Smart devices are everywhere, including wrist-based monitors. These wearables promise to count our steps, remind us to move and offer insight into our sleep. But can we trust that they accurately measure our sleep?
Most wrist-based devices are based on an accelerometer, which measures wrist movement. The data collected with the accelerometer – how often the wrist moves and how powerful that movement is – is coded as sleep or awake. In some cases, programs will also label sleep as light or deep, which seems to imply that sleep is good or bad. Some devices also monitor the heart rate. Small variations in the timing of the heartbeat – which occur naturally in certain situations with a regular heart rhythm – can also give some clues about the sleep stage. During a stable deep sleep, breathing is usually very regular, just like the heartbeat.
After this wrist device collects data about our movements and / or heart rate during sleep, it is transmitted wirelessly to our phone or computer and software programs analyze these to create graphs and charts that allow us to “see” our sleep.
Sleeping is believing, right?
It can be almost magical to go to bed, sleep, and then immediately get a graph that shows what we did while we slept. Was it a good night? How much deep sleep have we got? A few taps on the phone will show the truth. A graph tells us how we have spent the past hours, with an overview of the time awake, time in deep sleep and light sleep. We can even get a general “score” for the night. This is data based, so it must be correct, right? The answer is much more uncertain.
How well do these devices measure sleep?
First, it is worth noting that the software algorithms that decide what sleep and what wake is, are a bit of a “black box.” These are owned by the different companies that make the devices, which means that doctors and researchers don’t know exactly how the programmers decided to make these decisions. The software code and therefore sleep interpretation can vary between different brands, or even different devices within a brand.
Remember that perhaps a wrist device determines that you are awake after a lot of vigorous movements – think of brushing your teeth – while for another device a single small arm twitch can be considered awake. How many movements mean that we have woken up? A? Ten? Over which period, one movement per minute? Ten movements in two minutes? How powerful should those movements be? How does the software decide that we are ready for good or that we fall asleep after exercise? How well can the device even absorb movements – does it know if the wrist device is too loose? With all these factors, the possibilities for coding the data and interpreting the data are endless.
Secondly, there is little to no data that compares portable devices with research or clinical measurements. Actigraphs are small wrist-based devices that sleep providers and researchers use to measure sleep over longer periods of time. Just like devices available to consumers, they use accelerometers to sort sleep versus wake. However, actigraphies have been extensively studied and compared with sleep logs, sleep studies and other data. Sleep providers have a fairly good feeling for their strengths and weaknesses and therefore for the use of the data. Consumer devices are changing fast – newer monitors, frequent software upgrades. In general, the studies suggest that these wrist devices overestimate sleep duration (how long we sleep) and how much of the night is spent asleep (sleep efficiency).
Finally, there is even less data on how these devices are affected when there is a concomitant sleep or medical condition, or through medication. Consider a patient with insomnia who meditates when he cannot sleep and is still in bed. This absence of movement and steady breathing can easily be interpreted as sleep by a wrist-based device.
What is the gold standard for measuring sleep?
A sleep study, also called polysomnography, measures brain waves, muscle tension, breathing and heartbeat, while a technician supervises, often in a hospital. The information of the brain waves determines awake versus sleep and the stage of sleep. This is considered the gold standard for determining sleep characteristics in most circumstances. However, it is time and labor intensive and expensive (and not always covered by health insurance).
Should we pay attention to these devices?
Devices with wrists seem to stay here and people will be curious about their sleep. I generally recommend that my patients view their sleep data with a pinch of salt. It is only a part of the image to record, and is not a substitute for a high-quality sleep log or other forms of sleep assessments. The advantage is that the collection of this data is fairly passive and can be done for a longer period of time to gain insight into patterns.
When we consider how these wearables measure sleep, we know what they can and cannot do. In general, the devices probably give us a rough idea of the time we spend in bed (which may or may not be the same as sleep time), and the regularity of the sleep zone (the time we slept or tried to sleep between going to bed and get up). Gaining insight into these two factors can be very useful and difficult for some patients to determine in other ways. Although the wrist devices do not replace a medical opinion or sleep examination in a hospital, if they help us think about our sleep and how much (or little) we get, they can play a role if they are used with care.