The need for multiple night testing is well recognized in sleep medicine because of a considerable and relevant night-to-night variability. In a study with multiple recordings using WatchPAT®, the OSA severity of 24% of patients was misclassified when using one night compared to the average of three nights. On average, pAHI varied by 57% from night-to-night. The variability of pAHI could partially be explained by the variability of time spent in the supine position with more time supine leading to a higher pAHI (Tschopp et al 2021). Smith (2007) suggested that the AHI should be indicated with a confidence interval to indicate the uncertainty regarding its true value. The Minimal Detectable Difference (MDD) is of special interest in sleep medicine, especially when assessing treatment effects. MDD was found to be 12.8/h and the standard error of measurement was 4.6/h for 4 nights of polysomnography (Aarab et al. 2008). For WatchPAT®, measuring two and three nights showed a small reduction in MDD from 19.1/h to 18.0/h (Tschopp et al. 2021, in press). Only one study using pulse oximetry assessed the night-to-night variability over 14 days (Stöberl A. et al 2017). The study confirmed the enormous variability and focused mainly on its impact on OSA severity.

While the night-to-night variability has been extensively studied for polysomnography, respiratory polygraphy, and WatchPAT®, little is known about the optimal number of nights to be recorded. There is convincing evidence from the literature, that the recording of multiple nights is the only way to assess the severity of the patient's disease with clinically reasonable accuracy. Moreover, the MDD with only one night's recording is astonishingly high. The question is how many nights should be recorded to achieve acceptable diagnostic accuracy. The precision of the OSA measurement depends on the clinical situation. For example, to diagnose severe OSA, a higher variability might be acceptable without influencing the treatment decision. However, when comparing treatment effects, the MDD should be as small as possible. The recording of multiple nights might be cumbersome for patients (e.g. with polysomnography or respiratory polygraphy) as well as costly. These factors have to be taken into consideration for the clinically feasible number of recordings.

Sleepiz One Connect offers the unique opportunity for a contactless recording of breathing combined with conventional pulse oximetry and is a minimally invasive diagnostic tool that allows measurements over several nights. Studies with multiple night recordings will offer a basis for diagnostic recommendations in future guidelines.

The study aims to investigate the variability of obstructive sleep apnea at-home sleep apnea testing. By investigating the variability, we want to quantify the improvement in diagnostic accuracy by additional measurements.

The hypothesis is that additional recordings offer a significant improvement in diagnostic accuracy by reducing the variability. The reduction in variability will diminish with each additional recording.

Inclusion Criteria:

• Age ≥ 18yrs
• Obstructive Sleep Apnea (OSA) defined as AHI ≥ 5/h
• Consecutive recordings are possible over 10 nights within 3 weeks
• Informed consent

Exclusion Criteria:

• Age <18yrs
• No obstructive sleep apnea AHI < 5/h
• Central sleep apnea events > 25%
• Patients with implanted active devices (e.g pacemakers, neurostimulators)
• Multiple night recordings are not possible
• Insufficient language comprehension in German
• Pregnancy

Patients with suspected obstructive sleep apnea will be eligible for this study.