USE OF THE EPWORTH SLEEPINESS SCALE, THE NOSAS, AND THE STOP-BANG QUESTIONNAIRE TO IDENTIFY PATIENTS WITH MODERATE-TO-SEVERE OBSTRUCTIVE SLEEP APNEA

S. Yin,H. Xu,C. Zhang,J. Zou,J. Guan,H. Yi
DOI: https://doi.org/10.1093/SLEEP/ZSAA056.598
IF: 6.313
2020-01-01
SLEEP
Abstract:Abstract Introduction A variety of scales and questionnaires regarding sleep and sleep-related disorders have been widely used in scientific research and clinical practice, as important tools for differential diagnosis and rapid screening of complex sleep disorders, especially obstructive sleep apnea (OSA). However, the diagnostic efficacy of different scales and questionnaires for patients with different severity of OSA and of different demographic characteristics has not been clearly described. In this study, we evaluated the ability of the most popular scales, including the Epworth Sleepiness Scale (ESS), the NoSAS, and the STOP-BANG questionnaire in predicting moderate-to-severe obstructive sleep apnea (OSA) by gender. Methods This cross-sectional study screened 2,031 consecutive subjects referred with suspected OSA from 2012 to 2016. Anthropometric measurements, polysomnographic data, ESS, NoSAS scores and STOP-BANG scores were recorded. Receiver operating characteristic curve analyses were performed, and the final predictive models were verified in a validation cohort. Results A total of 1,840 adults were finally included. The STOP-BANG questionnaire afforded a better diagnostic accuracy than did the ESS, with different cutoffs for the two genders: 3 in males and 1 in females. A predictive model based on STOP-BANG yielded an area under the curve (AUC) of 0.918 (0.897-0.935), a sensitivity of 79.89%, and a specificity of 89.19%, in males; and an AUC of 0.951 (0.914-0.975), a sensitivity of 80.52%, and a specificity of 95.92%, in females. In the validation cohort, the sensitivity and specificity were respectively 85.44 and 93.00% in males and respectively 83.02 and 87.60% in females. Conclusion The STOP-BANG questionnaire was moderately effective when used to screen for moderate-to-severe OSA. A STOP-BANG-based predictive model afforded excellent diagnostic efficacy, which could be applied in clinical practice. However, gender differences must be considered. Support This study was supported by Grants-in-aid from Shanghai Municipal Commission of Science and Technology (Grant No.18DZ2260200).
What problem does this paper attempt to address?