Background: The gold standard for diagnosis of OSA is polysomnography which is labor intensive, and even simpler polygraphic approaches require staff resource. It is estimated that most patients with severe OSA are undiagnosed and thus untreated in many countries including China. It is important to detect patients with severe OSA in a cost-effective way.

Methods: We recently developed a simple questionnaire: BMI ?30 kg/m² (2 points), BMI 25-30 kg/m² (1 point), often snoring (1 point), apnea during sleep (1 point), morning or daytime fatigue (1 point), dry mouth in the morning (1 point). A total score ?5 was suspected as severe OSA.

Results: The questionnaire was evaluated in 484 patients who had completed an overnight polysomnography study for clinical reasons.  We found that AUC was 0.697(95%CI 0.654-0.738) for severe OSA. A prospective study was further performed in 195 subjects referred for polysomnography in Guangzhou China. The AUC was 0.797 (95%CI 0.731-0.863) for diagnosis of severe OSA with specificity of 93.9% (95%CI 88.0-97.2%) and sensitivity of 39.7% (95%CI 27.8-52.8%).

Conclusions: The simplified questionnaire could be a useful tool for detecting severe OSA patients who need immediately treatment in community in a Chinese population.