Home-testing devices for obstructive sleep apnoea show acceptable diagnostic accuracy in adults

A systematic review funded by the National Institute for Health and Care Research examined whether newer, smaller portable devices designed for home use offer reliable diagnosis. Reviewers searched databases through September 2023 and identified multiple studies comparing six novel devices to standard tests in adults with suspected OSA. Sensitivity (ability to correctly identify people with the condition) was generally high, ranging from 80–100% in most studies, though it fell below 80% in two. Specificity (ability to correctly identify people without the condition) was more variable, ranging from 25–100%, with many estimates clustering in the 70–80% range. Test failure rates ranged from 0–18%. An economic model found that all six devices were less costly and slightly less effective than respiratory polygraphy, but notably more cost-effective than oximetry alone, with incremental costs below £10,000 per quality-adjusted life-year gained—well within standard NHS thresholds. However, the analysis notes substantial uncertainty: most included studies were conducted in clinic settings rather than actual home environments, and only three studies involved children (all in clinic, not home). The authors caution that definitive relative comparisons between devices remain difficult.

The systematic review identified considerable variation in specificity estimates between and within studies depending on the severity threshold chosen for diagnosis—a finding that underscores the importance of understanding which device and cut-off a clinician selects. For instance, some devices showed high sensitivity but lower specificity, meaning they may miss fewer cases of OSA but falsely flag more people without the condition. Interestingly, the economic model showed that AcuPebble and Sunrise devices generated incremental net monetary benefits exceeding cost-effectiveness thresholds (at £20,000–£30,000 per quality-adjusted life-year) compared to respiratory polygraphy, suggesting these specific devices might offer better value. The 0–18% test failure rate is clinically important: even a small failure rate delays diagnosis and may require retesting. The authors emphasize that most effectiveness data come from controlled clinic settings, not genuine home use, which may overestimate real-world diagnostic accuracy since patient comfort and adherence differ at home. This gap means the true cost-effectiveness in actual home deployment remains uncertain. No health economic comparison to home-based pulse oximetry was identified, leaving a key clinical pathway unevaluated.