格拉斯哥抗精神病药副反应评定量表中文版信效度分析

Validity and reliability evaluation of Chinese version of Glasgow antipscyhotics side-effect scale based on the experience of schizophrenia patients

目的分析精神分裂症患者自我体验的格拉斯哥抗精神病药副反应评定量表(Glasgow antipsychotics side-effect scale,GASS)中文版的信度和效度。方法选取200例完全符合DSM-IV诊断标准的精神分裂症患者为研究对象,进行GASS中文版量表评定,信度考察通过计算全样本克朗巴赫α系数(Cronbach'sα)评价量表内部一致性。效度考评采用探索性因子分析方法考评量表结构效度;采用临床总体印象量表中的副反应子量表(clinical global impression scale,CGI)和临床副反应量表(treatment emergent symptom scale,TESS)进行评估,通过GASS与TESS、CGI副反应子量表评分进行相关分析来评价GASS量表的校标效度;以临床标准为参考,计算不同GASS分值的灵敏度和特异度、ROC曲线下面积(area under curve,AUC)来划定量表划界分和验证量表的实证效度。结果信度评价:克朗巴赫α系数为0.926,表明量表内部一致性好。效度评价:通过探索性因子分析,共提取到四个主因子,分别命名为锥体外系副反应因子、生殖泌尿系统副反应因子、抗胆碱能副反应因子、过度镇静和心血管副反应因子。GASS量表的19个条目都在主因子上有较高的负荷值(0.531~0.972),说明量表结构效度良好;GASS总分与TESS和CGI副反应子量表评分严重程度呈显著正相关(r=0.90,P<0.05,r=0.87,P<0.05)。以临床评估标准为参考在区分轻度药物副反应时划界分为1分时灵敏度、特异度最好,分别为96.3%和70.7%,AUC为0.78;区分中度药物副反应时,GASS总分≥20分灵敏度、特异度最高,分别为86.3%,84.7%,AUC为0.92;区分重度药物副反用时GASS总分≥41分灵敏度、特异度最高,分别为92.1%,96.3%,AUC为0.96。结论GASS中文版具有较好的信度和效度,能够反应患者抗精神病药物副反应情况。

Objective To evaluate the reliability and validity of Chinese version of Glasgow antipsychotics side-effect scale(GASS).Methods Totally 200 patients diagnosed as schizophrenia following the criteria of DSM-IV were enrolled in the study.Cronbach’sαwas used to investigate the internal consistency of GASS.Exploring factor analysis was used to examine the construct validity.Pearson correlation analysis between GASS and TESS and CGI sub-scale of side effect were conducted to evaluate the criterion validity.Based on clinical criteria,the sensitivity and specificity and ROC curve was calculated so as to establish the cut-off point of the scale and test discrimination validity.Results For reliability,Cronbach’sαws 0.926,indicating that GASS had a good internal consistency.Four main factors were extracted by exploratory factor analysis which were extrapyramidal side effect factor,genitourinary side effect factor,anticholinergic side effect factor,over sedation and cardiovascular side effect factor.For validity,exploring analysis demonstrated that nineteen item of the scale has relatively higher factor load on the principal factor(0.531-0.972),suggesting that the scale had a good construct validity.The total score of the GASS was positively correlated with that of TESS and CGI subscale of side effects(r=0.90,P<0.05;r=0.87,P<0.05).The cut-off point of experimental validity of GASS for mild antipsychotics side effects was determined as≥1.With this cut-off point,GASS had both high sensitivity(96.3%)and high specificity(70.7%),area under curve(AUC)were 0.78.The cut-off point of GASS for moderate antipsychotics side effects was determined as≥20,with the sensitivity and specificity of GASS were 86.3%and 84.7%,AUC were 0.92.When the cut-off point of GASS for severe antipsychotics side effects was determined as≥41,GASS had both high sensitivity(92.1%)and high specificity(96.3%),AUC were 0.96.Conclusion GASS has good reliability and validity and can reflect the side effects of antipsychotics,and it can be adopted as a useful instrument to evaluate the severity of side effects of antipsychotics.