Hi Geoff,
Thanks for taking the time to watch the video. Following some of your points I thought this might be interesting: I stand to be corrected but I think there is a gap between the current understanding of 'user error' and 'device error' that people don't readily look at. The default position appears to be that if there is a mistake and the device has done what it says in the manual then the nurse needs more training.
Psychological studies show that training is effective in reducing mistakes and knowledge based errors, but not effective in reducing slip errors. Knowledge based errors are things related to not knowing how to work something, e.g. not knowing how to operate a new blood glucose monitor. Slip based errors are errors associated with pressing the wrong button because they look the same, mixing things up like putting your orange juice in your cornflakes, forgetting where you have left the patient notes or controlled drugs cupboard keys etc.
The example (which is now dated so it wouldn't happen now but it shows the gap I am referring to): A nurse was in a very busy emergency situation where there are multiple audio alerts going off and she has to do too much. One of the infusion pump alerts is easily silenced by pressing the alarm silence button. The reason the pump alert was going off was that it wanted more information (I can't recall the reason). Whilst dealing with all the other things the nurse silenced this alarm three times as it repeatedly went off. However, she was actually raising the rate of the infusion each time she silenced the pump because she was hitting the up button rather than the silence button - this also meant the alarm was silenced.
The view of the expert reviewing this case was that it was clearly a user error i.e. the nurse should have been more careful about the buttons they pressed and should always check the device after interacting with it. Therefore there is no issue with the device and the nurse is blamed and needs more training (if she still has a job).
I thought that the design was at least partly at fault. The expert considered my view and conceded that the buttons could have been placed further apart. However, I wanted him to go further than this, why did the up button silence the alarm in this way which led to the mistake? A good design would have prevented this error. To me this interaction set the nurse up to fail.
I can't recall where I read it but we can use what is called a substitution rule in these instances. We don't know enough about the incident, but from what we do know, if we substitute that nurse with another nurse could we reasonably expect them to act in the same way. I would say 'yes' in this busy situation. This to me points to the design being at fault so it shouldn't only be classed as a user error.
So I would argue that this is not just a 'user error,' and it is not a 'device error' as it follows the instructions. I think there is a gap - a 'usability error' or 'design error' or 'interaction error' whatever you want to call it. I think these sorts of errors are confidently and wrongly categorised similar to the expert above, and that people don't readily look for usability issues. I'd be interested to hear people's views on this.
