Orthosomnia — when your sleep tracker is the reason you cannot sleep

What orthosomnia is

The term entered the literature in 2017. The phenomenon predates the term by at least a decade.

Baron, Abbott, Jao, Manalo, and Mullen published "Orthosomnia: Are Some Patients Taking the Quantified Self Too Far?" in the Journal of Clinical Sleep Medicine in 2017. The paper described patients arriving at a sleep clinic with insomnia complaints centered on tracker data: my sleep score is bad, I am not getting enough deep sleep, my recovery is low. In several cases the patients' actual sleep, measured in the lab by polysomnography, was unremarkable. The tracker data was the problem. Not the sleep.

The etymology is plain. Ortho- means correct. -somnia means sleep. The pursuit of correct sleep. The clinical content of the term is that the pursuit is the disease — that the patient's attempts to achieve a quantitatively perfect night produce, mechanically, an actually worse one.

Prevalence data is still thin. Baron's original report estimated roughly thirty percent of tracker-using patients presenting with insomnia complaints showed orthosomnia features. Consumer surveys in 2023 and 2024 put subclinical orthosomnia features at fifteen to twenty-five percent of consistent Oura, Whoop, and Fitbit users. The clinical, work-impairing version is rarer. The mild, anxiety-amplifying version is common enough to matter at the population scale these devices now reach.

Baron coined the term in 2017. The condition she described existed three years earlier, when the first Oura shipped.

The mechanism — performance anxiety with a dashboard

Orthosomnia runs on the same machinery as performance-anxiety insomnia, with a single addition: a number on a screen that gives the anxiety a focal point.

The arc is consistent across patients. The user starts wearing a tracker for self-knowledge — biohacking interest, recovery curiosity, athletic optimization, mild concern about sleep. The tracker assigns a daily score, branded as Readiness, Recovery, Sleep Score, or some equivalent. The first weeks are informative. The user notices that the score correlates loosely with how they feel. Then the score becomes the goal.

Once the score is the goal, behavior changes to chase it. Bedtime moves earlier to push total sleep time up. Caffeine cuts deepen beyond what is necessary. Alcohol exits the diet even when the patient does not particularly want it gone. Supplements arrive. Sleep hygiene checklists get followed with a rigor the original recommendations never intended. The bed becomes a place where you go to score well.

Performance anxiety follows by the standard route. The act of scoring well is itself a vigilant state. Vigilance is sympathetic activation. Sympathetic activation suppresses the parasympathetic transition into sleep. Onset latency lengthens. The score drops. The user, looking at a worse number, increases the optimization effort. The loop closes. The mechanism is the same one we covered in paradoxical intention — sleep effort is the problem — with the difference that here the effort is structured around a daily metric and reinforced by an interface engineered to maximize engagement.

If you have had a good night and the tracker disagrees, you will trust the tracker for about three days. Then your body wins the argument.

What sleep trackers actually measure, and what they don't

This is the technical critique that earns the rest of the article. The devices are not useless. The devices are also not what they appear to be on the dashboard. Worth being specific about which is which.

What they measure reasonably well

Total time in bed, motion-based, around 95 percent accurate against polysomnography. Sleep onset and offset within 10 to 20 minutes. Resting heart rate, accurate to within a beat or two. Heart rate variability, consumer-grade — directionally useful, not absolutely accurate against ECG. Skin temperature deviation from personal baseline, Oura's strongest variable, and probably the most useful single signal any consumer wearable currently produces.

What they measure badly

Sleep stages — REM, deep, light, awake — are inferred from movement and heart-rate variation. Validation against polysomnography puts consumer-tracker stage classification accuracy at fifty to seventy percent across devices and populations. For a four-class problem, that is meaningfully above chance and well below diagnostic. The stage minutes shown on your dashboard are statistical guesses with per-stage error bars of fifteen to thirty minutes, not printed next to the numbers. A chart that says thirty-eight minutes of REM probably means somewhere between eight and sixty-eight minutes of REM.

What they miss

Brief arousals — the two-to-thirty-second micro-awakenings normal in any night — are missed almost entirely. Apneic events are missed almost entirely. Periodic limb movements are missed. Consumer trackers do not diagnose sleep disorders. They occasionally flag patterns that warrant a real workup; that is the limit of their diagnostic utility.

What they cannot measure at all

Sleep quality. There is no objective measurement of it — quality is subjective by construction. The "sleep score" is a composite of measured variables weighted by an algorithm that is opaque, brand-specific, and changes with firmware updates. Oura has changed its scoring twice in five years. Whoop has changed it once. Each change retroactively invalidated user trend data without retroactively re-scoring the historical days.

Treating your Oura sleep-stage breakdown as a medical chart is treating a horoscope as a blood panel. The total sleep time is real. The thirty-eight minutes of REM is a guess with a thirty-minute error bar that is not printed next to it. Stop optimizing the number. The number is not the thing.

Sleep stage minutes on a wrist or finger device are statistical guesses dressed as measurements. The error bars are real and they are wide.

The score is the problem

Snerva's position on consumer sleep tracking is unusual among sleep-content sites and it is not subtle. The data is fine. The score is the harm vector. Most of the orthosomnia that turns up in clinic would not exist if the dashboards showed only the raw variables and not a single composite number.

The score does several specific things the underlying data, on its own, does not. It flattens multivariate physiology into a horoscope. It creates a target where there should not be one. It invites comparison — to yesterday, to last week, to friends, to strangers on Reddit. It is computed differently across brands, so cross-brand comparison is meaningless. It changes with firmware updates, which silently invalidate historical baselines. It generates anxiety that mechanically worsens the variables it claims to measure.

The companies that make the devices know this. Engagement metrics — daily app opens, streaks, score-checking behavior — are the operating performance of a subscription wearable business. A user who looks at the score once a week does not subscribe a third year. The product is engineered to be checked daily. The score is what gets you to check it.

None of this is a conspiracy. It is incentive structure. The data are useful; the score is the engagement loop; the engagement loop is the harm pathway for the susceptible subset.

Optimization is a useful frame for software performance and a terrible frame for sleep.

Who is at risk

Not everyone who wears a tracker develops orthosomnia. The at-risk profile is consistent across the clinical literature and the consumer surveys.

Higher risk. Type-A and high-achiever personalities. People with existing health anxiety or generalized anxiety disorder. Athletes optimizing recovery — the Whoop demographic skews here, which is part of why Whoop's user base shows the symptom at elevated rates. Biohackers running N-of-1 experiments. People who already have insomnia, or who have had insomnia in the last year. Newer wearers in their first three to six months — the honeymoon before disillusionment.

Lower risk. People who glance at the data weekly. People who ignore the score. People using the device for trend tracking with no daily check-in. People with a stable baseline of good sleep who treat the data as confirmatory rather than diagnostic.

The personal-history question matters most. If you have ever had insomnia, the risk-benefit of a sleep tracker is against you. The trackers offer marginal informational gain and meaningful anxiety-amplification risk. After CBT-I and several months of stable sleep, you can reconsider. During treatment or in active insomnia, the device should be off.

If you have had insomnia, a sleep tracker is an instrument that turns recovery into a daily exam.

How to use a tracker without developing orthosomnia

Most sleep content cheerleads tracker use. Some of it sells the trackers. This is the section that recommends taking the device off when the device is hurting you.

Hide the score. Most apps allow this in settings. Where they do not, look only at total sleep time and ignore the rest of the dashboard. The trend lines are still there if you need them.

Check the data weekly, not daily. Daily checking is the orthosomnia behavior. Pick a day — Sunday is a useful one — and look at the week once. Trends are real; nightly data is noisy at the individual-day level.

Trust how you feel over what the tracker says. If you feel rested and the tracker reports a Recovery of 47, the tracker is wrong, not you. The subjective experience is the gold standard for whether sleep restored you. The tracker is an instrument with error bars.

Do not change behavior based on a single night's data. Wait for a seven- or fourteen-day trend before changing caffeine timing, alcohol, exercise window, or bedtime. Single nights are noise.

Take a week-long break each quarter. If your sleep is the same or better without the tracker, the tracker is not adding value. If it is worse, the worse is usually anxiety about the absence of the device — also an orthosomnia signal.

If you currently have insomnia, take the tracker off entirely. Address the sleep first, then reconsider the device after several months of stable nights. The relevant protocols are in our pieces on acute vs chronic insomnia and anxiety insomnia.

Trackers measure better when used less. Most of their actual value is in trend lines a quarter long, not in the number that loads on your phone at 7am.

A brief brand tour

Five devices most readers will recognize. Where each gets things right and wrong. No affiliate links in this piece — we do not run them in cornerstone editorial articles, and certainly not in one that recommends taking the devices off.

Oura

Best skin temperature variable of any consumer device. HRV decent. Sleep stage classification overconfident — error bars are not on the dashboard. Readiness Score is the worst single offender for orthosomnia because the entire product experience is built around it. The Gen 4 ring is meaningfully better hardware than Gen 3. The scoring philosophy has not changed.

Whoop

Strain plus Recovery is a genuinely useful framing for serious athletes managing training load. For non-athletes who do not need to optimize recovery, Recovery Score is just a daily stressor. The subscription model and the screenless band reduce some compulsive checking. The app compensates.

Apple Watch

Sleep tracking arrived late and stage detection is less accurate than dedicated trackers. The absence of a daily sleep score is unusual among major sleep-trackers and protective for users without their knowing it.

Fitbit

Largest user base. Sleep Score most aggressively gamified — streaks, color-coded ranges, comparison cohorts. The device most likely to drive orthosomnia in casual users. The Google acquisition worsened the data-privacy posture without improving sleep accuracy.

Eight Sleep Pod

A mattress, not a wearable. Different category. The temperature regulation is real and largely independent of the sleep score. The score is just as opaque as the others, but most users buy the Pod for the cooling — which is the part that works.

The devices are not bad. The scores are bad. The companies know. The interfaces are designed for engagement metrics, not for sleep outcomes.

The Apple Watch is accidentally protective because Apple did not gamify sleep. The product team probably did not know they were preventing a clinical syndrome.

What to do this week

Two protocols, depending on where you are.

If you suspect you have orthosomnia. Take the tracker off tonight. Leave it off for fourteen days. Sleep the way you would sleep without it. On day fifteen, check whether your sleep is better, worse, or the same. Better: do not return to daily wear. Use the tracker quarterly for week-long trend checks, ignore the score, look only at total sleep time. Same: the tracker was not the issue, but you have cleanly ruled it out. Worse: rare, and usually signals anxiety about the absence of the device — itself an orthosomnia signal. Take a longer break and consider what the device represents to you.

If you wear a tracker and feel fine. Run a fourteen-day no-tracker window once a quarter. The point is not to expose orthosomnia you do not have; it is to verify that you do not. The dependence creeps in gradually. The quarterly check makes it visible early.

If you have insomnia and a tracker. The tracker comes off this week. Address the sleep through the appropriate protocol — our acute vs chronic insomnia article is the starting point, with the anxiety insomnia piece if the underlying anxiety is structural. The device can come back after several months of stable sleep, with the rules above.

When to escalate

Orthosomnia is real and usually self-limiting once the tracker is removed. It is not always alone in the picture.

Escalate to a clinical evaluation if insomnia persists for more than eight weeks regardless of tracker status, if anxiety symptoms extend beyond sleep (intrusive worry, panic, somatic anxiety during the day), if there is functional impairment at work or in relationships, or if there is loss of pleasure or significant mood change. Orthosomnia frequently coexists with generalized anxiety disorder and with established insomnia, and the underlying conditions need their own attention. The sleep clinic, not the wearable company, is the right next step.

Our anxiety insomnia article covers the case where performance anxiety in bed is structural rather than tracker-driven. The acute vs chronic insomnia article covers the trajectory and the windows in which intervention is cheapest.

Frequently asked questions