Wearables: the Apple Watch, Whoop, Oura ring - lifesaving devices or useless data collectors?
Will they save your life?
What gets measured gets managed - it’s true in business and it’s true in healthcare.
For most of us, our doctors only measure things at discrete intervals - when you’re in for a visit, when you have bloodwork done, when you go for a test.
And that leaves a lot of our individual data uncollected - your resting heart rate, your daily blood pressure, your heart rate variability, body temperature, and the list can go on.
Enter wearables - computers that we wear that can measure these things, providing a window into our health between bloodwork and between doctor’s visits.
The argument for these devices is that by collecting all of this data, we get unique health insights that can’t be replicated in the traditional way, an alleged way forward to prevent disease before it starts and “take control of our own health.”
But I’m not fully sold on the hype from these companies.
Let’s go through the major cardiac promises of these devices when it comes to heart health.
Atrial fibrillation detection
The most hyped feature of the Apple Watch (at least in my clinical practice) is that it can detect an important irregular heartbeat early.
Atrial fibrillation (or Afib) is the most common life-altering irregular heartbeat. Some estimates suggest as many as 10 million Americans have Afib.
The reason it’s life-altering is that Afib increases the risk of stroke, and we often treat patients with Afib with blood thinners to reduce this risk.
While many people have symptoms from Afib, not everyone does, which is where the Apple Watch detection comes in.
In theory, finding asymptomatic episodes of atrial fibrillation could lead to more people on blood thinners, which could lead to stroke prevention.
The problem is that finding Afib isn’t exactly the same as preventing strokes.
The Apple Watch Study found that patients who were notified of irregular heartbeats by their Apple Watch had a higher incidence of strokes, heart attacks, and heart failure:
But this isn’t a trial that can assess whether finding Afib via an Apple Watch notification actually changes outcomes for people.
Look at that table a little bit closer - these aren’t the same groups of people. The ones who received notifications are quite a bit sicker than the ones who didn’t.
It almost feels dumb to type that sicker people are sicker, but I think it’s an important way to frame the above findings.
It doesn’t mean that these detections aren’t life saving, it just means that we don’t really know.
In my day to day practice, for my patients that have notifications of Afib (in those who don’t have a prior diagnosis), there are probably 3 false positives for every true positive - meaning that a lot of people who are notified by their Apple Watch aren’t really having Afib when we look at a better quality monitoring device.
The upside of Afib detection is unproven - a theoretical benefit in stroke reduction by detecting short episodes of asymptomatic irregular heartbeats.
The downside here is stress, additional testing, and more downstream costs for people who don’t have anything to be concerned about.
Heart rate variability (HRV)
I was at a dinner a few years ago and had a chance to spend time talking with one of the medical advisors for FitBit.
He spoke a lot about how valuable an insight heart rate variability could be and how it was the future of the wearable industry.
If you look at the websites for Oura, Fitbit, Whoop or any other major wearable, the promise of heart rate variability is one of the central parts of their value proposition.
Heart rate variability is a window into the balance of our nervous system - the relative contributions of the “fight or flight” (sympathetic) versus “rest and digest” (parasympathetic).
The more stressed we are - the higher our adrenaline levels and the more revved up our sympathetic nervous system is - the lower our heart rate variability.
The theory here is that a lower HRV means that we’re more stressed out and more likely to get sick.
There’s data to back it up - lower HRV is found in people more likely to develop heart disease in multiple different populations.
But the data on HRV strikes me pretty similar to the data on Afib detection with the Apple Watch - sicker people are sicker.
It’s not that low heart variability itself is causing you to have a higher risk of heart disease, it’s that low heart variability is a marker for being more chronically ill.
The other problem here - there’s too much individual variation in the readings to learn much from your absolute HRV. I’ve had a number of patients report their average HRV to me with questions like “Is 16 too low of a heart rate variability?” or “what can I do to get my HRV to go up?”
The same things that make us healthier increase our HRV:
Better sleep
Stress reduction
Less alcohol
More aerobic exercise
Meditation
Rather than being concerned with the absolute HRV number, if you’re into this variable then looking at your own changes over time is going to be more useful.
There’s too much variation from person to person to make sense of whether 50 is a “good” number or a “bad” number. Rather, your changes overtime may reflect your behaviors.
Honestly, for most people, HRV doesn’t give you additional information that you can’t get from just looking at your resting heart rate.
And I’m actually pretty skeptical that it tells you anything additionally beyond just evaluating how you feel, unless you’re a really elite athlete whose exercise performance truly needs to be optimized rather than most of us who just need to move more.
Oxygen levels - SpO2
I’m not going to spend too much time on the pulse oximeter readings from these devices because I think they’re a total scam.
If you don’t have emphysema, aren’t sick with a nasty infection, and aren’t feeling short of breath, there’s essentially zero reason to be measuring this.
I’m open to the concept that I’m wrong here, but I see very close to zero utility of having this information for a regular, fairly healthy individual.
How these things really make a difference - if they help you enact behavior change
Behavior change is the holy grail of medicine - it’s the answer to all of our problems, but no one can figure out how to make it happen.
For every patient I have that successfully changes their diet, increases their exercise, and loses weight, there are probably two dozen people who fail.
In my opinion, the most important impact that wearables have is that they give you the ability to Hawthorne effect yourself - if having a wearable capturing this information makes you more likely to change your behavior, then it’s a slam dunk
If having to be confronted with the way that drinking a lot of alcohol impacts your Oura readiness score or lowers your heart rate variability ultimately leads you to drink less, then this device is a good investment.
If looking at the numbers from your wearable makes you more likely to exercise, it’s worth it’s weight in gold.
But to think of having this information as a window into your health to catch disease early or discover that you’re secretly sick strikes me as hype over substance.
There isn’t much question that the Apple Watch can detect atrial fibrillation that otherwise would have gone undetected - but whether this leads to any important change in outcome remains totally unknown.
In other words - we have no idea whether finding atrial fibrillation because you’re wearing an Apple Watch will actually lead to a lower risk of stroke.
And if that’s the reason you’re wearing it, you have to be willing to accept that the watch signaling you that there’s something wrong may not actually indicate a real problem - there’s a fair chance it’s just a false positive reading.
I’m also pretty skeptical that it will make you feel any better - after all, if you weren’t having symptoms of the atrial fibrillation and it was just detected because of your watch, what is there to actually improve?
The bottom line - if having this information changes your decision making then it’s a good investment and if it just makes you more stressed you should take it off.