Müller M J, Marx-Dannigkeit P, Schlösser R, Wetzel H, Addington D, Benkert O
Department of Psychiatry, University of Mainz, Germany.
J Psychiatr Res. 1999 Sep-Oct;33(5):433-43. doi: 10.1016/s0022-3956(99)00018-7.
A German version of the Calgary Depression Rating Scale for Schizophrenia (CDSS-G) approved by the author of the original scale is presented comprising a semi-structured interview for 9 items to sensitively and specifically assess depression in schizophrenia and related disorders. The process of translation is outlined and the finally derived CDSS-G was investigated with respect to interrater reliability in three studies. To keep comparability with the CDSS source version a standard procedure was used. Two trained raters jointly assessed ten schizophrenic patients (study I). In a second study, videotapes with the CDSS-G were presented to clinically inexperienced raters (study II, N = 14/15) to test the agreement on the CDSS-G in this sample. Finally, in a third study clinically experienced researchers participated in a rater training (study III, N = 34). They carried out CDSS ratings on three patients with mild depressive symptoms. The dependence of interrater reliability on depression severity was investigated for all studied patients. Both intraclass correlation coefficients (ICC) and weighted kappa coefficients (kappa(w)) were calculated. The results revealed a high ICC = 0.97 in study I for the total CDSS-G score. Single item ICC values were all above 0.70. The results of study II revealed somewhat lower agreement on CDSS-G items and total scores in psychiatric novices with however acceptable values of kappa(w)>0.50 for the total scores. Study III yielded satisfactory results (0.66<kappa(w)<0.76) for clinically experienced psychiatrists in schizophrenic patients with mild depression. The results demonstrate adequate to excellent interrater reliability of the CDSS-G corroborating the results of the original version in different settings.
本文呈现了经原量表作者认可的德语版《卡尔加里精神分裂症抑郁评定量表》（CDSS-G），它包含一个针对9个条目的半结构化访谈，用于敏感且特异的评估精神分裂症及相关障碍中的抑郁症状。文中概述了翻译过程，并在三项研究中对最终得出的CDSS-G进行了评分者间信度的调查。为保持与CDSS源版本的可比性，采用了标准程序。两名经过培训的评分者共同评估了10名精神分裂症患者（研究I）。在第二项研究中，向临床经验不足的评分者展示了CDSS-G的录像带（研究II，N = 14/15），以测试该样本中对CDSS-G的一致性。最后，在第三项研究中，有临床经验的研究人员参与了评分者培训（研究III，N = 34）。他们对三名有轻度抑郁症状的患者进行了CDSS评分。针对所有研究患者，调查了评分者间信度对抑郁严重程度的依赖性。计算了组内相关系数（ICC）和加权kappa系数（kappa(w)）。结果显示，在研究I中，CDSS-G总分的ICC高达0.97。单个条目的ICC值均高于0.70。研究II的结果显示，精神科新手对CDSS-G条目和总分的一致性略低，但总分的kappa(w)>0.50仍在可接受范围内。对于患有轻度抑郁症的精神分裂症患者，研究III为有临床经验的精神科医生提供了满意的结果（0.66<kappa(w)<0.76）。结果表明，CDSS-G具有足够至优异的评分者间信度，在不同环境下证实了原始版本的结果。
