Prediction of outcome at discharge from acute care following traumatic brain injury

Objective: To compute outcome probabilities for persons with traumatic brain injury at discharge from acute care. Participants: Three hundred thirty-nine patients with traumatic brain injury (239 mild, 48 moderate, 52 severe). Setting: Level I trauma center. Main Measures: Predictor variables considered were age, education, Glasgow Coma Scale score, duration of posttraumatic amnesia, cerebral imaging results, and need for neurosurgical intervention. Outcome measures were Extended Glasgow Outcome Scale and discharge destination. Results: Logistic regressions showed that a shorter posttraumatic amnesia decreased the probability of moderate to severe disability. Moreover, discharge home was less probable for patients with positive cerebral imaging. Conclusion: This model can help predict rehabilitation needs upon discharge from an acute care hospital.