As the health-and-fitness world becomes more “evidence-based,” we all seek the one thing we know will lead us to our fitter, stronger, healthier selves: data.
We want to study our daily calories, sleep, macros, water, fiber, and bathroom breaks. And as our craving grows, the end-all-be-all of human data collection has risen to the top of the sweat-covered, step-counting fitness mountain: the wrist-bound activity tracker.
While activity trackers have been around for decades in the form of pedometers (your classic step counter) and portable heart rate monitors, we’ve experienced a recent surge of wearable trackers. Juniper Research reports that in-use fitness wearables have almost tripled since 2014 [1]. Information you could only get at your doctor’s office or in a research lab is now literally at your fingertips. But popularity comes with doubt: How do these trackers work, are they accurate, and what’s the best way to use them?
The Ins-and-Outs of Your Tracker
Jump, hop, sleep, walk… your activity tracker is watching you. While you can decide for yourself how creepy that might be, the nuts and bolts of an activity tracker are based on basic technology: accelerometers and altimeters [2]. Your movements are recorded by the accelerometer, which plugs those motions into a math equation to see if they are steps or just you trying to flap your wrist around until you hit 10,000 “steps.” Using your height, bodyweight, and gender, the device will estimate your stride length and estimate how many calories you burn while walking or running. (If your device measures heart rate, this can help estimate energy burn, too. But more on that later.)
Your tracker may also measure sleep with actigraphy, which determines sleep quality by how much you move during the night. And most products claim to use photoplethysmography, an impossible-to-pronounce-the-first-time word that means “shining a light against your skin and recording how much is reflected to the machine or absorbed by the pigment of your blood.” This lets the watch read your pulse. (They could’ve just said that.)
What’s most important here is that when these trackers look at activity they (1) rely on very similar technology to work and (2) need to make mathematical assumptions to gauge how many calories you burn. And where there are assumptions, there are often errors and bad jokes about the spelling of assu… you get it.
How Reliable Are They?
Many tracking products will make lots of claims about how they measure heart rate, total daily energy expenditure (TDEE), sleep, and steps, but don’t mention how accurate they’re supposed to be. Most companies perform their own “reliability studies” that go unpublished, with the excuse that they “do not disclose [their] specific validation study methodologies and results.” Or at least that’s what a spokesperson from Fitbit had to say [3].
As the science catches up to the sales, it looks like capturing human data isn’t as easy as it seems. That’s what researchers from Stanford University found when they decided to compare the reliability of the most popular trackers: the Apple Watch, Basis Peak, ePulse1, Fitbit Surge, Microsoft Band, MIO Alpha 2, PulseOn, and Samsung Gear S2 [4]. While trackers have been around for decades, most studies on these devices have been simple, looking at heart rate and energy expenditure while the subject sits in a chair. This study was the first to look at wrist-worn trackers that monitor steps, heart rate, and TDEE in people while they walked, sat, and cycled. At the same time the participants were connected to an EKG to measure their true heart rate and a calorimeter to measure their true energy expenditure. These were compared to the values on the activity trackers.
The trackers did a good job of measuring the participants’ heart rate: the Apple Watch had the least error (2%) and the Samsung Gear S2 had the highest (about 7%). That’s not too bad. But these devices basically failed when it came to measuring energy expenditure. Take it this way, the researchers said that more than 5% would be considered a “significant error.” The TDEE error was all over the place, with 27% error in the Fitbit Surge and 92% in the PulseOn!
Let’s put that into perspective. With that much error, your tracker could show you burned 500 calories when you actually burned anywhere between 250 and 750 calories. That’s bad. And other researchers have found similar amounts of error for energy expenditure over 24-hour activity trackers during sedentary, light, and moderate activity [5].
What Should We Do About It?
So, is your Fitbit just high-tech snake oil? Should we condemn it to the pile of unused technology— right next to CD players and flip phones? Not quite.
Activity trackers can still be very useful. As the study above mentioned, they did a good job of measuring heart rate, which could be very useful for aerobic training and general health markers. We also know that people who use activity trackers, or at least pedometers (which are also pretty accurate) are 27% more active than those who don’t [6]. Also, when it comes to measuring sleep in healthy populations, it appears that actigraphy measures aren’t too bad compared to the brainwave-measuring gold standards in sleep quality [7]. And products like the Fitbit Charge HR show promising accuracy when compared to standard actigraphy tools [8]. (That said, subjective sleep quality and objective sleep measurements aren’t as similar as we would like to think [9]. The bottom line is: If you feel great after a full night’s rest but your watch says you slept like a new parent, you’re probably better off assuming the sleep was good and going on with your day.)
The Tracking Life for You
If you’re okay just knowing how many steps you take in a day, you should stick to the simple and less expensive pedometers. They’ll be accurate and all that you need.
But, if you want more than that out of your tracker, heed this advice: Buy nice, not twice.
If you’re a fitness enthusiast, invest in something that can track your heart rate and has a user-friendly interface. Maybe look for one with GPS technology to track your run times (if you’re into that sort of thing). Luckily, most trackers today have easy-to-use apps that sync with your smartphone to show all your data.
Here’s the biggest takeaway: Even though trackers don’t give you a very accurate picture of calories burned, they can provide a solid picture of your daily activity. Your maintenance fitness and food intake should be based on your normal activity level (such as 10,000 steps)—not a rough (inaccurate) estimate of your “daily energy burn.” If your tracker shows you’re low on daily steps, you can add a quick ten-minute walk into your schedule and feel more confident that you’re hitting your activity minimum.
There you have it. Your activity tracker will (probably) make you more active, give you some good data, and looks pretty nice on you! But don’t expect it to tell you much about the calories you burn, and definitely don’t have that extra donut because you “earned it” in spin class. Activity trackers have their place—counting steps, heartbeats, and shut eyes—but it looks like that’s about it so far.
Now, strap that fancy thing on and get out there.