Article

OBJECTIVES: (1) To examine predictive validity of global neuropsychological
performance, and performance on timed tests (controlling for manual motor
function) and untimed tests, including attention, memory, executive function, on
return to productivity at 1 year after traumatic brain injury (TBI). (2) To
compare predictive validity at 8 weeks versus 5 months postinjury. (3) To examine
predictive validity of early degree of recovery (8wk-5mo postinjury) for return
to productivity. DESIGN: Longitudinal, within subjects. SETTING: Inpatient
neurorehabilitation and community. PARTICIPANTS: Patients (N=63) with moderate to
severe TBI. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Primary
outcome: return to productivity at 1 year postinjury. Primary predictors:
neuropsychological composite scores. Control variables: posttraumatic amnesia,
acute care length of stay (LOS), Glasgow Coma Scale score, age, and estimated
premorbid intelligence quotient. RESULTS: Return to productivity was
significantly correlated with global neuropsychological performance at 5 months
postinjury (P<.05) and showed a trend toward significance at 8 weeks. Performance
on the untimed composite score, and more specifically executive and memory
functions, mirrored this pattern. Logical Memory performance significantly
predicted return to productivity, but not other memory tests. Timed tests showed
no significance or trend at either time point. Early degree of recovery did not
predict return to productivity. Among control variables, only acute care LOS was
predictive of return to productivity. CONCLUSIONS: Findings validate utility of
early neuropsychological assessment for predicting later return to productivity.
They also provide more precise information regarding the optimal timing and test
type: results support testing at 5 months postinjury on untimed tests (memory and
executive function), but not simple attention or speed of mental processing.
Findings are discussed with reference to previous literature.

Langue : ANGLAIS