Objective : We aimed to identify clinico-radiological risk factors that may predict unfavorable neurological outcomes in traumatic brain injury (TBI), and to establish a guideline for patient selection in clinical trials that would improve neurological outcome during the early post TBI period. Methods : Initial clinico-radiological data of 115 TBI patients were collected prospectively. Regular neurological assessment after standard treatment divided the above patients into 2 groups after 6 months : the Favorable neurological outcome group (GOS : good & moderate disability, DRS : 0-6, LCFS : 8-10) and the Unfavorable group (GOS : severe disability-death, DRS : 7-29 and death, LCFS : 1-7 and death). Results : There was a higher incidence of age $geq$35 years, low initial GCS score, at least unilateral pupil dilatation, and neurological deficit in the Unfavorable group. The presence of bilateral parenchymal lesions or lesions involving the midline structures in the initial brain CT was observed to be a radiological risk factor for unfavorable outcome. Multivariate analysis demonstrated that age and initial GCS score were independent risk factors. The majority of the Favorable group patients with at least one or more risk factors showed improvement of GCS scores within 2 months after TBI. Conclusion : Patients with the above mentioned clinico-radiological risk factors who received standard treatment, but did not demonstrate neurological improvement within 2 months after TBI were deemed at risk for unfavorable outcome. These patients may be eligible candidates for clinical trials that would improve functional outcome after TBI.