Night-to-Night Variability in Sleep Apnea: What 1,184 Home Sleep Tests Show

By Dr. Chelsie Rohrscheib, Wesper

Most sleep apnea diagnoses rest on one night of data. A Wesper study of 149 adults and 1,184 home sleep tests looked at how much can change from night to night.

Key findings

• 69.7% of participants had results that fell into different severity categories on different nights.¹
• Among participants with sleep apnea, 52% had a normal result (AHI under 5) on their first night of testing.¹
• On average, it took 3.2 nights to record a first abnormal result and 3.6 nights to reach a patient’s highest recorded severity.¹
• 26% of participants with a prior diagnosis had home results more severe than their original diagnosis.¹
• The authors suggest four or more nights of testing for a reliable picture of severity.¹

Why sleep apnea severity changes from night to night

Sleep apnea is usually diagnosed from a single night of testing, either in a sleep lab or at home. The main measure, the apnea-hypopnea index (AHI), counts breathing pauses and shallow breaths per hour of sleep. That count can shift from night to night with sleep position, how much REM sleep a person gets, alcohol or sedatives, how tired they are, and how well they sleep with sensors on.

Other research has found the same pattern. In a study of 10,340 adults tested on three consecutive nights, about 20% of patients with mild-to-moderate sleep apnea would have been placed in the wrong severity category based on one night alone.² A study that followed 67,278 people for an average of about 170 nights found the chance of misdiagnosis from a single night ranged from about 20% to 50%.³

What we studied

This retrospective study looked at 149 adults across the U.S. who tested at home with Wesper Lab, an FDA-cleared Type III home sleep test built for testing over multiple nights.¹ Together they completed 1,184 nights of testing, an average of 7.9 nights each. Anyone who reported using a sleep apnea treatment, such as CPAP, an oral appliance or positional therapy, was excluded.

Participants were 58% male, with an average age of 51. AHI was scored using AASM criteria, and severity followed the standard AASM categories: under 5 is normal, 5 to 14.9 is mild, 15 to 29.9 is moderate, and 30 or higher is severe.

How often severity changed

Seven in ten participants (69.7%) crossed at least one severity category across their nights of testing. Only 30.3% stayed in the same category every night.¹

Among participants who had sleep apnea on at least one night, 51% also had at least one night in the normal range. The chance of a normal night depended on how severe a patient’s sleep apnea was at its worst:¹

• 26.8% of patients whose highest severity was mild had at least one normal night
• 18.1% of patients whose highest severity was moderate had at least one normal night
• 6% of patients whose highest severity was severe had at least one normal night

Patients with mild to moderate sleep apnea, the group with the least obvious symptoms, were the most likely to have a normal night.

How big the swings were

The average gap between a participant’s lowest and highest AHI was 12.4 events per hour.¹ Participants who stayed in one category had an average spread of 3.2 events per hour. Participants who crossed categories had an average spread of 16.4 events per hour, which is larger than the 10-point span of the mild category itself.

A patient with an AHI of 12 on one night could be classified as mild, even if other nights show values above 30.

What the first night showed

Among participants who were positive for sleep apnea across their testing, 52% had a normal result (AHI under 5) on their first night.¹ On average, participants needed 3.2 nights to record a first abnormal result and 3.6 nights to reach their highest recorded severity.

What we saw in patients with a prior diagnosis

A subgroup of participants reported that they had already been diagnosed with sleep apnea. Their prior diagnosis was self-reported and could not be independently checked.¹

• 58% had home results in a different severity category than their original diagnosis
• 26% had results more severe than their original diagnosis
• 11% of those diagnosed with mild sleep apnea had at least one night in the severe range
• 38% of those diagnosed with moderate sleep apnea had at least one night in the severe range

What was linked to the variation

Time spent sleeping on the back (supine) had a moderate positive correlation with AHI: nights with more supine sleep tended to have higher AHI. Time in REM sleep had a weaker positive correlation.¹ Both can change a lot from one night to the next.

What this means for testing

The AASM recommends in-lab PSG when a single home sleep test is negative, inconclusive or technically inadequate in a patient suspected of having sleep apnea.⁴ This study adds a practical point: when home testing continues over several nights, severity becomes clearer. The authors suggest four or more nights as a working target for a reliable picture, while noting that patients with borderline results, positional or REM-related sleep apnea, or a high clinical suspicion may need more.¹

Wesper includes multi-night testing as standard with every study, so clinicians can see how a patient sleeps across several nights.

Limitations

This was a retrospective study of one device, without an in-lab PSG comparison on the same nights. Participants were self-selected users of a commercial home testing platform rather than consecutive clinic patients, and the group was mostly male and middle-aged. Prior diagnoses were self-reported. The four-night figure is a practical estimate that needs to be confirmed in prospective studies.¹

Frequently asked questions

Does sleep apnea severity change from night to night?

Yes, for many people. In this study of 149 adults and 1,184 home sleep tests, 69.7% of participants had results in different severity categories on different nights.

Can a single sleep study miss sleep apnea?

It can. Among participants with sleep apnea in this study, 52% had a normal result on their first night of testing. Other large studies have found that one night can put 20% or more of patients in the wrong severity category.

How many nights of home sleep testing are recommended?

There is no formal standard yet. In this study, it took an average of 3.2 nights to record a first abnormal result and 3.6 nights to reach the highest severity, and the authors suggest four or more nights for a reliable picture.

What causes sleep apnea to vary between nights?

Sleep position is a major factor. Nights with more time spent on the back tended to have higher AHI. Time in REM sleep, alcohol, sedatives and how well someone sleeps during testing can also play a role.

What should happen after a negative home sleep test?

The AASM recommends in-lab PSG when a single home sleep test is negative, inconclusive or technically inadequate in someone suspected of having sleep apnea. Your clinician will decide the right next step.

What’s next

For an earlier summary of this study, read Repeat, longitudinal sleep testing in a diverse population demonstrates large night-to-night variability in sleep apnea severity. For what happens when a single home test fails, read Two Home Sleep Tests, Zero Data, and Still No Diagnosis. To see multi-night testing in your clinic, book a demo.

Sources

1. Rohrscheib C. Longitudinal Testing with a Home Sleep Apnea Test Reveals Large Night-to-Night Variability in Sleep Apnea Severity. Wesper white paper. Data on file.
2. Punjabi NM, Patil S, Crainiceanu C, Aurora RN. Variability and Misclassification of Sleep Apnea Severity Based on Multi-Night Testing. CHEST. 2020. pure.johnshopkins.edu
3. Lechat B, et al. Multinight Prevalence, Variability, and Diagnostic Misclassification of Obstructive Sleep Apnea. American Journal of Respiratory and Critical Care Medicine. 2022. researchnow.flinders.edu.au
4. Kapur VK, Auckley DH, Chowdhuri S, et al. Clinical Practice Guideline for Diagnostic Testing for Adult Obstructive Sleep Apnea. Journal of Clinical Sleep Medicine. 2017;13(3):479-504. jcsm.aasm.org