Making medical devices safer – insights from the CHI+MED project (funded by @EPSRC)

CHI+MED (Computer-Human Interaction for Medical Devices, grant number EP/G059063/1) was a six-year, £5.8m EPSRC-funded multi-site project to look at ways of improving the safety of interactive (programmable) medical devices, such as infusion pumps.

We did this by considering medical devices from both social (people using them) and technical (designing interfaces, software) perspectives (‘sociotechnical model’) and worked with a variety of people who interact with them such as patients and nurses, people who train nurses to use these devices, people who decide which devices to buy (hospital procurement), people who decide which devices can be sold within the UK/ EU/ US (regulators), people who design the software that makes them work and people who design the interfaces that can make these devices safer, or less safe, to use.

We also shared our findings with schoolchildren and students (the next generation of designers and healthcare professionals), highlighting in engaging ways how better design can make a huge impact. One example was a very simple game: microwave racing – pitting two microwave against each other to see which could cook some microwaveable popcorn first. The ease of use of the device translated into time saved in cooking popcorn and we used this to talk about speed and accuracy tradeoffs, and whether the blame should fall on the end user if popcorn wasn’t immediately forthcoming.

Animated overview gif of our booklet

Our end-of-project publication ‘Making medical devices safer to use – insights from CHI+MED‘ (an animated overview is shown above) is now available to download as a free PDF (printed copies are available on request while stocks last) and each page gives a quick overview of an aspect of our research.

The booklet is divided into several thematic sections – each having its own landing page (main pages linked below) on our project website.

• Hazards and risks – we’ve worked with the FDA to find ways of reducing hazards arising from software and interfaces (some ways of entering numbers are safer than others). We also spotted opportunities to raise awareness about the dangers of button batteries in hearing aids
• Normal use – mostly things go smoothly without incident but we can learn from ‘normal everyday practice’ as well as from times when things go wrong. For example we’ve looked at the range of contexts in which devices are used (in hospital and at home) to help manufacturers design devices that fit the way people are actually using them.
• Entering data safely – a look at our work on number and data entry systems and what factors can reduce the risk of an error being made (and what can increase the likelihood that an error is spotted and recovered from)
• Learning from incidents – when something goes wrong a clinical incident report is recorded but improvements in the reporting form can make it easier to learn from mistakes.
• Tools – we have developed tools and methods for different purposes including: to support the design and procurement of new medical devices; to help in arguing that a device satisfies regulatory requirements, as well as to check such claims; and to document and to analyse medical incidents.
• Education, guidance and public engagement – our stories are helping people understand the problems and we have used magic shows to introduce issues around designing safer medical devices. Our citizen science project, Errordiary, collects funny, frustrating and fatal examples of errors and we use these to provoke people to think about human error, resilience and blame culture. We’ve also found that safer medical device design provides a rich context for teaching computing in school.

This booklet summarises key outcomes from the CHI+MED programme. It includes links to these web pages giving more detail about each area of our work as well as references to papers and other resources. Hard copies are available from Prof. Paul Curzon, p.curzon@qmul.ac.uk (while stocks last).