Inspired by our annual Industrial Design/Human Factors newsletter, our QA/RA. Project Management, EE, ME experts, President, and ID/HF Manager shared insights and examples of how ID and human factors affect medical commercialization. We hope you enjoy their stories and benefit from their insights.
Talk to users of existing versions of the product and observe them at work.
We were working on an intravascular ultrasound system. Our human factors people talked to nurses who actually worked in that area and learned an amazing amount about what they didn’t like about the current state of the art device. One of the devices was sitting wrapped up in the corner because the nurses didn’t like the way that it constrained their workflow. That was a pretty remarkable insight.
That product was successful in the end, but it took a lot of investment from the company developing it. Luckily there were other features that they took for an update that helped the new product gain market adoption. It took a lot of effort and was a very high risk that could have failed because of the workflow constraint issue identified by talking with users of the current version.
Another example was a doctor with a product related to demyelinating neurological conditions. The doctor asserted that he sits opposite his patients and he wanted the patients to see one side of the instrument while he views the other side of the instrument. Our trained observers spotted that he sat down beside his patient and worked collaboratively looking at the device from the same side. We ended up making a product with flexibility to work beside or opposite the patient. That’s the importance of going and seeing in the early stages before locking a design in.
Understand the Use Environment
We were working on an infusion device used while flying around in air ambulances to understand the constraints of that work environment. We realized that the things that the users were unhappy with were quite different from what the early discovery had been talking about. There are so many times we’ve gone out in the field and discovered big changes from the client briefing document. What we learn in the field is much more constraining, or has more potential than what was originally briefed.
We once designed a stereotactic device with a feature that used a screw that went clockwise at the recommendation of our client. When we delivered the prototype device and the client used it, they came back to us and said, “In all other devices that we use this feature actually goes the opposite way and not how we designed it.” That would be a huge risk for user error because surgeons instinctively would use the device incorrectly. We had to redesign the screw to turn the other way around in line with the user experience and expectation.
I just got off of a call where we were mining the client team for human factors and ergonomic information to do with a clinical study they’re on right now. They were talking about the information that they needed as they were placing a probe on the patient and how that might be displayed on the screen. We talked about this for a little bit of time. But later in the conversation, they pointed out that you actually can’t see the screen in an O.R. because it’s too far away. Even people who are using a device a lot are able to carry all these contradictions in their heads. We realized that we are going to have to solve this problem a different way than just putting a bar graph or a number on the screen. We may have to invent something new in order to allow them to check their placement on the patient at a distance.
If you don’t have a detailed conversation about human factors and industrial design, it’s easy to just get lulled into something and then later find out that all that great stuff you put in isn’t actually useful in its form.
Value of formative testing
We conducted summative testing on a device. Every procedure that we had run through a formal formative testing, passed. Everything that we changed which didn’t go through formative testing, failed. When you follow a process that includes formative evaluation of early concepts, detailed design and alpha betas, you will have a very successful summative testing. This is a perfect example of how Human Factors and ID impact medical device commercialization.
It makes Good Business Sense
A startup really cares about being valuable. Generally, adoption speed is dramatically affected by human factors and industrial design. That adoption curve is going to be one of the major drivers of value for a startup company exit. 25 to 50 percent of the exit value of a company is driven on that adoption curve and in turn by the human factors and industrial design. Industrial Design ensures delivering a product that people want to use that fits into their workflow. This helps the product adoption go smoothly and successfully. It’s gigantically disproportionate.
Human factors is a regulatory constraint
You need to make sure you have identified user risks due to usage and mitigated them so the product doesn’t fail regulatory approval or due to usage risk after launch.
With Class II Devices, you need to show regulators that the device is just the same as the predicate device. At the same time, it has to have a revolutionary element to be worth changing. There is a huge aspect of Industrial Design and Human Factors in bridging those divides.
Human Factors and Industrial Design for safety purposes is required by the regulatory bodies. We had a plan for how to clean the device and we took that through formative testing. It passed and we made some adjustments, but it worked. And then we had some issues with verification testing, so we changed the process and didn’t go back through formative testing again to address those changes. When it hit summative, there were usability issues related to that change. We then had to resolve and justify all those issues so it was not a barrier to regulatory approval.
One device had a really interesting journey, probably one of the most complex ones we’ve had. There aren’t many medical devices that you have to worry about operating during winters on an airport runway as well as a surgical suite. But this was certainly one of them. For example, we were trying to push the device into emergency at Health Science Center in Winnipeg and slush on the sidewalk was jamming up the wheels.
Our ID and HF team collected lots of client and KOL feedback and created a prototype made of plastic, cardboard and lots of plywood. Sometimes seeing a physical embodiment is really important. We often work virtually with lots of soft tools. But until key opinion leaders see a physical thing that actually demonstrates what the device will have to do, it’s hard to know what’s going to work and what’s not. Luckily, our KOL was explicitly honest about all the features and functions of the prototype device and the workflow that were not going to work.
Often somebody says something like “it’s a transplant device” and you think “I have to have surgeon input.” That’s the first thing you think of. We got input from the ambulance drivers because they’re a part of the workflow. We went to Calgary and engaged with the guys flying little Lear jets. Their jets are tiny. They’re actually microscopic and it’s really hard to fit large items into those spaces. We actually had to dismantle the device into a bunch of different pieces. It’s so easy to get distracted, “it’s got to be like the right height for the surgeon”. It’s got to fit into this sort of institutional environment really well, which it does. But it’s quite short sighted to just think about that environment exclusively.
I like this example because there were so many different stakeholders in the process. We actually went over to the airport before and brought them some donuts so they let me do stuff to their ambulance, which was fun. That was a pretty big eye opener, too, in terms of what we can and cannot do from a device perspective, like securing it when they drive like maniacs through town, which they love to do. I remember on our way to the hospital in Edmonton, I saw the nurse that was with me, and her eyes were wide open as we were just ripping down a highway full tilt. And I was like, “Do they always drive like this?”. She’s like, “No, we don’t need to drive this fast.” The realism of it all was kind of fun.
That’s just one of many different stories that brings the importance of it’s not just enough to take people’s comments at face value. You’ve really got to get in there and understand the insight and how human factors and ID will impact the medical device success
Key Opinion Leaders (KOLs)
We went into procedural observations for epidural installations on women that are pregnant. Generally, delivery is the primary use case for epidurals, where the drug that reduces the amount of pain during labor is put directly into the CNS. This method avoids pumping drugs that can slow a baby’s heart down.
It was really funny because the KOL was telling us how they execute the LOR (loss of resistance procedure) to find the epidural space. And so we went in there. “Our first patient is really good. She’s a nurse from Squamish. Totally cool with you guys being there. Everything should be great because she’s used to medicine.” And she’s screaming at her husband and swearing at him. And he was just getting his hand crushed by her. And I remember thinking, “this is quite intense”. And then the KOL just jumped in there, didn’t even do the LOR because he’s done thousands of these things. He just gets the drug and comes back to us, “Sorry, guys. I kind of had to jump in there. I promise you, the next one will be the way that you want me to do it.” It was sort of an eye opener in terms of how much people follow rules and do their thing.
This example led to a short discussion within the group: “It’s such a dynamic. We’re going to develop your product and it’s going to work just the way it’s supposed to. The “supposed to way” comes up from time to time. But we’re never actually going to use it because your device in the “supposed to way” of doing it takes too long.
And even after you observed him doing something where he specifically did it counter to the way he did. This little thought that the way he did it, he normally does it, would be the next time. And was the next time the normal way he did it?”
He did. I think he was careful about who he picked for the next one. She wasn’t really in hard active labor at the time. “This one’s for sure going to be good. And so we went up there and she was relaxed.” It’s the opposite of most product companies’ problem. How do you get the KOL out of pleasing mode?
That’s the challenge when you’re just asking people pointed questions or doing cognitive walkthroughs. Cognitive walkthroughs are great to give the lay of the land, but they’re not often exactly representative of a real life situation. And you have to be able to understand both scenarios in order to make good decisions.
KOLS are very important. It’s very important to select the right KOLs. It’s also good to talk to more than one KOL because they may do things in a way that is not adopted by the rest of the medical community. This is hard to figure out if you haven’t talked to more than one person. Make sure you actually capture the real world performance and experience and not just somebody’s very particular view on how things should work.
We were working on a proof of technology study and were focused on the technology only to determine if it physiologically works. To accomplish this, we needed to build a little device to get out and conduct some clinical studies. We quickly threw together a prototype with a handle. There was no ID or HF design work. We thought, that’s okay, usability isn’t really important at this stage”. The first person to use it said, “we’d need three hands to use this”. That’s an example of a situation where we skipped ID/Human Factors intentionally and clearly the device couldn’t be commercialized as is because it was technically unusable.
The ID part of the ID/HF equation
The ID part of the ID/HF equation is really important from a commercialization process perspective. Making stuff look good, feel good, giving it the right form early on, or at least understanding what that will be. Going through that thought process of what should the embodiment look like, even at a proof of concept level so that you start planting the seeds in people’s heads. Lots of other companies think of the ID process as just the skin. And that’s what it ends up being if you don’t think of it early enough. Because that’s all that’s left. All the guts have been figured out, the structure has been formed and now all you can do is pick the colors and the reds for the corners.
But if you think about it early enough, then you have all the flexibility you need to make a form that is perfectly suited for the use cases. And usually when you think about it early enough, you end up with something that is highly attractive, highly useful, inviting to the user, limits use errors and so on, purely by its form. It looks exactly how the user interfaces with it. And these are all things that only come from a good detailed ID HF approach to product development.
We hope you enjoy our stories of how Human Factors and ID Impact Medical Device Commercialization. We would love to hear from readers about your experiences.
Astero StarFish is the attributed author of StarFish Medical team blogs. We value teamwork and collaborate on all of our medical device development projects.