How robots will replace doctors
Farhad Manjoo has a provocative thesis on the future of medicine: Robots will take the jobs. And not just the bottom-rung of jobs. “The doctors who are the juiciest targets for automation might not be the ones you’d expect,” he writes. They’re not the nurses or the primary-care docs. “They’re specialists like my wife — the most highly trained, highly paid people in medicine.”
I’m not convinced. Manjoo’s argument is that robots tend to be good at doing a very specific thing over and over again. That’s pretty much the definition of specialization. Ergo, specialists had best welcome their new robot overlords. Conversely, robots aren’t very good at having a long conversation with a human to figure out, in a subtle and sensitive way, what’s really wrong with them. But — and I recognize this will be an unpopular sentiment — are doctors?
After all, who becomes a doctor? High-achieving type-A folks who are really, really good at science. Some of them also happen to be very humane, very empathetic people. But not all of them. Perhaps not even most of them.
But then, we’re not sitting in that room wrapped in a garment made of the finest recycled sandpaper because we were hoping for a good conversation. We’re there because we’re sick, or worried we might be sick, and we’re hoping this arrogant, hurried, credentialed genius can tell us what’s wrong. We go to doctors not because they’re great empaths, but because we’re hoping medical school has made them into the closest thing the human race has developed to robots.
As Atul Gawande wrote in “The Checklist Manifesto,” “the ninth edition of the World Health Organization’s international classification of diseases has grown to distinguish more than thirteen thousand different diseases, syndromes, and types of injury.” And that doesn’t take into account all the possible symptoms and recommended lab tests and side effects of, and interactions between, various medications. That’s complexity beyond any human’s capacity to handle. But it’s not beyond a computer’s ability.
Manjoo suggests robots are built for surgery. But most doctors right now are thinking about robots built for diagnostics. They’re thinking about a version of IBM’s Watson that can cross-check symptoms with medications with a patient’s history and come up with an array of possible diagnoses ranked by likelihood. They’re thinking about that so much that, on Tuesday, the famed Cleveland Clinic is hosting an innovation conference in which clinicians will be competing against Watson.
My prediction is that long before robots undermine specialists, they will undermine primary-care doctors. But here’s where Manjoo is right: They will do that by specializing. They will allow the medical profession to break the conversation, the human element of the practice, from the technical diagnosis. Because doctor’s offices already have people who are really, really good at talking to patients, and they’re also the people who, more often than not, actually know how to use the computers.
They’re called nurses, and patients like them quite a bit. They’ll speak with the patient and then they’ll go to the computer and type up what they have heard and refer the patient to the appropriate specialists or medications. The software will “know” more and catch more than most doctors could on their own, and the nurses, by virtue of being less expensive, will have more time to talk with the patients and inform the computer of what they said.
There’s precedent for this. Librarians, for instance, now work with the public to find information on computers. Loan officers now speak with eager borrowers and plug their financial system into a computer program. Accountants now plug their client’s income information into tax-preparing software. There are plenty of professions that used to mix technical skills and conversational skills and now mix conversational skills and computers. Eventually, medicine will be one, too.