EBME Forums Forums Free discussion Out of Hours Technology Medical Device Usability and Human Error

Hi,

I am working on a project looking at usability issues with medical devices and human error that is influenced by device design. For example, sending an email without an attachment, recording the wrong TV program, and assembling flat packed furniture wrongly are examples of human error which could potentially be improved through better design.

I am a researcher new to the medical domain and so would love to hear more about:

1) Device design interaction issues you are aware of.
2) Stories of human error and near misses that you think were influenced by device design.
3) If you consider usability, Human Factors or HCI in purchasing? And if so how?

Apologies if some of these questions appear a bit naive... I look forward to learning more about these issues from practitioners over the course of this work, and if the project generates more discussion about these issues then all the better .

The project is called CHI+MED and a link to a short film I've made for it can be found here: Man-Machine Nightmares

Thanks

Welcome to the forum, Dominic.

All that sounds very interesting. And yes, I reckon you've come to as good a place as any to secure feedback on the subject (ie, this forum).

It must be nice to have funding for a six-year project!

I've just enjoyed the video. It's nice to see at least that infusion pumps are (generally) better designed than microwave ovens (what does the Chaos button actually do, I wonder)!

I'm pleased to see that you've homed in on infusion pumps.

There was some work done by the US Air Force a few years ago which looked into the possibility of standardizing pump interfaces. Alarm tones in all medical equipment, but especially pumps and patient monitors, is another obvious area. Not to mention alarm fatigue!

Defibrillators are another area worth looking at. I would say that most modern ones have the design problems more or less sorted out. And so they should, by this time!

On a slightly different tack, I see a lot of older equipment myself, wherein it is often easy to see (once you open them up) where better design decisions could have been taken. Broken parts that could (should) have been more robust, etc. But (as my long suffering colleague has been known to remark) ... isn't hindsight a wonderful thing?

Also, there have been a number of items (of medical equipment) which have truly evolved over the years. That is, with each new version building upon the solid design of the original, and adding sensible features as the technology developed. The Ivac (Alaris, whatever) P-series syringe pumps are an example which spring readily to mind. And, conversely, many others for which time has seemingly stood still (the MS-16A ambulatory syringe driver, for example ... a subject for "research" if ever there was one)!

Dominic,
Apart from general equipment management one of my roles is medical device training, so people using equipment correctly is high on my agenda. I also have a personal interest in ergonomics in general so I am glad to see some serious work being done on this.
Have you seen the recent publication aimed towards pump manufacturers telling them what we, the end users, want and what we think is a good idea? I thnk it was the NPSA who published it.
Happy to talk more about this subject.
Contact details in my profile.
Robert

Dominic, bear in mind also (and as I have said a few time before - but you probably haven't seen it) that most medical equipment is already well designed (it has to pass through the FDA, and all the rest), and that clinical staff are not inherently stupid (far from it, in fact). But it is the conditions* under which the stuff is used that is the real "problem"!

Think "hospital design", "the hospital as a system", "design in the round" etc. Whilst, of course, avoiding consulting architects at all costs!

However, if your focus must be kept within the narrow framework of equipment per se ... then surely the keyword, as far as operator->equipment interfacing is concerned, must be:- "intuitive"!

And, lastly, you may not realise how rugged medical equipment needs to be. In short, stuff gets dropped (sometimes a lot)!

* Stressed out, chaotic, dramatic, rushed, over-worked etc., etc.

Dominic, You may want to take a look at the NPSA document 'Design for patient safety - A guide to the design of Electronic Infusion devices' Ed1 2010. It can be accessed from their website. I think this may the document that Rojo was writing about. Lots of info' in there that is relevant to what you need.

Link

See also the further links offered at the References section towards the end of this nice document. Including (for instance) this one and this.

Thanks for the tips and pointers - really useful .

We have a project meeting at the beginning of October and one of the items on the agenda is to discuss the NPSA's new Design for Patient Safety booklet. We have been looking at this already and hope to be in a position to contribute to a version two sometime in the future (maybe?) (we have a representative from the NPSA on our steering committee ). It seems like they have done a good job, both in content and clarity. Thoughts on this would be useful to feed in to the meeting if you have them. I'll have a look at the other refs.

Glad you liked the film Geoff. Here is an explanation of the 'chaos button', which was found by another friend who was intrigued when they found one on their hotel room's air conditioner! Apparently the latter uses choas swing technology . All sounds very clever but perhaps not immediately 'intuitive' to a user .

Pointers to other equipment to look at e.g. MS-16A ambulatory syringe driver and defribulators are really useful... I'll have a look at the other threads and the one on alarms.

I am definitely interested in the evolution of devices. I've heard of patients sharing knowledge of how to silence their own infusion pumps after an occlusion alarm sounds (typically from bending their arm) even though they're not allowed to use the pumps themselves. They watch the nurses and copy them which is bad. Also annoying for nurses having to manage these often trivial alarms. Apparently the Asena pumps sound, then retest the line, then stop the alarm and continue infusing if it has cleared. This uses less nurse time and means patients don't break the rules. Clever.

I'm glad the role of context was raised as investigating clinical situations is more my part of the project. Other areas include device design, cognition and stakeholders. There is always a focus on device interaction but I also look at understand the work around them. I have recently finished a small stint at an Oncology Day Care Unit and still doing the analysis. One nurse said it's annoying when a pump runs out of battery part way through a long treatment. This rarely happens as nurses are aware of this, and the interface shows the battery level. The device alarms and counts down when the battery is low so it can be changed. However, I wondered why the pump doesn't tell you it won't last if you program a 2hr infusion and it only has an hour of power left?

Our perspective definitely doesn't consider people as stupid. My film tries to convey that we all make mistakes, everyday: To Err is Human. A nurse I spoke to acknowledged this and said her mantra was "Check, check, check." People are often very clever at monitoring their own activities and errors, and finding workarounds and strategies to make less errors e.g. leaving the umbrella by the front door. I'd like to find examples like this in the medical context.

I like this example of a design glitch - not so clever : Opticlick design vulnerability

All good stuff, Dominic. Be sure to keep us updated.

I'm not convinced that we're quite ready yet for a Chaos button on infusion pumps, though.

"Check and Test" used to be one of our (unofficial) mottos in the army too. When we had nothing else to do (and sometimes even when we had) we would check our kit* - again and, er, again (to the point where it would probably appear a bit anal to an outsider).

Which Asena pump do you have in mind there, I wonder? We need to be clear about the distinction between volumetric infusion pumps and syringe pumps. In general, infusion pumps deliver fluid volumes, whilst syringe pumps deliver drugs.

Certainly some of the newer pumps (of both type) "pull back" a bit on occlusion, but whether they cancel the alarm (upon sensing a return to normal line pressure) ... well, I'm in doubt about that. Hopefully someone on here can clarify that one (otherwise I'll need to run some tests on the next Asenas I come across)!

I like the idea of "predictive battery capacity technology". Now that we have so-called "intelligent" battery packs, the sort of thing you describe should not be outside the realms of possibility, I should have thought.

And, as for copying the nurse. Many pumps have a "lock out" function, so all should be well as long as the nurse shields the key-press sequence involved. If a higher level of security is required in our New Spec ... then we shall need to think about a swipe card, chipped key fob, or some such thing. But then ... what happens when the nurse looses the key?

Lastly:- "umbrella by the front door". What about the endless argument in New Hospital Design about the so-called Doctors' Hand Basin? I have been in hospitals that have this (small) basin positioned (at great expense) near the door in every patient room. Yet some - many - of the "blighters" (yes, I'm being polite today for some reason) still didn't use them to wash their hands between visiting patients! What's the answer there - cameras? (no, due to patient privacy). Please discuss!

* I'm referring here to the "green kit", of course ... rather than medical equipment (although we certainly checked and tested that enough, as well)!

That's a brilliant document!

Anyone know where we can purchase Infusace pumps?