Diagnosis: Beyond the Overnight Sleep Study
Polysomnography (PSG) — an in-laboratory overnight sleep study monitoring brain waves, eye movements, muscle activity, heart rhythm, blood oxygen, chest movement, and airflow — remains the gold standard for sleep apnea diagnosis. The key metric is the Apnea-Hypopnea Index (AHI): the average number of apneas and hypopneas per hour of sleep. Mild OSA is defined as AHI 5-14, moderate as 15-29, and severe as 30 or more. However, AHI has recognized limitations as a sole severity marker: two patients with identical AHI can have very different oxygen desaturation profiles and cardiovascular consequences depending on the depth and duration of each event. Oxygen desaturation index (ODI) and cumulative time below 90% oxygen saturation (T90) are increasingly used alongside AHI to capture true physiological burden.
Home sleep testing (HST) — portable monitoring devices that record airflow, respiratory effort, and oxygen saturation without the full array of PSG channels — has become the standard diagnostic approach for moderate-to-high probability OSA in most developed countries. HST is substantially less expensive, more convenient, and produces AHI measurements within clinically acceptable agreement with PSG for typical obstructive sleep apnea. However, HST has important limitations: it records total monitoring time rather than confirmed sleep time, leading to AHI underestimation; it cannot diagnose central sleep apnea patterns; and it misses non-respiratory sleep disorders (periodic limb movements, parasomnias) that may be contributing to daytime symptoms.
Clinical prediction tools can guide pre-test probability assessment before formal testing. The STOP-BANG questionnaire — assessing Snoring, Tiredness, Observed apneas, blood Pressure, BMI, Age, Neck circumference, and Gender — identifies high-risk individuals with 94% sensitivity for moderate-to-severe OSA. A neck circumference above 40cm in women and 43cm in men is particularly predictive. The Epworth Sleepiness Scale quantifies subjective daytime sleepiness on a standardized scale — scores above 10 indicate significant hypersomnia warranting investigation. Using these tools as part of a systematic screening protocol would substantially reduce the years-long diagnostic delay that currently characterizes most OSA diagnoses.
Wearable technology has significantly changed consumer-level sleep apnea detection. Devices including certain Apple Watch models, Withings ScanWatch, and ResMed’s WHOOP integration now provide SpO2 monitoring and sleep breathing disturbance alerts. While these devices are not diagnostic and do not replace formal testing, they serve as valuable screening tools that have prompted formal sleep study referrals for many previously undiagnosed individuals. Consumer polysomnography devices including Dreem and WatchPAT are approaching clinical-grade diagnostic accuracy, with WatchPAT achieving FDA 510(k) clearance as a prescriptible diagnostic device. The democratization of sleep monitoring represents a major opportunity to address the vast underdiagnosis problem.