BACKGROUND: Deep venous thrombosis (DVT) is a major cause of mortality and morbidity after traumatic brain injury (TBI). There is no consensus regarding appropriate screening, prophylaxis, or treatment during acute rehabilitation. METHODS:: This prospective observational study evaluated prophylactic anticoagulation during rehabilitation in patients with TBI aged 16 years or older admitted to 12 TBI Model Systems rehabilitation centers (July 2004-December 2007). After propensity score stratification within center, the odds ratio associated with incidence of symptomatic DVT or pulmonary embolism (PE) for patients who did and did not receive prophylactic anticoagulation was estimated using conditional logistic regression in patients who were not screened for DVT on rehabilitation admission or who screened negative; the analysis was repeated in these two subgroups. RESULTS:: Patients with identified DVTs at rehabilitation admission (n = 266) were excluded, leaving 1,897 patients: 1,002 screened negative, 895 unscreened; 932 received prophylactic anticoagulation, and 965 did not. Symptomatic DVT/PE was detected in 32 patients (15 of 932 [1.6%] with prophylaxis, 17 of 965 [1.8%] without). After propensity score adjustment, the odds ratio (95% confidence interval) for symptomatic DVT/PE with prophylaxis versus no prophylaxis was 0.80 (0.33-1.94) in the full analytic population and 0.46 (0.12-1.84) in the screened-negative subgroup. The only probable venous thromboembolism-related death occurred in the prophylactic anticoagulation group. Fewer new/expanded intracranial hemorrhages occurred among patients who received prophylactic anticoagulation. CONCLUSIONS:: Prophylactic anticoagulation during rehabilitation seemed safe for TBI patients whose physicians deemed it appropriate, but did not conclusively reduce venous thromboembolism. Given the number of DVTs present before rehabilitation, screening and prophylaxis during acute care may be more important. Copyright © 2010 by Lippincott Williams & Wilkins.