The significance of STOP-BANG, Epworth and Berlin questionnaires in predicting obstructive sleep apnea confirmed by respiratory poligraphy

Introduction: Obstructive sleep apnea (OSA) is a common sleep-disordered breathing disorder whose diagnosis requires polysomnography, which sometimes limits availability. The aim: To determine the diagnostic value of the STOP-BANG questionnaire, Epworth Sleepiness Scale (ESS), and Berlin Questionnaire in predicting OSA severity confirmed by respiratory polygraphy. Methods: A retrospective analysis included 105 patients with diagnosed OSA (AHI ≥ 5) treated at the Pulmonology Clinic of the University Hospital Center Niš. Results: The mean age was 52.95 ± 11.25 years, BMI 34.85 ± 6.68 kg/m², and the mean AHI 34.20 ± 21.13 events/h. Severe OSA was present in 53.3% of the subjects. All three questionnaires showed a statistically significant correlation with AHI (Spearman r = 0.361-0.370, p < 0.001). For the prediction of moderate to severe OSA (AHI ≥ 15), STOP-BANG with a threshold of ≥ 5 had the best AUC of 0.759, sensitivity of 90.5% and specificity of 38.1%. For the prediction of severe OSA (AHI ≥ 30), ESS showed the best AUC of 0.704 with balanced sensitivity (66.1%) and specificity (61.2%). The Berlin questionnaire had high sensitivity (92.9%) but low specificity for both severity levels. The optimal thresholds according to the Youden index differed from the standard ones: ESS ≥ 7 for moderate to severe, ≥ 12 for severe OSA; STOP-BANG ≥ 7 for both levels. Conclusion: STOP-BANG questionnaire with a threshold of ≥ 5 is the most useful screening tool for the identification of moderate to severe OSA, while ESS has the best performance for the prediction of severe OSA.