Night-to-Night Variability
Why we testthree nights, not one
Sleep apnea severity swings from night to night. A single night can miss it or understate it. Published research puts the misclassification rate as high as 1 in 5 patients. Every Nocturne Health home sleep test runs three nights in your own bed.
Medical content reviewed by the sleep medicine physicians at Nocturne Health · Last updated July 2026
It's not just a nicer experience: it changes the answer
If you’ve compared home testing services, you’ve probably noticed that most score a single night while Nocturne Health always tests three. That’s worth more than convenience. Obstructive sleep apnea doesn’t show up the same way every night. How much you drink, which position you sleep in, how much REM sleep you get, how you slept the night prior, and even how strange the recording device feels on night one can all move your apnea count up or down. A single night is a snapshot of one of those nights. Three nights is closer to the truth.
The research below isn’t marketing copy. It’s what happens when researchers test the same patients for multiple consecutive nights and compare each single night against the multi-night average.
What multi-night research actually found
Two independent studies, using different methods and patient populations, arrive at the same conclusion: a single night materially misclassifies sleep apnea severity for a meaningful share of patients.
| Study | Sample | What they found |
|---|---|---|
| Punjabi et al., 2020 (Chest) | 10,340 adults, 3 consecutive nights of home testing | About 1 in 5 patients with mild or moderate sleep apnea on night one were misclassified, either as normal or as a milder category, compared with the average across all three nights. |
| Stöberl et al., 2017 (Journal of Sleep Research) | 77 patients with confirmed OSA, tracked over 2 weeks | 84% had a night-to-night oxygen desaturation swing of more than 10 events/hour, 78% shifted severity category between nights, and a single night carried up to a 60% chance of missing moderate OSA. |
Why one night can mislead
None of this means a single-night test is broken. It means one night is a sample of something that genuinely varies, and several ordinary things push that sample around:
- Body position: obstructive apnea is typically worse lying on your back, and how much time you spend supine varies night to night
- Alcohol: even one standard drink measurably relaxes the airway and raises the apnea count
- REM sleep proportion: apnea is often most severe in REM, and how much REM you get shifts night to night
- Nasal congestion, sleeping environment, and how deeply you sleep in an unfamiliar recording device (the so-called first-night effect)
Any one of these can make a single night look better or worse than your typical sleep. Averaging three nights doesn’t eliminate that variation. It keeps one unusual night from deciding your diagnosis.
Why most home sleep tests are still one night
Mostly cost and logistics. A one-night device is cheaper to build, ships lighter, and the patient mails it back the next morning. That’s a faster, cheaper product to run at scale. A three-night protocol needs a device built to run that long, and a round-trip shipping window built around it. Nocturne Health built the three-night protocol in from the start: it’s the standard test, not an upsell, and it’s included in the flat $169 price.
What three nights gets you
- A severity classification (normal, mild, moderate, severe) that reflects your typical sleep, not one unrepresentative night
- Fewer false negatives: a bad first night doesn't erase sleep apnea that shows up on nights two and three
- A treatment decision grounded in an averaged result, not a single data point
- More confidence in a negative result: three clean nights is stronger evidence than one
- A materially lower chance of needing to repeat testing because one night of data was inconclusive
If you want the fuller picture of how the test itself works, see our home sleep test page. If you’re weighing home testing against an in-lab study, read home sleep test vs. in-lab, or if you’re comparing testing services generally, see how to choose an online sleep apnea test.
Get tested the way the research says works
Start with a virtual consultation ($199) with a sleep medicine physician. If a home sleep test is indicated, it's $169 for three nights, always, with results typically within 72 hours. Available to patients in Arkansas, California, Massachusetts, Ohio, and Pennsylvania.
Three-night home sleep testing: common questions
Why do most home sleep apnea tests only record one night?
Cost and logistics. A single-night device is cheaper to ship, and the patient mails it back the next day. Recording three nights means the device, the battery life, and the return shipping all have to be built for a longer at-home stay, making it a more expensive test to run. Nocturne Health builds the three-night protocol into the $169 price rather than charging extra for it.
How much does sleep apnea severity actually vary night to night?
More than most people expect. A 2020 study of over 10,000 patients (Punjabi et al., Chest) found that roughly 1 in 5 patients with mild or moderate sleep apnea on their first night were misclassified compared with a three-night average. A smaller 2017 study (Stöberl et al., Journal of Sleep Research) found more than three-quarters of patients shifted severity category between nights, with up to a 60% chance a single night would miss moderate OSA entirely.
Can a one-night sleep test give a false negative?
Yes. Apnea severity is influenced by body position, alcohol, how much REM sleep you get, and other factors that vary from night to night, so a single quiet night can score below a treatment threshold even when your typical sleep would not. That's the core finding behind the multi-night research on this page, and it's why the AASM's own diagnostic guideline flags night-to-night variability as a reason to consider repeat testing when a negative result doesn't match the clinical picture.
Does testing three nights cost more?
No. The three-night home sleep test is $169, the same flat price whether you'd otherwise expect a one-night test to cost less. We built the protocol around three nights from the start rather than upselling extra nights on top of a one-night product.
What if one of the three nights doesn't record well?
Testing across three nights makes an unusable study far less likely in the first place. A sensor slipping on one night doesn't cost you the whole test. If your study still doesn't produce a usable, interpretable result, our Test Success Guarantee means you get your money back.
Is a three-night home test as good as an in-lab sleep study?
They answer different questions well. An in-lab polysomnogram captures one night under full technologist supervision with far more sensors; a home test captures multiple nights in your own bed, which the research above suggests matters at least as much for getting an accurate severity read. See our home sleep test vs. in-lab comparison for the full picture of what each does and doesn't cover.
References
- Punjabi NM, Patil S, Crainiceanu C, Aurora RN. Variability and Misclassification of Sleep Apnea Severity Based on Multi-Night Testing. Chest, 2020;158(1):365-373.
- Stöberl AS, Schwarz EI, Haile SR, et al. Night-to-night variability of obstructive sleep apnea. Journal of Sleep Research, 2017;26(6):782-788.
- Kapur VK, et al. Clinical Practice Guideline for Diagnostic Testing for Adult Obstructive Sleep Apnea. Journal of Clinical Sleep Medicine, American Academy of Sleep Medicine, 2017.