Traumatic brain injury (TBI) can occur concomitantly with spinal cord injury (SCI). Much of the initial work in this area has focused on identifying coincidence rates and risk factors; less has focused on possible long-term implications of TBI when it occurs with SCI. In this study, SCI/TBI and neurologically matched SCI-only groups were formed on the basis of clinicians' ratings of neuropsychologic test scores. SCI/TBI and SCI-only groups were also formed using Halstead Category cutoff scores, presence/absence of loss of consciousness, and clinicians' ratings of severity of TBI. Dependent measures assessed an average of two years postinjury measured personal, social, and family adjustment of the individual with SCI and that of a significant other. Loss of consciousness and nonconsensus clinical ratings of presence/absence of TBI predicted postdischarge adjustment poorly. Severity ratings in the moderate to severe range, and Category cutoff scores did have some predictive value, with patients defined as impaired being more likely to report adjustment difficulties or being described as having adjustment difficulties by a significant other. The difficulty of making unequivocal diagnoses of TBI in this population is discussed and implications for future research delineated. © 1991.