did it go far enough?
Milton Yarberry: Did it hit the back of your head? Exactly.
Jon Speer: Because if it didn't, you think," Oh, it was in there long enough, it was inserted far enough." And then it goes to get analyzed and it's like," Oh, no big deal."
Milton Yarberry: Right, right. But if it wasn't in far enough, because there's no clear indicator other than, as a friend of mine mentioned who administers these, she said," You can tell that you hit the right spot if the person wants to punch you."
Jon Speer: I was going to say because there's brain tissue that's been detected on the swab.
Milton Yarberry: Well, you can't count on that. May not be present conditions.
Jon Speer: All right. So that is the twist, I think, from an IVD perspective, that I'll say the classic medical device just doesn't, it's not in that realm. But I think there's some areas that, I guess maybe we should explore this a little bit. Where are they the same? Right? And you've mentioned some equipment and things like that. I'm like," Oh, well that sounds like a classic electromechanical device from a medical device perspective." But I know this is your world, you deal with design and development and that sort of thing, but how are they the same from a manufacturing standpoint?
Jon Speer: Absolutely. And so a couple of things there. So a quality system, and I think this is confusing to a lot of people that I've talked to over the past several years, who are IVD companies. They seem, we'll go with confused, with respect to why they need a quality system. And I think your point is really well stated is an IVD is a medical device. It's just a certain type of medical device. An IVD as analogous to an orthopedic implant.
Milton Yarberry: Right.
Jon Speer: They're both medical devices, they just have different modes or functions and intended use and method of delivery and that sort of thing, but they're both medical devices. So from an FDA perspective, that says 21 CFR, part 820 is applicable to both of you. And if you're outside the US, ISO 13485 is applicable to both of you. And I think that that's not clear to a lot of IVD companies. crosstalk suspicions as to why, but go ahead, I'll chime in on that here.
Milton Yarberry: Oh no, let's hear it, suspicions are good. Let's hear that.
Jon Speer: I think a lot of IVD, and this is probably an overgeneralization, but I think a lot of IVDs probably originated in a laboratory somewhere, maybe by a researcher in some way, shape or form. And sometimes, folks who have that skillset, they're amazing people, but they're not always commercialization type people, they're always regulatory type people necessarily. They just are trying to figure out a way that is less invasive to do a task through some sort of, I'll say magic i. e. chemistry, to be able to take a small fluid sample of some sort, analyze it and give these results. I actually happen to have a chemical engineering degree. I think you do too, don't you?
Milton Yarberry: No, I'm double E, so electrical engineering. But yeah.
Jon Speer: I took a lot of chemistry in school and I remember taking a lot of analytical chemistry and it seems like analytical chemistry and IVD, they're partners in some way shape or form.
Milton Yarberry: Yeah, definitely.
Jon Speer: You're taking it through all this analysis, and you're like," Oh, look at this peak on this FTIR, and that means that there's this thing." And then I'm like," Okay, if you say so." But I think it's a little more like there's definitely science involved, but to the lay person, it seems like a little bit more black magic. But I think it's hard to translate some of that science, that research, that clinical, that laboratory side of things into something that's commercialized and into something that's regulated. I think there's just a disconnect there. I think it's just an origin of how those products can be. That's my take. I could be way off, but I don't know, do you have any thoughts?
Milton Yarberry: Well, yeah. Just in terms of, I have been surprised to see people who have been in this space a long time, sizable companies, also also smallish companies, but people who have worked in the industry a long time, have them momentarily struggle with it. It's funny the way you mentioned confused. Is that the term you used?
Jon Speer: Yeah.
Milton Yarberry: It almost does seem like," Okay, well maybe we're class one," because there's no patient involved or something, some sort of optimism around that. And it's like, no, no, no, you just have a different risk- based system that's based on the assay. But then, this is a bit of the ambiguity. This is the part where maybe I'm not even sort of clear after working in both of these types of projects is the focus on the design controls for the software in a normal medical device is kind of intense. I think what happens is that on the IVD side, there's a lot of focus on clinical. So inaudible percent of IVDs must have a clinical trial.
Jon Speer: Yes.
Milton Yarberry: Whereas 10 or 15% of 510(k) s do. So all the focus is on how are we proving it and what the analysis demonstrates, what sort of samples have we curated, which patient population, are we in flu season, all those factors dominate their submission. And so maybe there, it's just less attention on the software side.
Jon Speer: Yeah. I hadn't thought of it in that perspective before, but it makes a lot of sense. So to play back what I'm hearing you say is the emphasis is more on the outcome, the result, the clinical-
Milton Yarberry: And the orchestration, the orchestration of that big clinical trial. Yeah.
Jon Speer: Right. And so, in that spirit, the controlled design and development methodology just, I don't want to imply that there's not some methodology from a design and development perspective, but it's just not done in a design control type construct that a software as a med device or a classic medical device might have followed or adhered to during that process.
Milton Yarberry: Yeah. And I see it reflected in things like when you go to verification, I'm used to, with a medical device going to verification, the features are done, period. You're not going to verification with anything that's incomplete. On the IVD side, I think it's somehow more forgivable because the features there don't affect a patient. You add a button, you add a feature, it doesn't necessarily affect. It affects the person administering the test, but it doesn't affect the assay. Anything having to do with the assay and the calculation, the analysis behind that, that's hollow ground.
Jon Speer: And it's almost like the equipment used to analyze the assay is almost commoditized.
Milton Yarberry: Right. Exactly. It's de- emphasized, yeah, in that regard.
Jon Speer: All right. Well, I want to take a brief break. I want to remind everyone I'm speaking with Milton Yarberry. Milton is the Director of Medical Programs at Integrated Computer Solutions. You can find out more, a ton more about Integrated Computer Solutions by visiting ics. com. That's right folks, it's a three- letter domain. So that tells you a little bit about how long ICS has been around. But Milton, I thought you might have a few words to share with folks about some of the types of products and services that you do and help companies with.
Milton Yarberry: Sure. Yeah. Thanks for pointing out the three- letter domain thing. That hadn't ever really occurred to me. I wonder if IBM was taken at the time that they filed for that. So ICS, yeah, it's a company that's been around since the eighties, late eighties. Currently, we're software consultants where we do full stack development. The majority of our business is in medical regulated space. We have an offshoot of the company that specializes in usability and UX design. But our Zen, our methodology here, has to do with UX first, has to do with when you're creating a product, you start with user interface, you do that ahead of design controls with prototypes in as much as you can. And then that shortens the entire development cycle if you have a fast means to go from UX to backend software. So that's sort of our area of specialty.
Jon Speer: Yeah. And I think as it relates to the topic today, specifically with med device it's, I don't want to say it's a no brainer because clearly there's still, as an industry, we have work to do with respect to user interface and human factors and usability. But I think it's really, in my opinion, an underserved areas specifically with IVD. And I think it gets back to what you and I are talking about, that we don't realize that an IVD is a medical device. And I think a lot of times, because of that, people haven't focused on the user interface and the user experience and the usability and human factors thereof. So definitely check out ICS, www. ics. com. And I want to remind you too, all of you listening, that Greenlight Guru, we're here to help as well. We have the only medical device access platform in the industry. It's been designed specifically for the medical device industry by actual medical device professionals. Workflows to help you during design, development, risk, as well as post- market and all your documents and records. So check it out, www. greenlight. guru to learn more. And we just launched a really exciting program called the Greenlight Guru Academy. And this is a program where we have many courses on a variety of topics and more to come, lots more in the queue, but we have a course, it's free, it's on design controls and another course that's free on risk management, some on regulatory submissions. But we'll be adding more courses to that. So check that out as well, just search for Greenlight Guru Academy on our website and check it out. All right. So getting back into the conversation, Milton, we've talked actually a lot about, we've done more than just dance around it, but I think the other thing that creates some challenges and I hinted at it a little bit and you did a good job of explaining the clinical aspect, but I think things like lab developed test, LDTs, even I've done some work over the years with a couple of different IVD tech companies and they were, I don't know if it was the try to be CLIA waived if that was the motivation or something else. But they're like," Oh no, we're RUO," or research use only. It's like they were trying to find the perceived simpler path from a regulatory perspective. Do you have any thoughts?
Milton Yarberry: Yeah. I think the term I'm used to hearing is investigative device.
Jon Speer: Yeah, that's another term. I think these are all somewhat synonymous with one another.
Milton Yarberry: Right, right. And investigative device, I think there, a lot of the focus is just around treating the people in your clinical trial well and treating their data well.
Jon Speer: Right.
Milton Yarberry: But yeah, that's a different route. It gets away from the design control auspices, if you will, in that the thing that you're testing for might be less, it might be proactive in terms of what it is you're testing for it and the end result that you're trying to get to. Whereas, if you have an assay for a particular disease, it's very clear that you're going for a positive or negative on that.
Jon Speer: Yeah. And some of these products, I think historically, and I think this adds to the overall confusion, a lot of these products are in that gray area where it's not clear how or if they are regulated. And I remember back in the United States probably about, let's go with about five or six years ago, something like that, these investigational products or I think some people refer to them as these LDTs, or lab developed tests, many of them are like physician's offices, they develop an assay or test for whatever purpose or use. But that has largely, or historically not been super regulated. And there was a movement back, several years ago, where FDA was going to pull them under regulation and then backed off on that. So that's been an on and off again. I don't know if you've been paying attention over in the EU, but we're a few months away, or soon the EU MDR is live, and then the IVDR goes into effect in May, 2022. And what I've heard in the EU is now that those new regulations are pulling a lot more things under regulatory control, and I think this is creating quite the stir in the industry.
Milton Yarberry: I am aware of that, I have been reading some of that and I keep looking for the analogous thing to happen on the FDA side. And I'm not quite seeing it yet.
Jon Speer: Yeah, the other foot will drop, I think. I think the other foot will drop. And so, now when?
Milton Yarberry: Do you think that they are waiting to see the outcome or is it intentional or is it just happenstance?
Jon Speer: I don't know. When they were talking about this from the FDA perspective several years ago, I don't know what was motivating that.
Milton Yarberry: Okay.
Jon Speer: I think in some respects, some could say," Oh, it's financially motivated," which maybe.
Milton Yarberry: It's always easy to be cynical with that, yeah.
Jon Speer: The more companies regulated, the more establishment registration fees and that sort of thing. But I think what's happening in the EU is, sadly there've been some product issues and events in the marketplace that have caused some problems.
Milton Yarberry: The rollout there has been less than smooth. Can you point a finger back to that though?
Jon Speer: Yeah. So what's going to happen on this side of the pond, nobody wants any sort of negative adverse event to drive changes in regulations.
Milton Yarberry: Right. It always makes terrible policy, right?
Jon Speer: It does make terrible policy. But conversely, if it's too Wild West and there's not enough oversight and regulation, then that's problematic. But then again, you have other provisions in place, I think in the United States, like CLIA, which is in a matter of speaking I guess, a form of oversight.
Milton Yarberry: Control. So then of course you can't say all those terms and words without tripping on EUA. So how will EUA change the complexion of regulation after this obvious wave subsides?
Jon Speer: Yeah. And I think that's still the big unknown, and maybe the first question is how much longer will we be under EUA? No one knows the answer to that either because the moment that that goes away, then all of these companies, these products, these tests, et cetera, et cetera, that were brought to market under the EUA, now the clock's ticking. To stay in the market, they're going to have to do something like a 510(k) or PMA or something of that nature. So I don't know what's going to happen, but it's going to be interesting. And I'm not sure if interesting is fascinating, exciting or terrifying, but nonetheless, there will be something.
Milton Yarberry: Right, right. I think in the beginning we were seeing more companies reaching for the EUA with intent for being temporarily in the market. Here we are, a year later, and we're seeing more companies say," Okay, we're going for EUA immediately, but we're following right up with a 510(k), so that we're in the market."
Jon Speer: I think those are the smart companies, the companies that are taking that approach.
Milton Yarberry: Yeah. Yeah. At this point, I can't envision launching an EUA only. The timeline just seems too uncertain.
Jon Speer: I don't know this, I'm speculating, but I can't imagine the FDA is going to, in current state and time, take a EUA only application without there being a more defined strategy," We're going to do this first, 510(k) and then this, and then that." I just can't imagine FDA being that way. But then again, FDA doesn't call me for advice, at least not yet.
Milton Yarberry: Not yet, not yet. Give them time. Also, the evolving nature of the EUA. Initially, it felt like it was so Wild West in terms of email, initiate, submission. God, if they would only do that on the 510(k) side well, as opposed to the, you must print it in duplicate ship, they'll never take a look at it, plus digital form.
Jon Speer: Right. Absolutely. All right, so Milton, we've covered quite a few nuances and intriguing aspects from an IVD perspective. I guess before we wrap things up today, any other final thoughts, tips, pointers, advice?
Milton Yarberry: Yeah, wrapping back to the differences, again, what intrigued me about this topic was that there was a partitioning between the two. I think I saw it or read it someplace that people typically work in medical devices or IVDs, but they don't cross over. And it made me wonder, why don't people cross? They have so much in common. IVD is a medical device, all that should be directly transplantable. The answer I came to was that basically it's a mindset of the manufacturer that I think is the main difference. Whereas on the IVD, again, I'm a medical device guy looking at IVD or who has done IVD now, but I see that on that side of the fence, the clinical trials, those things, it revolves around the analytical fidelity of the assay. They're looking for valid evidence, protocol design, hypothesis formulation, sample curation, stored versus live versus simulated, performance goals, confidence intervals. Let's say CT data integrity, for example, like their clinical trials, they need to be performed across at least three geographically different sites. That brings a whole nother level of epidemiology awareness that you don't get in most medical devices. For most medical devices, it's usually a company that specializes in a certain therapy and it's about applying that therapy and patient safety. Whereas an IVD, it's sample collection, transport handling, sample preparation, and a unique link between safety and effectiveness, so where there's no patient interaction. And safety relates the false positives or negatives. So I think that's the primary thing that separates these two and it makes it difficult for the crossover. It's similar enough to be mistaken for one another and they're different enough to be confusing.
Jon Speer: For sure. And I think part of it is there are some overarching regulations that govern both. But then there's some nuances that couldn't be further apart from one another.
Milton Yarberry: Exactly.
Jon Speer: It's a really insightful look at that. So Milton, I appreciate all your time and the thought provoking conversation and diving in a little bit deeper on this. Again, I want to remind folks Milton Yarberry, Director of Medical Programs, Integrated Computer Solutions. He has a ton of expertise on user interface and user experience, as does the ICS team. So a great resource to have in your corner. Check them out, www.ics.com. As always, thank you for being listeners and, maybe even now, viewers of the Global Medical Device Podcast. If you're listening to the Global Medical Device Podcast where you normally have, that's fantastic. We've started to incorporate video as well. So be on the lookout for future episodes. Thank you for keeping the Global Medical Device Podcast as the number one podcast in the medical device industry. And as always, this is your host and founder at Greenlight Guru, Jon Speer. And we'll talk to you again real soon. Thank you.
The Global Medical Device Podcast powered by Greenlight Guru is where today's brightest minds in the medical device industry go to get their most useful and actionable insider knowledge, direct from some of the world's leading medical device experts and companies.
Nick Tippmann is the Chief Marketing Officer for Greenlight Guru, a MedTech Lifecycle Excellence Platform (MLE) that provides an industry-specific solution to help medical technology innovators around the world use quality as an accelerator to move beyond baseline compliance and achieve True Quality. Tippmann is...