Hospital bosses love AI. Doctors and nurses are worried. (Washington Post)

[highplainsdem]

https://www.washingtonpost.com/technology/2023/08/10/ai-chatbots-hospital-technology/
Archive page at https://archive.ph/E7STB

This article mentions both good and bad things AI can do in hospitals - though the first paragraphs attribute something good to AI that WASN'T from AI, as I'll explain below this excerpt.

Michelle Mahon, quoted below, is assistant director of nursing practice at the National Nurses United union.

Though AI can analyze troves of data and predict how sick a patient might be, Mahon has often found that these algorithms can get it wrong. Nurses see beyond a patient’s vital signs, she argues. They see how a patient looks, smell unnatural odors from their body and can use these biological data points as predictors that something might be wrong. “AI can’t do that,” she said.

Some physicians interviewed by Duke University in a May survey expressed reservations AI models might exacerbate existing issues with care, including bias. “I don’t think we even really have a great understanding of how to measure an algorithm’s performance, let alone its performance across different race and ethnic groups,” one respondent told researchers in the study of caregivers at hospitals including the Mayo Clinic, Kaiser Permanente and the University of California San Francisco.

At a time of severe nursing shortage, Mahon said hospital administrators’ excitement to incorporate the technology is less about patient outcomes and more about plugging holes and saving costs.

“The [health care] industry really is helping people buy into the all the hype,” she said, “so that they can cut back on their labor without any questions.”

Re what I was saying about AI wrongly being given credit for a doctor intervening in time to help a patient:

The hospital's new AI showed a 74-year-old lung patient was at risk of dying. It didn't show why he was at risk, but the doctor who saw that alert from the AI on a computer screen rushed to see him and found that his oxygen levels were dropping because of a kinked chest tube stopping fluid from draining from his lungs.

The lowered oxygen levels had been picked up by the AI because of pulse oximetry, which has been around a long time, and is usually monitored by NURSES. This hospital had apparently shunted that monitoring to the AI, through which a doctor checking a computer screen was informed there was a risk of death, but not exactly why.

This is NOT an improvement in patient care, though that hospital apparently hopes to spend less on nurses.

And that same AI gives false positives about which are the sickest patients and has sent doctors to check out patients who are doing well, wasting the doctors' time.

This is a practically brain-dead stupid misuse of technology, sold to the hospitals as a cost-cutting and life-saving technology.

But the Washington Post starts the article with four glowing paragraphs about this technology, which they admit could be an advertisement for AI. The bit about doctors being sent on wild goose chases after AI alerting them to nonexistent dangers is buried in the last three paragraphs.

And nary a word about why that hospital didn't have nurses monitoring oxygen levels, as usual.

Or why their super-smart AI didn't tell doctors exactly why the AI felt certain patients were in danger.

The Post failed its readers by not asking questions about those last two points...and by slanting the article with that intro, when they have to know that many people read only the start of articles.