Smartphone study reveals the world’s sleeping habits

Researchers in the USA have used a smartphone app to see how people's sleep habits vary around the world. More specifically they've investigated how much the timing of sunrise and sunset affect people's sleep times or if social and cultural factors are more important. "Quantifying these social effects is the next frontier in sleep research," they write in the paper in Science Advances.

The study involved the ENTRAIN smartphone app which helps people recover from jet lag by recommending ideal levels of light exposure based on a user's typical sleep routine. Users have the option to make their information available for research. Olivia Walch and her colleagues at the University of Michigan began collecting data from the app in 2014 and the new analysis is based on information sent in by 8070 users around the world during the first year.

Overall the data showed that a later sunrise goes hand in hand with later waking-up times, and that a later sunset is associated with people going to bed later, just as predicted based on how light affects the suprachiasmatic nucleus – the bundle of neurons behind the eyes that controls our sleep cycle, also known as the circadian rhythm. But crucially, the link between sunset and bedtime was weaker than biological explanations would predict.

Put differently, the time we get up is strongly influenced by the timing of sunrise, but the time we go to bed is not as strongly influenced by sunset, suggesting other social and cultural factors are involved. Consistent with this account, most of the cross-cultural differences in sleep – for example, the Dutch reported the most sleep and Singaporeans the least – were explained by later bed times in the countries getting less sleep.

The difference between the countries with the most and least sleep wasn't huge: just under 7.5 hours for Singapore and just over 8.1 hours for The Netherlands. But the researchers emphasised that even a 30-minutes difference is meaningful, especially when you consider that sleep debt can have a cumulative effect over time.

Users of the app from the UK averaged about 8 hours sleep (a healthy amount) with average wake time just after 7 am and average bed time just before 11.15.

The researchers were also able to use the smartphone data to compare sleep habits by age, gender, and time spent exposed to natural light. Age was the most important factor with older people tending to go to sleep earlier. There was also much less variability in the sleep times of older users, which could because of biological mechanisms that narrow the window of opportunity for when it's easy for older people to fall asleep.

This age-related finding could have everyday relevance – "being careful about how much light affects your circadian clock could be more and more important to sleep as you get older," the researchers said. If your body's only willing to sleep between fairly limited hours, you're best off listening to it and switching off that TV.

Meanwhile, women were found to get more sleep than men – 30 minutes more, on average – thanks both to going to bed earlier and waking up later. The gender difference was greatest in mid-life so that middle-age men are the demographic group getting the least sleep, on average.

In terms of exposure to outdoor, natural light, app users who had more of this tended to report going to sleep earlier and sleeping more, which is as you'd expect based on the effect of daylight hours on the brain's circadian clock.

The researchers concluded that their results "point to the suppression of circadian signaling at bedtime as an important target for clinical sleep intervention; and suggest that age-related differences in the window during which sleep can occur are evidenced on a global scale". Aside from these specific insights into sleep, the group also said their findings show the power of modern smartphone technologies as a research tool. ""This is a cool triumph of citizen science," said co-author Daniel Forger in a press release.

Further reading

—A global quantification of "normal" sleep schedules using smarphone data