Global Functional Outcome and Multidimensional Neuropsychological Morbidity at 12-Months After Paediatric Traumatic Brain Injury

Background: The Glasgow Outcome Scale Extended for Pediatrics (GOSE-P) is widely used as a global outcome measure after paediatric traumatic brain injury (pTBI), including as an endpoint in neurocritical care research. However, the extent to which it reflects multidimensional neuropsychological morbidity is uncertain.

Objectives: To examine the relationship between GOSE-P and multidimensional neuropsychological outcomes 12 months after pTBI.

Methods: This secondary analysis used prospectively collected data from the multicentre STARSHIP study. Of 135 participants, 41 had an observed 12-month GOSE-P and at least one additional neuropsychological assessment measure. Eight measures assessed executive, participation, emotional/behavioural, health-related quality of life, anxiety/depressive, and communication domains. Associations with GOSE-P were examined using Spearman rank correlations and logistic regression.

Results: At 12 months post-injury, clinically significant impairments were identified across all domains, most frequently in communication (11/17, 64.7%) and executive functioning (20/32, 62.5%). Poorer GOSE-P outcome was associated with greater executive (ρ=0.44, p=0.017) and emotional/behavioural difficulties (ρ=0.35, p=0.049), poorer quality of life (ρ=0.63, p=0.001), and reduced communication (ρ=0.52, p=0.03). Each one-level increase in GOSE-P was associated with higher odds of emotional/behavioural impairment (OR 1.72, 95% CI 1.10-2.72) and impaired quality of life (OR 1.77, 95% CI 1.03-3.01). Despite favourable global recovery (GOSE-P 1-2), 11/19 (57.9%) participants had impairment in at least one assessed domain. Extended-assessment completion was variable, limiting precision.

Conclusions: GOSE-P showed meaningful associations with several domain-specific outcomes, supporting its value as a pragmatic global measure after pTBI. However, favourable global recovery did not exclude clinically important multidimensional morbidity. Targeted domain-specific assessment may provide additional information when detailed characterization of recovery is required.