Abstract INTRODUCTION Traumatic brain injury (TBI) is a leading cause of morbidity and mortality in the US. The effects of TBI on quality of life may not become apparent for years after the injury. There are conflicting reports in the literature regarding long term outcomes. Physicians are often asked to predict long term functional and cognitive outcomes, with limited data available. METHODS Patients with severe TBI (GCS = 9) who previously participated in a clinical trial during the 1980s were followed up with and compared to healthy controls without history of TBI. A health questionnaire, sports concussion assessment tool version 3 (SCAT3) and the Telephone Interview for Cognitive Status-modified (TICS-m) were completed over the phone and compared with controls using t-test. GCS at admission and 12-month GRS were used to predict to TICS-M at 30 years using linear regression. RESULTS >45 of the initial 168 subjects were confirmed alive, and 37 (13 females; mean age: 52.43 years S.D. 10.7) consented. Controls (n = 58; 23 females; mean age = 54 years, S.D. 11.5) had lower symptom severity score (6.7 S.D. 12.6 versus 20.6 S.D. 25.3; P = 0.005), lower total number of symptoms (3.4 S.D. 4.7 versus 7.12 S.D. 6.5; P = 0.006), higher standardized assessment of concussion score (25.6 S.D. 2.8 versus 21.2 S.D. 6.9; P = 0.001), and lower corrected MPAI-4 (22.3 S.D. 17.0 versus 43.7 S.D. 12.8; P < 0.001). GCS at admission did not predict cognitive status at 30-years assessed using TICS-M (P = 0.345). The Glasgow Outcome Scale score at 12-months was correlated to TICS-M at 30 years (R = 0.548, P < 0.001); each point decrease in GOS decreasing the score at TICS-M by 5.6 points. CONCLUSION Remote history of TBI disrupts the lives of survivors long after injury. Admission GCS does not predict cognitive status 30 years after TBI. The GOS at 12-months predicted the cognitive status assessed using TICS-M score at 30 years.