The objective of this study was to conduct a systematic review and meta-analysis to evaluate the performance of smartwatches in detecting obstructive sleep apnea compared with polysomnography.

Polysomnography (PSG) is the gold standard for diagnosing obstructive sleep apnea (OSA). Smartwatches with integrated sensors have been explored as convenient, non-invasive tools for OSA detection but their diagnostic accuracy compared with PSG remains uncertain.

The literature search was conducted on PubMed, Scopus, and Cochrane Library through September 23, 2025 for studies evaluating the diagnostic accuracy of wristwatches for OSA. Statistical analysis was performed using R Software version 4.4.0 and OpenMeta[Analyst]. Pooled analyses of sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were determined along with their 95% confidence intervals (CI). The quality of studies was analyzed using the QUADAS-2 tool.

Across 12 included studies, a total of 233 normal adults and 959 patients with an apnea-hypoapnea index (AHI) ≥ 5 were identified. The pooled mean age of subjects was 41.64 ± 15.23 years, and 74% of the population were males. Smartwatch-based detection of obstructive sleep apnea demonstrated a pooled sensitivity of 88.6% (95% CI: 82.1–92.9%; I² = 72.9%) and specificity of 73.0% (95% CI: 66.9–78.3%; I² = 0.0%), indicating high detection ability with moderate heterogeneity across studies. The overall diagnostic accuracy was 85.5% (95% CI: 79.7–89.9%; I² = 80.2%), while the diagnostic odds ratio was 16.4 (95% CI: 9.9–27.3; I² = 38.4%), reflecting strong discriminatory power. The area under the ROC curve (AUC = 0.71; 95% CI: 0.67–0.87) further supported the good overall diagnostic performance of smartwatches compared with polysomnography or home sleep apnea testing.

Smartwatch-based detection of OSA showed high sensitivity and good overall diagnostic accuracy, demonstrating strong potential as a screening tool. However, variability across studies suggests that smartwatch-based assessments should complement, rather than replace, standard PSG.