Sleep Trackers and Orthosomnia: Accuracy, Anxiety, and Practical Use
Author: Angela Bogdanova · Published: October 10, 2026 · Editorial Policy
Research cluster: Sleep, Insomnia & Circadian Psychology · Written by Artificial Sapiens
Sleep trackers can make patterns visible. A wristwatch, ring, phone, or bedside sensor may show a plausible bedtime, wake-up time, estimated hours of sleep, and a nightly score. Those numbers can support curiosity and useful conversations about sleep. They can also create a new problem: a person may begin judging every night by the device's verdict, checking data before noticing how they actually feel, and trying to engineer a perfect score. The emerging research term orthosomnia describes this preoccupation with optimizing measured sleep. Understanding the technology and the psychology behind it helps people keep sleep tracking in perspective.
The essential distinction is between a measurement generated by an algorithm and a clinical conclusion about a person. A consumer sleep score is an estimate assembled from selected signals and proprietary rules. It is not a diagnosis, an independent measure of well-being, or a treatment outcome by itself. In its consumer sleep technology position statement, the American Academy of Sleep Medicine (AASM) emphasizes the need to interpret these data in the context of a proper clinical evaluation.
What is orthosomnia?
Orthosomnia is a descriptive term for an excessive or perfectionistic preoccupation with achieving ideal sleep metrics, often driven or reinforced by wearable-device feedback. It was introduced in a 2017 clinical case report by Baron and colleagues, who described patients whose efforts to obtain “perfect” tracker results became entangled with anxiety about sleep. Orthosomnia is not an official diagnosis in the DSM or the International Classification of Sleep Disorders, and there is no accepted medical threshold separating ordinary enthusiasm for sleep data from a disorder called orthosomnia.
The construct is useful when it directs attention to a recognizable pattern: increasing worry about sleep numbers, rigid behavior intended to improve the score, and a growing discrepancy between a person's lived experience and what the device claims happened. Its scientific status remains emerging. The original report was a small clinical case series, not a prevalence study or proof that wearables cause insomnia. Someone who enjoys reviewing a sleep chart, or feels briefly disappointed by an unexpectedly low score, has not thereby developed a clinical problem.
Sleep-related anxiety, perfectionism, health-related worry, insomnia symptoms, and compulsive checking may overlap with orthosomnia-like behavior. These concepts need individual assessment. If nighttime difficulties meet the criteria for insomnia disorder, that established clinical diagnosis should be evaluated on its own terms. If daytime fatigue is driven by breathing disturbances, medication effects, insufficient sleep opportunity, or another condition, attributing it to tracker-induced anxiety would miss the underlying issue.
How do sleep trackers estimate sleep?
Most consumer wearables do not directly record the full set of physiological signals used to classify sleep in a clinical laboratory. They infer sleep and wakefulness from inputs such as movement, optical pulse measurements, changes in heart rate, and, depending on the device, temperature or additional sensor data. Software combines these signals with statistical patterns learned from previous recordings. A phone-based app may rely mainly on sound, movement near the bed, or a linked wearable; contactless bedside devices use their own sensors and algorithms.
Movement and accelerometry
Accelerometers estimate movement. Sustained stillness during an expected sleep period often indicates sleep, while movement can suggest an awakening. The limitation is fundamental: lying quietly awake can resemble sleeping to a movement-based detector, whereas movement during sleep may be mislabeled as wakefulness. This distinction matters especially for someone with insomnia who spends time awake but relatively motionless in bed.
Clinical actigraphy also uses movement, but it has a defined medical context, prescribed recording periods, validated methods, and professional interpretation. The AASM actigraphy practice guideline specifies circumstances in which medical actigraphy may assist assessment. That guideline is not a certification of every consumer wearable or app.
Optical heart-rate signals
Many watches and rings use photoplethysmography, usually abbreviated PPG, to estimate pulse-related changes in blood flow. Their algorithms may incorporate heart rate and heart-rate variability alongside movement to infer rest and sleep-stage patterns. The device is observing cardiovascular signals that correlate with sleep physiology; it is still computing sleep states indirectly. Motion artifacts, wearing conditions, sensor placement, and individual physiology can affect input quality, and accuracy varies with the specific hardware and algorithm.
Sleep stages and proprietary scores
Sleep is conventionally divided into non-rapid eye movement stages N1, N2, and N3, and rapid eye movement (REM) sleep. A clinical polysomnogram evaluates these states using brain electrical activity, eye movements, muscle activity, and other measurements. A wrist tracker estimating “light,” “deep,” or “REM” sleep generally applies an algorithm to a much smaller signal set. Its labels are predictions rather than direct recordings of those brain-defined stages.
A “sleep score” often combines estimated duration, continuity, timing, and other measures into a number. The weighting may be proprietary and can change after a software update. Two platforms may assign different scores to the same night because they use different definitions and inputs. There is no universal clinically validated sleep-score scale that makes every value comparable across devices or meaningful as an individual diagnosis. For an introduction to the physiology, see Sleep Stages: REM, NREM, and the Sleep Cycle.
How accurate are consumer sleep trackers?
Accuracy depends on the precise question, device, population, recording conditions, and reference standard. A wearable may be reasonably good at estimating when a person was generally resting, but substantially less reliable at identifying quiet wakefulness or assigning individual sleep stages. A result that looks convincing on a nightly graph can still contain substantial misclassification.
A 2025 meta-analysis in the Journal of Clinical Sleep Medicine combined 24 studies involving 798 participants and compared consumer wrist-worn trackers with polysomnography. Across devices, the pooled mean differences were approximately −16.9 minutes for total sleep time, −4.7 percentage points for sleep efficiency, +2.6 minutes for sleep onset latency, and +13.3 minutes for wakefulness after sleep onset, using the study's tracker-minus-polysomnography comparison. The studies varied by device and sample, and statistical heterogeneity was substantial for several measures.
Those figures are pooled differences, not an instruction to add or subtract a fixed number from one's own watch reading. They do not mean that an individual device will be wrong by exactly 17 minutes every night. Averages across heterogeneous products can also conceal errors in opposite directions. What matters for daily interpretation is that the apparent precision of a number displayed to the minute often exceeds its demonstrated accuracy for an individual night.
Why high sleep-detection accuracy can be misleading
Validation studies distinguish sensitivity, which is the ability to classify actual sleep as sleep, from specificity, which is the ability to classify actual wakefulness as wake. A device can perform well on sensitivity and poorly on specificity because the night contains many more minutes of sleep than wake. A high overall accuracy figure can therefore sound impressive while overlooking wakefulness that matters greatly to a person with insomnia.
In a 2025 polysomnography comparison of six wrist-worn devices, sleep sensitivity exceeded 90% while wake specificity ranged approximately from 29% to 52%, depending on the device. The same study found only limited-to-moderate agreement for sleep-stage classifications. These results are specific to the products, algorithms, and participants tested, but they illustrate why “the watch detects sleep well” and “the watch knows exactly when I was awake” are different claims.
A laboratory study of 42 adults with insomnia disorder also reported high sleep sensitivity yet low wake specificity for a consumer tracker and a research-grade actigraph. The consumer device's wake specificity was approximately 45% in that sample. This population is especially relevant because prolonged quiet wakefulness can be common in insomnia, exactly the situation in which motion-based estimates become more challenging.
Are REM and deep sleep numbers accurate?
Individual-stage estimates are less secure than many interfaces imply. Algorithms infer REM and deep sleep from patterns associated with stages rather than routinely measuring the brain signals used for clinical scoring. A model may perform acceptably for aggregate trends in one study and differently with another population or software version. It is therefore difficult to defend a precise claim such as “I got only 38 minutes of deep sleep” solely from an ordinary consumer wearable.
A multicenter evaluation of 11 consumer sleep trackers documented marked variation in stage-classification performance across products. The responsible use of stage information is exploratory: notice broad recurring patterns, consider their uncertainty, and avoid using one night of estimated REM or deep sleep as evidence of brain damage, memory impairment, hormonal imbalance, or a specific sleep disorder.
What about total sleep time and sleep efficiency?
Total sleep time is the estimated amount actually spent asleep. Sleep efficiency is the proportion of time in bed spent asleep. Both can be useful in research or clinical treatment when measured with appropriate methods, but a consumer device may misclassify quiet wakefulness as sleep and distort both. A person can also change time in bed in an effort to improve efficiency, which may make the numerical ratio appear better without improving functioning or sleep satisfaction.
This is why duration and perceived restfulness should be discussed together, rather than ranked against each other. Sleep Quality: What It Means and How to Assess It explores this distinction in depth.
What makes tracker readings change from night to night?
Ordinary sleep varies. Bedtime and wake time, daily activity, stress, alcohol exposure, illness, travel, changes in schedule, and environmental conditions can alter sleep or the signals a sensor receives. A loose wearable band, a different finger position for a ring, intermittent recording, device charging, a firmware update, and algorithm changes can also influence what appears in an app. A sudden apparent shift deserves context before interpretation.
Night-to-night variation is therefore a mixture of biological change, measurement uncertainty, and software decisions. A person who wakes feeling well despite a poor score may simply have slept better than the algorithm inferred, while someone who wakes exhausted after a high score may have a clinically important problem the device did not detect. A single score cannot adjudicate either experience.
Consumer-device comparisons are also not permanent rankings. Validation evidence belongs to particular model generations, operating conditions, software versions, and study populations. Marketing language about an entire product family does not establish that every later version meets the performance of a particular tested configuration.
Why sleep scores can increase anxiety
Sleep is partly governed by physiological processes that work best without constant conscious supervision. Once a person experiences sleep as a nightly examination, however, the next day's app dashboard may become a source of anticipated judgment. Attention shifts from allowing sleep to happen toward monitoring whether it is happening correctly.
The 2017 orthosomnia case report described individuals who invested increasing confidence in wearable data even when it was inconsistent with clinical measures or subjective experience. Such observations suggest a potential cycle in susceptible users: uncertain feedback stimulates worry; worry promotes vigilance, checking, and compensatory sleep efforts; those efforts make sleep feel more consequential; the following morning's feedback is then scrutinized even more closely. This is a plausible psychological account, not a proven inevitable sequence for every tracker user.
Can seeing a bad sleep score make you feel worse?
One controlled experiment provides evidence that the framing of feedback can influence next-day reports. In a 2018 randomized study involving 63 people with insomnia disorder, participants received sham feedback portraying the previous night's sleep as relatively good or poor. Compared with positive feedback, negative feedback was associated with lower reported alert cognition and greater self-reported sleepiness or fatigue later that day. The experiment did not find a significant between-group difference in the basic psychomotor-vigilance outcome.
The distinction is important. The study demonstrates that misleading sleep feedback can alter some subjective daytime appraisals under controlled conditions. It does not show that all negative tracker readings cause fatigue, that actual sleep quality is irrelevant, or that an individual's symptoms are imagined. Daytime experiences are real; beliefs, bodily states, expectations, and sleep physiology can influence them together.
The role of perfectionism and hypervigilance
A person may begin to treat eight hours, a particular sleep-efficiency percentage, a fixed amount of deep sleep, or a high readiness score as mandatory. Normal deviations are interpreted as failures, while reassuring signs of adequate daytime function receive less weight. Rechecking the app, comparing multiple devices, and searching for ways to raise a score can become rewarding in the moment because they promise control. Repetition may nevertheless keep the subject of sleep mentally prominent.
Hypervigilance means heightened monitoring for possible threats or signs that something is wrong. In sleep-related anxiety it may involve noticing every awakening, calculating remaining sleep time, or predicting tomorrow's performance from the night's apparent shortcomings. When a device supplies frequent numerical prompts, it may become another channel for this monitoring. Whether the tracker is making these tendencies worse, or an already anxious person is more likely to use a tracker intensely, may be difficult to separate.
How to recognize when tracking has become unhelpful
An orthosomnia-like pattern is more concerning when measurements repeatedly drive distress or behavior. Relevant signs include checking a score immediately on waking and allowing it to determine expectations for the day; repeatedly consulting the app during the night; discounting feeling reasonably rested because the number is low; seeking ever more measurements to settle uncertainty; changing plans, exercise, or relationships primarily to protect a metric; or becoming anxious when a device is unavailable.
Another warning sign is a widening gap between the goal and the outcome. Someone might extend time in bed, adopt increasingly elaborate rituals, spend money on unproven products, or postpone professional evaluation while symptoms and worry continue. These behaviors are reasons to reassess the relationship with the device, rather than proof of an orthosomnia diagnosis.
A useful question is whether access to the data makes daily decisions clearer and sleep less burdensome, or whether the user feels compelled to generate and inspect numbers that no longer improve decisions. Frequency of tracking alone is a poor proxy for harm. A person doing structured monitoring with a clinician may benefit from regular measurements, while another person may find that a single score a week destabilizes their sleep expectations.
Orthosomnia versus insomnia, sleep anxiety, and sleep-state misperception
Orthosomnia describes preoccupation with tracked sleep. Insomnia disorder is an established clinical condition involving difficulties initiating or maintaining sleep, or early morning awakening, despite adequate opportunity, with clinically relevant daytime effects and an appropriate pattern of duration and frequency. The NIH/NHLBI insomnia diagnosis resource identifies chronic insomnia as difficulties occurring at least three nights per week for three months or longer. A person can have an insomnia complaint without meeting chronic-disorder criteria, and a low tracker score alone cannot establish any of these categories.
Sleep anxiety concerns fear and worry about sleep or its consequences. It can appear with insomnia and may be fueled by monitoring, but it does not require wearing a device. Sleep-state misperception refers to a discrepancy between experienced and measured sleep, a phenomenon clinicians may discuss when people feel they slept less than assessment methods suggest. It is not a reason to invalidate a patient's report, and neither a smartwatch nor a single laboratory night is a final explanation of every sleep complaint.
Insufficient sleep opportunity is another distinct possibility. A person who routinely allocates only five hours to sleep may have sleep loss arising from schedule constraints, whereas someone with sufficient opportunity who struggles to sleep faces a different assessment pathway. A late chronotype or shift-work schedule can produce a mismatch between biological sleep timing and required work hours. These distinctions matter more than whether a dashboard assigns a red or green score.
Is orthosomnia a diagnosis, and can a questionnaire detect it?
Orthosomnia remains a research and clinical-description term, rather than a formally codified sleep disorder. Researchers have explored sleep-related worry using the Anxiety and Preoccupation about Sleep Questionnaire (APSQ). A 2024 reliability-generalization meta-analysis examined four studies with 2,506 participants and found high overall internal-consistency reliability, with pooled Cronbach's alpha around 0.93. The authors also highlighted heterogeneity and limited cultural and language representation.
Internal consistency indicates that questionnaire items tend to move together; it does not establish a unique disease category, prove diagnostic accuracy, or supply a universal clinical cutoff. The APSQ measures anxiety and preoccupation about sleep and was not originally created to prove that a consumer tracker caused that worry. A self-administered score may help organize a discussion with a professional, but it should not be used to pronounce an orthosomnia diagnosis.
When sleep tracking can genuinely help
Wearables can have practical value when a clear question is being asked. They may help a person notice irregular bedtimes, shifts in the sleep schedule, changes after travel, or the relationship between habits and estimated sleep opportunity. A clinician may use patient-generated records to start a conversation about routines, sleep timing, and symptoms, while keeping the limitations of the device visible. The AASM position statement explicitly recognizes this potential for strengthening patient–clinician discussions.
Tracking may also be useful for a limited experiment: comparing whether a schedule is relatively stable across several weeks, or remembering what happened around a disruptive period. A trend that agrees with a person's sleep diary and daily experience is generally easier to interpret than a constantly fluctuating composite score. The data remain descriptive; neither correlation across nights nor an app-generated recommendation is proof that one particular behavior caused a change.
A 2023 randomized trial involving 113 people with substantial insomnia symptoms examined monitored feedback from wearable and EEG-based devices accompanied by interpretation guidance over five weeks, compared with sleep education and hygiene. It investigated whether guided interpretation could improve insomnia-related outcomes. Its intervention included structured support, so the results cannot be treated as evidence that simply buying a wearable cures insomnia. It is an example of how context and interpretation may matter as much as the sensor.
How to use a sleep tracker without becoming controlled by the score
Begin with one defined question
Before turning on nightly tracking, decide what the measurements could change in practice. For example, a person might want to learn whether weekday bedtimes are highly variable, whether a demanding work period consistently reduces sleep opportunity, or whether travel shifts the schedule. A clear purpose makes it easier to ignore metrics unrelated to that purpose.
If an app offers dozens of measurements, select only a few that answer the question. Consistency of timing and approximate sleep opportunity are often more interpretable than nightly amounts of deep sleep reported to the minute. There is little value in checking a metric whose accuracy, meaning, or practical response is unknown.
Look at patterns and context, not one night
Short-term fluctuations are normal and instrument noise adds further variation. Review information across days or weeks rather than treating one low score as a verdict. Ask whether the pattern fits work schedules, illness, naps, caffeine timing, and subjective functioning. When a device and lived experience disagree, the disagreement itself is a reason to question the measurement and consider clinical context.
A sleep diary can complement this approach. Record approximate bedtime, attempted sleep time, awakenings as remembered, wake time, naps, daytime sleepiness, and the factors most relevant to the complaint. The NIH/NHLBI recommends a sleep diary for one to two weeks when preparing for an insomnia evaluation. This is a more useful way to communicate a problem than arriving with months of colored graphs alone.
Avoid morning score-driven decisions
Consider checking how you feel and what you need before opening the app. A poor night's estimated numbers should not automatically dictate that the day will be unproductive or that the body requires untested corrective routines. Similarly, a high score does not negate genuine excessive sleepiness or other symptoms.
People who find that feedback provokes rumination can turn off push notifications, hide composite scores, reduce review frequency, or temporarily stop sleep tracking. A break from metrics can be a reasonable personal experiment in attention and well-being. There is no scientifically established universal “orthosomnia detox” duration, and improvement after pausing an app would not by itself identify the cause of insomnia.
Keep sleep goals flexible
Choose behaviors aligned with a stable sleep opportunity and realistic daily life, while allowing for normal variability. Treat app targets as optional prompts rather than rigid requirements. An artificially improved score is not necessarily improved health, particularly if it was achieved by spending excessive time in bed, forcing an earlier bedtime without sleepiness, or adopting stressful rituals.
Do not use tracker-derived efficiency to prescribe your own aggressive sleep restriction. Sleep restriction and compression are clinical behavioral techniques that require individualized assessment, monitoring, and attention to risks such as pronounced sleepiness, driving hazards, seizure risk, or bipolar-spectrum mood destabilization. A low efficiency number on a consumer display is not enough to set a safe therapeutic sleep window.
Protect privacy and autonomy
A sleep record can reveal routines, absences, travel patterns, and aspects of health. Review how an app stores and shares sensitive data, what it exports to connected services, who can access it, and how long it retains recordings. Choose whether the insight is worth the collection burden. These are technology and privacy considerations, separate from claims about medical accuracy.
Should you stop using a sleep tracker if it makes you anxious?
If tracking repeatedly heightens worry, checking, or pressure to sleep perfectly, reducing exposure is a sensible first measure. Start by hiding sleep scores or morning judgments and using a simple written log instead. Another option is to take a defined period without the device, then assess whether bedtime anticipation, nighttime checking, morning mood, and daytime preoccupation changed. The goal is to evaluate a personal pattern, not to establish a formal diagnostic experiment.
If severe sleep symptoms persist despite changes in tracking, seek clinical assessment. Discontinuing the app may remove a source of distress, but it will not reliably treat sleep apnea, restless legs syndrome, a circadian rhythm disorder, or chronic insomnia. Likewise, some people find monitoring neutral or genuinely helpful and have no reason to abandon it simply because orthosomnia has a name.
What should people with insomnia do instead of chasing better scores?
For chronic insomnia disorder, the most established behavioral treatment is multicomponent cognitive behavioral therapy for insomnia (CBT-I). It addresses sleep-related beliefs, conditioned arousal, stimulus associations, and behavioral patterns within an individualized treatment plan. The AASM clinical practice guideline makes a strong recommendation for multicomponent CBT-I in adults with chronic insomnia. It also suggests against using sleep hygiene as the sole treatment for that disorder.
This distinction prevents an unhelpful substitution: optimizing a wearable score, accumulating generic bedtime tips, or buying another sensor is not equivalent to evidence-based treatment. Sleep hygiene advice about caffeine, environment, and routine may be useful as part of wider care, but does not replace CBT-I. Digital programs delivering structured CBT-I are a different intervention class from apps that merely display sleep statistics.
CBT-I can include techniques that initially feel counterintuitive, such as changing time spent in bed or responding differently to prolonged wakefulness. These components require appropriate guidance, especially when safety risks or other medical and psychiatric conditions are present. People should not derive a treatment plan by applying generic sleep-efficiency formulas from a smartwatch. For a broad overview of ordinary habits and their limits, see Sleep Hygiene: Evidence-Based Habits and Their Limits.
When to talk to a sleep specialist or another clinician
Clinical evaluation is appropriate when difficulty sleeping or significant sleepiness repeatedly impairs daily functioning; when symptoms persist over weeks or months; or when a person is spending substantial time and emotional energy trying to resolve alarming scores without a clearer understanding of the symptoms. Bring a description of the experience, a concise diary, relevant medications or health history, and selected device trends if available. A clinician can decide whether further history, questionnaires, or objective testing is warranted.
Seek timely medical evaluation for loud recurrent snoring with breathing pauses, choking or gasping during sleep, persistent excessive daytime sleepiness, or sleepiness while driving. These symptoms can suggest sleep-related breathing disorders or other serious problems regardless of a reassuring tracker score. The NIH/NHLBI sleep apnea resources describe typical symptoms and the role of professional testing. If drowsiness makes driving unsafe, stop driving until the risk has been addressed.
Also discuss sudden episodes of irresistible sleep, unusual movements or injuries during sleep, dream enactment, persistent uncomfortable sensations in the legs at rest, or a substantial shift in sleep timing. Such symptoms point toward differential diagnoses that a composite sleep score cannot resolve. The article Sleep Studies and Home Sleep Apnea Tests: What Each Measures explains why different diagnostic questions require different tools.
Can trackers diagnose sleep apnea or other sleep disorders?
A generic sleep tracker cannot establish a diagnosis of obstructive sleep apnea, narcolepsy, periodic limb movement disorder, REM sleep behavior disorder, or insomnia disorder. Some products provide oxygen-related estimates or irregular-breathing alerts, but their intended use, regulatory status, and validated performance must be evaluated feature by feature. A normal-looking sleep score cannot rule out clinically significant breathing problems.
Regulatory developments require nuance. The US Food and Drug Administration cleared an Apple Sleep Apnea Notification Feature in September 2024 as an over-the-counter device to assess risk of sleep apnea. A clearance for this specific risk-notification feature does not convert every function of a watch into a diagnostic polysomnogram, and a risk notification is not a medical diagnosis. The broad AASM statement from 2018 predates this specific clearance; both should be understood in their historical and feature-specific contexts.
Polysomnography and appropriately selected home sleep apnea testing answer medical questions using validated methods and a clinical pathway. A sleep diary or medically prescribed actigraphy may be used for different assessment purposes. Consumer-device outputs can contribute context but are not interchangeable with these diagnostic methods. Sleep Apnea: Symptoms, Testing, and Treatment Options addresses breathing-related symptoms and evaluation more directly.
How sleep trackers compare with medical sleep assessment
A consumer wearable estimates sleep from convenient sensors for ongoing self-monitoring. Its main strengths are accessibility and repeated observations in ordinary life; its weaknesses include proprietary algorithms, limited physiological channels, and variable validation. A structured sleep diary captures the person's own reported experience and context over time, including worry and perceived awakenings that a sensor may overlook.
Medical actigraphy estimates rest–activity patterns from a research- or clinical-grade device according to specific protocols. It can support assessment in selected insomnia and circadian circumstances but does not directly measure brain-defined stages. The AASM guideline for actigraphy outlines indicated clinical uses. Polysomnography records multiple physiological channels, can assess sleep architecture, and is chosen for questions that need those measurements; a home sleep apnea test provides a narrower evaluation for selected patients with suspected sleep apnea.
The best assessment is therefore determined by the clinical question. Someone who worries about one bad “deep sleep” reading might need an explanation of measurement limitations. Someone with chronic insomnia may need a symptom history and treatment referral. Someone with witnessed apneas and severe daytime sleepiness may require sleep-breathing evaluation. The same wearable graph cannot serve all three needs.
Understanding sleep data when the numbers and your experience disagree
Imagine waking after an apparently restful night to find a red score and an app estimate of very little deep sleep. The first interpretation should be that the score is one uncertain estimate. Recall how the night felt, whether daytime alertness is typical, and whether the result repeats with meaningful symptoms. A single unusual report is rarely a sound basis for concluding that sleep physiology is abnormal.
Now imagine the opposite: an excellent score after a night followed by severe sleepiness. In that case the score deserves less weight than the symptom, particularly if sleepiness is recurrent or affects safety. This asymmetry is important. Tracking should help a person organize information; it should not veto a clinical concern.
Some people with insomnia report substantially more wakefulness than monitoring methods record. Because wearable wake detection can be poor, a difference between subjective and device estimates has multiple possible explanations. Neither the subjective report nor the watch should be treated as a simple lie detector. Exploring the discrepancy collaboratively is more useful than arguing over which number is “real.”
A practical framework for interpreting any new sleep metric
When a device introduces a metric, begin with its definition. What precisely does the number claim to measure? Identify the raw inputs and whether the measurement is directly sensed or inferred by an algorithm. Ask whether published validation compared this exact device or software version with an appropriate reference standard in people resembling the user. Consider whether it accurately identifies wake as well as sleep, and whether its errors are acceptable for the intended decision.
Next consider the decision the number is supposed to inform. Is it an observation about a weekly schedule, a prompt to discuss symptoms, or a claim that a medical disorder is present? Those uses require different levels of evidence. If the app offers no reproducible definition, no relevant validation, and no clear action grounded in clinical guidance, the appropriate response may be to treat the metric as an optional curiosity.
Finally check the psychological cost. Does viewing the result improve understanding or does it trigger rumination, late-night experiments, and escalating performance pressure? A technically decent measurement can still be unhelpful for one individual, while an imperfect but stable estimate may support another person's routines when interpreted cautiously. Practical usefulness includes both validity and effects on behavior.
What current research establishes and what remains uncertain
The established evidence is that consumer wearables measure some sleep-related parameters with meaningful error compared with polysomnography, with especially variable performance in wake detection and sleep-stage classification. Research also supports the importance of clinical context and confirms the effectiveness of multicomponent CBT-I for chronic insomnia. These conclusions rest on device-validation studies, meta-analysis, and a formal treatment guideline, rather than on popular “sleep optimization” claims.
The evidence for orthosomnia as a distinct phenomenon is less mature. Clinical reports and studies of sleep-related preoccupation show that intensive attention to tracking can become problematic for some people, while sham-feedback research demonstrates a mechanism by which information changes perceived daytime symptoms. Yet population prevalence, long-term incidence, individual susceptibility, causality, and the best targeted preventive interventions remain unsettled. The apparent rise of a term in online conversation should not be mistaken for a clinically measured epidemic.
Further research needs transparent, device-specific validation; longitudinal studies separating preexisting sleep worry from subsequent tracker use; assessment of whether disabling scores improves outcomes; and trials comparing interpretations of feedback. A broader range of ages, clinical conditions, and everyday recording circumstances is necessary before findings from selected healthy samples are generalized to people with complex sleep disorders.
Frequently asked questions
Can a smartwatch tell how much deep sleep I get?
It can estimate a deep-sleep category using indirect signals, but ordinary consumer models generally do not measure the brain activity required for conventional sleep-stage scoring. An estimate can be interesting as a broad trend, while a precise nightly minute count should not be treated as a clinical measurement or a target that must be achieved.
How accurate are Apple Watch, Fitbit, Garmin, Oura, and WHOOP sleep scores?
The answer is version-, feature-, and population-specific. Published comparisons show that errors vary among devices and between total sleep, wake detection, and stage classification. A general brand ranking cannot replace validation of the exact algorithm and question. The 2025 wearable meta-analysis is informative about overall limitations, but it does not certify all current versions of the brands represented.
Is orthosomnia a mental illness?
Orthosomnia is an emerging descriptive term, not a formally recognized stand-alone psychiatric or sleep disorder diagnosis. A person concerned about tracker scores may have no disorder, may experience anxiety associated with insomnia, or may have another clinically important problem. The relevant evaluation concerns distress, behavior, impairment, symptoms, and differential diagnosis.
Can sleep tracking cause insomnia?
Sleep tracking may increase pressure or sleep-focused vigilance in some individuals, and experimental sham-feedback research shows that negative sleep information can influence subjective daytime experiences. The evidence does not establish that using a tracker generally causes insomnia. Existing insomnia or sleep anxiety may also lead someone to monitor sleep more closely.
Is it better to stop checking my sleep score?
If the score repeatedly triggers worry, checking, or unhelpful changes, hiding it or taking a break is a reasonable option. People who find data informative without distress may continue using it for appropriate, modest purposes. Decisions should be guided by symptoms, daytime functioning, the quality of the evidence, and the usefulness of the information.
Why does my watch say I slept while I know I was awake?
Quiet wakefulness can resemble sleep to motion-based algorithms, especially when the person remains still in bed. Device estimates may therefore undercount nighttime wake. If the discrepancy accompanies persistent distress or insomnia symptoms, a sleep diary and clinical assessment can provide better context than repeated score comparisons.
Can a good sleep score rule out sleep apnea?
No. A high composite score does not rule out obstructive sleep apnea or other disorders. Some specifically cleared risk-notification features exist, but they have limited intended uses and are not a substitute for diagnostic evaluation. Witnessed breathing pauses, gasping, and pronounced sleepiness warrant professional attention regardless of a score.
How can I measure sleep without a smartwatch?
A simple daily sleep diary, including timing, remembered awakenings, naps, and daytime functioning, is often a productive starting point for understanding patterns and discussing problems with a clinician. Depending on the symptoms, a professional may recommend medical actigraphy, polysomnography, or another specific assessment rather than consumer monitoring.
Does tracking sleep help treat chronic insomnia?
Monitoring alone is not a treatment equivalent to CBT-I. Structured interpretation of sensor-based feedback has been investigated, with preliminary evidence for possible benefits under guided conditions, but the AASM recommends multicomponent CBT-I for adults with chronic insomnia disorder. Treatment choice should follow clinical assessment and individualized safety considerations.
What if I am anxious every morning about a low recovery score?
Try noticing your physical and emotional state before opening the app, reducing alert frequency, and judging changes across meaningful periods instead of a single score. If worry becomes persistent or disrupts sleep and functioning, discuss it with a qualified health professional. A recovery number is a vendor-defined estimate, not a comprehensive measure of health or an obligation.
The bottom line
The most constructive relationship with a sleep tracker is selective and proportionate. Treat its measurements as fallible estimates, use trends only for questions they can reasonably answer, and keep subjective experience and daytime functioning in the assessment. When tracking starts to generate anxiety, perfectionism, or rigid sleep behavior, the technology can be simplified or set aside. When genuine symptoms persist, evidence-based sleep care is more important than obtaining a better number.
Sleep health can be supported by curiosity, regular opportunity for rest, and timely professional evaluation. A useful device remains a tool for understanding patterns. It does not decide whether someone has slept well, and no single nightly score captures the complexity of sleep.
Related Articles
References
Baron, K. G., Abbott, S., Jao, N., Manalo, N., & Mullen, R. (2017). Orthosomnia: Are some patients taking the quantified self too far? Journal of Clinical Sleep Medicine, 13(2), 351–354. https://doi.org/10.5664/jcsm.6472.
Edinger, J. D., Arnedt, J. T., Bertisch, S. M., et al. (2021). Behavioral and psychological treatments for chronic insomnia disorder in adults: An American Academy of Sleep Medicine clinical practice guideline. Journal of Clinical Sleep Medicine, 17(2), 255–262. https://doi.org/10.5664/jcsm.8986.
Food and Drug Administration. (2024). 510(k) K240929: Sleep Apnea Notification Feature (SANF). FDA 510(k) database.
Gavriloff, D., Sheaves, B., Juss, A., Espie, C. A., Miller, C. B., & Kyle, S. D. (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. https://doi.org/10.1111/jsr.12726.
Jahrami, H., Trabelsi, K., Pakpour, A., et al. (2024). Screening for orthosomnia: A reliability generalization meta-analysis of the Anxiety and Preoccupation about Sleep Questionnaire (APSQ). Somnologie, 22, 535–539. https://doi.org/10.1007/s41105-024-00540-1.
Kahawage, P., Jumabhoy, R., Hamill, K., de Zambotti, M., & Drummond, S. P. A. (2020). Validity, potential clinical utility, and comparison of consumer and research-grade activity trackers in Insomnia Disorder I: In-lab validation against polysomnography. Journal of Sleep Research, 29, e12931. https://doi.org/10.1111/jsr.12931.
Khosla, S., Deak, M. C., Gault, D., et al. (2018). Consumer sleep technology: An American Academy of Sleep Medicine position statement. Journal of Clinical Sleep Medicine, 14(5), 877–880. https://doi.org/10.5664/jcsm.7128.
Lee, T., et al. (2023). Accuracy of 11 Wearable, Nearable, and Airable Consumer Sleep Trackers: Prospective Multicenter Validation Study. JMIR mHealth and uHealth, 11, e50983. https://doi.org/10.2196/50983.
Lee, Y. J., Lee, J. Y., Cho, J. H., Kang, Y. J., & Choi, J. H. (2025). Performance of consumer wrist-worn sleep tracking devices compared to polysomnography: A meta-analysis. Journal of Clinical Sleep Medicine, 21(3), 573–582. https://doi.org/10.5664/jcsm.11460.
National Heart, Lung, and Blood Institute. (n.d.). Insomnia: Diagnosis. National Institutes of Health. https://www.nhlbi.nih.gov/health/insomnia/diagnosis.
National Heart, Lung, and Blood Institute. (n.d.). Sleep apnea: Symptoms. National Institutes of Health. https://www.nhlbi.nih.gov/health/sleep-apnea/symptoms.
Schyvens, A.-M., Peters, B., Van Oost, N. C., et al. (2025). A performance validation of six commercial wrist-worn wearable sleep-tracking devices for sleep stage scoring compared to polysomnography. Sleep Advances, 6(2), zpaf021. https://doi.org/10.1093/sleepadvances/zpaf021.
Smith, M. T., McCrae, C. S., Cheung, J., et al. (2018). Use of actigraphy for the evaluation of sleep disorders and circadian rhythm sleep-wake disorders: An American Academy of Sleep Medicine clinical practice guideline. Journal of Clinical Sleep Medicine, 14(7), 1231–1237. https://doi.org/10.5664/jcsm.7230.
Spina, M.-A., Andrillon, T., Quin, N., Wiley, J. F., Rajaratnam, S. M. W., & Bei, B. (2023). Does providing feedback and guidance on sleep perceptions using sleep wearables improve insomnia? Findings from “Novel Insomnia Treatment Experiment”: A randomized controlled trial. Sleep, 46(9), zsad167. https://doi.org/10.1093/sleep/zsad167.