The topic of human factors can be confusing: How does it relate to design factors and risk? Why is it included in the regulations and how can this factor be handled?
On today’s episode, we have Bryant Foster, the Vice President of Research & Design at Research Collective. Bryant applies the principles of cognitive psychology to the design of new technologies and is a great resource on the topic.
Some highlights of this episode include:
What Bryant does as the VP of Research & Design at Research Collective and what his main focus is.
The differences, overlaps, and intersections between human factors, design controls, project development, and research and development as these processes fit into the medical device industry.
How companies should address the human factors standpoint.
Common mistakes to avoid when considering the human factors.
Human factors in usability and ergonomics as well as the difference between formative studies and validation.
What a task analysis is and how that relates to the overall risk assessment.
The importance of considering the various users of the product or device.
What human factors has to do with post-market usability.
Memorable quotes by Bryant Foster:
“Human factors is where we apply the knowledge of humans to the design of things.”
“Many times a task analysis will point out things that should have been in the risk assessment... it backs up the claim that these work together.”
“The goal of the regulations is to create products that will improve people’s lives.”
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 in companies.
Jon Speer: I don't know about you, but sometimes this topic of human factors can be confusing. How does this relate to design controls and risk and all the other things that we're already doing? Is this yet another thing that has been created by regulatory agencies and standards organizations that is just making life more confusing for med device? Perhaps. But I'm real encouraged because on this episode of the Global Medical Device podcast, I have Bryant Foster, the VP of Human Factors from Research Collective and he does an excellent job of bringing some clarity to this topic of human factors.
Jon Speer: Hello and welcome to the Global Medical Device podcast. This is your host, the founder and VP of Quality and Regulatory at Greenlight Guru, Jon Speer. Pretty excited about this one today, folks, lovely topic of human factors. It has certainly come up so many times in the medical device industry, from FDA guidance documents to standards from IEC, and so on and so forth, and yeah it can create a lot of confusion. Well, good news, we have an expert on the call today, on the podcast today, and that expert is Bryant Foster and he's the VP of Human Factors at Research Collective. Bryant, welcome to the Global Medical Device podcast.
Bryant Foster: Thank you, I'm excited to be a part of it. I found this podcast earlier this year and have just loved listening to it, and kinda rounding out my knowledge about the whole regulatory landscape. It's been great, so I'm glad to kinda be a part of it now.
Jon Speer: Great, so Bryant, what does a VP of human factors do? I guess I wanna understand that a little bit before we dive too deep, but also tell us a little bit about Research Collective as well.
Bryant Foster: Yeah, so primarily we do human factors research, and our main focus these days, is in the medical device space. So we do a lot of usability studies which are kind of the thing that are talked about a lot, and I wanna talk a little bit more about that and how that differs from human factors, or how it fits in the human factors. So we do the usability studies and then all sorts of other research involving humans and people using any type of technology, so it could be nurses or people with different types of conditions and really trying to understand what it's like to be them, and be as knowledgeable as we can about their role and what they're dealing with, to help inform the design of the products they're using.
Jon Speer: Alright, and Research Collective as a company, this is your expertise, this is all that you do?
Bryant Foster: It is. So, we're all a bunch of kind of psychology nerds, if you will.
Jon Speer: Yeah.
Bryant Foster: So human factors people, we kinda come... I won't speak for everyone, but I'd say a lot of us come from two different ways, we're either really interested in technology, so a lot come from engineering or design and then they get really excited about humans and the brain, and so they find that human factors kind of blend these two. Or they're really into psychology, cognitive psychology, or even clinical and then they find that, wow, it's interesting to see how people interact with all the new technologies that are coming out. And so, human factors is really kind of that blend of technology and psychology.
Jon Speer: Yeah, it's interesting, I guess I hadn't thought about it that way, but it totally makes sense. So let's dive into some of the nuances, if that's okay with you. Let's start to give our listeners a few nuggets of information that they can take home today and use in their design and development efforts, manufacture, whatever the case may be. Well, let's really start to understand some of the differences between human factors versus design controls versus product development versus risk management and all those sorts of things where they are different where they intersect and an overlap. First thing that I'm a little bit curious about, from your perspective, is you talked about doing a lot of work in the human factor space with the medical device industry, so there's obviously some sort of relationship between human factors and design controls, but in your mind, how do you see these things being related, or how should they be incorporated with one another?
Bryant Foster: That's a great question, and actually, after I answer, I'd like to hear what you think too.
Jon Speer: Okay. [chuckle]
Bryant Foster: Maybe we look at it similarly.
Jon Speer: Okay.
Bryant Foster: But the human factors really, as I'm listening to your podcast, I hear your conversation about or a couple of conversations you've had about it, and really, it sounds like you see it and a lot of people do as just good design practice. If you're making a product, it's inevitable that someone is going to need to use that and so whether it's a medical device or an automobile someone is going to have to interact with that product. And so in the medical device world and with standard practices for documenting your design process, human factors is really just a matter of showing that you are taking the user into account and hopefully doing that early on, kind of, through maybe some contextual research where you're going and learning about where they work and what they have to work with kind of the space limitations they have, what are they good at, what do they have difficulties with? And then coming up with a product that what meets the needs that you've uncovered, and then getting to the usability testing where you're actually testing that product and making sure that it does what you've intended it to do and that users can use it in the way you intend them to use it.
Bryant Foster: So I think in terms of the design controls and things, it's just one part of it, it really... In fact earlier this year, AAMI released a new document, Technical Report Number 59, and that talks about how to incorporate human factors into design controls. And so I think to... If anyone's interested in learning more about that, that gives some really good specific instructions for where human factors fit in the design control.
Jon Speer: Yeah, that's a really great way to summarize that. Folks, I'll make sure that we provide a link to where you can learn more about the AAMI Technical Report 59, we'll provide a link to that, so you can review that. Again, just... It's always great to try to get context and more information, and figure out how these things are related, and how they flow with one another, 'cause I think the process is really, really important. I think sometimes when people think about this, Bryant... I'll give you my perspective on this. Sometimes I think people think of design controls as this vacuum or this bucket, if you will, and then they think of things like risk management in this separate bucket, and they think of something like human factors in maybe a third bucket, but if that's your approach, folks, frankly you're wrong. You need to figure out how to incorporate these things 'cause they're all related, they're all one flows to the other. And if you heard what Bryant just said, and let me put it in more of a design control view of the world rather than maybe a human factors view of the world, but user needs are so important to the design control process. Your intended use is so important to your product development efforts because these are things that you identify or what you say your product is going to do, and these are things that you capture from what is important from the perspective of the clinician, the end user, the patient, all these sorts of things.
Jon Speer: And as you go through that entire design and development process and you start to define your inputs and your requirements and so on, you need to eventually demonstrate through things like design validation that the product does what you say is going to do. This is really the whole essence of design controls, and human factors is a huge part of that because the days are gone or they should be gone, where we just slap some information and a user manual or instructions for use, and think that we've addressed "usability" just because it's in the IFU. It's really about making sure that people who are gonna use and interact your products that they know how and what to do.
Bryant Foster: I love hearing your more global view of with the design controls and the bigger process I think as human factors people we sometimes fall into the trap of just seeing our piece of it and it really is a small piece. But I think, as you mentioned, it's very intertwined in the big picture and kind of an example that would be something like the risk assessment when it comes to human factors, if someone picks up the guidance who's not too... The FDA Human Factors guidance if they're not too familiar with human factors, it talks about task analysis. And so it seems like we're always having conversations about what is the task analysis, how does it differ from the risk assessment and where... How do they pair up? And so really, as you mentioned, Jon, they should work together. So, the risk assessment is more of an engineering document where you look at all of the potential hazards and...
Jon Speer: Exactly.
Bryant Foster: Bad things that could happen, and then the task analysis is what do people have to do with this thing and then you find where those things meet up and you say, "Okay, if one of the things in the risk assessment is that the device gets water on it, and then one of our tasks is that people understand that they shouldn't put it in the water then what is the potential hazard of that?" and so they really are living documents that should work hand in hand.
Jon Speer: Yeah, exactly, exactly. That's a great way to think about it. So let's... Let's think about a medical device company or manufacturer, and they're already up to their eyeballs, so to speak, in their product development process. Maybe they're preparing for a 510(k) submission, or getting ready to start some sort of clinicals or something like that. With one caveat: They haven't really done anything from a human factors standpoint, what should they be doing, how should they address that?
Bryant Foster: I think to... It's really just to not wait any longer. [chuckle] That would maybe be the first recommendation. And I think one of the things we do quite a bit is kind of a gap in assessment, specifically related to the human factors. And so someone like me could come in and give you an idea of what... Kind of talk about what the end goal is and then decide where should we go in terms of getting ready, but ultimately what we wanna get to is a product that can pass the validation, human factors or usability study that says everyone can use this device as it's intended. And so that usually doesn't happen on the first attempt. And so what can we do? Maybe setting up some small studies, where we start getting some users in earlier in really small, what we call, formative evaluation, where we're just getting it in front of people and really kind of learning, "Okay, what are we at now, what's our baseline?" And then looking for ways to change, either the interface of the device or if it is instructional materials, revising those hopefully way before the product is trying to be submitted to FDA or something like that, because by then it's too late.
Jon Speer: Yeah, and I think that's really important. Sometimes I, Bryant, I share some stories with folks on the podcast of things that happened early in my career where I probably didn't... I probably made some mistakes. And we've all made mistakes and that's fine, but it's what we learn from those. And I remember one of the early mistakes that I made as a medical device product developer, and here it is in a nutshell. I literally got a cocktail napkin sketch from a physician once who said, "Hey I got this design, I got this idea, and here it is," and then he sketched it out and handed it to me. He said, "Can you do this?" And of course, I'm an engineer and a lot of engineers believe that they can solve any problem. And then I'm like, "Of course, I can solve it." And so I did some sketching and I quickly tried to get hands-on and build a prototype and that sort of thing, and I iterated back and forth with this doctor, and we spoke, I guess so to speak, through prototypes. And we got toward the tail end and I believe this particular project resulted in a 510(k) submission and clearance and we went to market, and let's just say it... When we launched it in the market, it landed with a resounding thud. [chuckle]
Jon Speer: And I'm scratching my head, I'm like, "What in the world? We spent so much time, we iterated." And then there was this light bulb that went on, I was like, "Oh crap. I designed a product for one person and I didn't consider the entire user base or enough of the user base to incorporate what others might be using. I got data and input from one person and that's a big mistake." I didn't follow good human factors now, did I?
Bryant Foster: Yeah, and I think that's hopefully what we try to avoid. And another piece of advice, if a manufacturer is going out and trying to learn about their product... So if we could go back to younger Jon and redo this whole thing, maybe you would bring in some representative users and let them use it and then you would follow what we lovingly called a shut up rule, which is just to let them use the device as they see fit, and then learn as much as you can from them, 'cause they truly are the experts in their field, and so even though they don't know your device, they're the most knowledgeable people about what they want that device to do, what they need it to do, what their patients need it to do. And so I think just getting in front of them and shutting the mouth, we can learn quite a bit from those users
Jon Speer: I liked that, the shut up rule, folks. I mean it's hard sometimes because as a person who's passionate about what you do, and maybe knowledgeable about design principles and things like that, sometimes we're very quick to want to provide solutions, but sometimes it's just best just observe and listen, and shut your mouth. So I like the shut up rule. You mentioned something a moment ago, you said the word "formative". And I know there's these different terms that I'm gonna confess, I'm not 100 percent clear on these, but can you maybe take a few moments, and talk a little bit about differences between formative versus validation versus summative usability studies?
Bryant Foster: Absolutely, my favorite topic. So I'm actually gonna back up just a little bit more even and talk about human factors and usability, and even ergonomics. So on one of your previous podcasts, you talked about how they're used interchangeably. And unfortunately for our industry, we have not done a very good job of defining these or marketing ourselves. I think a lot of times, people don't even know how to refer to what we do because of the different names. And so the way I see it, human factors is the overarching practice, and it's where we apply knowledge of humans to the design of things. Ergonomics, and this is different in the US versus Europe, so that makes it more difficult, but ergonomics tends to typically be, at least in the US, more related to physical products and physical design. So things like posture and the way we interact with our hands and things like that. So that's kind of more ergonomics. And then usability is simply a method that is used within human factors to understand how people interact with a specific product or technology.
Jon Speer: Yeah.
Bryant Foster: Usability testing. So we get it in front of them, have them use it and then try to make it better
Jon Speer: Okay.
Bryant Foster: So usability testing then has formative tests or evaluations and those are what we do early on. So you might do that with a prototype, it could be a very simple prototype, maybe if you have a graphical user interface, you might put that on a tablet, and just mock it up really quick, and have people go through tasks with that and hopefully learn what their ideal model is for that interface. And so, really formative evaluations or formative tests are to learn. We're getting things out and trying to learn and iterate to make the product better each time. At the end of a formative or at the end of all the formatives and all the iteration, you get to a validation usability test and sometimes that is also called a summative. And summative really means, it's kind of like the culmination of all your formative studies, and so FDA has started using validation more, but really, validation and summative are used... Kinda interchanged, but in a validation or a summative, we are not looking to learn anymore, we're looking truly to validate that the device can be used by those users in representative environment of use. And so in a study like that, we're just really setting... Setting up a scenario and then asking them to use the product and hopefully, if our design is what it should be, people can use it correctly. And we were able to validate its design.
Jon Speer: That, that was a...
Bryant Foster: The usability of its design, I should say.
Jon Speer: Yeah, that was just a really great way to explain that. So thank you for that. And folks, just to remind you, I'm talking to Bryant Foster from Research Collective today on the topic of human factors, just to give you some perspective, you should go check out their website, website is research-collective.com, you can learn a great deal about the organization. But Research Collective does a great deal of work of hands-on experience, specifically for medical device companies, their team of researchers are master degrees, PhD degrees in human factors and cognitive science, they have significant experience with research design fielding and advanced statistical analysis and reporting. Their team, the team at Research Collective, provides user research in the field, you can do it online, in their in-house usability labs. They also do a lot of work with recruiting and research participant management and they work with some of the most innovative medical device companies in the world.
Jon Speer: So you wanna check out Research Collective. Also a note: Research Collective will be doing a webinar on the topic of human factors in January 2018 with Greenlight Guru, so we'll provide information about that as well, but do check them out. So Bryant, let's... You talked a little bit about this a moment ago, It might be worth spending a moment or two going into a little bit more about this but you talked a moment ago about task analysis and risk assessment, and all those sorts of things. But let's maybe spend a moment or two going into that, provide a little bit more explanation about that. What is a task analysis and how does that link to the overall risk assessment?
Bryant Foster: Sure, so the task analysis is a user-centered approach to the design of the product. So what are all of the tasks that a user or multiple users would need to do in order to complete or use the device, correct? So that would involve anything from unboxing it, maybe unpackaging, setting it up, using the device, and then any tasks they need to do and then ultimately, maybe disposal or storing it; maybe there's reprocessing involved or cleaning and so all of the tasks that users need to complete should be accounted for. And what we do is we kind of document all of those tasks and then we look into... We go deeper. And so simple tasks like, if I can think of one that would be a good example. Maybe it's turning the device on. So maybe there's a power button you need to turn it on. Well, that seems like a really simple task, but maybe there's some subtask that people need to do in order to do that, so maybe they have to first pull it, pull out of the box, they might need to access the power button and then actually press the power button. They have to know where the power button is, they have to know that it has to be turned on. And so there's the elements within each subtask that we refer to as the PCA elements which are the perception, cognition, and action elements of any task.
Bryant Foster: So like I was mentioning with the power button, you have to perceive the button, you have to see it, you have to cognitively understand that that is the power button and then you have to physically be able to press it or turn it on. And so, it seems really overkill maybe just as I'm talking about it. But when you look at a task in that way, and that's a really simple task, other tasks are obviously much more difficult. But when you break a task up like that into its perceptual and cognitive and action elements, you can then see when you watch people perform that task and say they have difficulty or they're unable to power on the device, you can then say, "Okay, why were they unable to power on the device? Was it that they didn't see the button there? Was it that they didn't know that that was the power button? Maybe the label on the button was not something they were used to, and so they didn't know that was the power button? Or maybe they physically weren't able to do it, maybe it was kind of recessed in their finger, they have a bigger finger than normal, and they couldn't get it inside to press the button." And so then, as researchers, factors... Scientists will look and say, "Okay, what is the root cause of that error or difficulty? Was it one of those PCA elements, the perceptual, cognitive, and action?
Bryant Foster: And then wink it back to the design and in that way, hopefully inform the design to be better, so maybe bring the button out so that everyone with all finger sizes can actually press it or change the label so that the power symbol looks like what people would expect for a power symbol. And ultimately, each of those tasks has a hazard associated with it, so people don't need to, don't know how to turn on the power button or aren't able to turn on power button. There's some associated hazard and that's where we pull in the elements of the risk assessment to say how critical is it that people are able to do this task? And so it's kind of a lot, but...
Jon Speer: No, but a lot... It makes good sense, because I think sometimes people table their risk assessment activities, again, I think this is a conventional wisdom or conventional approach that I think we, as product developers need to change, sometimes we save that risk assessment for a later stage activity but that risk assessment needs to be... And I've... Folks, I've written about this, we probably had a few other podcasts, we've probably done webinars on this topic, but risk assessment is so invaluable to what you're doing in the product development step, it should be started early on in the product development process, even if you don't have a prototype, but you know what you're trying to achieve, that risk assessment can add a lot of value, and as Bryant has just described it's very important as from a usability, from a human factors standpoint, so that it's... You're basically addressing design features and using that as a tool to mitigate those risks throughout the entire process.
Bryant Foster: Yup. And we definitely... The risk assessment is what we use to then fill out how important these different tasks are, and it's unfortunate how many times our task analysis will point out things that should be in the risk assessment that aren't there. And so... And I guess that is unfortunate, but it also, I guess backs up the claim that these things work together. So if you do a good task analysis, you'll be able to make sure your risk assessment is complete and vice versa, your risk assessment can help your task analysis.
Jon Speer: Yeah, and I love that approach too, because sometimes I think when we do risk, we're down in the weeds, we're at the sub-sub-sub-subcomponent level sometimes looking at this individual item and what happens if that fails, or does this or does that? But we need to, in my opinion, we need to bring this, our risk assessments up to a much higher level and think about how this product is being used. I love when you describe that from the moment that the product arrives, at point of views and it's taken out of the box and this happens, and that happens, and so on. The other thing that resonated with me as you were describing that is the person who is going to use the device for the procedure or whatever it may be, might be different than the person who receives the product and takes it out of the box. And so there's different users that we need to consider as part of human factors, depending on what the tasks are.
Bryant Foster: Absolutely. And each of those users has a different mental model, so the person in receiving, in the hospital has a different kind of perspective than the position who's using it upstairs, and then even different from the person who's in the reprocessing department, who's cleaning the device after use. And so making sure that those users are kind of thought of as the process that they need to complete is being designed is really important.
Jon Speer: A really good point, really good point. So far, most of what we've talked about is human factors really in the scope of design and development, but... And that's important, don't mishear me, but there's also important aspects of human factors and usability that should be considered once we go to market, once we launch. Talk a little bit about what we should be doing with post market... Monitoring post market usability.
Bryant Foster: Yeah, and actually, you had made a point of this when I saw you earlier this year and I... It's having me thinking about it ever since. So a lot of our, I guess, not a lot of our clients, but sometimes we hear people or manufacturers just hoping to get through their usability validation and what they're doing with us but they really just wanna get their device on the market, and then that means that everything will be okay from there, but that's obviously not necessarily true. And so the manufacturers are definitely responsible for keeping an eye on all of the post market kind of risk or hazards that are happening, and some of those will link back to usability. And so I think it's just definitely a good idea to keep that in mind and if those things come up, I would recommend getting out ahead of it and not wait until FDA recognizes it and asks you to do it or potentially pulls your device off of the market.
Jon Speer: Yeah, and it lends itself very well that this mentality, this approach lends itself very well to the intent behind risk management. Risk management is intended to be a total product life cycle process, so it doesn't stop, once you launch, it continues on throughout that entire product's life cycle. I also am a big advocate of what I refer to as living design controls, I think sometimes we get... We have a preconceived misconception that... And part of it is the industry or the regulations don't do us any favors because they use terms like design history file, so that implies moment in time, something that we're gonna archive, but I'm a big fan of living design controls. Your product, it will evolve over time, and it also... The other area where I think that Greenlight is very focused on through the postmarket workflows of our software, this being able to manage and monitor quality events, but not just reacting to situations, waiting for things to happen. Another way to thinking about it is sometimes we get stuck in a rut where we wait for something and then we correct it afterwards. Really what we're talking about here is being proactive and preventive in nature, so that we can solicit feedback and information about our products and learn about what's working well, what's not working well, opportunities for improvement before they rear their ugly head and become real big problems.
Bryant Foster: That's great. Yeah, I think the overall intent of these regulations. And this is a theme I hear in your podcast and I really love it, but it's... They're the base rule, the whole goal is to make products that are improving people's lives and whether that's through usability or anything else, we're just trying to ensure that what we're giving people is truly doing what it's supposed to do, and either making them better, healing them or helping them with a better life, and so once it's on the market, that's when we can really see if it's doing that and should keep an eye on that.
Jon Speer: Yeah, for sure. Well, Bryant, we've covered a lot of ground on our conversation today on human factors. Any parting thoughts or words of wisdom before we wrap up today's podcast?
Bryant Foster: No, I think this has been great. I've enjoyed talking to you. And I like your perspective on this, it's kinda helpful in seeing where we fit within the larger process, so it's been a good conversation.
Jon Speer: Alright. I wanna thank my guest, Bryant Foster, the VP of Human Factors at Research Collective. Again, go check them out research-collective.com and be on the lookout for that webinar that's coming up with Research Collective and Greenlight Guru in January. Let me tell you a little bit about how all this topic of human factors, how it relates to Greenlight Guru and our exciting pre- and post-market workflows that are helping medical device companies over the world manage their Quality Management System as well as get products to market faster and safer. Yes, we have workflows to help manage design controls and risk management. And as you've learned today from the information that Bryant has shared, those things, human factors dovetails and really flows with company's product development process and becomes critical information that we, as product developers, absolutely should be on the quest to have and find and know before we launch our product to market.
Jon Speer: And then with Greenlight Guru's postmarket workflows are there to help you stay ahead of things, manage information and events that you're learning about your products and processes so that you can always improve and make your products even safer and even better than they already are. If you wanna learn more about the Greenlight Guru software platform, I would encourage you to go to www.greenlight.guru, click on the blue button, request more information and someone from our team will be happy to share with you more about our software platform.
Jon Speer: Again, this is Jon Speer, the host, the founder and VP of Quality and Regulatory at Greenlight Guru, and you have been listening to the Global Medical Device Podcast.
ABOUT THE GLOBAL MEDICAL DEVICE PODCAST:
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.