When Sleep Tracking Backfires: Orthosomnia and a Healthier Way to Use Your Data

A sleep tracker is supposed to help you sleep better, and for most people it quietly does: it makes bedtime visible, rewards consistency, and catches slow drifts you would never notice one night at a time. But sleep clinicians have started describing a smaller group of people for whom the relationship inverts. They lie awake worrying about the number they will wake up to. They feel fine until the app informs them they shouldn’t. Researchers gave the pattern a name — orthosomnia, the perfectionist pursuit of ideal sleep data — after meeting patients whose trackers had become one of the things keeping them awake.

This article takes that failure mode seriously — more seriously than you might expect from a blog attached to a tracking app. Tracking is a tool, and any tool can turn in your hand. The evidence below covers what orthosomnia looks like in the clinic, a striking experiment showing that a sleep score can change how you feel rather than merely report it, what a consumer tracker actually knows about your night, and the habits that keep data working for you — including the one that no one selling an app enjoys writing down: sometimes the healthiest move is to stop looking at the metric for a while.

Orthosomnia: When the Score Overrules Your Own Experience

In 2017, sleep researchers published a case series — descriptive reports from clinical practice, not a controlled trial — about patients whose use of sleep trackers had turned perfectionist. The coinage “orthosomnia” was a deliberate echo of orthorexia, the unhealthy preoccupation with eating perfectly. One patient was spending progressively more time in bed trying to push a sleep figure higher. Others arrived convinced by their devices that something was deeply wrong with their sleep — and here is the detail that made the paper famous: when an overnight sleep-lab study, the reference standard that records the brain’s actual electrical activity, contradicted the tracker, some patients sided with the tracker. The wearable’s tidy number felt more real than the laboratory and more real than their own days.

Baron, K.G. et al. (2017). “Orthosomnia: Are Some Patients Taking the Quantified Self Too Far?” Journal of Clinical Sleep Medicine, 13(2), 351–354.

A case series cannot tell you how common this is, and the honest answer is that nobody has a precise number — most people who track sleep never develop anything like it. What the paper established is that the pattern exists, that sleep clinics now recognize it, and that its signature is always the same: the score becomes more trusted than the person’s own experience of their nights and days.

The Sham-Feedback Experiment: A Fake Score Changed Real Days

If orthosomnia sounds like a quirk of a few anxious personalities, the second study should give you pause, because it demonstrates the underlying mechanism experimentally. Researchers gave people with insomnia a wearable sleep monitor and, each morning, showed them feedback about their night. The feedback was sham: participants were randomly told they had slept well or slept poorly, regardless of what the device actually recorded. The fake number did real work. On mornings they were told their sleep had been poor, people reported lower mood, more daytime sleepiness and fatigue through the day — even though their actual sleep was no different from the mornings they were praised. Told the score was good, they felt and functioned better.

Gavriloff, D. et al. (2018). “Sham sleep feedback delivered via actigraphy biases daytime symptom reports in people with insomnia: Implications for insomnia disorder and wearable devices.” Journal of Sleep Research, 27(6), e12726.

This is a nocebo effect, and it reframes what a sleep score is. The number does not simply describe your day; it helps write it. A glance at a bad score before breakfast can manufacture some of the very grogginess it claims to have measured — which means the ritual of checking first thing every morning is not neutral. For people whose sleep is already fragile, it can be an active ingredient.

What Your Tracker Actually Knows About Your Night

Part of the cure is remembering what the number is made of. Consumer sleep trackers do not measure sleep; they estimate it. Sleep is defined by the brain’s electrical activity, and a wrist device records none of it — it infers sleep from movement and heart-rate signals. A clinical review of consumer devices concluded that they are reasonable at detecting when you were asleep versus awake, but that their output is an estimate with real limits, not a ground-truth recording of what your brain did.

Kolla, B.P. et al. (2016). “Consumer sleep tracking devices: a review of mechanisms, validity and utility.” Expert Review of Medical Devices, 13(5), 497–506.

Human–computer-interaction research reached a matching conclusion from the user’s side. When researchers studied how people actually live with sleep-sensing technology, they found the double edge directly: the same tools that support awareness and healthier routines can undermine sleep health when users chase scores or treat rough estimates as objective truth about their bodies.

Ravichandran, R. et al. (2017). “Making Sense of Sleep Sensors: How Sleep Sensing Technologies Support and Undermine Sleep Health.” Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems, 6864–6875.

The practical hierarchy follows from the mechanics. Bed and wake times, total duration, and multi-week trends are the most trustworthy things a tracker reports. Stage breakdowns — the deep-sleep and REM percentages — are the roughest estimates in the report, precisely because stages are a brain phenomenon inferred from a wrist. It is a small irony of the whole genre that the least accurate number on the screen is the one most likely to upset people — a night-to-night swing of a few percentage points in “deep sleep” sits comfortably inside the estimation error, and deserves none of the worry it routinely causes.

The Signs Tracking Is Working Against You

None of the following makes you broken or gullible — they are reasonable human responses to a confident-looking number. But each one is a signal that the data has started running the relationship:

• You check the score before deciding how you feel. The morning glance has become the verdict, and your own sense of restedness has been demoted to a second opinion.
• Bedtime has acquired stakes. You feel a pulse of anxiety at lights-out about what tomorrow’s number will say — the exact arousal that makes falling asleep harder.
• You stay in bed to farm the score. Lying awake longer to nudge a sleep figure upward is the tracker-era version of trying too hard — and clinicians treating insomnia often work to reduce aimless time in bed, not extend it.
• Stage percentages can ruin your morning. Distress over a “low deep sleep” readout means the least reliable part of the report has taken the biggest emotional role.

If you recognize yourself in one or two of these, the fix is rarely dramatic. Most people do not need to throw the device in a drawer forever; they need to change the cadence of checking and the authority they grant the number — which is what the next section is about. If you recognize all four, and bedtime has become something you genuinely dread, skip ahead to the section on getting help, because that conversation is worth more than any settings change.

A Healthier Way to Hold the Data

The research above points to a set of defaults that keep tracking useful. None of them requires abandoning your data — they change when you look and how much weight the number carries:

• Review weekly, not every morning. A weekly look at your sleep preserves everything trends can teach and removes the daily nocebo ritual the sham-feedback study exposed.
• Judge trends, not nights. A single bad night is noise — normal biology, owned by everyone. Two or three drifting weeks are a signal worth acting on.
• Rank your own experience at least as high as the score. If you feel rested and the app disagrees, you are allowed to win that argument. The lab-validated definition of a good night includes how your day goes.
• Take deliberate breaks from any metric that hooks you. A week or two without looking — at sleep scores or any other number that has started tugging at you — is a reset, not a failure. If the break feels like relief, that is information.
• Remember what the data is for. The goal was never a perfect chart; it was more energy, better moods, good days. If those are improving, the numbers are doing their job even when they look untidy.

There is a physiological reason the gentle stance wins. Sleep is one of the few bodily processes that responds badly to effort: trying hard to fall asleep raises the very arousal that prevents it. A metric that raises the stakes at bedtime pushes in exactly the wrong direction. Treating a rough night as unremarkable — because it is — is not lowering your standards. It is the operationally correct move for a system that works best when you stop gripping it.

Where Lamplit Fits In

Lamplit’s design starts from the finding that runs through all of this research: how you feel is data too, and it deserves top billing. Your mood check-in, energy and journal are first-class entries in the app — free, and deliberately placed ahead of any number — so the record of a night includes the part only you can report. Sleep itself can be logged manually for free, and there is no proprietary sleep score waiting to pass judgment on your morning. If you want your wearable’s sleep data alongside that, it syncs automatically via Apple Health or Health Connect with Lamplit Pro, and Pro’s weekly AI recap reads the week as a whole — the weekly, trend-first cadence this article has been arguing for.

Honesty requires the next sentence too: no app design makes anyone immune to over-tracking, and that includes this one. Any number a screen shows you — a sleep trend, a streak, a chart that has started dictating your evenings — can hook the same machinery. If one does, take a break from that screen; your journal and mood entries keep working without it, and the record you return to will be the richer for the words. The measure of a tracking habit is never the tidiness of the data. It is whether your life is going better.

When It’s More Than a Data Problem

The evidence here has limits worth naming: the orthosomnia literature is young — case reports and small experimental studies rather than large trials — and how many trackers help versus harm is genuinely unknown. What is well established is what to do when sleep itself stays broken. If insomnia persists for months, the first-line treatment is cognitive behavioural therapy for insomnia (CBT-I) — a structured, well-evidenced programme, not a better gadget or a higher score. A tracker also cannot diagnose a sleep disorder: loud snoring, gasping awake, or unrefreshing sleep despite adequate time in bed belongs with a clinician. And if anxiety about health data itself becomes persistent and distressing, that too deserves professional support rather than another settings tweak.

The Bottom Line

Sleep tracking helps most people and quietly harms a minority, and the difference lies less in the device than in the relationship. The score is an estimate, not a verdict; it can shape how you feel as much as report it; and the parts most likely to alarm you are the parts it measures worst. Read your data weekly, trust trends over nights, let your own experience keep a casting vote, and walk away from any metric that has started to own you. The tracker works for you — never the other way around.

Start free on Lamplit — journal, mood and sleep in one place, with how you feel treated as data that counts.