Victoria Betton, author of Towards a Digital Health Ecology: NHS Digital Adoption through the COVID-19 Looking Glass, on the importance of user-centred design in EPR implementation.
It may surprise the casual observer to read that NHS data can still largely be found in paper notes, ring binders and filing cabinets.
However, the inexorable beat of the digital drum is slowly but surely shifting this analogue state of affairs. By 2023, 90% of NHS Trusts are expected to have an electronic patient record (EPR) in place. This is an awe inspiring leap from the 40% status quo and has to be achieved in a short period of time.
Buying and implementing an EPR is a colossal expenditure — an NHS hospital EPR costs between £20 million and £100 million over ten years (and can be even higher). The majority of that cash goes on technical infrastructure, with relatively little to help clinicians adapt to new systems and working practices. It is therefore unsettling to learn that lack of clinical engagement and inadequate resources for change are frequently cited challenges.
A new EPR should help clinical and operational workflows and even how the physical space is organised. There are many interdependencies, not least WiFi and having the right devices. All of this requires the opportunity and headspace to think about things differently. Evidence tells us that EPR implementation should provide a sustainable platform for ongoing development rather than a single event.
So could a user-centred design approach provide an opportunity to anticipate and respond to issues of EPR engagement, usability and improvement early and continually over time?
Non user friendly EPRs contribute to frustration, stress and burnout on the part of health professionals. When observing clinicians in a hospital ward, I’ve seen them bounce around tabs to enter data, have multiple options to enter the same data, and witnessed the system log them out and lose all their work. The user experience is often poor, requiring tabs to be opened and closed as well as data cut and pasted.
Poor user experience is not only annoying, it has impacts on patient safety, time spent with patients, data quality and governance. It’s all round bad for everyone. So why don’t we spend more time focusing on it? With tight budgets and short implementation windows, it’s the important but not urgent factor that is the most easily dropped.
Re-designing an EPR for Usability
At the Service Design in Government conference (2022) myself and Phillippa Rose gave this challenge to a group of designers. Using flash cards, we asked them to take a typical EPR implementation and redesign it to focus on those all important human factors. Here are four key takeaways from the session:
Start early with user needs — take time to build user needs and goals into your thinking from the start of your business case and keep them at the core of your requirements. Use observation, interviews & analysis of data (e.g. clinical incident reports) to give you the insights you need.
Bake in adoption from the get-go — make sure there is sufficient resource and time in the business case to engage and involve EPR users at each stage of the process — from defining needs through to procurement, implementation and ongoing optimisation over time.
Get it right before you configure — use wireframes and simulation to test out before you start to configure the EPR. Make it as easy as possible for users to enter data in the right place the first time.
Iterate — create a process that allows for ongoing iteration, learning and optimisation of the EPR. Don’t send floor walkers in for two weeks and ask them to leave. Ongoing adaptation and improvement are key.
It became clear to us during our session that user researchers, service designers and UX designers have a lot to offer an NHS Trust implementing a new EPR. Combined with a Trust’s Quality Improvement team, a new EPR could be reframed as an improvement project rather than simply a technology one.
And finally, EPR usability is not a trivial concern — it is a patient safety, staff satisfaction and data quality imperative.