Hafkenscheid A
Sinai-Centre, Jewish Mental Health Services, Amersfoort, The Netherlands.
Acta Psychiatr Scand. 1993 Nov;88(5):305-10. doi: 10.1111/j.1600-0447.1993.tb03464.x.
This study aimed to determine the replicability of the interrater reliability coefficients obtained with a standardized and expanded Brief Psychiatric Rating Scale (BPRS-E) in a 1991 psychometric evaluation. Furthermore, intrarater reliability was assessed. At item level, interrater concordance turned out to be satisfactory for most of the BPRS-E items. However, only a few of the items reached acceptable chance-corrected coefficients. In contrast to the previous study, the anxiety-depression subscale met the standard of acceptable interrater reliability in the present study. As in the 1991 study, the 10-item psychotic disintegration scale as well as BPRS-18 global scores met (or closely approximated) this standard. The 6 additional items of BPRS-E did not contribute to the scale's reliability. Joining the samples of the 1991 and replication studies (to cover the range of symptoms' severity and heterogeneity more fully) did not improve interrater reliability. Intrarater reliability coefficients were globally comparable to interrater reliability coefficients. In all, the results of this replication study suggest that only the anxiety-depression subscale, the 10-item psychotic disintegration scale and the BPRS-18 global scale can be used reliably in unselected groups of psychiatric inpatients in acute distress.
本研究旨在确定在1991年的心理测量评估中，使用标准化且扩展的简明精神病评定量表（BPRS-E）所获得的评分者间信度系数的可重复性。此外，还评估了评分者内信度。在项目层面，对于大多数BPRS-E项目，评分者间的一致性结果令人满意。然而，只有少数项目达到了可接受的机会校正系数。与之前的研究不同，在本研究中，焦虑抑郁分量表达到了可接受的评分者间信度标准。与1991年的研究一样，10项精神病性解体量表以及BPRS-18总分达到（或接近）了这一标准。BPRS-E的另外6个项目对该量表的信度没有贡献。将1991年研究和重复性研究的样本合并（以更全面地涵盖症状严重程度和异质性范围）并没有提高评分者间信度。评分者内信度系数总体上与评分者间信度系数相当。总体而言，这项重复性研究的结果表明，只有焦虑抑郁分量表、10项精神病性解体量表和BPRS-18总量表可在处于急性痛苦中的非选择性精神病住院患者组中可靠使用。
