Y K Wing1, R H Li, C K Ho, S Y Fong, L Y Chow, T Leung. 1. Sleep Assessment Unit, Department of Psychiatry, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong, China. ykwing@cuhk.edu.hk
Abstract
OBJECTIVE: To validate the Chinese version of Ullanlinna Narcolepsy Scale (CUNS). METHODS: A total of 234 subjects [163 male (69.7%) and 71 female (30.3%)] including 17 patients with narcolepsy, 21 normal controls and 196 patients with various sleep and psychiatric disorders were studied. The diagnoses of these patients were independently ascertained with sleep laboratory confirmation whenever indicated. All the subjects were interviewed through the telephone by a trained lay interviewer who was blind to the diagnosis. The questionnaire included demographic information, sleep habits and CUNS. RESULTS: Narcoleptic patients had a significantly higher CUNS score (18.6+/-4.7; 95% confidence interval (CI) 16.2-21.0) and differentiated well from all the other groups (F(6,227)=28.4, P<0.001). The CUNS has a satisfactory internal consistency with Cronbach's alpha of 0.75. The principal component analysis with varimax rotation revealed two factors, namely sleepiness and cataplexy factors, which accounted for 45.5% of the total variance. The best cut-off point for the CUNS scale was found to be at 13/14 with high specificity (93.5%), sensitivity (94.1%), negative predictive value (NPV, 99.5%) and modest positive predictive value (PPV, 53.3%). The AUC of receiver operating characteristic (ROC) analysis was 0.97 (95% CI 0.95-0.99). CONCLUSIONS: The CUNS was validated with satisfactory psychometric properties. The cross-cultural validation of UNS scale suggested that it could be used across the ethnic groups.
OBJECTIVE: To validate the Chinese version of Ullanlinna Narcolepsy Scale (CUNS). METHODS: A total of 234 subjects [163 male (69.7%) and 71 female (30.3%)] including 17 patients with narcolepsy, 21 normal controls and 196 patients with various sleep and psychiatric disorders were studied. The diagnoses of these patients were independently ascertained with sleep laboratory confirmation whenever indicated. All the subjects were interviewed through the telephone by a trained lay interviewer who was blind to the diagnosis. The questionnaire included demographic information, sleep habits and CUNS. RESULTS: Narcoleptic patients had a significantly higher CUNS score (18.6+/-4.7; 95% confidence interval (CI) 16.2-21.0) and differentiated well from all the other groups (F(6,227)=28.4, P<0.001). The CUNS has a satisfactory internal consistency with Cronbach's alpha of 0.75. The principal component analysis with varimax rotation revealed two factors, namely sleepiness and cataplexy factors, which accounted for 45.5% of the total variance. The best cut-off point for the CUNS scale was found to be at 13/14 with high specificity (93.5%), sensitivity (94.1%), negative predictive value (NPV, 99.5%) and modest positive predictive value (PPV, 53.3%). The AUC of receiver operating characteristic (ROC) analysis was 0.97 (95% CI 0.95-0.99). CONCLUSIONS: The CUNS was validated with satisfactory psychometric properties. The cross-cultural validation of UNS scale suggested that it could be used across the ethnic groups.
Authors: Lei Chen; S Y Y Fong; Ching W Lam; Nelson L S Tang; Margaret H L Ng; Albert M Li; C K W Ho; Suk-Hang Cheng; Kin-Mang Lau; Yun Kwok Wing Journal: Sleep Date: 2007-07 Impact factor: 5.849