Post by James Bagrow, Sigga Svala Jonasdottir, and Sune Lehmann.
We all know that stressed-out feeling we get during a trip. From nerves about making a tight connection to lost luggage, for many of us, stress comes along with travel. It’s easy for travel to blow up your sleep schedule. Struggling to fall asleep in a new place from the dreaded jet lag, it seems like everything is against us when it comes to catching sleeping on the road.
Or is it?
Sleep science has long been interested in exploring the effects of jet lag, travel at distances long enough to cross time zones and affect our circadian rhythms. And the first night effect is a well-documented hindrance to achieving rest in a new setting. Combining these effects, the general expectation is that travel should negatively affect sleep, making it harder for travelers to achieve and maintain a healthy sleep schedule.
Recently, big data has come to the world of sleep science, thanks to popular sleep tracking “wearable” devices. Millions of individuals across all walks of life and every time zone on earth are now using such devices to track and hopefully improve their own sleep quality. These data inspired us to dig deeper into the question of how travel and sleep interact.
Using a large-scale dataset describing the sleep of wearables users, along with their approximate home location, we studied the typical sleep pattern of individuals at home and away from home. At first, expecting to find that travel generally lowered sleep quantity, to our surprise, we found something more nuanced: for people who usually sleep excessively, yes, less sleep was achieved when traveling. But for people who usually get less than the recommended quantity of sleep, we found that more sleep was had when on the road. In other words, travel tends to have a balancing effect on the quantity of sleep.
Let’s get into the nitty-gritty. Our study explores the change in sleep duration during travel relative to typical sleep duration at home. Overall we find a clear dependence between the change in sleep duration due to travel and the quantity of typical nighttime sleep individuals obtain at home. Our findings are evident from the raw data and further supported via statistical models. To use a concrete example, starting from a short sleeper, an average male residing in Europe or North America, sleeping 4.5 hours at home on weekday nights would gain 1.6 hours of sleep on a travel night. A similar individual who obtains the population average amount of sleep (7 hours per weekday night), gains about 0.5 hours when traveling away from home. Finally, an individual with the same demographic characteristics who sleeps a lot (on average of 9.5 hours per night at home) loses 0.75 hours when traveling.
More generally, when controlling for demographic characteristics, weekend-weekday differences, distance traveled and time zone changes, we consistently find that sleep during travel tends to have a balancing effect providing respite to underslept individuals, while the deleterious effects are reserved for those who tend to be well-rested at home. Of course, distance and time zone changes do play a role in longer journeys, and people tend to lose more sleep with longer travel distances with an added negative effect when a trip crosses time zones. Understanding the effects of such journeys will be key to future studies.
In terms of differences across demographics, we saw substantial differences across regions of residence. On average, individuals residing in Asia tend to lose more sleep during travel, while those living in North America and Europe tend to gain sleep during travel nights relative to home behavior. This difference is reduced when we consider weekend nights only, a pattern which could be due to strict work schedules on weekdays in the Asian countries we consider; such schedules could result in more regular sleep schedules in those regions .
Looking forward, how can we sharpen these results in the future? Crucially, it’s worth disentangling the type of travel being made - something we cannot do here. Someone going on vacation may experience very different stress than if they were on a work trip, although both can be stressful or stress-free. Typical wearables data are not ideal at distinguishing such travel intent, but one dimension we can characterize is whether travel occurred during a weekday or weekend period. Weekdays and weekends do not completely align with work and pleasure travel, but it is reasonable to assume that, in the aggregate, all else being equal, weekend travel is more likely to be for pleasure than weekday travel. Interestingly, we found the same effects regardless of weekday or weekend travel, meaning that the same balancing effect occurs.
What might be the source of this balancing effect? One hypothesis could be that it’s related to sleep homeostasis. If we’re underslept due to chaos at home, a nice hotel room could be just the place for a night of recovery sleep. Similarly, if we’re getting lots of sleep at home, then maybe those pesky and unfamiliar details - that weird light from the TV or the strange thumping noise from the neighboring hotel room - are more likely to keep us awake.
Right now, with the COVID-19 pandemic continuing to smolder, travel is not at everyone’s top of mind. But someday, hopefully soon, travel will again be considered a commonplace activity, and with it, sleep when traveling will once more become a hindrance. Or a help.
 Hours Worked (Indicator) (Organisation for Economic Co-operation and Development, 2021); https://doi.org/10.1787/a452d2eb-en