Home Sleep Apnea Test (HST)

What a home sleep apnea test is

A home sleep apnea test — sometimes called a home sleep test, out-of-center sleep testing, or by the abbreviation HST — is a portable diagnostic device a patient takes home, self-administers for one or more nights, and returns for analysis by a sleep physician. It is designed specifically to detect obstructive sleep apnea by recording respiratory signals and oxygen levels during sleep. It is not a general-purpose sleep study and cannot diagnose the full range of sleep disorders.

The American Academy of Sleep Medicine (AASM) classifies sleep monitoring devices into four levels:

• Type I — full attended in-lab polysomnography (PSG), the diagnostic standard, recording at least seven channels including EEG (brain activity), EOG (eye movement), EMG (muscle tone), and full respiratory and cardiac signals.
• Type II — full unattended portable PSG with the same channel set as Type I but used in the home. Rare in routine clinical practice.
• Type III — limited-channel portable monitors with at least four channels, typically airflow, respiratory effort, oxygen saturation, and heart rate. This is the standard HST used in adult clinical care.
• Type IV — single- or dual-channel devices, usually oximetry-only or airflow-only. Limited diagnostic utility; not generally recommended as a primary diagnostic tool.

When clinicians and articles say “HST,” they almost always mean a Type III device. The rest of this page uses that convention. Since AASM endorsement of HST as an acceptable diagnostic tool for selected adult patients in 2007, and reaffirmed and refined in the AASM 2017 clinical practice guideline (Kapur et al.), HST has substantially expanded access to sleep apnea diagnosis — particularly in regions with limited in-lab capacity. Most newly-diagnosed adults today receive their diagnosis via HST rather than in-lab PSG.

Not all HSTs are created equal — and PSG is

The label “HST” hides a wide range of device quality. Within the AASM type classification described above, two devices can both be marketed as home sleep apnea tests and yet measure substantially different things. This matters for patients trying to interpret their own results, and it matters for clinicians deciding where to refer.

The single most consequential channel — the one that separates an adequate HST from an inadequate one — is direct airflow measurement, typically via a nasal cannula or a thermistor placed at the nostril. The American Academy of Sleep Medicine’s 2007 Portable Monitoring Task Force established that any HST used for the diagnosis of OSA must, at minimum, record airflow, respiratory effort, and blood oxygen saturation. A device that omits any of these three falls below AASM-recognized standards. Some commercially marketed devices use chest impedance as an indirect airflow proxy, or rely on oximetry alone — and these do not meet the standard.

What this looks like in practice across the device classes:

• Standard Type III HST. Records all three minimum channels — nasal airflow, respiratory effort via chest and abdomen belts, and oximetry — plus typically heart rate and body position. This is the device most adult sleep apnea workups use, and what AASM-accredited sleep services dispense. Meets AASM standard.
• HST Type II. Full portable montage that records the same channels as in-lab PSG (EEG, EOG, EMG, respiratory channels, ECG-derived heart rate) but unattended in the home. Type II is rare in routine clinical practice — the equipment is expensive and the self-application of EEG sensors is error-prone — but when used correctly it produces near-PSG-quality data. Meets AASM standard.
• Type IV oximetry-only devices. Record only blood oxygen saturation, sometimes with a heart rate channel. These devices fall below AASM standards. They cannot distinguish obstructive from central apnea, cannot quantify airflow events, and are no longer accepted as standalone diagnostic tools. CMS Medicare coverage of any Type IV device requires that it measure at least three channels including airflow — most consumer oximetry-only devices do not meet even this lower CMS bar.
• Peripheral arterial tonometry (PAT) devices. Measure actigraphy (movement), oximetry, and PAT — a finger-vasoconstriction signal that serves as an indirect proxy for the sympathetic nervous system activation that accompanies apnea-related arousals. PAT devices are a separately covered Medicare category and have peer-reviewed validation studies. The AASM and the American Academy of Neurology have both qualified their endorsement, citing the indirectness of PAT-based respiratory analysis compared to direct airflow measurement. PAT devices are not the standard of care; they are an alternative pathway with specific use cases.
• Consumer-marketed devices that do not meet AASM minimums. Include some wristband-only devices, some smartwatch-based "screening" tools, and some finger-clip devices that record only oximetry. These are not diagnostic devices. They may produce reports that look superficially similar to a clinical HST, but the signal quality is materially different and they do not produce a clinically actionable diagnosis. The term "HST" on a marketing website does not, on its own, tell a patient what they are getting.

Polysomnography, by contrast, is highly standardized. AASM-accredited sleep laboratories follow a uniform Type I montage — at least seven channels covering EEG, EOG, EMG, ECG, airflow, respiratory effort, and oximetry, with attended technologist supervision throughout. The signal-acquisition variability between two AASM-accredited PSGs done in different cities is far smaller than the variability between two devices both marketed as “HST.” This is one of the practical reasons a borderline or negative HST in a high-suspicion patient should escalate to PSG rather than to a repeat HST or to an alternative consumer device.

The practical takeaway for someone considering an at-home apnea test: ask which specific device the provider uses, confirm it meets AASM minimum channel requirements (airflow plus respiratory effort plus oximetry), and confirm that interpretation is performed by a board-certified sleep physician affiliated with an AASM-accredited program. Device class matters more than device brand, and “home sleep test” without those two confirmations is a marketing label, not a diagnosis.

What an HST measures — and what it doesn’t

The fundamental difference between an HST and a full polysomnogram is which channels each records. HST captures the respiratory signals needed to detect apnea and hypopnea events. PSG captures all of those plus the neurophysiological channels needed to score sleep itself.

To set up the comparison: polysomnography (PSG) is the in-lab overnight sleep study, conducted in an accredited sleep center/lab with a sleep technologist on duty. It records a comprehensive panel of physiological signals — typically twelve or more channels covering brain activity, eye movement, muscle tone, breathing, oxygen, and cardiac rhythm — and remains the diagnostic gold standard for the full range of sleep disorders. An HST records a much narrower channel set in the patient’s own bed, designed specifically to detect obstructive sleep apnea. The table below shows what each test captures.

The clinical implication of this channel difference is significant. Without EEG, an HST cannot tell when the patient was asleep versus awake. The apnea-hypopnea index (AHI) computed from a PSG uses total sleep time as the denominator. The respiratory event index (REI) computed from an HST uses total recording time. Because patients are inevitably awake for some portion of the night while wearing the device, the recording-time denominator is larger than the sleep-time denominator — and the resulting REI tends to be a few units lower than the AHI the same patient would get on a PSG. This is one of several reasons HST tends to underestimate severity.

Beyond the AHI versus REI distinction, the absence of EEG means an HST cannot detect arousal-based hypopneas — events where breathing partially obstructs and the brain transiently wakes the patient up to restore the airway, but oxygen saturation never drops enough to flag the event by oximetry alone. PSG counts these events because the EEG sees the arousal. HST does not.

The absence of EOG and EMG also means an HST cannot score REM versus non-REM sleep. Many people have REM-predominant sleep apnea — apnea events concentrated during REM, when respiratory drive is naturally less stable. A patient with otherwise mild apnea by total-night metrics can have severe REM apnea masked by their non-REM stretches. HST cannot disaggregate this.

When an HST is the right test

The AASM 2017 clinical practice guideline recommends HST in adults meeting all of the following criteria:

• High pretest probability of moderate-to-severe obstructive sleep apnea. Pretest probability is assessed by clinical interview and validated screening tools — most commonly the STOP-BANG questionnaire and the Epworth Sleepiness Scale. A patient with loud habitual snoring, witnessed apneas, daytime sleepiness, hypertension, and elevated BMI fits this profile cleanly.
• Uncomplicated clinical picture. The symptoms point to obstructive apnea and not to one of the comorbid scenarios listed in the next section. The history doesn’t suggest central apnea, periodic limb movements, narcolepsy, parasomnia, or significant insomnia.
• Adult. HST is not validated for use in children and is not recommended for pediatric assessment.
• Capable of self-administering the test. The patient can place the sensors correctly and tolerate them through the night without supervision.

The two screening tools cited above warrant brief description. The STOP-BANG questionnaire is an eight-item yes-or-no screener — Snoring, Tiredness, Observed apneas, Pressure (hypertension), BMI greater than 35, Age over 50, Neck circumference greater than 17 inches in men or 16 in women, and male sex. A score of 3 or higher indicates elevated risk; 5 or higher correlates strongly with moderate-to-severe OSA. The Epworth Sleepiness Scale is an eight-item self-report that asks the patient to rate the likelihood of dozing off in eight everyday situations, on a 0–3 scale. A total score above 10 suggests excessive daytime sleepiness; above 15 suggests severe sleepiness warranting investigation regardless of other findings. Neither tool is diagnostic on its own, and a positive STOP-BANG without clinical correlation does not establish the case for HST. They are blunt but useful instruments for triaging which patients warrant which test.

For the right patient, HST has real advantages. The test happens in the patient’s own bed, which produces sleep architecture closer to a typical night than what most people get in an unfamiliar lab environment. Turnaround tends to be faster because in-lab capacity is not a bottleneck. Cost is generally lower for both patient and payer. Patients who would have refused or delayed an in-lab study sometimes accept HST without the same friction, which means earlier diagnosis and earlier treatment.

The trade-off is that the smaller channel set raises the bar on patient selection. HST works well when the pretest probability is high enough that a clean positive result confirms what the clinician already suspected, and a clean negative result is unlikely. HST works less well when the picture is ambiguous and the test is being asked to do diagnostic work the channels can’t reliably do.

When an HST is not the right test

The AASM guideline is explicit about scenarios where HST should not be the primary diagnostic tool. In each of these cases, the missing channels matter clinically, and the patient should go to in-lab polysomnography instead.

• Children. Pediatric apnea differs substantially from adult apnea: it is more REM-predominant, the AHI thresholds for diagnosis are lower (an AHI of 1.5 in a child is meaningful; an AHI of 1.5 in an adult is not), and the polysomnographic findings are more nuanced. PSG is the standard for any pediatric sleep evaluation. See pediatric sleep apnea for the broader context.
• Suspected central sleep apnea. Central apnea events look similar to obstructive events on respiratory channels alone. Distinguishing them reliably requires watching for the absence of respiratory effort during the airflow loss — which an HST records, but with less sensitivity than PSG. For patients with heart failure, opioid use, prior stroke, or otherwise elevated central-apnea risk, PSG is preferred.
• Significant cardiopulmonary comorbidity. Heart failure (especially NYHA III or IV), severe chronic obstructive pulmonary disease, neuromuscular disease, and chronic hypoventilation all complicate the respiratory pattern in ways HST is not designed to characterize.
• Suspected non-respiratory sleep disorders. Restless legs syndrome, periodic limb movement disorder, parasomnia, REM sleep behavior disorder, and narcolepsy require channels HST does not have. If the clinical picture suggests these, PSG (sometimes with MSLT for narcolepsy) is the appropriate test.
• Significant insomnia. If the patient routinely sleeps very little — say, four hours or fewer — total recording time may produce so few measured respiratory events that the test result becomes statistically unreliable.
• Prior negative HST with persistent symptoms. A negative HST in a patient with a high pretest probability of sleep apnea is more likely a false negative than a true negative. The next step is PSG, not a repeat HST.

This list is not exhaustive. The general principle: if the channels HST can’t see are channels that would matter for a given patient, the test isn’t appropriate for that patient.

How accurate an HST is

For appropriately selected patients, HST is reasonably good at confirming moderate-to-severe obstructive sleep apnea. Across published studies summarized in the AASM 2017 guideline, HST sensitivity for an AHI of 15 or higher is in the range of 80 to 90 percent against PSG as the reference standard. Specificity is similar. Performance falls off in milder apnea, where small absolute differences in the event count produce larger relative differences in classification.

The most important practical limit, however, is not the headline sensitivity number. It is that HST tends to systematically underestimate apnea severity compared with PSG on the same patient. Two mechanisms drive this:

• The recording-time denominator. As discussed in the channels section above, REI uses total recording time and AHI uses total sleep time. Time spent awake during the recording dilutes the REI relative to the AHI the same patient would get on a PSG.
• Missing arousal-based hypopneas. Without EEG, HST cannot count hypopnea events that resolve via cortical arousal without a measured oxygen desaturation. In some patients, particularly those with predominantly upper-airway-resistance physiology, this is a substantial fraction of total events.

The aggregate effect is that HST-derived severity is typically 10 to 30 percent lower than the same patient’s PSG-derived severity, with wider variation in individual cases. A single-night HST also captures only one night of natural variability — a known issue in apnea testing more broadly, since AHI varies meaningfully night-to-night within the same patient.

The clinical implication is straightforward. A clean positive HST in a patient with a fitting clinical picture is reliable enough to act on. A negative HST in the same patient is not a reason to stop investigating. Multi-night HST protocols (typically three nights) reduce single-night variability and improve diagnostic confidence; some centers use these routinely. But the underlying ceiling on what HST can detect, set by the missing channels, doesn’t change.

What the test actually involves

Most HST workflows look broadly similar across providers. Specific device models and brand-specific procedures vary; the steps below describe the typical adult experience.

Before the test. A clinician — primary care physician, sleep specialist, or sometimes an integrated sleep service — orders the test based on screening, clinical interview, and pretest probability assessment. The patient picks up the device from a clinic or sleep lab, or has it shipped to their home with written instructions.

The night of the test. The patient self-applies the sensors before bed: typically a nasal cannula to measure airflow, an elastic belt around the chest to measure respiratory effort, and a fingertip pulse oximeter for oxygen saturation. Some devices also include a sensor for body position and a small microphone for snoring. The device records continuously through the night. Most devices ask the patient to confirm a successful setup with a simple light-or-sound check before sleep.

Single-night versus multi-night. A single-night recording is the standard, but two- or three-night protocols are increasingly used to reduce single-night variability. Multi-night testing also helps when the first night’s recording has technical issues — for instance, the cannula falling off — that would otherwise require a repeat appointment.

After the recording. The patient returns the device. The recording is downloaded and analyzed by a sleep physician — typically the patient’s referring sleep specialist, sometimes through a third-party scoring service. Modern software produces a draft analysis automatically, but the human read by a board-certified sleep physician is what AASM accreditation requires for clinical reporting.

Turnaround. Results are usually available within one to two weeks of device return, sometimes faster. The referring clinician then discusses findings with the patient: if the test is positive for moderate-to-severe sleep apnea, the conversation moves to treatment options, most often CPAP therapy; if the test is negative or equivocal but symptoms persist, the next step is usually in-lab PSG.

Where HST fits in the broader pathway

HST is one step in a larger sequence. For most patients, the path looks like this:

• Recognition. The patient, a partner, or a primary care clinician notices symptoms consistent with possible sleep apnea — loud habitual snoring, witnessed pauses in breathing, gasping awakenings, persistent daytime sleepiness, morning headaches, or refractory hypertension. See sleep apnea and snoring for the symptom spectrum.
• Screening. A primary care visit, occupational health screen, or sleep specialist consultation establishes pretest probability using validated instruments (STOP-BANG, Epworth Sleepiness Scale, occasionally the Berlin Questionnaire) alongside clinical history.
• Diagnostic test selection. If pretest probability is high and the picture is uncomplicated, an HST is ordered. If pretest probability is high but the patient profile fits any of the “not appropriate” criteria above, in-lab PSG is ordered instead. If pretest probability is low, the next step is usually further history and possibly addressing alternative explanations for the symptoms before any sleep study.
• Diagnosis. A clear positive HST closes the diagnostic question for moderate-to-severe OSA. A negative HST in a high-suspicion patient escalates to PSG. An equivocal HST in a borderline-severity patient also typically escalates to PSG.
• Treatment. Once the diagnosis is established, the conversation moves to therapy — most commonly CPAP, with oral appliances and surgical options reserved for specific clinical scenarios. CPAP titration is sometimes done via an auto-titrating home device, sometimes via in-lab titration PSG, depending on patient profile and clinic preference.

HST is a tool that works well within its design envelope and not at all outside it. The thing it does best is to confirm what an experienced clinician already strongly suspects in an otherwise uncomplicated adult patient. The thing it does worst is to rule sleep apnea out in a patient whose clinical picture says it’s probably there.

One practical note about the screening-to-test handoff: in many US health systems, primary care physicians can order HST directly without a sleep specialist consultation, particularly when the pretest probability is clearly high. In other systems, a sleep specialist visit is required first to confirm that HST (rather than PSG) is the appropriate test. The path varies by region, by health plan, and by the local availability of integrated sleep services. Patients who notice symptoms suggesting sleep apnea should mention them to their primary care clinician — that’s the entry point that determines which path applies.