This certainly isn’t a comforting statistic.
After heart disease and cancer, medical errors kill more Americans than anything else, claiming a quarter of a million lives a year, according to a study by researchers at Johns Hopkins University.
If bungles and safety lapses in the hospital were accounted for as deaths from disease and injury are, they would be the third most common cause of death in the U.S., leading to more fatalities than respiratory disease, the report in the British Medical Journal argues.
The new estimate, published today, draws on four studies of deaths due to errors that have come out since the 1999 report. The authors extrapolate from those findings to reach their estimate of 251,000 annual deaths. Even that figure, they say, probably underestimates the actual toll, because it includes only deaths in hospitals, not in out-patient surgery centers, nursing homes, or other health care settings.
While that article’s worth mentioning on its own, it’s particularly interesting in the context of a piece published yesterday at Wired titled, Why an Autonomous Robot Won’t Replace Your Surgeon Anytime Soon.
Here are a few excerpts:
The Smart Tissue Autonomous Robot could sew more evenly and consistently than even an experienced surgeon, according the report published in Science Translational Medicine. “It is a really nice piece of work. They’ve managed to push the envelope” says Ken Goldberg, director of UC Berkeley’s Center for Automation and Learning for Medical Robotics, who was not involved with STAR.
But in this case, STAR was still dependent on a surgeon to make the initial incision, take out the bowel, and line up the pieces before it fired up its autonomous suturing algorithm. “When you drive a car you use cruise control. The same logic would apply for surgical technology,” says Peter Kim, a pediatric surgeon on the STAR team. Just as cars have gained more autonomous functions—parallel parking, lane changes—STAR has been programmed to do other things like cut and cauterize, and Kim says they’re planning to do an entire supervised surgery like removing the appendix. But unlike Google’s autonomous car, which doesn’t even have a steering wheel, nobody is talking about a surgery robot with no human supervision.
If the technology behind STAR is going to make it into the hospital any time soon, according to Kim, it’ll probably be integrated into an existing platform. That could mean, for example, adding automated tasks to something like da Vinci, where the doctor still has final control. It makes sense, because the real advance behind STAR is software, not hardware. The robotic arm is just an arm from the German company Kuka, which makes robotic arms of all sizes for industrial use. What makes STAR unique is its ability to “see” inside the 3-D folds of soft tissue by using a 3-D lightfield camera—similar in concept to Lytro’s camera—that looks for fluorescent biomarkers injected inside the tissue. “The key to this paper is smart imaging technologies,” says W. Douglas Boyd, who specializes in robot-assisted heart surgery at the UC Davis Health System. “This is where the huge leap of advancement in these autonomous systems will be.”
But technical capability isn’t the only barrier to acceptance among surgeons. Mazor Robotics makes a robotic system that identifies where surgeons should insert bone screws into the spine. Their machine could have easily done the drilling too, but Mazor found that surgeons preferred to give the go ahead and hold the drill themselves. “We had the technical ability to do it, but you have to go one bit at a time,” says Christopher Prentice, Mazor’s CEO. “The key to robotics in surgery is to add value, and I believe it’ll be incremental value. It’s not someone who swoops in.” That was Johnson & Johnson’s mistake with its anesthesiology bot.
It’s the same with cars. No one is trying to sell you a fully autonomous car yet. But the cars we do drive are already becoming increasingly automated, first with cruise control and now with lane change and parking assist. You’ll be lulled into trusting the robot driver and robot surgeon. But lulling will be slow and incremental. You can’t put the cart before the robot.
We sure do live in fascinating times.