The first thing Annie and her partner do when they wake up in the morning is ask each other how well they slept. “And I literally say, ‘I’m not sure yet, let me check,’” – and Annie, a chief people and safety officer, reaches for her smartwatch.
Annie started monitoring because she worried she wasn’t getting enough good-quality sleep. Now she’s a self-confessed sleep data “nerd”, mining her sleep data for insights into her general health and wellbeing, using it to inform lifestyle decisions and even occasionally to guide how much she aims to accomplish in a day.
Sleep monitoring is a boom industry, mirroring what devices and apps such as Fitbits and Strava have done for physical activity. Market reports vary on the value of this industry, but it is clearly lucrative and growing rapidly. A quick search reveals a wide range of devices – rings, headbands, watches and other wrist-worn devices, under-mattress devices and bedside devices – all suggesting their use will unlock such quality sleep as to make Rip Van Winkle jealous.
An estimated 40% of Australians are not getting enough good quality sleep, and one in 10 experience chronic insomnia. “We do know there are a lot of people who do worry about their sleep and whether they’re getting enough sleep, particularly if they’re not meeting some of the recommended sleep duration guidelines,” says Dr Hannah Scott, a senior research fellow in sleep psychology at the Flinders Health and Medical Research Institute in Adelaide, and co-inventor of a wearable device that tracks and treats chronic insomnia.
Scott sees the rise in the use of sleep trackers as generally good news. “They’ve certainly improved awareness around the importance of sleep and around healthy sleep patterns so overall, I’d say they’ve probably had a positive effect.” But there’s a downside. “If you try harder to exercise, you’ll become fitter, but we have the opposite problem with sleep actually; that the harder you try, the harder it is to actually obtain sleep,” Scott says. “We can be creating some problems over people becoming too obsessive about trying to optimise it.” There’s even a term for it: orthosomnia, which describes an unhealthy preoccupation with sleep-tracking data.
The most accurate picture of sleep health is derived from what’s called polysomnography, which requires a person to spend the night in a sleep laboratory with their head and body covered with electrodes that monitor and measure brain wave activity, eye movement, breathing, heart rate, muscle movement and blood oxygen levels. That provides a wealth of information such as time spent in different stages of sleep, how many times someone wakes up and how long it takes them to fall asleep, says Prof Christopher Gordon, professor of sleep health at Macquarie University in Sydney.
“Wearables – and that’s lumping a lot of different devices in one word – but generally they’re not that accurate at being able to tell how long you took to fall asleep and how long you’re awake and asleep overnight, and that’s because it’s not measuring brainwave activity,” he says. That brainwave activity is used to determine time spent in different stages of sleep: stage one, two and three of non-REM sleep and REM sleep.
What wearables can detect and measure – in varying combinations and with varying degrees of accuracy – is heart rate, temperature, movement and blood oxygen levels, which are then fed into algorithms that determine whether the picture painted by that data is of someone sleeping soundly or restlessly awake. “It could be a device that’s specifically measuring movement only, and it’s looking at algorithms that say if your arms are moving a lot you’re awake, if it’s not moving a lot it’s sleep,” Gordon says. But “that has very little agreement with what happens in your brain in terms of the qualitative aspect of sleep”.
The other challenge is that there isn’t a clear understanding of exactly what good sleep looks like, says Associate Prof Jen Walsh, director of the Centre for Sleep Science at the University of Western Australia in Perth. “It’s an area that’s debated within our profession,” she says. There’s sleep quantity – simply the amount of time spent asleep – and sleep quality, which is more complex and takes into account time spent in different stages of sleep, whether sleep is broken, how often and for how long. “Sleep quantity is quite easy to define and calculate, whereas sleep quality is somewhat harder,” she says. Current guidelines suggest adults should aim for between seven and nine hours of sleep a night but there isn’t such clear advice on what type of sleep – how much of each stage – is optimum.
Sleep quality is also highly subjective and sometimes doesn’t match what even the most accurate lab-based monitoring says, according to Dr Maya Schenker, a postdoctoral researcher on trauma and sleep at the University of Melbourne. “If we feel like we slept very badly, it doesn’t matter what the watch is telling me,” she says. Even in people with chronic insomnia, sleep often looks a lot better on the polysomnography than what they subjectively report.
Consumer sleep trackers do have one advantage over sleep study polysomnography: they are used every night for long periods of time. “One night in a sleep lab just gives you a single snapshot of that one night, but it’s not necessarily a reflection of every single night while you’re at home in your own bed without wires strapped on you,” Walsh says.
Rachel says her sleep-monitoring ring has helped her to understand some of the factors that help her get a better night’s sleep. “If I do pilates in the evening, I seem to mostly sleep better,” the Canberra-based public servant says. And Annie has noticed that if she has a glass of wine at any time in the evening, her heart rate during sleep is about 10% higher.
This is where most experts see the usefulness of sleep trackers in a consumer setting: helping people understand how their lifestyle habits and behaviour affect their sleep, and making changes to improve it.
“A lot of people are interested in changing their sleep habits, but it’s hard to find a place to start,” says Dr Vanessa Hill, a sleep scientist at the Appleton Institute at CQ University in Adelaide, who also consults for Samsung Health. Data alone isn’t generally enough to change behaviour, but “if your watch can send you a notification where it’s like, ‘hey, yesterday, you went on a walk at this time and it improved your sleep’, or ‘yesterday you stopped drinking caffeine around this time’ or whatever, and that helps you fall asleep faster”, that can motivate people to change, she says, “I think that’s the best potential that these kinds of trackers can have.”
Hill uses a smartwatch and a ring to monitor her sleep, and says she does check her sleep scores – particularly her heart rate during sleep, which she says may predict oncoming illness – as soon as she wakes up. “I look at what they’ve been overnight, because if I am getting sick or getting a cold or something like that, my heart rate variability will actually tell me before I feel any symptoms myself,” she says. “If, for whatever reason, I have really bad heart rate variability one night, I’m just like, I need to take it easy today, something’s up with my body.”
Many experts stress that consumer sleep trackers are not diagnostic tools and have some important limits. “If you train an algorithm on a set population that is healthy, you’re not going to necessarily pick up the same signal out of a population with, say, peripheral vascular disease with reduced blood flow into the fingers,” says Dr Donald Lee, a respiratory at sleep physician at the Woolcock Institute of Medical Research in Sydney. Sleep habits also change over the lifespan, which may not be reflected by the algorithms used.
However sleep trackers do provide an opportunity to encourage healthier sleep habits, Lee says. “If we can engage people to … go to bed with a purpose, to turn out the light and go to sleep and improve their sleep habits by engaging in the conversation, it’s a good thing for the health trackers to be doing.”
