Since some sequelae of spinal cord injury (SCI) may be confused with symptoms of depression, the use of the Beck Depression Inventory (BDI) in individuals with SCI may be problematic. We hypothesized that seven BDI items would be poor discriminators between depressed and nondepressed SCI subjects. To test this hypothesis, 124 veterans with SCI were administered the BDI and were evaluated for the presence of a Major Depressive Episode. Discriminant function analysis revealed that three of the seven items were poor discriminators. Since clinicians are faced with the dilemma of deciding whether endorsement of BDI items is indicative of depressive symptomatology or the sequelae of spinal cord injury, we derived cut scores appropriate for SCI patients. Data on the sensitivity and specificity of these cut scores are provided.