MedTechUX — Conquering the Human Designing Medical Devices for How People Actually Think

And one… Hello and welcome, everybody, and thanks for joining today's MedTech UX webinar with, Meryl Zavad from Associate Director of User Experience at ClareMed, Inc. Merrill is joined by a few of his team members who may or may not come on camera at some point during this call. While we're getting started. I would like to make a few housekeeping announcements. This webinar is being recorded. Slides will be shared with attendees, tomorrow, along with a replay. So the recording and slides will be sent tomorrow to the email that you registered at. If you do not receive them, please

Feel free to reach out and let us know. For a certificate of attendance, which we will prepare manually, please put your email address and name as you would like it to appear in the chat function. Use the chat function to communicate with me during this call. and use the QA function to ask questions of Merrill and the Clarimed team, related to human factors. So, substantive questions go in the QA. Please include your name. In case we,

Don't answer anonymous questions if you want us to follow up. That's what I'm saying. Time permitting, we will try to get to all of those questions. There will be a short poll, about 15 minutes in, and a survey at the end. Please complete both. It really helps us quite a bit. understand you and your needs. While I'm talking, feel free to use the chat to say what country you're calling in from, and if it's in the US, please put the city or state or something like that, so that we know. This is just something You know, for the fun of it. And let's see, there will be, ClareMed Kit, usability kit, 24-page, document available, and I will put a link in for that, and a 5-day optional follow-up email course. So, you can either write,

Kit, or course. or both, in the chat, and we will make sure to send that to you, or I'll put the link in. Upcoming events… really quickly, this Friday in Focus, Cybersecurity Priorities for MedTech 2026 with Jose Barquequez from CyberMed. Thursday, January 22nd, Regulatory Roadmap, MDR and IVDR Compliance Priorities with Basil Akra, Eric Volbrecht, Tom Patton, and Julia Hoyer. And on the 28th, Real Life PMCS, practical lessons, and priorities for 2026 with Daniela Carrere. Matteo Moso and Diane Legere. So, I will put links to all of those in, and please do me a favor and like the latest issue of MedTech Leading Voice. And, we just crossed the 8400 subscriber mark, yesterday or this morning.

And so, if you're not a subscriber, please subscribe at MedTechLeadingvoice.com, which is a LinkedIn publication. That's it for my announcements. I want to turn it over to Meryl. I've been working with Meryl for a few weeks now on this presentation, and I'm extremely excited about this conquering the human. Merrill, why don't you take it over from here, and I will go away.

Sounds great. Thanks very much, Sean. And thanks, everyone, for joining. Again, I'm Merrill Zavod. I am the Associate Director of User Experience for Claramed. I received my degree in Human Factors Psychology many moons ago, have worked across several industries, but for the last 12 or so years, I've been working in pharma and medical devices. So, you're probably, or maybe, or hopefully, wondering why the talk is called Conquering the Human. After all, when we're designing for people, we're supposed to have empathy for them and view them as a partner in design, right? Well… Yes and no. You see, there are actually two parts to every person, and the way you treat each one of them during design is very different. So, you've got, basically, the conscious user and the unconscious human. So…

When we talk about user-centered design, and most of you are probably somewhat familiar with this term. We're talking about when you're designing a tool. you really need to understand the specific users that you are designing for. So, for example, you want to look at their characteristics, you want to know what kind of background knowledge they have, what kind of experience they have, their attitudes, whether they might have disabilities. You also have to pay attention to their environment and the context within which they'll be using the tool. So, for example, if you were designing a tool to help radiologists

detect abnormalities in an x-ray, you would likely take a very different design approach than if you were creating a tool to help elderly dialysis patients identify meals that were safe for them to eat. So it's really all about the individual characteristics of the specific users that you're designing for. On the other hand, there are also a set of universal traits, which all of us as humans, no matter how old we are and what our backgrounds are, that we all share. And these are, for the most part, subconscious traits. We really don't have a whole lot of, control over them. But as you'll see, they become very, very important in understanding. So it's not only about understanding your conscious user, what they tell you, it's about understanding what's going on underneath.

And what you're going to see is, for the most part, while your user will likely try to help you with design. the underlying human will do everything in their power to hurt you during design. So, when we're talking about user-centered design, you've probably heard time and time again the word empathize, that we should empathize with our users. As we design, and that's absolutely correct. When you look at the underlying human, however, there's really no empathy there, because they have no control over this. You really have to hit it head on. So… What you're seeing here is one proposed model of how humans process information during the completion of one or more tasks.

This specific model was proposed by Wickens. There are several other models out there, but they all tend to be pretty similar and share most of the same traits. So, what we're going to be doing, we're going to be walking through this step by step, and what you're going to see is, along each one of these steps, there are actually several things, that can prevent the step from going correctly. The problem is that this is really a chain, from viewing a stimulus to figuring out what you need to do with that stimulus and responding to it. And if there's a break anywhere along the chain, the entire process tends to collapse.

So, as I walk through these, you'll see on the right-hand side, I mentioned some of the particular biases or weaknesses that we all share. I'm not going to go through each one, individually, you can see that they're there, but I will be talking about a couple of them for each one of these. So, in order to carry out a task, the assumption is there's going to be some sort of stimuli that starts the task, whether it be a visual signal or an audio signal. So the first step is actually detecting, that stimulus. If we don't detect the stimulus, nothing else can happen in the chain, and you completely miss it.

So, why do we miss things? Well, the stimulus might not be salient enough. So, if you were expecting a person to find a blinking light among other blinking lights, for example, they might not be able to do that. They might be focusing somewhere else. The problem is that you can only really focus on one stimulus at any time. So… Someone might be focusing on something else, and again, just completely, missed that something happened. Fatigue can also have an effect here when we talk about the vigilance decrement. So over time, as we get tired, we do tend to miss things. We don't pay as much, attention.

So, assuming that we did detect the stimulus, maybe it's a blinking light, we have to figure out next what exactly that stimulus means. We have to be able to interpret that stimulus. And there's several issues that can occur here. Sometimes the stimulus is very similar to another, and we might misperceive it. There's also this pattern recognition bias, and I want to stop for a moment and spend a little time on this, because what you're going to see is this is actually critically important. And it can be a detriment, but it can also be used as a useful design weapon. So, the fact is that human beings rely on pattern recognition, in everyday life.

We don't tend to look at things in detail, especially if we recognize a pattern, because it would take too long and take too much resources to actually pay attention to every single thing. So if we see a pattern, we recognize it as that pattern, and we go on. So, for example, ask yourself, how many times have you walked up to an elevator and actually read the buttons, the up and down buttons, to make sure that you were going in the right direction? Probably never, because the simple fact is, as soon as you start walking towards there. you recognize the pattern. You see two buttons, one above the other, and there's no reason to read. You know that that's, something that you're very used to, and so you move on.

Now, the problem here is that sometimes we misrecognize a pattern. So, if there's something that violates that pattern, but we recognize the pattern, that's where errors can occur. So, as a design tool, however, it's always a good thing to use patterns, because it saves time, and it saves resources, and it saves the person from having to look at every single detail. So moving on, assuming that we do correctly identify what the stimulus is, we then have to decide what we're going to do with it. And here, once again, there are several biases, as you can see over on the right-hand side, that can affect our ability to do this correctly. So, for example, you've probably heard of confirmation bias, and this is

the human trait where we do tend to look for information that confirms what we were thinking. You've probably seen this on Law & Order a thousand times, because, again, it takes a lot less resources To look for information that confirms something than to look for information that might mean, something else. Another one that you've probably, seen in real life is the availability heuristic. And what this means is Things that have happened to us most recently tend to have a more vivid experience for us and tend to be activated more quickly than a memory that is older and that has undergone some decay.

Overconfidence bias, this, again, we've probably all seen, and this is overestimating one's own judgment or accuracy. So, a lot of times, we believe that we are experts at something, that we know what's going on, and we end up ignoring information that might violate that. So, the other piece of this, in both identifying what the stimulus is and deciding what to do with that, is memory. So, we have long-term memory, and we have working or short-term memory, and both of these are critical in making proper identification of a stimulus and deciding what to do with it.

Now, we all know that memory does have its limitations. So, for example, we all know that in working memory, you can only hold, they say, 7 plus or minus 2, chunks of information at any given time. So, if we provide the human with more than that, they're not going to be able to remember all of the information. So you can see there's also some other biases here. We may simply forget. So sometimes, even if we are working with something finite in working memory, if there's other stimuli coming in, working memory might get overloaded. Sort of the same thing that we just talked about on the last slide, there's also the recency effect, where recent memories tend to be activated much more easily and quickly than those that are older.

Normalization of deviance, this is an interesting one. Sometimes it's referred to as the cry-wolf syndrome, and basically the idea here is that, for example, if we have an alarm that's repeatedly false or it's incorrect, we become desensitized to it, and we simply stop responding to it. So, if at some point the alarm actually does mean something, we rely on our memory that says, well, this typically means nothing, and perhaps we actually don't take action. So, the last piece is, what is our response? So we've made a decision what we need to do with that stimuli, whether it's I need to press a button, for example, or pull a lever, or adjust a valve.

So even if you've decided on the proper response, now you have to actually execute on that response. And once again, there are several innate human traits that may prevent that from going, correctly. So, we've all seen slips and lapses. Perhaps you've… Reached for the Preparation H rather than your toothpaste when you're going to brush your teeth. things are similar. So, again, you can use the same analogy there, where if you have two tubes sitting on your sink and they look very similar, you may grab the wrong one, even though you know you should be grabbing the other. And then there's time pressure. So, again, we've all probably experienced this, where, especially if you're being rushed, you may do the wrong thing, or press the wrong button.

So, just to give you a personal example of this. I was once driving with my son at a pretty high speed when a deer jumped out in front of us. So, instead of braking, for whatever reason, I just leaned on the horn. And we ended up going pretty much through the deer. So, the proper response Would have been to apply the brakes. again, for whatever reason, I thought that sounding the horn might be a good idea, and that the deer would get out of the way, which, of course, he did not. And, my son still makes fun of me to this day, you know, asking me, what exactly were you thinking? You think the deer was going to pay attention to your horn and politely move out of the way.

So lastly, the one part that we haven't talked about is that as part of this theory, we have a set of attentional resources that drive each one of these steps. Now, that pool of attentional resources is finite. It's not unlimited. And every stage requires some level of these resources. if… more resources are required than what is available. The stage can either be erroneous or not take place at all. So this is where we have to be very careful, because you not only have to worry about the actual task that the human's supposed to be focusing on in terms of what level of attentional resource they need, but also what else is going on. You know, if they are trying to complete two tasks at the same time, well, both of those are going to require

resources. And again, if you go beyond that level of resources, you're going to end up, with failures. So, there are also other factors beyond our control that can negatively affect performance and the availability of resources at one or more of the processing stages. Fatigue, for one. You've probably all noticed this. When we are fatigued, those attentional resources tend to not be able to do their job. Sickness, again, can drain our attentional resources. Some medications can have an effect on our resources. But also, our state, our emotional or psychological state. If something happened to us in our life, and we are thinking about that, and going over that again and again, and trying to complete a task, well, a lot of these attentional resources are actually being

pulled away from the task at hand. So, we have to think about these when we're designing. So it's not only a matter of making sure that the way we design something stays within the realm of what humans can do with their attentional resources, but also what else might be Pulling from those resources. So… What can we do? What are the tools that we can use to battle against this underlying user who has set out to make our lives miserable in one way or another? So, a couple of traditional methods that have been used. The first is selection. So, even though these are universal traits, we do know that sometimes some people have stronger skills in one, in one or another.

So, we can try to select those humans who are most apt to be able to perform a task correctly. However, a couple issues with this approach. First of all, it's sometimes very difficult to determine exactly what the skills and abilities are required to complete a task. And it's also often difficult to accurately measure or test for whatever those skills or abilities may be. And lastly, even if we are able to identify suitable individuals, there might not be an adequate supply of those people. So, the other thing that had been used as a traditional approach is training. So, there is the possibility of helping people get over some of these biases through training.

However, we've probably all seen that even with the best of training, people don't always perform as they were trained. They may forget. If it's been a while since the training occurred, they may revert back to their old habits. Training can be expensive, especially depending on how intense that training is going to be. And lastly, as I mentioned, training can account for some of those biases. We can train people to overcome some of those biases, but typically not all of those. So training usually can't account for all of those. So, where we saw both of these fall a little flat, first was in World War II.

So, in World War II, we had new planes that fighter pilots were going to be using, and these were very complex devices. So, what the Army Air Force attempted to do was first select those individuals who showed the traits that they would be able to handle a complex machine like this. They also put those same pilots through intense training. Now, while it helped, it didn't account for everything, and there were still many, many crashes during World War II. Even though we selected the greatest pilots and gave them the best training, they still made mistakes. So, what's the alternative approach? Now, a more modern approach is to try and handle these biases and these limitations through proper design.

So, there are a few ways during design that you can help prevent errors from occurring. So, we realize that humans will make errors. I think we've already seen that that tends to be the case. It's what humans do, it's what we're good at. We make errors. So, what can we do about it? Well, the best thing to do is design something so it actually excludes the error from ever happening, also known as locking out, the error. So, just to give a simple example of this, polarized outlet plugs, AC outlet plugs. one of the fins, or the blades, is thicker than the other, so that it cannot be inserted the wrong way. It has to be inserted the correct way. There's literally no way, unless you absolutely forced it, to make that error of plugging it in the wrong way.

So… it's not always the case that we can do that. Sometimes there has to be flexibility, and that means that the human can make an error. So, what can we do in that case? Well, in cases that errors are still possible, we can try to make them difficult. to commit. So, using a simple example, in Microsoft Word, for example, if you are about to close out a document that you haven't saved, you'll usually receive a message that says. something to the effect of, are you sure you want to close this without saving? So, the user can still make the error, they can press yes, go ahead and close it, but at least we're raising the awareness that they're about to make an error, making it more difficult to commit, rather than if they were closing it, just closing it out and not saving it for them.

Lastly, once again, as hard as we try, people are going to make errors, and so what do we do in that case? In that case, what we try to do is minimize the consequences of those errors occurring. So, in this case, we can use another example from Microsoft Word, and that's autosaving. So, in modern versions of Microsoft Word, our documents are saved for us. We don't even need to tell the system to go ahead and save the document. It does it for us. Now, we may not need that saved document, but it's there just in case we make an error that we can recover from it. So, you're probably asking yourself, or have asked yourself, well, if humans are so fallible and have all these opportunities to make errors, why don't we remove them from the loop altogether and turn it over to technology?

Well, first of all, if the great warrior Sarah Connor taught us nothing else, she taught us that you never just hand the keys over to the machines. And the reason for that is machines do still make errors. And if we completely automate something, the human ends up losing their skills. And if something were to happen, and the machine does have an error, the human won't be able to take over adequately. So, you know, we've seen this, for example, the fear with self-driving cars. We've seen that they do make mistakes. So, if we get to the point where everyone has self-driving cars, and no one has to steer, or merge into traffic, or use the accelerator or the brake, we will lose that skill.

And we know that even self-driving cars are not perfect, and that at some point, the human may need to take over. So, if the automated car is all of a sudden swerving towards another lane, the human has to be able to take over and be able to drive. Now, if they've been relying on this for so long, they won't have that skill anymore. So, the point here is, we don't want to completely eliminate the human from the loop. However, on the other side of the coin, there are opportunities here for automation, especially with the advent of AI. So really, automation should be used to augment human capabilities, not replace them.

So, for example, we already talked about that humans have very limited memory stores. Well, in those cases, we can use machines to be able to capture loads of data that the humans simply can't store on their own. The other thing that we can use AI for is figuring out how to accommodate for some of the biases and the failures that we've already talked about. So, it's not that there's no opportunity for automation, there most certainly is. It's that we have to use it carefully. And you have to think about, what you want to give to the human, and what you want to give to the machine, because machines are certainly better at some things.

But there are still some skills that humans are actually better at than machines. So, for example, dealing with fuzzy or incomplete information. This is something that humans tend to still be better at. Now, that gap seems to be closing rapidly. But for example, you've all probably experienced those annoying CAPTCHA pop-ups when you're trying to access a website that ask you to read degraded writing and put in the text. So this is something that humans do very well. We can read text Whether it's in different colors, different shapes, has lines through it, we can still read it. That's something that computers, again, at least for now, don't do particularly well. So, the point is, you really have to pay attention to task allocation.

So, this is the… the entertaining portion, of the talk for today. So, I'm going to walk you through a personal case study of how my subconscious human decided to take on some medical product designers, and I believe it won pretty thoroughly. So… We're going to be talking about some contact lens solutions, so those who wear contacts will appreciate this all the more, but just as a little background, I've personally been wearing contact lenses for about 45 years now. I've been wearing them ever since I was in about 3rd grade. Now, it used to be, back in the day, taking care of these contact lenses

Was really a pain. So, every night, you had to take out your lenses, you had to wash them with some saline solution. Then you had to put them in a case, and put that case in a heating device, which would basically boil the contact lenses, which would disinfect them. And then in the morning, you would spray them off with some more solution, and then you would put them in your eyes. Also, once a week, you would have to soak them in a special solution, because proteins tend to build up on contact lenses over time. So, once a week, you would have to soak them with this special solution that would get rid of the protein, and then you had to wash them off from there.

Well, at some point, Bausch & Lam came along with this solution called Renew, and it was touted as an all-in-one solution. And sure enough, what this solution did was basically everything. So, it would wash the lenses, it would sterilize the lenses, and it would also remove proteins from the lenses. So, basically, from this point on, all the person had to do was, at night, pour some of this into their lens case, place the lenses in there. Eight hours later in the morning. pop them out, put some more solution on just to wash them off, and pop them right back in your eye. So, this stuff was a godsend for those of us who wear contact lenses. It was really a game changer.

So, I had used this solution for many, many years, as it was… it was really great. But one day, I was at the grocery store, because I needed some more of it, and I noticed that it was empty on the shelf. There were no more bottles of it. However, right next to it was another bottle that looked very similar. It was called ClearCare. And as I read it, it sounded to be exactly the same as Renew. So you can see that it says that it cleans, it disinfects, it removes proteins, it's one solution. So I figured, okay, this is simply a competitor to Renew. So I grabbed it off the shelf. And I took it home, and in the evening, I opened up the box, and I saw that a little lens case fell out, which…

wasn't unnormal. Usually Renew comes with a brand new lens case when you buy it as well. This lens case looked a little bit different, but I didn't think anything of it. I tossed it aside and used the lens case that I normally use. I put my contact lenses in there with some of the solution, and in the morning, I opened up the case, I sprayed it off with a little bit more of the solution, and I quickly popped the lenses into my eyes. Within about 5 seconds, both of my eyes clamped shut, and it felt like there were being… there were razor blades being pulled across my eyes. And I literally could not open my eyes, and it burned like nothing I have ever experienced.

So, about… it took about 15 minutes until I was actually able to open my eyes again, and they still absolutely killed. Every time I blinked, it, again, felt like razor blades. So, I was thinking to myself, what the heck is going on here? What did I do? I thought maybe I accidentally got soap into the container with the contact lens solution. I really couldn't figure out what was going on. Well, I finally, grabbed the bottle, and I looked at the back. and I noticed… Some writing on here. So, as you can see, there's a whole lot of writing on here, but there are a couple things written in red. Same font, but written in red. It says, do not put

AOC clear or disinfecting Solution directly into the eye. Okay, I didn't do that. Then it says, ClearCare must be used with a specially designed lens case. So, I thought, okay. What does that have to do with anything? So, I kept reading, and down below, you can see it also says, do not put directly an eye. I didn't. Use only the lens case provided. Well, it doesn't say why, it just says use it. Well, I figured. That just meant, for whatever reason, maybe it does a better job of disinfecting within that particular landscape. But the fact was, there were really no warnings on here at all. So, I made an appointment that day with my ophthalmologist.

where she discovered that I had suffered chemical burns on, both of my corneas. And when she asked me what had happened, I told her that I used this solution called ClearCare, and she immediately recognized it and said that she had actually had several patients who had done the exact same thing. And she said the reason for what happened is ClearCare, unlike Renew, uses hydrogen peroxide in its solution to disinfect. Now, I found out that hydrogen peroxide, when applied to the eye, tends to burn the eyeball. So, the special lens case… That was provided with this actually contained a neutralizing agent, so it neutralized the hydrogen peroxide.

Well, again, if you look at the back of this, first of all, I of course didn't even bother reading it, because I assumed that this was exactly the same as Renew. But even if I had read it. I began thinking to myself, there's nothing here that really would have told me, you know what, I should probably use this lens case, or I could end up burning my eyes pretty badly. So, the interesting thing about this, as I said, my ophthalmologist mentioned that she had seen this with several other people. Well, I did notice, a few months later, I was in the grocery store again, and I saw that ClearCare was still on the shelf, but they had actually redesigned their bottle.

So, they did a couple things here, but still not adequate. They did make a very clear portion of the bottle where it lists out warnings. It does say, do not use a flat lens case. It only works with the special lens case provided. So again, this was my initial thought. was that, okay, you know, maybe it just won't work quite as well, but it's not a big deal if I just use a regular, case. Don't remove lenses. It needs time to neutralize. Okay, that doesn't really mean anything to me. Don't rinse with ClearCare. Okay, that's probably something that I did incorrectly. Don't squirt it directly into your eyes. Okay, that's not something that I would do.

So, there are still warnings, but once again, we as humans, this is another bias, oftentimes we don't pay attention to warnings unless we understand exactly what the consequences, are going to be. So, if you notice here, nothing mentions that you better really pay attention to this, because you're going to burn your eyeballs out, if you don't. You can see on the top picture on the right, they also put a little wrapper that says, use only lens case provided. Once again, they don't tell you a reason why. Now, the other thing they did, which actually is a very good design principle, was they actually put a collar around the top of the bottle, which, again, has the inadequate, but they are warnings.

So basically, the user, is forced to… read those, or hopefully at least see them when they are opening the bottle. So this, again, is actually a very good design principle, and can be used as a powerful tool. Basically, you're forcing the user to see that, because they probably realize that a lot of people will not read the back of the bottle, so let's hit them with it, right in the face. However, once again, the warnings themselves are, inadequate. So, looking back at my situation, you can see that my underlying human, used several of the weapons that we've talked about to overcome the medical product designers here.

So, the first is, inattentional blindness. I didn't even look at the back of the box. even if I hadn't, you know, equated this to Renew, I probably wouldn't have looked at the back of the box, because it's contact lens solution, and there's really nothing to it. Now, the other part of that is, the obvious one is pattern recognition bias. So, if you recall, this is what the two bottles looked like. So, I recognized the pattern. You know, once again, it saved myself resources having to read every single thing. This appeared to be… it appeared to follow the same exact pattern. If you notice, it says basically the same thing. So I immediately recognized the pattern and assumed that it was actually the same. However, I found out the hard way, that it, in fact, was not.

So, the lens case. Once again, really pattern recognition for me. So, even though the shape was different, it did look different from a regular lens case. I assumed all lens cases were the same. I had never heard of a lens case that had any sort of neutralizing agent within it, or why a lens case would even need a neutralizing agent. And the other part you can see is I certainly did have the overconfidence bias. I know how to use lens cleaning solutions, so why would I have to read, the directions? So, as you can see, in this case, my underlying human did battle with those

medical product designers, and I think I really, showed them and, and, won pretty thoroughly, in this case. So, that is my personal, case, study in how these innate human traits can really take over, in any stage along the human, processing line. So the point of all this is. Yes, we should be designing for specific users. We should understand what their specific traits are, what the environment is that they'll be using the device in, what knowledge they bring to the table, specific knowledge, what training they might bring to the table. However, there are these universal traits, and by knowing them, we may not be able to overcome them completely, but by being aware of them, we can…

Designed so as to, once again, either lock the error out, from ever occurring, or, at the very least. Try to warn the user, that it's going to happen, so that maybe they can catch it, or realizing that humans, one way or another, will always make errors, trying to minimize the consequences of those errors. So… That is what I have. I guess we will, open it up.

Fourth.

Questions in the last, minutes that we have here.

So, I've gone ahead and launched the poll question, which I'm sorry to get to so late, but if everybody would take a moment to respond to the poll, that would be helpful, and while we're doing that. you know, a couple things came to mind. First of all, there were a couple questions that were submitted. One was about cultural differences, and you just got done talking about the universal traits. But I would love to open it up to you and your colleagues, to talk about regional differences, cultural differences, and how to, respond to that, because I don't know.

Yeah, so I'll let my colleagues weigh in as well, but just real quick, that's absolutely critical. So we know that across cultures. sometimes different things can have different meanings. So, colors, for example, may have a different meaning to us in the United States as opposed to someone in China, for example. Other things, you know, realize that here in the United States, we read text from left to right, but that's not always the case. In places such as Israel, they read the opposite way. There's also cultural biases that come into play that we need to think about as well. So these are, you know, different perspectives that are brought to

the table, so absolutely, knowing the culture, is huge. Now, that being said, there are still several of these that you can view, as universal, where it really doesn't matter. The culture may have more with designing, you know, the specific around the user, I guess more towards user-centered design, where we have to understand exactly how to design for them. Whereas, again, universally, we know that whether you're from the United States or from the Middle East. are from China, that you are only capable of holding 7 plus or minus 2 Chunks of information in your working memory at any given time.

So, I'll turn it over to my colleagues if they have any, examples or thoughts on that.

I'm unable to hear Anne-Marie. Are you able to hear her, Meryl?

I'm not…

Try coming off mute, Anne-Marie.

Can you hear me now?

Yes.

Oh, oh, great, sorry about that. I just wanted to say, yeah, great, I think as a human-centered designer and human factors engineer, listening to other people's examples, you always learn something new, so, yeah, love your examples there, Meryl. In terms of the cultural question, such a good question, and one for us to think about is early on in the design process from a human-centered perspective. And I wanted to give just a very quick example. So, it's not medical, unfortunately, but I used to work in consumer goods, and we were designing products, you know, for Japan, for, like, Asian countries, also for the US and UK and Europe.

And one thing we identified when we were developing the Dyson Airwrap. was that this tool was being developed to create curly hairstyles, and you style from wet to dry. And when we took the prototype to Japan, we actually started to learn that the participants didn't, didn't like it as much because they actually wash their hair in the evenings. So having a tool that allows them to style wet to dry wasn't really something that was desirable to them. So, you know, we had to think about, well, what does that mean? Because Asian's a huge… Asia's a huge market! And the other thing was learning about how the different hair… actually, the hair type works very differently, Asian hair, compared to the US and European Caucasian hair. So, yeah, really great question, and I would say the earlier you can understand who your users are, what their cultural habits are, what

Their routines are, can actually really impact how you then design the product.

Great example.

Hello, can you hear me?

Yes.

Nope.

I'll… I'll just check beforehand. So, yeah, that's… that's actually, I agree completely with… with both Merrill and Anne-Marie. I… I had, some experience with different markets, and this is more… maybe more pragmatic to human factors. So when… when testing the same product in different markets, you can see differences that can be from very subtle on how people perceive different things, or the importance they give to different Messages the device may give, but also The… the difference can go… very deep into the actual user that uses the device. So, in some culture, in some countries, a user may use a device. Some other countries, that user may

partially use it, may use it differently, may interact with other people to use it, or may not even be their role to use it at all. That means that something that you design for a user with a specific background and a specific knowledge, and maybe you avoid giving training because you expect that person to have a certain kind of knowledge, applies to one country, doesn't apply to another. So, that is something that is very important, you know, when we go into designing devices for a healthcare professional, where you expect maybe a nurse to do it, and in another country is a technician. That does it, or a surgeon that does it. So, these differences are very important. And another one that I had recently was about names of products, and that is also very important, knowing in,

In a country, or in a culture, what a name means.

And I, I had this, this experience in France, where there was a name, and everyone, every participant said, I don't like this name because it reminds me of, Verhoover. While in UK, or even for me, I'm Italian, I don't… I wouldn't associate it with a Hoover, but everyone in France would. So knowing that. it's very important that you cannot just apply everything to everyone and… and expect the same… the same response. So yeah, I'm an advocate to have different… different… even different evaluations in different markets, so yeah, absolutely.

So, I just want to jump in. There is another question, and it is about the burden of responsibility, from a regulatory perspective. Does it lie on the manufacturer versus the healthcare system, ensuring exclusion, prevention, etc? Should, the manufacturer preempt de-skilling of clinicians and run a refresher on a device from time to time? Or should the health system take charge of that? Who's… who's… responsibility is it? I'm gonna… Open that up, too. The panel, and let you guys decide who wants to answer it. That was one of the submitted questions from the audience.

I mean, from my perspective, that's a very, it's a kind of hard one to answer generally, because you need to really understand what the specific product is and what the risks are attached to it. From a regulatory standpoint, it really is… there's a lot of responsibility on the manufacturer to mitigate the risks as far as they can through design and labelling, and also training. But as we know, the FDA have a really strong emphasis on labeling and training being your secondary measures, and really what can we do as designers to design out any inherent risks, you know, whether that's through how, how risks are communicated on the graphical user interface on the device, what alerts are being indicated, and how to inform the user on what they need to do if that, you know, if there is a risky situation that occurs.

So, it's a hard one to answer. I would say my… always my recommendation is what can be done as far as possible from a design perspective, and then looking at, okay, from a labelling and then training perspective. you know, making sure that, you know, if it is a complex device, let's say hemodialysis system, right? You're probably going to want to refresh the training with your nurses, or especially if there's new nurses that come in, if you have a You know, change over in terms of, you know… but it's hard because then there's an element of the the clinical side that has to manage that risk, because the manufacturer's not going to know if there's going to be a huge turnover of staff, for example. So, yeah, tricky one to answer, and yeah, you sort of need to get down to the nuances of what the risks of the device are.

Sorry, not a very good answer to the question, but it's, one of the things is always, it depends. That's one of the things we say a lot of the time.

Certainly, certainly, and I… and I appreciate your insight on that. You know, the question that brings to mind is the importance of. post-market surveillance, and product feedback once the product is actually on the market, monitoring that information. Do you have any best practices or ideas that can be shared, you know, in terms of signal monitoring, or… I don't know, how are companies to understand what happens. You do your best planning, you do your testing, you put the thing out there, and then you discover How far wrong people can go.

Yeah, it's true. You know, you can do all the testing you want, and you can simulate the environment as well as you can, but once again, once a product is in the wild, humans will figure out a way to mess it up. It's just… a fact. So, you know. testing certainly allows you to eliminate a lot of those errors, but you have to be able to look at things post-market once they're actually live, and potentially have to do some tweaking to the design. You might have thought that you caught every single thing, but you can pretty much be guaranteed

It doesn't matter if you ran 60 people through testing, once again, once it's out there, there will be additional, issues.

There's, a couple of examples I can just quickly add to that. So, number one, I've worked with a manufacturer who have, like, a… they basically have a… they work very closely, the human factors team, with the quality, quality team and also the customer service team. So they, every three months, do an audit where they pull in what the complaints have been, you know, directly to the manufacturer, but also on the mod database in the US and and other platforms. And they actually do a known use problems analysis every… every 3 to 6 months to see what new use problems, specifically to the user, not mechanical failures, that could be contributing to harmful situations.

So that's one thing that companies can do to try and keep on top of that. The second thing is actually contextual inquiry. A lot of people think that doing contextual inquiry is something you really want to do at the start of the design process, but there's a lot of benefit to doing it once the product's in market, because you can go on-site, you can observe, you know, whether it's in a patient's home or in a clinical environment, you can go and observe that being used in its real context. Out, you know, not in a human factors lab, and really start to learn what's the environmental conditions are impacting how this is being used, what are the particular other tools that the healthcare professionals might be using with the device that you didn't even think about that could be impacting how it's being used.

And then, obviously, trying to identify if there's any risks there that… from a safety perspective, but also what could be done to actually make it better. So even if it's not causing any harm or risk, what could be optimized in the next generation of that product to make it an even better user experience?

Yeah, just to… just to second that, Anne-Marie, where the strengths of contextual inquiry in terms of identifying these types of issues is is undeniable, as Anne-Marie mentioned and alluded to, both in terms of going out into the field, observational methods, but also in-depth interviews, many a time conducted focus groups. Really a perfect environment in which to identify and unpick these, and also problem solve. And also use of surveys, so we have a… there are a whole raft of tools that we have at our disposal, which, often used in combination, provide the greatest insights, the greatest strength.

Thank you for that. The last question we're probably gonna have time for, is… Oh, good lord. The, looking at the wrong thing. I love this question. If medical device design is fairly long established, and there is no… not much significant evolution in technology, eGBTCA catheter, PTA catheter, balloon expandable stent system, etc, If the devices are used by trained, experienced medical professionals, and the product is on the market since many years, do we still need to do usability study for EUMDR compliance, or FDA 510K, or PMA approval in the United States? Certainly PMA approval in the United States you do.

Thoughts? Comments? Anyone?

The question is always, and this is the question that every regulatory is going to ask is. Whatever you are doing with your device, even if it's just a slight change from whatever is in the market, or whatever is already yours, and you're just changing something. Are you adding any risks, or are you modifying any risks where preexisting before? By modifying, it doesn't always mean adding them. Are you modifying, for example, a mitigation that was in place? And the question is, how can you be sure that now that device is as safe as before, and that mitigation is as… relevant, and as, Efficient as before. You… if you… Can do that, then… You know, probably…

that there is a justification there, but most of the cases, what regulatory bodies want to see is that your device is as safe as it was before with the changes you made. So yeah, it's true that some things that stay the same, they don't necessarily have to be re-evaluated. But in general, when you make a change on a device, a design change on a device, that changes also how the flow, how that device is used. It changes the risks, it changes the interaction with the user, and that means that Most probably that device has to be retested, because you have to revalidate that everything you did before is still valid, or everything that you did afterwards is still, safe as it was before.

Yeah.

Oh, brilliant.

Data…

I would also say that even if you don't make a change, if we're talking about something that has been in the field for a very long time. even if you don't make a change to the device, you have to realize that the users are likely changing, again, especially if it's something that's been out for many, many years. So, a medical professional coming up today has very different training, or at least somewhat different training than someone who came up through the field, you know, 20 or 30 years ago. They're more used to using things like technology, today, so… Even if the device itself doesn't change, you know, I don't know what the exact regulatory rule is, but probably a good idea to test it out every now and then, because your users may be changing.

Absolutely.

I know we're at time, I'm just going to add a quick thing to that around, a device of unknown provenance is something that's talked about in 62366, so that's where a device has been, validated prior to, well, 62366 and the FDA guidance come into fruition. So, you can take a look at 62366, which gives some, information about what you sort of can and can't do, if it is a device that has been on the market. Let's say for 15 years, and you want to make some improvements, so, yeah. That's certainly something you can look at.

Well, I would like to thank all of you for joining us today. I would like to thank Claramed, for this opportunity, and the excellent information presented. On the final point, a couple people have commented on, state-of-the-art changing constantly, as well as clinical evidence requirements in the EU, so certainly take a look under EUMDR. Requirements, and what is expected. And be aware that the state of the art is always changing, even if you think your product is static. I, was thrilled. Again, if anybody would like, the 24-page usability prep kit, or the course, just put that in the chat, or message me offline, or, and thank you so much for your time, everyone, today. Merrill, do you want to sign off? Say goodbye?

Sure. Yeah, thank you all for your time, and I wish you godspeed in your battle against the human.

Thank you so much. Thank you, Anne-Marie, Davide, Alper. Everyone have a great day. Thank you all for attending. Bye-bye.