Recent research suggests that the severity of obstructive sleep apnea (OSA) might change dramatically night-to-night, resulting in significant consequences for diagnosis and treatment. The study used multi-night in-home recordings to assess OSA prevalence and the impact of night-to-night variability in OSA severity on diagnostic classification in a large, global, non-randomly selected community sample drawn from a consumer database of people who purchased a novel, validated, under-mattress sleep analyzer. 

Between July 2020 and March 2021, 67,278 patients aged 18 to 90 undertook in-home nightly monitoring for an average of ~170 nights per participant. To define OSA, they used a nightly mean apnea-hypopnea index (AHI) greater than 15 events/h. The outcomes were multi-night global prevalence and risk of OSA misdiagnosis based on single night AHI scores. Main Results and Measurements Data from over 11.6 million nights were collected and evaluated. OSA was shown to be prevalent in 22.6% (95% CI: 20.9-24.3%). In patients with OSA, the chances of being misdiagnosed based on a single night ranged from 20% to 50%. Misdiagnosis error rates dropped as the number of monitoring nights rose (e.g. 1-night F1-score=0.77 vs. 0.94 for 14-nights) and remained steady after 14 nights. 

According to multi-night in-home monitoring employing revolutionary non-invasive under-mattress sensor technology, the global prevalence of moderate to severe OSA of ~20%, and ~20% of persons diagnosed with a single night study may be misclassified. These findings underscored the need of taking night-to-night variation into account when diagnosing and treating OSA.