Validity and reliability of Chinese version of alcohol withdrawal scale(AWS)
10.3760/cma.j.issn.1674-6554.2010.07.029
- VernacularTitle:酒精戒断状态评定量表中文译本的效度和信度测试
- Author:
Chuanjun ZHUO
;
Yueqin HUANG
;
Yi TANG
;
Lei YANG
;
Jun GENG
;
Jitao LI
;
Xiangyang GAO
;
Bing LI
- Publication Type:Journal Article
- Keywords:
Alcohol Withdrawal State;
Alcohol Withdrawal Scale (AWS);
Validity;
Reliability
- From:
Chinese Journal of Behavioral Medicine and Brain Science
2010;19(7):661-663
- CountryChina
- Language:Chinese
-
Abstract:
Objective To evaluate the validity and reliability of the Chinese version of Alcohol Withdrawal Scale (AWS). Methods Totally 175 patients diagnosed as alcohol dependence according to the criteria of ICD-10 were studied. Intraclass correlation coefficient (ICC) analysis was applied for examining interrater consistency and Cronbach' s α for internal consistency. Factor analysis was used to examine the construct validity. Correlation analysis between AWS and CGI,Revised Clinic Institute Alcohol Withdrawal Syndrome Scale(CIWA-Ar) were conducted to evaluate the criterion validity. Based on clinical criteria,ROC curve was calculated so as to test the discriminative ability and establish the cut-off point of the scale. Results ( 1 ) Reliability: ICC value was 0.93,and Cronbach's α was 0.83,which indicated good interrater and internal consistency. (2) Validity:the correlation coefficients of the two subscale with the total scale score were 0.78,0.83 respectively. The correlation coefficients between the subscale were 0. 81 and factor analysis revealed that each item of the scale had relatively high load on the primary factor (0.409 ～0.926). At the time of admission,the total score of the AWS was positively correlated with that of CGI ( r = 0.71, P < 0.05 ). The total score of the AWS also was positively correlated with that of CIWA-Ar ( r = 0. 86, P<0. 05). As treatment went on,total score of the AWS showed a downward trend,at the end of the first week,the total score of the AWS was positively correlated with that of CGI ( r = 0.62, P<0.05). (3)The cut-off point of AWS for mild alcohol withdrawal state was determined as ≥3. With this cut-off point,AWS had both high sensitivity (92.1% ) and specificity (73.5% ) ,and the area under curve (AUC) was 0. 91. The cut-off point of AWS for moderate withdrawal state was determined as ≥7, and the sensitivity and specificity of AWS were 94.3 % and 89.7 % respectively, with the AUC of 0.94. The cut-off point of AWS for severe withdrawal state was determined as ≥ 10. With this cut-off point AWS had both high sensitivity (94. 9% ) and high specificity (92.6% ) .with the AUC of 0.93. Conclusion AWS has good reliability and validity and can reflect the change of the disease and the efficacy of treatment.
