OBJECTIVES: The Berlin Questionnaire (BQ) has been used to help identify patients at high risk of having sleep apnea in primary care. But it has not been validated in a sleep clinic for Korean patients. The aim of this study is to evaluate the usefulness of the BQ as a screening tool for obstructive sleep apnea (OSA) for Korean patients in a sleep clinic. METHODS: The BQ was prospectively applied to 121 subjects with OSA suspicion who visited to our sleep clinic. All subjects performed overnight polysomnography. OSA was defined as an apnea-hypopnea index (AHI) > or =5. We investigated the sensitivity, specificity, positive and negative predictive values of the BQ according to severity by AHI. RESULTS: In 121 subjects, 73.6% were males, with a mean age of 48.8+/-13.0 years. Twenty-five (20.6%) patients did not have OSA (AHI<5), 30 (25%) patients had mild OSA (AHI> or =5 and <15), 26 (21.4%) had moderate (AHI> or =15 and <30), and 40 (33%) had severe OSA (AHI> or =30). The BQ identified 69.4% of the patients as being at high risk for having OSA. The sensitivity and specificity of the BQ were 71.9% and 40%, for AHI> or =5, 75.8% and 38.2% for AHI> or =15, 77.5% and 34.6% for AHI> or =30, respectively. The positive and negative predictive values of the BQ were 82.1% and 27.0% for AHI> or =5, respectively. Positive and negative likelihood ratios were 1.2 and 0.7, and the overall diagnostic accuracy of the BQ was 65.3%, using an AHI cut-off of 5. CONCLUSION: Due to modest sensitivity and low specificity, the BQ does not seem to be an appropriate tool for identifying patients with obstructive sleep apnea in a sleep clinic population.