Orthosomnia: How Sleep Trackers Are Making Us Anxious About Sleep — and What Good Design Does Instead

There’s a particular kind of modern irony in lying awake at 1 a.m., anxious that your sleep tracker is about to record a bad night — and, by worrying, guaranteeing it does. The device you bought to improve your sleep has become the reason you can’t fall asleep.

This isn’t a rare quirk. It’s common enough that sleep researchers gave it a name: orthosomnia. And while it’s easy to read as a story about anxious users, it’s really a story about design — about what happens when a complex, deeply human process gets compressed into a single number and handed back to people as a daily grade. For anyone building products on sleep data, it’s one of the most important cautionary tales in the field.

Orthosomnia is real, and it now has a name

The term was coined in 2017 by Kelly Glazer Baron and colleagues at Rush University Medical Center, in a Journal of Clinical Sleep Medicine paper titled “Orthosomnia: Are Some Patients Taking the Quantified Self Too Far?” [1]. They described patients arriving at sleep clinics not with a diagnosable disorder, but with a fixation: their trackers said their sleep was inadequate, and the pursuit of better numbers had itself become the problem.

The name is deliberate. “Ortho” means correct or straight — the same root as orthorexia, the unhealthy obsession with perfect eating [1]. Just as orthorexia turns a healthy interest in nutrition into a disorder, orthosomnia turns a reasonable interest in sleep into a perfectionistic quest that backfires.

What started as three case studies has since been quantified. A 2024 cross-sectional study of 523 adults estimated the prevalence of orthosomnia at 3% to 14%, depending on how strictly it’s defined, and found that 35.8% of participants regularly used sleep-tracking devices [2]. Critically, those identified with orthosomnia had higher insomnia scores than those without — the anxiety and the poor sleep travel together [2]. The phenomenon is now established enough that researchers published a validated measurement instrument, the Bergen Orthosomnia Scale, in 2025 [3]. A clinical curiosity has become a measurable public-health pattern.

The cruel irony: the numbers causing the anxiety are often wrong

Here’s what makes orthosomnia especially unfair. People are losing sleep over data that is, frequently, an estimate dressed up as a fact.

No consumer wearable measures sleep directly — they infer it from movement, heart rate, and temperature. And while they’re good at telling sleep from wake, they’re far shakier on the stage breakdowns people fixate on — two devices on the same wrist routinely disagree on the same night. In a 2024 polysomnography study, the Apple Watch overestimated light sleep by 45 minutes and underestimated deep sleep by 43 minutes per night; even trained human technicians scoring the same lab data agree only about 80% of the time [4]. So the “you only got 38 minutes of deep sleep” verdict that ruins someone’s morning may be off by more than the number itself.

The result is a feedback loop with no floor: an anxious person checks a number, the number is imperfect, the imperfection feeds the anxiety, and the anxiety degrades the very thing being measured.

The nocebo trap: believing you slept badly makes it true

The deepest mechanism behind orthosomnia is one of the best-documented effects in psychology — and it should give every product designer pause.

In studies of “placebo sleep,” researchers gave participants false feedback about how well they’d slept, then tested them. The striking finding: people’s belief about their sleep quality predicted their cognitive performance and mood more strongly than how they had actually slept [5]. Tell someone they slept poorly and their mood, alertness, and test scores drop — regardless of the truth. Tell them they slept well and the opposite happens.

A sleep score is exactly this kind of feedback, delivered every single morning. When a tracker announces a poor night, it isn’t just reporting on the day — it can cause a worse one, priming the user to feel tired, foggy, and underslept. The score becomes a self-fulfilling prophecy. And sleep is uniquely vulnerable here, because it’s a performance you can’t force: the harder you consciously try to sleep, the less it comes. Anxiety is the one input guaranteed to make the output worse.

So is tracking the villain? No.

It would be easy to conclude that sleep trackers are bad and people should throw them away. That’s the wrong lesson, and the evidence doesn’t support it. For the large majority of users, trackers are benign or genuinely useful — they nudge earlier, more regular bedtimes, build awareness, and surface real problems like chronic sleep restriction or signs of sleep apnea. Only a minority tip into clinical preoccupation [2].

The villain isn’t measurement. It’s a specific, lazy style of measurement. Orthosomnia is overwhelmingly manufactured by three design sins:

1. The single authoritative number. Compressing a complex night into one score — “62, poor” — with no context strips away meaning and leaves only judgment. A number with no explanation is just an anxiety delivery mechanism.
2. False precision. Presenting noisy, estimated sleep-stage minutes as if they were lab-measured fact invites users to obsess over figures the device can’t actually stand behind.
3. The daily high-stakes grade. Gamifying sleep into a test you pass or fail every morning turns rest into a performance — and performance pressure is the enemy of sleep.

Each of these is a choice. And each can be made differently.

What good sleep feedback looks like

The same data that fuels orthosomnia can, presented well, do the opposite — informing without alarming. Four principles separate the two:

Explain, don’t just score. A number alone provokes anxiety; a number with its reasons provokes understanding. “Your sleep was lighter than usual, largely because you went to bed two hours later than your norm” gives the user something to act on instead of something to fear. Explainability is the single most important antidote.

Trends over single nights. One night is noise. The honest, useful signal is the pattern over weeks. Designing the experience around trends — and explicitly telling users that any one night barely matters — defuses the daily-grade trap at its source.

Behavior over outputs. Steer attention to what people can actually control. A user can’t will themselves into more deep sleep, but they can keep a consistent bedtime — and sleep regularity is both controllable and independently predictive of health. Pointing users at controllable behaviors replaces helplessness with agency.

Honesty about uncertainty. A feature that frames sleep stages as estimates, not verdicts, both tells the truth and lowers the stakes. Confidence calibrated to what the sensor can actually know is a feature, not a weakness.

The throughline: a good sleep experience is built to guide, not grade.

What this means for builders

Orthosomnia is a warning and an opportunity in the same package. The warning: if your sleep feature ships a bare nightly score with false precision and no context, you are not neutrally informing users — for a measurable slice of them, you are manufacturing anxiety and degrading the outcome you claim to improve. As the term enters the public vocabulary, users and the press are increasingly alert to products that make them feel worse.

The opportunity is the flip side. Sleep is already the most engaged-with metric in consumer health — so as a backlash builds against anxiety-inducing health tech, the products that win trust and retention will be the ones that leave users calmer and more capable, not more neurotic. That’s a design philosophy as much as a data problem: scores that come with their contributing reasons, an emphasis on trends and controllable behaviors over perfect nightly outputs, and honesty about what the data can and can’t say. It’s the approach we take with scoring at Sahha — explainable scores with the factors behind them, built to point users toward behaviors they can change rather than numbers they can only stare at.

The deepest insight orthosomnia offers is almost philosophical: in health, how you say something can change the thing itself. A sleep score isn’t a passive readout — it’s an intervention that shapes the next day. Build it like one. The goal was never the perfect number. It was a person who sleeps a little better, and worries about it a lot less.