What is the oxygen desaturation index (ODI) on a sleep study?

The oxygen desaturation index (ODI) is the average number of oxygen desaturation events per hour of study duration.¹ Depending on local protocol, an event is a drop in saturation of at least 3% or at least 4%.¹ OSA severity itself is determined by AHI, RDI or REI, so ODI is read alongside those indices.²

By Wesper. Published November 2, 2026.

What is the oxygen desaturation index?

The oxygen desaturation index is the average number of oxygen desaturation events per hour of study duration, as defined in LOINC code 90567-9.¹ It comes from the pulse oximetry channel, and each event is a drop in oxygen saturation that meets the threshold set by local protocol.¹

Oximetry is part of both minimum sensor sets that the 2017 AASM guideline on diagnostic testing for adult obstructive sleep apnea (OSA) describes for a technically adequate home sleep apnea test (HSAT) device.³ Those sets are nasal pressure, chest and abdominal respiratory inductance plethysmography and oximetry, or peripheral arterial tone (PAT) with oximetry and actigraphy.³ In-lab polysomnography also tracks blood oxygen levels,⁴ so ODI can be calculated for either kind of study.

Why are there 3% and 4% versions?

There are 3% and 4% versions because the LOINC definition lets local protocol set the event threshold at either level.¹ The same two thresholds appear in the hypopnea definitions of the AASM Scoring Manual.³

The recommended hypopnea definition requires that changes in flow be associated with a 3% oxygen desaturation or a cortical arousal, and the alternative requires association with a 4% desaturation without consideration of cortical arousals.³ The guideline notes that the AHI may be considerably different in a given individual depending on which definition is used.³ Our explainer on the 3% and 4% hypopnea rules covers that effect in detail.

The threshold matters for ODI in a similar way. A drop of 4% also clears a 3% threshold, so on the same recording a 3% ODI will generally be equal to or higher than a 4% ODI.

How is ODI different from AHI and REI?

ODI differs from AHI and REI in what it counts and in the time it divides by. AHI and REI count apneas and hypopneas, while ODI counts oxygen desaturation events.^5,1 Each index also has its own denominator, as the table shows.

The two kinds of count overlap in part. A hypopnea scored under either AASM rule can involve a desaturation that also registers as an ODI event,³ and a desaturation with no scored breathing event still counts toward ODI by definition.¹ In the other direction, Medicare's CPAP policy defines an apnea as a cessation of airflow for at least 10 seconds,⁶ and hypopneas scored under the recommended rule can qualify through a cortical arousal alone,³ so some AHI events may have no qualifying desaturation. Our explainer on the respiratory event index covers the AHI and REI denominators in more detail.

Does ODI set OSA severity?

ODI does not set OSA severity in the AASM quality measure for severity assessment at initial diagnosis. The measure states that severity is determined by AHI or RDI when polysomnography is performed, or by REI when out-of-center sleep testing is performed.² Its severity ranges are written for AHI and REI, from under 5 events per hour as normal for adults to 30 or more as severe.²

Medicare's CPAP coverage criteria are also written in terms of AHI or RDI.⁶ Because these severity ranges and coverage criteria are written for AHI, RDI and REI, ODI is read as a supporting measure next to the index that grades severity. For patients, our guide on how to read sleep study results is a plain-language starting point.

How do clinicians use ODI alongside other oxygen measures?

Clinicians can read ODI alongside other oxygen measures and the oximetry signal itself, because a count alone leaves out how deep and how long each desaturation was. ODI shows how often saturation dropped by the chosen threshold, and it gives equal weight to every event that clears it.

Hypoxic burden is one measure that looks at the depth and duration of desaturations as well as their frequency. Our guide to understanding hypoxic burden explains the metric.

Whatever oxygen measures a report includes, the AASM position statement on HSAT states that diagnosis, assessment of treatment efficacy and treatment decisions must not be based solely on automatically scored HSAT data.⁷ It also states that the raw data must be reviewed and interpreted by a physician who is board-certified in sleep medicine or overseen by a board-certified sleep medicine physician.⁷ Reviewing the oximetry signal lets the interpreting physician confirm that the desaturations counted in ODI come from a clean signal.

What should a report state?

A report should state the desaturation threshold used for ODI, since the definition allows either 3% or 4% depending on local protocol.¹ It should also make the time base clear. The LOINC definition expresses ODI per hour of study duration,¹ while AHI uses total sleep time and REI uses monitoring time.⁵

When the report also gives an AHI or REI, naming the hypopnea rule used for those indices shows whether the breathing and oxygen measures share a desaturation threshold.³ The amount of usable oximetry is worth noting too, since the AASM guideline describes a technically adequate diagnostic test as including at least 4 hours of technically adequate oximetry and flow data.³

Wesper develops FDA-cleared home sleep testing devices, and multi-night testing comes standard with Wesper studies. The Wesper portal trends ODI and hypoxic burden across nights, along with AHI at 3% and 4%, and gives clinicians full access to the raw data for review. You can read more about home sleep testing for clinics on our professional page.

Frequently asked questions

What does ODI measure on a sleep study?

ODI measures the average number of oxygen desaturation events per hour of study duration.¹ Each event is a drop in saturation that meets the threshold set by local protocol.

What is the difference between a 3% and a 4% ODI?

The percentage is the minimum drop in oxygen saturation that counts as an event, chosen by local protocol.¹ On the same recording, the lower threshold will generally count at least as many events.

Is ODI the same as AHI?

ODI counts oxygen desaturation events per hour of study duration, and AHI counts apneas and hypopneas per hour of total sleep time.^1,5 The two can differ on the same night because their events and denominators are defined differently.

Can ODI be used to grade OSA severity?

The AASM severity quality measure determines OSA severity with AHI, RDI or REI, and its severity ranges are written for AHI and REI.² ODI is reported alongside them as a supporting oxygen measure.

Which ODI threshold should a report list?

The report should list whichever threshold the lab used, since the definition allows either one depending on local protocol.¹

See how ODI and hypoxic burden trends look across nights in home sleep testing for clinics from Wesper.

This article is for general education and is not medical advice. Talk with a licensed medical provider about your symptoms and testing options.

Sources

1. LOINC 90567-9: Oxygen desaturation index. loinc.org
2. American Academy of Sleep Medicine. Quality Measure #277: Sleep Apnea Severity Assessment at Initial Diagnosis. Version 9.0, December 2024. aasm.org
3. Kapur VK, Auckley DH, Chowdhuri S, Kuhlmann DC, Mehra R, Ramar K, Harrod CG. Clinical practice guideline for diagnostic testing for adult obstructive sleep apnea: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2017;13(3):479-504. aasm.org
4. American Academy of Sleep Medicine. Sleep Study. Sleep Education patient page, reviewed September 2020. sleepeducation.org
5. American Academy of Sleep Medicine. AASM style guide for sleep medicine terminology. Updated November 2015. aasm.org
6. Centers for Medicare & Medicaid Services. National Coverage Determination 240.4: CPAP Therapy for Obstructive Sleep Apnea (OSA). cms.gov
7. Rosen IM, Kirsch DB, Carden KA, et al. Clinical use of a home sleep apnea test: an updated American Academy of Sleep Medicine position statement. J Clin Sleep Med. 2018;14(12):2075-2077. aasm.org