Tech —

Sleep it off: What you can and can’t learn from sleep tracking wearables

Most wearables can track sleep, but will you learn anything useful from them?

Fitness trackers didn't always monitor sleep, but the feature is now a sought-after staple in most devices, as sleep is just as important as exercise to a healthy lifestyle. Most wristbands monitor sleep now, and there are even specialized devices that go on your head or bedside table that can also keep track of how long and how well you sleep each night.

But sleep tracking isn't as simple as step tracking, and you need more than a simple accelerometer to measure it accurately. While motion is an indicator, it's not the only metric you should track to get a full picture of how well you slept.

What makes a good night’s sleep?

Understanding how well you slept on any given night depends on your time in light, deep, and REM sleep. Light sleep occurs soon after you fall asleep and your heart rate and body temperature go down. Your body falls into deep sleep typically after 10 to 15 minutes of light sleep, and this is when the body repairs itself from the day before. You'll spend most of your night in light and deep sleep, as REM sleep periods tend to be shorter.

REM sleep is the stage we hear about most often because it's known as the stage in which we dream. Since the brain is more active during REM sleep than the other two stages, your body changes during REM, increasing your heart rate and brain activity while your eyes move rapidly as you dream. The longest REM sleep phase can last an hour, but the body builds up to that amount of time. The first time you enter REM sleep after going to bed could last just ten minutes, but as the night goes on, your body enters longer periods of REM sleep.

It's important to experience all three sleep stages, which is why getting enough time to sleep is crucial. Typically, we experience light, deep, and REM sleep as a cycle, and we go through multiple sleep cycles each night. If you only have three or four hours of rest, you may not fit in enough sleep cycles for your body to fully repair itself from the day before. While the recommended number of hours of sleep per night varies by age, the National Sleep Foundation recommends seven to nine hours of sleep per night for regular adults ages 26 to 64.

Just as important as sleep stages is the number of times you wake up during the night. While some of us are dead to the world when we sleep, only to be awakened by the blaring sounds of an alarm, others wake up frequently. In fact, even the heaviest of sleepers can be technically awake at points during the night and not even know it. However, waking up for long periods of time, or waking up multiple times each night, can be disruptive to the entire sleep process and your productivity the next day. Aside from diagnosable sleep disorders, a number of things can wake you in the middle of the night, including too many alcoholic drinks before bed, not enough exercise throughout the day, and spending too much time on your phone or tablet before turning off the lights.

How can you track your sleep at night?

Considering the nuanced differences between sleep stages, it's tricky for modern wearables to accurately track sleep. Let's first talk about devices like Fitbit's Alta HR wristband. Most such devices use accelerometers to track movement both when you're awake and when you're asleep. However, if you're a naturally restless sleeper or someone who moves around a lot during the night, nighttime movement data can be deceiving. Wearables assume that if you're moving around at night, you're probably not sleeping as soundly as you could. For some people, that might be true, as many sleep disorders are characterized by excessive movement during sleep. However, not all movement during sleep is bad, and a certain level of it is natural.

Medical professionals and users have been vocal about movement being only one factor that determines a good night's sleep, so companies have recently incorporated information from other sensors to inform sleep data. For example, the Alta HR is the first device Fitbit released that uses nighttime heart rate data to estimate light, deep, and REM sleep time. A lot of devices that track sleep have optical heart rate monitors, but it's often unclear if that data is used to bolster sleep-tracking data. The Alta HR released with improved sleep tracking as a highlighted feature (and that feature came to other Fitbit devices with heart rate monitors in an OTA update soon after), and it uses heart rate and heart rate variability, or the variations between individual heart beats, to estimate changing sleep stages.

Then there are other types of smart health devices that claim to capture data similar, and in addition to, wearables like the Alta HR. Beddit, which was recently acquired by Apple, makes a strip-like mattress cover that captures movement, heart rate, humidity, and temperature information while you sleep. While it might seem like a more comfortable option than a wristband, that could be a disadvantage: for those who move while asleep, it won't have as much close and consistent contact with your body the entire night like a wristband would. That could also be a problem for bedside devices like the ResMed S+, a box-like device that uses non-contact, respiratory, and bio-motion sensors to track your sleep habits as well as the light, temperature, and noise levels in your bedroom.

While some of these devices use medical research and science to inform their sleep tracking methods and algorithms, it's important to note that none of them are certified medical devices. As a 2015 study published in the Dove Press Nature and Science of Sleep journal states, "The claims are not intended to be 'medical' in nature, such that the validation standards normally applied to medically used devices are not required." Since a Fitbit isn't a medical device, users shouldn't expect to be able to diagnose a sleep disorder from its data, much less use it as a replacement for seeing a medical professional. However, some sleep trackers may be better than others at obtaining accurate information about how we sleep.

Channel Ars Technica