Sleep Study: In-Lab PSG vs Home Test (and How It Sets Your Pressure)

A sleep study is the test that turns "I think I have sleep apnea" into a diagnosis, a severity number, and — if you need treatment — a CPAP pressure. But "sleep study" covers two very different experiences: a full night wired up in a lab, or a small kit you wear at home. This guide explains what each one measures, when the in-lab version is required, how the result becomes your prescribed pressure, and why the single night you sleep in a lab is only the beginning of your data story.

This is part of our Understanding Sleep Apnea pillar. If you're still establishing the basics, start with What Is Sleep Apnea? for types, AHI, and severity.

What a sleep study measures

A sleep study records what your body does while you sleep so a physician can count breathing problems and grade their severity. The headline number from any study is the AHI (Apnea-Hypopnea Index) — the average number of apneas (breathing pauses) and hypopneas (shallow-breathing events) per hour of sleep.

A comprehensive study captures far more than breathing, though. A full polysomnogram measures:

• Breathing airflow and effort — to detect apneas and hypopneas, and to tell obstructive events (airway collapses while you still try to breathe) from central events (your brain briefly stops sending the breathe signal). See Obstructive vs Central vs Mixed Sleep Apnea.
• Blood-oxygen saturation (SpO2) — how far your oxygen drops with each event.
• Brain waves (EEG), eye movement, and muscle tone — to score actual sleep stages and detect arousals (brief awakenings you don't remember).
• Heart rhythm (ECG), leg movement, body position, and snoring.

That EEG channel matters more than most people realize. Because a lab can see arousals, it can score a hypopnea that ends in a brain arousal even without a big oxygen drop. This is why lab-scored AHI and the AHI your CPAP machine reports later are not the same measurement — your machine estimates events from airflow and pressure alone, with no EEG and no way to detect arousals. We unpack that gap in AHI vs RDI and in How to Read Your CPAP Data.

A quick word on hypopnea scoring rules

Two scoring definitions exist, and which one a lab uses changes your number:

• AASM Rule 1A (recommended): a hypopnea counts if airflow drops with either a 3% oxygen desaturation or an arousal.
• AASM Rule 1B (acceptable / used by Medicare): a hypopnea counts only with a 4% oxygen desaturation.

The same night of sleep can produce a meaningfully higher AHI under Rule 1A than under the stricter 1B rule. If your AHI sits near a treatment cutoff, ask which rule your lab used.

In-lab PSG vs home sleep apnea test (HSAT)

The two paths trade thoroughness for convenience.

The key limitation of an HSAT is what it leaves out. Without EEG it can't confirm you were actually asleep, so it may divide events over time you spent awake, diluting and under-reporting your true AHI. A home test is good at confirming clear obstructive sleep apnea; it's weak at ruling it out.

When PSG is required (per AASM)

The American Academy of Sleep Medicine's diagnostic guideline (Kapur et al., 2017) draws a clear line. A home sleep apnea test with a technically adequate device may be used to diagnose OSA in uncomplicated adult patients who show signs and symptoms indicating an increased risk of moderate-to-severe OSA.

In-laboratory PSG is recommended instead of an HSAT for patients with any of the following:

• Significant cardiorespiratory disease (serious heart or lung conditions)
• Potential respiratory muscle weakness from a neuromuscular condition
• Awake hypoventilation, or suspicion of sleep-related hypoventilation
• Chronic opioid medication use
• A history of stroke
• Severe insomnia

PSG — not an HSAT — is also required to evaluate non-obstructive sleep-disordered breathing, including central sleep apnea and hypoventilation. And there's an important safety net: if an HSAT comes back negative, inconclusive, or technically inadequate, an in-lab PSG should be performed. A "normal" home test in someone with strong symptoms is a reason to escalate, not to stop looking.

If you're at the front of this process and unsure whether to even start, When to See a Doctor About Snoring and our sleep apnea symptoms guide can help you decide.

From study to titration to prescribed pressure

Diagnosis answers do you have apnea and how bad is it. It does not, by itself, set your CPAP pressure. That's the job of a separate step called titration.

There are two common routes to a pressure prescription:

1. In-lab titration study (or split-night study). You sleep in the lab on a CPAP machine while a technologist gradually raises the pressure to find the level that eliminates your events across sleep stages and body positions (apnea is often worse on your back and during REM sleep). In a split-night study, the first half diagnoses and the second half titrates — all in one night.
2. Auto-titrating (APAP) at home. Many patients are sent home on an auto-adjusting machine that hunts for the right pressure each night. A clinician then reviews the data — often the 95th-percentile pressure, which is the level your machine met or stayed under 95% of the night — to choose a fixed prescription, or simply leaves you on an auto range. The differences between these machine modes are covered in APAP vs CPAP vs BiPAP.

Titration determines the prescribed pressure. Its goal is to drive your residual events down to a healthy range.

What counts as "good enough"? A residual AHI below 5 events per hour is the widely used benchmark for effective therapy. The AASM defines an optimal PAP titration as reducing AHI to fewer than 5 — and below 5 is also the normal, non-apneic range. Many clinicians aim lower (for example, under 1-2) when that's comfortably achievable, but there's no formal guideline establishing "below 2" as a standard target, and goals are individualized with your provider — someone with severe OSA who lands at a residual AHI of 5-10 may still be doing well. See What Is a Good AHI on CPAP? for the full picture.

One titration caveat worth knowing up front: CPAP is designed to splint your airway open and reliably treats obstructive apneas, but it does not directly correct the unstable breathing drive behind central (clear-airway) events. Sometimes central events actually appear or rise once obstructive ones are fixed — a pattern called treatment-emergent central sleep apnea (TECSA), or complex sleep apnea, seen in roughly 5-15% of titrations. The reassuring part: it resolves on its own in most people (reported spontaneous resolution around 60-80%) within weeks to a few months of continued CPAP. Persistent cases warrant a clinician's evaluation and may call for BiPAP, ASV, or other measures. The wrong move is to raise your own pressure to chase central events — higher pressure doesn't fix them and can sometimes provoke them. We go deeper in Central Apneas Showing Up on CPAP.

Why home-data tracking complements the one-night study

Here's the catch with any sleep study: it's one night. A lab night is also an unusual night — strange bed, wires, a technologist down the hall. Your real therapy happens hundreds of nights later, in your own bedroom, where conditions change constantly.

Several things drift after your study that a single titration can never see:

• Leaks. A mask that sealed fine on night one may leak months later as cushions wear. A large leak doesn't just feel bad — it can invalidate or under-report your AHI, because the machine can't reliably score events through a flood of escaping air. (For reference, ResMed flags excess leak above roughly 24 L/min at the 95th percentile; Philips reports total leak, a different baseline.) See CPAP Leak Rate: What's Acceptable.
• Night-to-night variation. AHI naturally swings with alcohol, congestion, sleep position, and illness. A single number is noise; the trend over weeks is the signal. (Why Does My AHI Change Night to Night?)
• Things AHI never captured. Flow limitation and RERAs can leave you tired even when your AHI looks great — see CPAP Flow Limitation and Still Tired on CPAP With Good Numbers?.

This is exactly the gap ongoing data tracking fills. Your study gives you the starting prescription; your nightly data tells you whether that prescription is still working. SomniCharts imports data from ResMed, Philips Respironics (including the encrypted-by-default DreamStation 2), and Löwenstein prisma machines, then explains your AHI, leak rate, pressure, and event types in plain language — automatically, every night, in one place. A sleep study is a single snapshot; SomniCharts lets you keep reading your therapy data long after the lab night is over.

To be clear about the line between data and doctoring: the value of all this tracking is an informed conversation, not self-treatment. If your numbers drift, bring the trend to your sleep clinician rather than adjusting your prescription yourself. A clean, plain-English CPAP compliance report you can hand to your provider often does more for your therapy than any single night in a lab.

Frequently asked questions

Can I just do a home test and skip the lab? Often, yes — if you're an uncomplicated adult likely to have moderate-to-severe OSA. But if you have significant heart or lung disease, neuromuscular weakness, suspected hypoventilation, chronic opioid use, a stroke history, or severe insomnia, the AASM recommends in-lab PSG. And a negative or inconclusive home test should be followed by a full lab study.

Does my sleep study set my CPAP pressure? Not the diagnostic study by itself. Pressure comes from a separate titration — either an in-lab titration/split-night study or a clinician reviewing your at-home auto-CPAP data.

What pressure will I get? Whatever level controls your events down to a residual AHI generally below 5 per hour across sleep stages and positions. It's individualized; two people with the same diagnosis can end up on quite different pressures.

Why does my home CPAP report a different AHI than my lab study? Your machine estimates events from airflow and pressure with no EEG, so it can't detect arousal-based hypopneas the way a lab can. Different methods, different numbers — both can be valid. (CPAP Event Types Decoded)