Sleep apnea tests can absolutely be wrong, and the error rate is higher than most people realize. A single-night study misclassifies the severity of sleep apnea in roughly one out of four patients, and about one in five people who actually have the condition get a false-negative result from a single night of testing.1PubMed Central. Night-to-night variability in obstructive sleep apnea using peripheral arterial tonometry: a case for multiple night testing The reasons range from how your body naturally fluctuates from one night to the next, to the type of device used, to scoring rules that different labs apply differently.
Your Breathing Changes More Than You Think From Night to Night
The single biggest reason a sleep apnea test can mislead you has nothing to do with the equipment or the technician. Your breathing simply varies a lot from one night to the next. A study using multi-night home monitoring found that the key metric used to diagnose sleep apnea, often abbreviated AHI, varied by an average of 57% between nights in the same person.1PubMed Central. Night-to-night variability in obstructive sleep apnea using peripheral arterial tonometry: a case for multiple night testing That is not a small wobble. If your score on one night puts you in the “moderate” category, the next night might land you in “mild” or “severe.”
A large multi-night study confirmed this problem from a different angle. Among people who genuinely had sleep apnea based on their average across many nights, about 21% would have received a false-negative result on any given single night, meaning the test would have told them they did not have the condition at all.2American Journal of Respiratory and Critical Care Medicine. Multinight Prevalence, Variability, and Diagnostic Misclassification of Obstructive Sleep Apnea Severe cases were correctly classified on about 85% of individual nights, which sounds reasonable until you consider that mild and moderate cases were correctly classified on only about half of their individual nights.3PubMed Central. How Many More Nights? Diagnosing and Classifying Obstructive Sleep Apnea Using Multinight Home Studies If you have mild or moderate sleep apnea, a coin flip has roughly the same odds of giving you the right answer as a single night of testing does.
How Home Sleep Tests Can Undercount
Home sleep apnea tests, sometimes called out-of-center tests, are the most common first step when a doctor suspects sleep apnea. They are convenient and far cheaper than spending a night in a sleep lab. But they have a built-in tendency to underestimate severity.
The core issue is that most home devices cannot tell whether you are actually asleep. In a sleep lab, brain-wave sensors track your sleep stages, so the final score reflects only the time you were asleep. Home kits lack those sensors, so they divide the number of breathing events by the total time the device was recording, including any time you spent lying awake, getting up to use the bathroom, or scrolling your phone.4PubMed Central. Potential Underestimation of Sleep Apnea Severity by At-Home Kits: Rescoring In-Laboratory Polysomnography Without Sleep Staging If you spent two hours awake during the recording, those two hours dilute the count and make your breathing look better than it actually was during sleep. Research has shown that overestimation of total sleep time in home tests drags the severity score downward.5PubMed Central. Effect of Manual Editing of Total Recording Time: Implications for Home Sleep Apnea Testing
There is also the practical matter of putting the device on yourself. In a pediatric home-test study, 11% of recordings failed entirely due to device-related problems like detachment, the child not tolerating the sensor, or setup errors, and another quarter of recordings captured less than two hours of usable data.6PubMed Central. A photoplethysmography-based home sleep apnea test compared with polysomnography in children 2 to 6 years of age Adults have lower failure rates, but a loose finger sensor or a nasal cannula that shifts during the night can still degrade data quality enough to change the result.
How Lab Tests Can Mislead in the Other Direction
If home tests tend to underestimate, you might assume the gold-standard lab test always gets it right. It does not. The lab introduces its own biases.
Sleeping in an unfamiliar environment with wires attached to your scalp, chest, and face is not normal. Researchers call this the first-night effect: your sleep architecture shifts during that initial monitored night, which can alter how many breathing events the study captures.7Sleep Research. Research progress on the first‐night effect in polysomnography studies In children and adolescents, the amount of REM sleep increased during a second night compared to the first, suggesting the first-night recording underrepresented the sleep stage where breathing problems are often worst.8Archives of Disease in Childhood. First night effect for polysomnographic data in children and adolescents with suspected sleep disordered breathing
Body position is another factor. In the lab, the wires and sensors tend to keep people on their backs more than they would be at home. One analysis found that patients spent 56% more time in the supine position during monitored nights with attached leads than on nights without them.9PubMed Central. Influence of Body Position on Severity of Obstructive Sleep Apnea: A Systematic Review Because sleeping on your back worsens airway collapse, this can inflate the severity score. Research on position effects showed that about 60% of patients had their score more than double when supine, and that position-related misclassification, specifically underestimation on nights spent more on the side, affected 20% to 40% of patients.10PubMed Central. The impact of body posture and sleep stages on sleep apnea severity in adults So the lab can push your score higher than what you experience in your own bed, while a home test where you happen to sleep mostly on your side might push the score lower.
The Scoring Rules Are Not Universal
Even if the recording itself is perfect, the way breathing events are counted can shift your diagnosis. The sticking point is how much your blood oxygen level needs to drop during a partial breathing pause for it to count as a scored event. Some labs and guidelines use a 3% drop, others use 4%. That one percentage point sounds trivial, but it reshuffles patients between severity categories dramatically. In one study, changing from the 4% threshold to the 3% threshold increased the proportion of patients classified as having at least moderate sleep apnea from about 29% to about 74%.11PubMed. Effect of different oxygen desaturation threshold levels on hypopnea scoring and classification of severity of sleep apnea
This is not just an academic distinction. The stricter 4% threshold can miss patients who go on to develop real health consequences. Research found that people whose breathing events were only caught by the 3% rule, not the 4% rule, still had meaningfully higher rates of developing high blood pressure.12PubMed Central. Incidence of hypertension in obstructive sleep apnea using hypopneas defined by 3 percent oxygen desaturation or arousal but not by only 4 percent oxygen desaturation In other words, the 4% cutoff can produce a “normal” result while your breathing is still disrupted enough to damage your cardiovascular system over time. If your test report seems surprisingly mild and you know which scoring rule was applied, that context matters.
Software Glitches and Sensor Artifacts
Automated scoring software does the initial pass on most sleep studies, and it makes mistakes. A well-documented case illustrates the problem vividly. A patient’s initial report showed severe sleep apnea with a score of 38 events per hour. On closer review, the scoring software had flagged periods where the nasal airflow sensor suddenly registered zero flow, even though the patient’s chest and abdomen were moving normally and their blood oxygen stayed stable. What actually happened was a transient nasal obstruction that blocked the sensor, not true apnea. After a trained physiologist manually reviewed the data and removed the artifacts, the score dropped from 38 to 28, shifting the classification from severe to moderate.13PubMed Central. Misleading sleep-diagnostic results in a patient with variable nasal airflow obstruction
That is a difference that changes treatment decisions. A score of 38 might fast-track someone to pressurized airway therapy, while 28 opens the door to other options. Not every lab does a thorough manual review, and some home-test services rely more heavily on automated scoring. If your results seem out of step with how you feel or what your bed partner observes, asking whether the study was manually reviewed by a sleep technologist is a reasonable question.
What You Did Before the Test Matters
Alcohol is one of the most potent short-term modifiers of how you breathe during sleep, and it can swing a test result in a meaningful way. A meta-analysis found that drinking alcohol before bed raised the average severity score by about 4 events per hour and lowered the minimum blood oxygen level by almost 3 percentage points compared to sober nights.14PubMed. Impact of Alcohol Consumption on Snoring and Sleep Apnea: A Systematic Review and Meta-analysis In a case report, one man’s score nearly tripled, jumping from about 8 to 22 events per hour, on a night when he had been drinking.15PubMed Central. A Case of Mild OSA Exacerbated by Alcohol Use Alcohol relaxes the muscles that keep your airway open, depresses the part of the brain that responds to breathing disruptions, and often leads you to sleep more on your back.
One of the earliest studies on this topic found that the worsening in oxygen levels persisted for an additional night even when no alcohol was consumed, suggesting the effects are not limited to the night of drinking itself.16The American Journal of Medicine. Alcohol increases sleep apnea and oxygen desaturation in asymptomatic men Sedative medications can have a similar effect. The practical takeaway is straightforward: if you drank the night of your test, your result likely overstates how bad your breathing is on a typical night. If you specifically avoided alcohol for the test but normally drink in the evenings, the test may understate your real-world exposure.
Why Women Are More Likely to Get a False Negative
Sleep apnea has long been thought of as a condition that primarily affects overweight, older men who snore loudly. That stereotype shapes both who gets referred for testing and how the tests are designed to catch problems. Women with sleep apnea often present differently: they report insomnia, fatigue, mood changes, and morning headaches rather than the classic loud snoring and witnessed breathing pauses. Their breathing disruptions during sleep also look different on a recording.
Women, especially before menopause, tend to have more partial obstructions rather than complete blockages, more events concentrated in REM sleep, and more subtle arousals that fragment sleep without producing large drops in oxygen. Standard severity scoring, which counts mainly full or near-complete breathing pauses with significant oxygen dips, can miss these patterns entirely.17PubMed Central. Advances in the Diagnosis and Treatment of Obstructive Sleep Apnea in Women A woman can have severely disrupted, unrefreshing sleep and still come back with a “normal” or “mild” test result because the scoring system was not designed to flag her pattern. Sleep fragmentation indices and REM-specific analysis can help, but not all labs report those measures by default.
Wrist-Worn and Newer Portable Devices
Peripheral arterial tonometry devices, worn on the wrist and finger, have become popular because they are even simpler to use than traditional home sleep test kits. They measure blood vessel tone changes, heart rate, and oxygen levels rather than airflow directly. In patients with coexisting lung disease, one such device showed no significant difference in average severity scores compared to the lab gold standard.18PubMed Central. Accuracy of WatchPAT for the Diagnosis of Obstructive Sleep Apnea in Patients with Chronic Obstructive Pulmonary Disease That sounds encouraging, but the picture gets more complicated at the individual level.
A meta-analysis found clinically meaningful disagreement between these wrist-based devices and lab studies, with significant rates of severity misclassification.19PubMed Central. A meta-analysis of diagnostic test performance of peripheral arterial tonometry studies The accuracy problems are worst at the mild end: among patients classified as having mild sleep apnea by one widely used wrist device, only about half actually had mild disease when checked against a lab study. Roughly 20% turned out to have moderate or severe disease, and 30% had no disease at all.20Frontiers in Sleep. The role of the WatchPAT device in the diagnosis and management of obstructive sleep apnea
Because these devices rely on detecting changes in blood vessel tone, conditions that stiffen arteries, such as cardiovascular disease or atherosclerosis, can interfere with the readings. In patients with stiff arteries, one study found essentially no correlation between the wrist device results and lab results.20Frontiers in Sleep. The role of the WatchPAT device in the diagnosis and management of obstructive sleep apnea People taking certain heart medications, those with pacemakers set to atrial pacing, or those with sustained irregular heart rhythms are not candidates for these devices at all.
Children, Heart Failure, and Altitude
Certain populations face even steeper accuracy challenges. In children, out-of-center sleep tests perform worse than in adults, with diagnostic reliability dropping further in preschool-age kids compared to school-age children.21PubMed. Comparison of diagnostic reliability of out-of-center sleep tests for obstructive sleep apnea between adults and children Children’s breathing patterns during sleep are different enough from adults’ that devices calibrated for adult thresholds can both miss real problems and flag normal breathing as abnormal.
Heart failure introduces a separate complication. Fluid that pools in the legs during the day can shift up into the chest and neck when a person lies down at night. This redistribution swells tissues around the airway and can trigger both obstructive and central breathing events that may not be present on every night or at every stage of the disease.22PubMed Central. Home Sleep Apnoea Testing: Advances, Challenges and Considerations in Heart Failure Home tests that are designed to detect obstructive events may mischaracterize or miss central events altogether.
Altitude is a factor most people never think about. A study of patients who lived at higher elevations and then traveled to lower ones found that their severity scores dropped substantially with descent. The average score fell from about 49 events per hour at home elevation to 37 at a mid-level altitude, and patients who continued to sea level dropped further. Two of eight patients with moderate-to-severe scores at their home altitude fell below the diagnostic threshold entirely at lower elevation.23PubMed Central. The effect of altitude descent on obstructive sleep apnea If you were tested during a trip to a mountain resort and got diagnosed, or tested at sea level and got a clean bill of health but live at elevation, altitude could have shifted the result.
What to Do When You Suspect Your Result Is Wrong
The American Academy of Sleep Medicine has a clear recommendation: if a home sleep apnea test comes back negative, inconclusive, or technically inadequate, a full in-lab sleep study should be performed before ruling out the condition.24PubMed Central. Clinical Practice Guideline for Diagnostic Testing for Adult Obstructive Sleep Apnea: An American Academy of Sleep Medicine Clinical Practice Guideline That recommendation is rated as strong, meaning the evidence behind it is solid enough that it should be the default next step, not an optional one.
Beyond that guideline, a few practical questions are worth raising with your doctor if something feels off:
- Which scoring rule was used? If the 4% oxygen-drop threshold was applied, ask whether results would change under the 3% rule, particularly if your symptoms are significant but the numbers came back low.
- Was the study manually reviewed? Automated scoring can flag artifacts as real events or miss subtle ones. A trained technologist’s review catches errors that software misses.
- What did you do the night before? If alcohol, sedatives, nasal congestion from a cold, or sleeping in an unusual position were part of the picture, the result may not reflect a typical night.
- How did you sleep? If you barely slept during the test, a home device will dilute the score with all that awake time, and a lab study will have captured an atypical sample of your sleep stages.
A negative test in the presence of strong symptoms like daytime sleepiness, witnessed pauses in breathing, or waking up gasping does not close the case. The evidence on night-to-night variability alone suggests that a single snapshot misses the real picture for a meaningful fraction of people, and the testing method, scoring rules, and circumstances of the night all add additional layers of uncertainty on top of that biological variability.
When a Positive Result Might Overstate the Problem
Most of the conversation around testing errors focuses on false negatives, but false positives and severity inflation happen too. If you were tested in a sleep lab and spent an unusual amount of the night on your back because of the sensor wires, your score could be higher than what you experience at home. If you had a bad cold or allergic flare-up that swelled your nasal passages, an automated scoring system might have tagged sensor artifacts as real apneas, just as in the case where nasal obstruction inflated a score from moderate to severe territory before manual correction caught the error.13PubMed Central. Misleading sleep-diagnostic results in a patient with variable nasal airflow obstruction If you were prescribed treatment based on a single test that felt like an unusually rough night, a repeat study, ideally in a different setting or across multiple nights, can clarify whether the original finding holds up.