Getting a Headstart in Human Factor Principles with HE75

In this interview Dr. Edmond W Israelski, Director, Human Factors, Abbott Quality and Regulatory, Abbott, joins Andrea Charles from Pharma IQ, shares his top tip for successfully applying human factors in design controls during the design of a medical devices and how to avoid common pitfalls in design.  


Pharma IQ: Human factor principles and practices are playing an increasingly important role in medical device design, as manufacturers are recognising the benefit. What would be top of the agenda for the Association for the Advancement of Medical Instrumentation Human Factors Committee in 2011/2012?

E Israelski: The Committee just met recently and we planned our 2011/2012 activities, and so I can tell you very timely information. At the top of the list is we as a Committee, as human factors professionals, will be submitting comments to streamline the international standard IEC 6366 on usability to the IEC and the joint working group on ISO and IEC that is responsible for that standard. It’s coming up for period review since it was issued in 2007, so now is good timing to look at the experience that people have had using that. And we’ve had a few comments from around the world, but mostly members of our AAMI Committee and from the FCA who are looking to streamline and update those standards. So that’s number one. Number two is we’re looking to issue standards in a few other areas that are new. One is applying human factors to the design of Healthcare Information Systems, Healthcare IT, or e-Health. A third area is looking at standards or Catholic Information Report, or post-market surveillance, looking at human factors complaints, and other issues regarding the user at the face of a product and how to analyse trends, collect that data and then use it to make improvements in subsequent releases of that product, or to make on market improvements, so basically can plan analysis from a human factors perspective. And then we have a third activity after that, or I guess I should say a fourth activity, which is looking at the upfront user research area, so-called conceptual encanteria, that will give industry more guidance on how to do that more effectively, and so again that might be a standard or a technical information report in the area of conceptual encanteria or upfront user research. Those are the top items with that AAMI Committee.

Pharma IQ: What has been the impact of the latest standard?

E Israelski: Well, it’s very early. You’re referring to HE75, the standard that is basically design principles for designing medical devices which is this very long standard, almost 500 pages, which is physically an encyclopaedia of information that experts on the Committee have put together, tried to condense and make in a digestible form for manufacturers. They have a good starting point in their design, so they don’t have to go in, go out and do all this digging and the research about that makes for a good display in terms of contrast or brightness, or what makes for a good alarm in terms of frequency and cadence. But all this basic information that we know about human capabilities and limitations, that’s been captured from the vast human factors literature, and put into HE75. So I think we’re seeing companies that don’t have professional human factors people being able to get a good head start by going to the standard and being able to extract from it useful information that will be good for design inputs. Again it’s only been out less than a year, so it’s early to be able to make any definitive statement, but what we’re seeing early on is that people are in fact taking advantage of this knowledge that we put into the standard, and that in conjunction with good human factors process, such as the standard we have that preceded the HE75 before, by combining the design principles with good process, that we are hoping that we will see better medical devices that are safer, easy to use, and have greater market success. That’s the basic answer, is that it’s a little early to tell.
[inlinead]
Pharma IQ: From your experience, what the common pitfalls associated with the design of medical interfaces, and how can they be avoided?

E Israelski: Well, I think some comment on pitfalls are depending on doing market research is the only way of understanding the user’s need in terms of the actual interaction with the device. Market research is of course very important: you need to know about features and what customers want, but human factors of course is studying the behaviour, and studying it very carefully and scientifically. And so the a pitfall is sometimes convincing yourself that as a design team you have a lot of experience in a given area, that working; that type of device for many years you know it all; you don’t need to go out and do additional research with customers, because you feel you understand things, so trying to short circuit good human factors by thinking well, we do market research, we have clinical evaluations, and the new regulatory environment, certainly in the United States I think, is starting to ratch it up around the world, is that that’s not enough any more: you really do need to do good systematic human factor. So a pitfall is convincing yourself you don’t need to do it, or only doing it in a very superficial way. Another pitfall is that people think well, if I just do one usability test before I release the product, I’ll capture everything and that’ll be good enough. And in the new world the regulators don’t see that as being enough.

Pharma IQ:Yes, the study and the sort of human behaviour and interaction is key, and you should never second-guess what a person will do.

E Israelski: Correct. And the only way to do that is to study it scientifically: do, you know, a whole lot of disciplined approach of gathering data through usability tests, right upfront user research task analysis, doing very careful risk analysis, systematically looking at each and every task and doing a Thalia modes and effect analysis on faults in terms of users not being able to complete a task, committing what we call use errors, and analysing the probability, looking up some bits of data you have, or from your knowledge, and then of course examining the impact of that on harm and hazards that result, and systematically prioritising them and designing to mitigate.

Pharma IQ: So how much time would you say is spent sort of minimising use errors?

E Israelski: Well, that’s the problem, some companies spend a lot of time if it’s a high-risk product, but many companies kind of give it an after-thought just before release. They realise they need to do something regarding usability and understanding a particular use error so that those companies unfortunately will get into more regulatory difficulty in the future. I don’t know but some about how much resources they spend on usability was about 10% of your RD budget. Many companies don’t achieve that, 5% would be a reasonable number. So if you look at who’s working on the user interface, who’s evaluating it, doing upfront research on it, a good percentage of the total resources you have in the development includes those people who are looking at user interface and usability-related issues. That should be 5% or 10%. That would be a goal to strive for. Of course if the product is in new ground, new technology, something you really haven’t done before, you may just spend more than that 10%, and of course if it’s a very mature technology and you’re just making incremental improvements on a well-known, proven to be safe user interface, then you of course don’t have to spend as much. So we like to say that the manufacturer’s process should be scalable.

Pharma IQ: And what would be your top tip for successfully applying human factors in design control during the design of a medical device?

E Israelski: Well, that’s a good question. Our top tip is integrating human factors throughout the process. So right from the beginning, analyse, upfront user requirements go out with your contextual enquiry, understand users’ needs to profile the use environment, all the tasks you’re going to do, do early risk analysis, so putting it altogether. And so I would say making sure you should have trained human factors people on the product team from the beginning, don’t just bring the consultants in at the end. Ideally you have trained professional human factors people on staff. If you don’t, at least have people who are sensitive to that area, understand what is good human factors in usability engineering, and then they can bring on external resources that are professionally trained in human factors. So either you do it in-house, or you use outside consultants, but the tip would be use them early and often.

Pharma IQ: Are there any sort of new, exciting technologies out there that are helping you and your colleagues create user interfaces for medical devices?

E Israelski:I think there’s a lot of new technology that can make these interfaces better, more intuitive, less prone to user errors. Touch-screen technologies that have been advanced and we now see in our smart phones and in tablet form, such as i-Pads or Zoom tablets. These technologies, you now, they make the user interface better, more intuitive. You’re interacting, you’re flicking, you’re using gestures to do things like magnify, like using two touch-points and squeezing or expanding. We see that children as young as one year of age can interact with an i-Pad very easily. It’s these interfaces that they’ve made intuitive, and they are. At the other end of the spectrum people who in the past have kept away from computers, senior people, are now seeing the advantage of using user interface technologies with gestures and touch, they interact with them even better. There’s a small role for voice technology speech recognition, but I know in the medical field there’s lots of challenges in terms of background noise, and how do you correct your mistakes. So we see a little bit of that now. But the fact that you can have portable information at your fingertips, applications that work on smart phones, and i-Pads integrating with medical devices that are connected, so the fact that we have such great connectivity, high-speed: those are all I think the user interface and of course baseline technologies that will just make the user experience with medical devices, and the network of medical devices that much better. They get the information they need on time, accurately, no guessing, intuitive, very straightforward, smart software that could correct mistakes, see issues, alert you to them, like drug overdoses, or concentrations that seem inappropriate risks. We never see this in infusion pumps now, they all have smart technology, but all of that I think has a bright future in terms of getting even better and smarter. And the issues affect technologies as we see the smart phones and tablets I think will enhance that and make it even better.

Pharma IQ: You’ve sort or answered my next question there in terms of what’s next in terms of tablet and touch-screen in the immediate future, but if you had a crystal ball, what would you see maybe, say, five, ten years down the line?

E Israelski:Well, I guess I would see smarter, more adaptive user interfaces that understood you and your capabilities by looking at your history and interactions, predicting where you might make a use error in terms of it. If you are predisposed to making digit transpositions, which some of us do, we don’t have the top skills in terms of keeping track of a string of numbers; some people are more prone to that than others. You can imagine a user interface being adaptive and smart, having a sort of a learning curve, a neuro network-like capability that would be able to see this, and then slow you down for those parts of the user interface work seeing that you have in the past made mistakes, and ask you to do those more deliberately, and to double-check a few times, give you additional confirmations; maybe have you do things verbally that you can’t do with gestures or visually because it sees from its history interacting with you that you do certain things in one modality better than in other modalities. So continuing power as it gets better and the memory gets cheaper, you can imagine that we know more of these things can apply. And then of course gestures interfaces. Now we’re talking about with i-Pads and i-Phones gestures on a surface, a touch surface, but as we’ve seen with the gaming industry which so often leads the way in user interface technology, gestures of the whole body and hand movements. You think of the Mbox and Microsoft Connect, which people love for gaming. But those sorts of things may seem a little farfetched as serious interfaces for products like medical devices, but some level of incorporation of that may be very helpful. It would sense frustration where you don’t settle, or it may not be clear you have high confidence in what you just put in. Look at your facial expression, and it might say are you sure this is the programme you want to start? Because it could be smart enough to analyze that there may be some uncertainty and confusion in the way that you interact with the system either in terms of facial gestures or body gestures. So shoulder-shrugging and things of that sort could be the test. So I’m going down the road ten years, but that’s looking. 

Pharma IQ: Yes.

H Israelski: Yes. And of course speech technology gets better and better incrementally all the time. There’s been no big breakthroughs, but speech recognition has been around 20 years or more, and it has got better. We all probably experiences of interacting with airline reservation systems, and others where in the past you would repeat things over and over, and that’s no longer the case. Now you’re getting smarter and better. So that’s a technology I can keep an eye on down the road by getting infinitely better. And you combine that with gesture analysis. You know computers could get smarter than us.

Pharma IQ: Finally, I know you’re going to be speaking at the 14^th Software Design for Medical Devices event, taking place the 23^rd to the 25^th May in San Diego. For anyone interested in attending, what would be you key take-home message?

H Israelski: Well, I think it’ll be a great conference, because you’ll learn a lot from people from a wide variety of disciplines, but in particular for human factors, which is the area that I know most about. I think you’ll hear that if you’re not familiar with all the intricacies of human factors, come, attend the session; you’ll learn more. I hope to say a little bit more about the evolving nature of the processes in human factors that regulators are now saying they want. I mentioned right at the beginning that we’re looking at upgrading and revising the usability engineering and the process standards. A lot of that is based on maturity in the field of human factors, and the fact that regulators are raising the bar and expecting a lot more. This has been happening slowly over the last 15 years, and I believe it’s accelerated in the last two, and that’s witnessed by the fact that recalls are up, the FDA in the United States and I think international regulators are following suit, are issuing more guidance and standards in this area. And more is coming. I know that the US FDA is re-writing its guidance on human factors. So I would say come and you’ll hear about some of this, and in that way you’ll be prepared and you won’t be surprised.