Once upon a time car drivers had to know how to change gear by double de-clutching. Then synchromesh gears were invented, and changing gear became easy. Then automatic gears were invented, and the driver didn’t need to change gear or press the clutch; in fact there was no clutch to learn how to use. Then electric cars were invented, and no gears were needed at all.
The point is, if we have primitive technologies, we need to learn a lot about how to use them correctly and safely. As anyone who learned to drive in the old days knows, it takes a long time to learn the skills.
That is today’s situation with digital healthcare. Today’s digital healthcare is very complicated and full of automation surprises, not because we don’t know how to use it, but because it is primitive. Unlike car gears, digital technology is easy to update and improve; we do not need to wait decades to get improvements to the technology — if we want to fix the problems.
Patient harms are caused by poor design. It makes some sense to learn how to use it , but a far more strategic approach would be to emphasise how to get better systems that are easier and safer to use. That would mean improving standards and regulation so, on the one hand, manufactuers make better tech, and, on the other hand, so the NHS knows which systems conform to which standards when they want to procure better systems. It would mean emphasising training competent developers rather than users.
Consider that the entire digital health system in Ireland was taken out for six months out by a simple cyberattack . Although training staff about cybersecurity will help, the real solution is not having fragile computer systems that all go down when a nurse reads one email.
It’s obvious: if you improve the digital systems, you improve everything. In contrast, training healthcare professionals to be more digitally mature only improves healthcare one person at a time (to say nothing of the opportunity cost of the training). Perhaps the digital skills assessment tool (DSAT), aimed at life long learning for healthcare staff , should be used to help assess how bad digital is: the more time that has to go into learning how to use a technology safely, the more dangerous and time-consuming that technology is.
Rather than taking time out to learn how to use something complicated surely, more strategically, then, we could just get better technology that does the job properly in the first place. It would mean challenging the reactive training plans for a “digital first” NHS to include a proactive approach to transforming the quality of digital used in the NHS.
Currently, manufacturers say their digital systems have all sorts of benefits, like they are easy to use, reduce errors, and save money. Often this is just marketing gush, but it misleads procurement, which results in healthcare having systems that are thought to be safer. When the inevitable errors happen, then, it is natural to blame staff and to blame their training . Reading the Topol Report , you might think the solution is better training for healthcare staff, but the real solution is better technology — and if we had that we would not be scapegoating staff for problems caused by primitive technology.
1 Torjesen I. Health Education England outlines training plans for “digital first” NHS, BMJ 2022;379:o2921, DOI 10.1136/bmj.o2921
2 PWC, Conti cyber attack on the HSE, 2021. https://www.hse.ie/eng/services/publications/conti-cyber-attack-on-the-h…
3 Thimbleby H. Fix IT: See and solve the problems of digital healthcare, OUP, 2021.
4 Topol E. The Topol review: preparing the healthcare workforce to deliver the digital future. Feb 2019. https://topol.hee.nhs.uk.