A comparison of substance abuse and violence in the prediction of long-term rehabilitation outcomes after traumatic brain injury

Objective: To determine the relative contributions of substance abuse histo ry and violent etiology to the prediction of outcomes for individuals who s ustained a traumatic brain injury (TBI) requiring inpatient rehabilitation. Design: Longitudinal study of outcomes 1 year postdischarge from rehabilita tion. Setting: Specialized TBI acute rehabilitation unit. Participants: Three hundred fifty-one individuals consecutively admitted fo r rehabilitation. Interventions: Gathered data from patients' medical records (including etio logy of injury, initial Glasgow Coma Scale scores, and FIM(TM) instrument s cores at discharge), demographic details, and history of substance abuse; p hone and mail survey data from individuals (Satisfaction with Life Scale [S WLS]; Community Integration Questionnaire [CIQ]). Main Outcome Measures: CIQ and SWLS; relative contributions of injury etiol ogy, demographic and injury-related dependent variables, and substance abus e history to predictive model. Results: Almost 80% of persons with injuries from violence-related causes h ad a history of substance abuse. Substance abuse was found to contribute to the prediction of life satisfacton and productivity, while violent etiolog y was not a significant contributor to predictive models. Conclusion: Substance abuse history proved to be a strong predictor of long -term outcomes, while violent etiology of injury was less influential. The results of this study emphasize the need to include substance abuse history in all studies of outcomes after TBI, and to increase prevention efforts t o limit the effects of such a history.