Validation of GOAL questionnaire as screening tool for clinical obstructive sleep apnea: A large sample study in China.

Abstract

Obstructive sleep apnea (OSA) is a serious disease with a high prevalence in the general population. The purpose of this study is to explore the effectiveness of the GOAL questionnaire in the clinical screening of OSA and compare it with other existing screening tools. Outpatients and inpatients who underwent polysomnography (PSG) examination at the Sleep Medicine Center of the First Affiliated Hospital of Guangzhou Medical University from January 2013 to November 2016 were analyzed retrospectively. The basic data such as demographic, medical history, etc., and PSG data of the patients were collected, and the sensitivity, specificity, positive predictive value, negative predictive value and area under the curve (AUC) of GOAL and five other screening scales (the NoSAS score, Epworth Sleepiness Scale, the Berlin questionnaire, STOP, and STOP-Bang questionnaire) were calculated. Data from 2,171 participants (1,644 male; 78%) were analyzed there were 1,507 OSA patients [Apnea Hypopnea Index (AHI) ≥ 5 events/h] among them, accounting for about 69.415%. No matter which cut-off point (AHI ≥ 5, 15 and 30 events/h), the AUC score reveals that GOAL questionnaire had comparable screening ability to the NoSAS and STOP-BANG, and performed better than the ESS, and the AUC scores of the STOP questionnaire and Epworth Sleepiness Scale (ESS) were both lower than 0.7. When the cut-off point of the AHI was 5 events/h, the AUC of GOAL was the highest at 0.799 (0.781-0.816), and its sensitivity was the highest at 89.1%. The sensitivity levels of the NoSAS score and STOP-Bang questionnaire were 67.4 and 78.8% respectively, while ESS and the Berlin questionnaire have higher specificity (70.2 and 72.3% respectively) but lower sensitivity (49.3 and 60.0% respectively). GOAL is a free, efficient and easy to manage tool with a screening ability comparable to NoSAS and STOP-Bang, and better than that of ESS.