RELATIONSHIP OF 2 MEASURES OF INJURY SEVERITY TO PEDIATRIC PSYCHOLOGICAL OUTCOME 1-3 YEARS AFTER ACUTE HEAD-INJURY

Objective: To compare the strength of relationship of the Glasgow Coma Scale (GCS) versus the ASCOT probability of survival (Ps), a multidim ensional trauma severity index, with a comprehensive range of core are as of psychological functioning assumed to influence quality of life i n pediatric head injury survivors. Design: Correlational relationships between the GCS versus the Ps and standardized outcome measures in si x psychological domains were compared. Ordering effects of level of se verity on outcome were studied through analyses of variance for the GC S versus the Ps. Setting: A pediatric level I trauma center. Patients: Eighty-six children with mild to severe traumatic brain injury retros pectively recruited from consecutive head injury admissions for a 30-m onth period. Measures: Wechsler Intelligence Scale for Children-Revise d, Trail Making Test, Tactual Performance Test, Fuld Object Memory Eva luation, Finger Tapping, Grooved Pegboard, Beery Developmental Test of Visual-Motor Integration, Detroit Oral Directions, Woodcock-Johnson T ests of Achievement, Vineland Adaptive Behavior Scales, Achenbach Chil d Behavior Checklist, Harter Self-Perception Scales. Results: Signific ant correlations were found between each of the severity indices (GCS and Ps) and the IQ and neuropsychological variables. Both Ps and GCS w ere able to distinguish the severe injury group from lesser injury. Ps yielded a stronger pattern of correlations and a clearer step-wise se paration of the three injury severity levels on neuropsychological var iables. Only the GCS was related to adaptive behavior. Neither severit y index showed strong correlational patterns with behavioral or self-p erception psychological domains. Conclusions: Multidimensional trauma severity measures such as the Ps merit further study as tools for pred iction of longterm neuropsychological outcome following pediatric head injury, especially in the case of children with moderate head injurie s. Findings must be considered preliminary due to sample limitations b ut do emphasize the greater degree of independence of psychosocial, as compared to neurocognitive, outcome relative to initial injury severi ty.