Understanding Your Sleep Study Report

You just got back from a sleep study, or your home sleep test results arrived in the patient portal. The report is several pages long. Most of it is charts you do not recognize, tables of acronyms, and a one-paragraph summary at the end. Your follow-up appointment is in two weeks, and you have questions now.

This guide walks through what is actually on a sleep study report, what the numbers mean, and which sections most patients skip without realizing they hold the answer to "what should I actually do about this." It is not medical advice. It will not diagnose you. It will help you read your own report.

What kind of test was this?

The first thing to identify on your report is what kind of study produced it. Two kinds are common in 2026.

In-lab polysomnography (PSG). You slept overnight at a sleep lab. Technicians wired you up with electrodes on your scalp to record brain activity (EEG), on your face to track eye movements and chin muscle tone, on your legs to detect periodic limb movements, on your chest and abdomen via effort belts that record breathing effort, in or near your nose to measure airflow, on your finger for oxygen saturation via pulse oximetry, and on your chest for heart rhythm (ECG). PSG is the diagnostic gold standard because it directly measures sleep stages and arousals as well as breathing.

Home sleep apnea test (HSAT). You slept at home with a simpler device, usually a chest belt or a wrist sensor paired with a finger pulse oximeter, sometimes a small nasal cannula. HSAT measures airflow, oxygen, and effort or movement but does not measure brain waves. It can detect breathing events but cannot directly score sleep versus quiet wakefulness.

This distinction matters because the central index on your report, the one labeled AHI, RDI, or REI, was calculated differently depending on which test you took.

On PSG, the denominator is total sleep time, the hours you were actually asleep, scored from your EEG.

On HSAT, the denominator is total recording time, the hours the device was running. HSAT cannot tell sleep from quiet wakefulness directly, so it typically slightly underestimates the index relative to PSG. The same person, tested both ways on similar nights, often gets a lower number on HSAT than on PSG. If your HSAT result is borderline, this is one of the reasons your physician may suggest an in-lab follow-up.

HSAT also cannot reliably detect central apnea (where the brain stops signaling to breathe, distinct from the airway being blocked), and it cannot assess sleep architecture, leg movements, or arousal events. A negative HSAT does not rule out obstructive sleep apnea (OSA). It means a negative result on a less sensitive test.

The numbers everyone looks at

Most patients land on three or four numbers and ignore the rest of the report. These are usually the indices below.

AHI: Apnea-Hypopnea Index

Apneas and hypopneas per hour of sleep on PSG, or per hour of recording on HSAT (where it is technically called REI). The AASM defines:

• Apnea: a 90% or greater reduction in airflow lasting at least 10 seconds.
• Hypopnea: a 30% or greater reduction in airflow lasting at least 10 seconds, accompanied by either a 3% or greater drop in oxygen saturation, or an EEG arousal. See note below on which version of this rule applies.

Severity tiers in adults, per AASM:

These thresholds are diagnostic anchors. They are not bright lines for treatment decisions. A patient at AHI 8 with significant daytime symptoms may need more aggressive treatment than a patient at AHI 18 who is asymptomatic.

The hypopnea scoring rule problem. AASM offers two ways to score hypopneas: the "recommended" rule (3% desaturation or EEG arousal) and the "acceptable" rule (4% desaturation, no arousal credit). Medicare still uses the 4% rule for OSA diagnosis under its current Local Coverage Determination, while many academic and hospital labs use the 3% rule. The same patient, scored by both rules, can end up with two different AHI values, sometimes one mild and the other moderate, depending on which rule applies. If your AHI feels surprisingly low or high, look at the methods section of your report for which rule was used. This is not a flaw in the test. It is a known scoring inconsistency that affects what insurance will pay for.

The same patient, scored by both rules, can end up with two different AHI values, sometimes one mild and the other moderate.

RDI: Respiratory Disturbance Index

RDI counts AHI events plus RERAs, respiratory effort-related arousals. A RERA is a breathing-related arousal that does not quite meet the airflow or desaturation threshold to count as a hypopnea but still fragments sleep.

By definition, RDI is greater than or equal to AHI. If your report lists RDI 22 and AHI 15, you have seven RERAs per hour adding sleep fragmentation that the AHI alone does not capture. Some clinicians weight RDI more heavily in patients with significant fatigue or cognitive symptoms despite a borderline AHI.

REI: Respiratory Event Index

REI is the HSAT-specific cousin of AHI, calculated against recording time rather than sleep time. It is interpreted with the same severity tiers and the same general clinical relevance as AHI, but with the caveat that it tends to under-call relative to what PSG would have shown on the same patient.

The numbers most people miss

Your report contains several oxygen and sleep-quality indices that often matter as much as AHI but get less attention.

ODI: Oxygen Desaturation Index

The number of times per hour your blood oxygen drops by a defined amount (usually 3% or 4%, matching the hypopnea rule used). ODI correlates with AHI but measures something different. It captures the cardiovascular stress signal of repeated desaturation, regardless of whether each event met the formal hypopnea criteria.

A patient with AHI 8 but ODI 25 is desaturating more frequently than the AHI alone suggests, usually because many events are causing oxygen drops without meeting full hypopnea criteria. This pattern is associated with elevated cardiovascular risk and is worth discussing with your physician.

T90: Time Below 90% Oxygen Saturation

The total minutes (or percentage of the night) your oxygen saturation (abbreviated SpO2 on most reports) spent below 90%. Sometimes reported as a percentage of total sleep time, sometimes in raw minutes. AASM does not set hard tiers here, but in clinical practice, a T90 of more than 10% of sleep time is generally considered meaningful, and more than 20% is a strong signal of sustained nocturnal hypoxemia. This can push treatment urgency forward independent of AHI.

Lowest SpO2 (nadir)

The single lowest oxygen reading recorded during the night. This number alarms patients more than it should. A nadir of 78% sounds frightening, but a single brief desaturation matters less than the cumulative time spent low (T90 and ODI). A nadir of 78% paired with a T90 of 1% is a different pattern than a nadir of 85% paired with a T90 of 30%. Read these together, not separately.

Mean SpO2

Average oxygen saturation across the recording. Healthy adults average above 95%. A mean SpO2 of 91% across the night indicates sustained hypoxemia and is a flag.

Sections of the report most people skip

This is the part of the report that pays off the most for careful reading. Several sections contain information that materially changes what treatment makes sense, and most patients never read them.

Positional data

Your report should break down AHI by sleeping position: supine (on your back), lateral (side), and sometimes prone (front). A surprising fraction of patients have an AHI that doubles, triples, or more in the supine position compared to lateral. The clinical term is "positional OSA," typically defined as supine AHI being at least twice the non-supine AHI.

If your report shows AHI 22 overall but AHI 38 supine and AHI 9 lateral, you may want to discuss positional therapy options with your physician, devices or techniques that prevent supine sleep. For some positional OSA patients, this becomes part of the treatment conversation alongside CPAP or oral appliance therapy. The point is not that the report dictates a positional therapy decision. It is that this data exists in your report, and the conversation it enables is different from "you have moderate OSA, here is a CPAP."

Many reports include this breakdown in a small table easy to miss. Look for "AHI by position" or a row labeled "supine" inside the indices section.

REM versus NREM breakdown

A second small table shows AHI during REM sleep versus during non-REM sleep. REM-predominant OSA, where breathing events cluster heavily in REM sleep, is more common in women than men and is associated with a different cardiovascular risk profile than the whole-night AHI alone might suggest. Some patients are scored "mild" on whole-night AHI but "severe" during REM, which can shift the treatment urgency conversation.

Obstructive, central, and mixed events

Sleep apnea is not one thing. Most apneas in adults are obstructive, meaning the airway collapses despite the brain still trying to breathe. Some are central, meaning the brain stops sending the breathe signal. Some are mixed.

If your report shows central apneas making up more than roughly 5 to 10% of your events, that is worth flagging. Central apnea can be a separate condition (central sleep apnea, often associated with heart failure, opioid use, or stays at high altitude), or it can emerge under CPAP treatment as treatment-emergent central apnea. The treatment pathway differs from straightforward obstructive sleep apnea, so this number changes which conversation you should be having.

Arousal index

Arousals per hour of sleep. These are brief awakenings, often unconscious, that fragment sleep architecture. Some arousals are caused by breathing events and feed into RDI. Others are spontaneous, position changes, or unexplained.

A high arousal index with a low AHI tells you something different than a high arousal index with a high AHI. The former may indicate primary insomnia, restless legs, or other sleep disorders that need their own workup. AASM considers an arousal index above 15 per hour as elevated.

Sleep architecture and efficiency

Percent of time in each sleep stage (N1, N2, N3, REM), total sleep time, sleep latency (time to fall asleep), REM latency (time from sleep onset to first REM), and sleep efficiency (percent of time in bed actually asleep). Most patients glance at this and move on. A pattern of fragmented N3 sleep and reduced REM is the signature of untreated OSA disrupting sleep architecture, and improvement in this section is part of how treatment effectiveness is later assessed.

What is not on your report

Sleep study reports are diagnostic documents. They describe what your breathing and sleep looked like for one night. They do not directly tell you:

How sleepy or symptomatic you are. Daytime sleepiness, fatigue, cognitive symptoms, and partner-reported snoring are usually captured on a separate questionnaire (the Epworth Sleepiness Scale is common) and are interpreted alongside the report, not within it.

Your personal cardiovascular risk profile. OSA elevates risk for hypertension, atrial fibrillation, stroke, and metabolic disease, but the report itself does not quantify your individual risk. Your physician integrates the OSA findings with the rest of your health history.

What treatment will cost or what your insurance will cover. This depends on which hypopnea scoring rule was applied, your specific insurer, and your geography. Not in the report.

Whether one night is representative. Sleep studies sample a single night. Some patients have meaningful night-to-night variability, especially when alcohol, sleeping position, sleep deprivation, or upper-airway congestion differ. If your symptoms strongly suggest OSA but your study was borderline or negative, mention this to your physician.

What to do with this

Read your report twice. The first read, you will fixate on AHI and the summary impression. The second read, look for positional data, REM and NREM breakdown, the obstructive versus central event split, T90, and the arousal index. Mark anything you do not understand and bring it to your follow-up appointment.

If you want a plain-English read on what your AHI specifically means, what severity tier it falls into, what AASM considers first-line treatment for that tier, and what the next decisions on the table typically look like, you can use the AHI Interpreter on this site. Enter your AHI and the tool walks through severity, what it implies, and the treatment landscape.

Your sleep study report is the start of a conversation, not the end of one. The numbers are real, but they are not your diagnosis on their own. They are inputs into a clinical picture that includes your symptoms, your medical history, your age, your weight, and your goals. Bring the report. Bring the questions. The work that comes next is collaborative.