Teens lose sleep on nights they use phones after 10pm, tracker study finds

Parents who suspect that a phone glowing after 10pm is costing their teenager sleep now have objective data pointing the same way. A study published on 19 October in JAMA Pediatrics used passive phone sensors and wrist trackers — rather than the self-reports most previous work relied on — to measure, night by night, how late-evening smartphone use lined up with how an adolescent actually slept [s1].

How the study worked

Researchers drew on the Adolescent Brain Cognitive Development (ABCD) Study, a large long-running US cohort, and focused on Android users who had at least 14 days of overlapping data from a phone-sensing app and a Fitbit [s1]. The design is what makes it unusual: instead of asking teenagers to recall their habits, the Effortless Assessment Research System logged phone activity automatically, while the Fitbit recorded sleep timing and interruptions [s1]. Data were collected between 30 November 2020 and 5 February 2024 and analysed in 2026 [s1].

The final sample was 425 participants, with a mean age of 14.8 years; 216 (50.8%) were female [s1]. Between them they contributed 7,591 person-nights of paired phone-and-sleep data [s1]. Crucially, the analysis used a "within-between" method that separates two different questions: whether teenagers who use their phones more than their peers sleep worse, and whether the same teenager sleeps worse on their own heavier-use nights [s1]. The second comparison is the stronger one, because each adolescent acts as their own control.

What the trackers showed

On nights when an adolescent used their phone more than their own evening average after 10pm, they slept a mean of 20.05 minutes less and fell asleep roughly 64.73 minutes later than on a typical night [s1]. Those are averages across many nights, with wide night-to-night variation, but the direction was consistent.

The finer-grained analysis put a number on the dose. After adjusting for other factors, each additional minute of phone use after 10pm — above that adolescent's own usual amount — was associated with falling asleep 0.87 minutes later (95% confidence interval, 0.84 to 0.90), waking 0.37 minutes later (0.33 to 0.41), and a net change in total sleep time of −0.43 minutes (−0.47 to −0.39) [s1]. All of these associations were highly statistically significant [s1]. In plain terms, the lost sleep at the start of the night was only partly recovered by waking later, so the balance was less total sleep.

Timing mattered as much as quantity. The size of the association grew steadily, or "monotonically," across the cutoffs from 8pm to midnight — the later the use ran, the stronger the link [s1].

The minutes just before sleep

The study also looked at "proximal" use: the phone activity in the 5 to 30 minutes immediately before sleep onset. Here the signal was about fragmented, not delayed, sleep. Use in the final five minutes before falling asleep was associated with the most wakefulness after sleep onset — an extra 0.74 minutes (95% confidence interval, 0.50 to 0.98) of being awake once asleep [s1]. That association faded to negligible when the window was pushed back to the final 20 minutes before sleep, and it was concentrated in the first 30 minutes after the adolescent dropped off [s1]. The picture is of a phone put down at the very last moment nudging the early part of the night toward restlessness.

How to read it

This is an observational study, and the authors frame their findings as associations, not proof that phones cause poor sleep. Even with each teenager serving as their own control, something else could rise on heavy-phone nights — a stressful evening, a looming deadline, or simply not feeling sleepy — and drive both the scrolling and the late sleep. The cohort was also limited to Android users with Fitbits, so it may not represent every adolescent, and the sleep and phone measures, while objective, are device estimates rather than laboratory gold standards.

What the design does add is weight. By measuring behaviour passively and comparing each adolescent against themselves across thousands of nights, the study sidesteps the recall bias that has dogged screen-and-sleep research and still finds a consistent, dose-dependent pattern [s1]. The authors conclude that cutting evening phone use — particularly after 10pm and in the minutes right before bed — may help adolescents' sleep [s1].

For families, that points to two concrete levers rather than a blanket ban: how late the phone stays in use, and whether it is the last thing in a teenager's hands before sleep. Our explainer on whether school phone bans improve teen mental health covers a related debate. This article is informational and not medical advice.