Brazo P, Ribeyre J M, Petit M, Dollfus S
Centre Esquirol, CHU Côte-de-Nacre, Caen.
Encephale. 1996 May-Jun;22(3):165-74.
Since Crow, Andreasen et al. have described schizophrenia in terms of negative and positive symptoms, the dichotomic approach has been well established. As a matter of fact, factor analyses, especially principal components analyses, led with symptomatic specific scales, have proved their validity. But they have shown their limits too : some authors think that the dichotomic model fails to explain all of the schizophrenic psychopathology and that a third dimension including formal thought disorders, most of the time called "disorganization", should systematically be taken into account. In this study, the authors have hypothesized that a categorial approach could describe this "disorganization". Using a cluster analysis they investigated the existence of subtypes in a population including 136 schizophrenic patients assessed with the PANSS (Positive and Negative Syndrome Scale, Kayet al., 1987). The results suggested at least five subtypes: a pure positive subtype, characterized by high scores on items delusions, hallucinatory behavior, suspiciousness/persecution, and by a low score on conceptual disorganization item; a disorganized positive subtype, characterized by high scores on positive items, including conceptual disorganization item, and also high scores on unusual thought content and autistic preoccupation items; a negative subtype, characterized by high scores on negative items and low scores on positive items, including conceptual disorganization item; a mixed subtype, characterized by high scores on the most positive, negative and general psychopathological items; a residual subtype, characterized by low scores on all the positive, negative and general psychopathological items. The good validity of this analysis was showed since differences on a number of clinical characteristics were observed between the five clusters. These results demonstrated the oversimplication of the positive-negative dichotomy and the relevance of a disorganized subtype.
自克劳、安德烈亚森等人根据阴性和阳性症状描述精神分裂症以来，二分法已得到广泛确立。事实上，通过症状特异性量表进行的因素分析，尤其是主成分分析，已证明其有效性。但它们也显示出了局限性：一些作者认为二分法模型无法解释所有的精神分裂症心理病理学现象，应该系统地考虑包括形式思维障碍（大多数时候称为“紊乱”）在内的第三个维度。在本研究中，作者假设分类法可以描述这种“紊乱”。他们使用聚类分析研究了一个包含136名精神分裂症患者的群体中是否存在亚型，这些患者使用阳性和阴性症状量表（PANSS，凯伊等人，1987年）进行评估。结果表明至少有五种亚型：一种纯阳性亚型，其特征是在妄想、幻觉行为、猜疑/迫害等项目上得分高，而在概念紊乱项目上得分低；一种紊乱阳性亚型，其特征是在阳性项目上得分高，包括概念紊乱项目，在异常思维内容和自闭症关注项目上得分也高；一种阴性亚型，其特征是在阴性项目上得分高，在阳性项目上得分低，包括概念紊乱项目；一种混合亚型，其特征是在最阳性、阴性和一般精神病理学项目上得分高；一种残留亚型，其特征是在所有阳性、阴性和一般精神病理学项目上得分低。由于在五个聚类之间观察到了一些临床特征上的差异，因此表明了该分析具有良好的有效性。这些结果证明了阳性-阴性二分法的过度简化以及紊乱亚型的相关性。
