Like every other journalist I know, I spent part of last week mucking around with ChatGPT, a new artificial intelligence chatbot that can write jokes, poems, student essays and, yes, newspaper columns.
When I asked it to “write a hilarious column”, it came back with:
“As I sit here, typing away on my laptop, I can’t help but feel like the world’s biggest idiot. And no, it’s not just because I spent half an hour trying to figure out how to get my cat off the keyboard (spoiler alert: I failed miserably).” There followed 344 more words that could easily have been written by a human. Not a very scintillating human, but still.
There was the occasional cliché, and I feel the chatbot’s definition of “hilarious” was stretched. Impressively though, it repeatedly managed to avoid one defect that continues to plague the human: jargon.
In fact, it took a dim view of some of the worst kind — business jargon — declaring that babble about synergies and KPIs can be “confusing or exclusionary”. “In general,” it said primly, “it’s best to use clear and straightforward language that can be easily understood by anyone who is listening or reading.” Quite.
I was still thinking about this a day later when I came across a new study from University of Minnesota researchers on a type of jargon I have not thought about much before, but should have: the medical variety.
It’s fun to laugh about people who talk of blue-sky thinking in the low-hanging fruit space going forward. It’s clearly worse to tell a patient something about their health that they fail to fully grasp.
Doctors have known this for years but, like their jargon-spouting corporate counterparts, they keep at it regardless.
Alas, this means that some still use phrases that ordinary people think mean the opposite of what is intended, especially when it comes to “positive” and “negative” test results.
Although 96 per cent of people understood that a negative cancer test result was good news, only 67 per cent knew the phrase “your nodes are positive” meant the cancer had spread, the Minnesotan researchers found.
Acronyms also confuse. Some doctors still use phrases such as NPO, from the Latin nil per os, or nothing by mouth, when they want to say a patient should not eat or drink anything for a stretch of time.
The study shows how risky this can be. When participants were shown the term: “You will need to be NPO at 8am”, only 11 per cent understood what was meant. But 75 per cent knew exactly what to do if told: “You are to have nothing by mouth after 4pm.”
Words with a different meaning in medicine also spell trouble. When doctors talk of an “occult infection” they mean an infection that is hidden. In the Minnesota study, more people thought it had something to do with a curse.
The researchers think their study is the largest of its kind and the first to compare how well people grasp jargon versus non-jargon phrases.
But it is by no means the first to expose the problem. Other papers from the US and Europe have shown that medical jargon has been confusing cancer patients, diabetes sufferers and the parents of ill children or premature babies for years.
The good news is that patient understanding may be starting to improve.
Back in 2001, a UK study showed that only 52 per cent of people understood the phrase “the tumour was progressing” was bad news, which is unsurprising given that progress generally means something good is happening. But 79 per cent of participants in the Minnesota study knew that this phrase in fact meant the opposite.
It is not entirely clear why understanding might be rising but the pandemic may have helped. Years of waiting to see if a Covid test is positive or negative might explain why there is now almost universal understanding that a negative cancer test is good news.
Either way, it is better to stick with something anyone can instantly understand, like “the test shows no cancer”. That way, as the AI chatbot says, there’s a better chance that doctors will be understood by their patients, their patients’ families — and each other.
Read the full article here