I remember the day Charlotte and I saved the A&E department. The evening before, I had been trying to convince her how useful computers could be in medicine. All very well, she said, but the solutions I suggested were all expensive and complicated. Now, on a Friday night, the A&E department was crowded with serious acute admissions from the city’s revellers. Overwhelmed by these, the last thing the staff needed was the computer system being down – but the tech boys were upgrading over the weekend. As medical students, we were equipped to do what was now the most important bottleneck to the department’s productivity – filling in all the forms for investigations. No sticky labels could be printed, so the task fell to us. I like to think that we did it well, but it left Charlotte with a new appreciation for the role of computers in relieving the tedium of medicine.
My next attachment was in paediatrics. I met an SHO who remarked how much she hated the computers. I chuckled to myself at her German naivety, and guessed it would take 2 days to change her mind. Instead, it took 2 hours to change mine. I just had to see her fill a form in on the hospital’s computer system. What would’ve taken a doctor 30 seconds by hand (and still remain clear to another doctor) now took 5 minutes of arcane computer codes. Chest X-ray = CXR, but abdominal X-ray = abdoxray. Go figure. Each of these codes had to be learned, and no deviation was acceptable by the beeping machine. It was clear these codes had been chosen by a 20-something programmer who had never entered a hospital (except perhaps on a Friday night into A&E, but that’s another story).
I think this is the ambivalence of IT in medicine. On the one hand, so much potential. On the other hand, with computer scientists by-and-large the only ones with control of the process, the “solutions” often seem ironically named.
Let me give you an example. Imagine that, in the beginning, the car had been designed by computer programmers rather than driving enthusiasts. When you want to turn right, you have to press a button with a right-pointing arrow on it. Instantly a dialogue pops up asking whether you really want to turn right. Indeed, it warns, this might lead to a “path deviation error”. You have to click “OK” to continue. Then another dialogue pops up with a self-satisfied paper clip asking whether you’d like further help on the topic of turning, or help with printing your maneouveres. You close this window and wait as your car slowly turns. A progress dialogue pops up in front of your windscreen as you desperately try to look at the road. When done, you click the button marked “Cancel” to return to straightness. I won’t even tell you how changing gears happens.
We need doctors to become involved at the design stage of computer software, rather than the complaints stage.
Published September 1999 as editorial in the newsletter
of the British Medical Informatics Society