The relationship between Glasgow Coma Scale ratings and the neuropsychological functioning in acutely head injured thirteen through twenty-six year olds

No Thumbnail Available
Layton, Donald Charles
Wenck, L. Stanley (Lewis Stanley)
Issue Date
Thesis (D. Ed.)
Other Identifiers

The purpose of this study was to determine if the Glasgow Coma Scale (GCS) provides a valid indication of severity after closed head injury. A second purpose was to determine the nature of the deficits seen with head injured persons shortly after injury.The subjects were 69 patients ages 13 through 26 selected from consecutive closed head injury admissions to a large acute care hospital. The subjects received comprehensive neuropsychological testing after having passed the Galveston Orientation and Amnesia Test.Two a priori hypotheses were specified for each of eight neuropsychological test groupings (i.e., motor, memory, visuographic, achievement, abstraction and concept formation, language, problem solving, and psychometric intelligence). For each of the eight test groupings two, one-way multivariate analyses of covariance (MANCOVA) were used to determine statistical significance. Premorbid IQ was used as a covariate in all of the analyses. Specifically, these a priori contrasts were GCS group (8 compared with GCS group 9-15 and GCS group 9-12 compared with GCS group 13-15. Each of the significant MANCOVA tests was followed with discriminant analysis.The results revealed that 15 out of the 16 null hypotheses were rejected (p<.05 or less) thus providing clear evidence for the usefulness of the GCS as an indicator of the severity of injury. Most notable of the results is that the GCS group comparison of 9-12 and 13-15 reached significance in seven of the eight comparisons. Given this finding, there seems justification for the GCS division of 9-12 denoting a moderately injured group and 13-15 denoting a mildly injured group as was proposed in previous research. Redundancy indexes of 6% to 25% were obtained which indicated small to relatively large effect sizes for the various criterion variates.With discriminant analysis it was determined that accurate classification into GCS groups could be achieved in 64% to 94% of the cases based upon a combination of premorbid IQ and the criterion variates. This represented an increase in classification accuracy of from 13% to 51%to be over what could be accomplished with knowledge of the premorbid IQ alone. The motor test composite, problem solving test composite and abstraction and concept formation composite revealed the highest rates of reclassification.