Bad software destroyed my doctor's memory
Design this bad should sicken developers – but it's the rest of us who end up feeling queasy
Column "I have a problem," my medical specialist said – before he saw the fear on my face and quickly corrected himself. "No, not with you – you're fine! With this." Both hands swept past the shiny new convertible laptop-slash-tablet that sat on his desk.
"It’s not working?" I asked.
"Oh, it's working," he replied. "We spent six months digitizing every bit of paperwork that I've ever created in the last twenty years I've had my practice. It's all" – he gestured toward the screen – "in there."
I asked, tentatively, if that was a good thing.
From his sigh, I knew it wasn't.
IT professionals try to radically alter the workflow of medical professionals, without their input
"I've always had a nice big patient file to work with. Yours has everything from the first time I saw you, through to the very last time I saw you."
I remembered that file. It got bigger and thicker as the years passed, crowded with handwritten notes, his copies of prescriptions issued, reports from other medicos – all of the various details collected by modern medicine.
That file was nowhere to be seen on my specialist's completely clean desk, which sported just a telephone and that convertible laptop-slash-tablet.
"I saw you six months ago," he said. "That's in the file. I could just open it up, flip back a page or two, and see everything that I wrote, everything I prescribed, all of it. I don't have any of that now."
I asked what he meant and was told I'd been reduced to a "patient record."
"Yes, all of the information is in there, and sure, I can find it. But it's completely disconnected from when and how and why it happened. In a paper file, I could know things were related because they were close together. Here, everything is … well, I don't even know where it is," he lamented.
"So you've lost your memory?" I asked.
Another sigh. "Well, it is good for one thing," my specialist said. "It's completely portable. I can bring this along, take it out, and wherever I am, I've got my patients' records. I couldn't do that before."
He paused, trying to articulate something he didn't really have words for. "I just wish that I could swipe backward and forward through a patient record and have that move backward and forward in time through that record. So that I could swipe back twice and look at what was going on a year ago, or a week ago. But that's not how this works. And it feels like I never quite know where I am."
I understood: digitizing data is only the first step. How you access that data once it's digitized is the important part.
Had he been consulted, my specialist likely would have opted for a highly skeuomorphic design that faithfully replicated the look and feel of his paper record keeping. That simulation of a connection between himself, his files and his patients would have given him a sense of place, of knowledge, and capability. Without it, he felt like someone adrift on a sea of information, lacking a craft to navigate it.
- Social media is too much for most of us to handle
- Open the pod bay doors, GPT, and see if you're smart enough for the real world
- Open source AI makes modern PCs relevant, and subscriptions seem shabby
- It's time to reveal all recommendation algorithms – by law if necessary
This sort of "design blindness" happens whenever system designers prioritize the needs of the IT system, rather than its users. My specialist is therefore forced to deal with a client interface barely pasted over some sort of database. The data is shaped to fit the design of that database, rather than the other way around.
The database – and however many translation layers on top that it takes to do the job – should present its content in a form that is so easily apprehensible my specialist need never give it a second thought. It should just work.
Instead, twenty years of his experience lie trapped inside an information schema that provides benefits no one except the cloud provider that powers his new application.
Those benefits may be that it makes it easier to share data between doctors. Perhaps it's easier to keep track of billings. But the cost is in the quality of patient care.
That reminded me of a recent visit to my regular doctor – who had somehow forgotten that he'd just sent me for an expensive, invasive test. At the time, I wondered if my doctor's powers had waned.
Suddenly, I saw this forgetfulness in an entirely new light. The practice's patient management system probably hid that information a few pages down. When I popped into his office, he hadn't seen it, hadn't had his memory refreshed, and naturally had forgotten all about it.
As I am one of many hundreds of patients my doctor sees on a recurring basis, it's not surprising that details are forgotten. Yet this is exactly the sort of lapse that a good tool should help to prevent. If we can't trust our computing machinery to maintain clinical memory for patients and practitioners, why use computers in medicine at all?
The whole history of IT and medicine has been littered with failed digitization projects. Almost always that boils down to IT professionals trying to radically alter the workflow of medical professionals, without their input. IT seems to be gaining the upper hand, forcing digitization through the medical system.
Can we make that a human transition? If IT works hard to make good tools for doctors, nurses and other specialists, they make it better – and safer – for the rest of us. ®