Statistical studies on evaluation of mild disturbance of consciousness. Study of a simpler scale for clinical use

Abstract

In an effort to express the grades of mild disturbance of consciousness (MDOC), or clouding of consciousness, 25 items were selected for application in clinical examination. Factor analysis of the 25 items revealed that MDOC has a two-factor structure; namely, the performance factor (F1) and the verbal factor (F2). This structure was maintained in the sequential examinations. Guttman's scale analysis showed that the scale using the 25 items may be considered to approximate a unidimensional scale. By item selection according to communality, the 25 items could be reduced to 12 items without losing the two-factor structure or the usefulness and reliability of the examination. The principal component analysis (PCA) score, the F1 score, and the F2 score, calculated from the 12 items, were found to be good scales to describe an overall picture and characteristics of the F1 and the F2 of MDOC, respectively. Furthermore, by means of the characteristic curve (ogive) method, it was revealed that MDOC consisted of four clusters of items, the F2 factor (mild), F1 factor (mild), F2 factor (moderate), and F1 factor (moderate), according to the MDOC severity, so that a simpler scale composed of a representative item from each cluster could be constructed which exhibited a high correlation coefficient with the original PCA score.