I do what my Apple Watch tells me to.
When it tells me to stand at 50 minutes past the hour, I get up from the comfortable couch where I work and stretch, dance, kick, do jumping jacks. When I know I moved around plenty in the past 50 minutes and it still tells me I haven't moved enough that hour, I am affronted: How dare my watch say I have been stationary when I've been up and down the stairs repeatedly! If I work out without having my watch on, like the delightful Aqua Zumba class I did a few times this past summer, I make sure to log it immediately afterward. And when I hit my move, stand, and exercise goals for the day, the sparkly little rings on the watch feel a little bit like a podium ceremony: I achieved something.
It's ridiculous, I know. But as a great First Opinion pointed out this week, it's also troubling, because it's simply unclear how good the algorithms in a smartwatch are when it comes to health — like detecting atrial fibrillation.
"In previous independent evaluations on the Apple Watch, the detection algorithm was unable to classify 27.9% of 30-second ECG heart signals, more than one-quarter of the collected data. It seems to particularly struggle during intensive exercise. This indicates ample room for improvement," Cynthia Rudin, Zhicheng Guo, Cheng Ding, and Xiao Hu wrote. The consequences of inaccuracy can go far beyond my irritation at being nudged to stand when I've already moved around that hour.
"[H]ow good are these algorithms generally? The truth is, we just don't know," they wrote. "Answering this question is a nightmare." That's because even FDA-cleared health algorithms can be a virtual black box: The approval rests on large datasets that are not available to consumers or researchers.
One answer, the authors suggested, might be "for a federal agency to evaluate algorithms for major health-related applications such as afib detection." While companies might balk, it may be one of the best ways to protect consumer health. Be sure to give the piece a read.
Also in First Opinion this week: On the podcast, I spoke with Jody Dushay, an endocrinologist who recently wrote a First Opinion essay on the Wegovy shortage, and Laurie Brunner, one of her patients, about GLP-1s.
Plus: Jennifer Beam Dowd, an epidemiologist, explains why, yes, everyone should get an updated Covid shot. (I got mine! Did you?) And kangaroo mother care could help millions of preterm babies worldwide every year.
Sean Scanlon, comptroller for Connecticut, explains how his state is approaching the challenge of covering expensive GLP-1s for state employees, retirees, and dependents. That article is part of a partnership between First Opinion and Tradeoffs, a podcast exploring our confusing, costly, and often counterintuitive health care system. To learn more about how the high price of this new wave of weight loss drugs is causing employers to confront their attitudes about obesity, listen to a deeper dive from Tradeoffs, and subscribe to never miss an episode.
Recommendation of the week: The new season of the podcast "The Dream" is looking at the shady business of life coaches, and I'm hooked. If you haven't heard them yet, the previous seasons on multilevel marketing and wellness scams are also must-listens.
P.S. I'll be at the STAT Summit this week. If you're there, come say hi!
No comments