An analysis of Motor Index score, pinprick sensory score, joint position sense score, somatosensory evoked potential (SSEP) grade in the ulnar (SSEPu) and posterior tibial (SSEPt) regions, and overall SSEP grade (mean SSEPu + t) was conducted in 36 patients with cervical spinal cord injuries to determine the relationship of these scores, both individually and in combination, to functional outcome (as determined using the Barthel Index) at 6 months after injury. The clinical and electrophysiological data were obtained on the same day within 2 weeks after injury. Nineteen patients underwent two SSEP tests 1 week apart within the first 3 weeks following injury in an attempt to identify mean SSEPu + t improvement. Somatosensory evoked potential grading was based on the presence or absence of the cortical evoked potential, the amplitude of the early cortically generated waveform (P22 or P37), and the interpeak latency across the lesion site. Mean SSEPu + t had the strongest individual relationship with outcome (R-square 0.75, p less than 0.0001) and mean SSEPu + t improvement over a 1-week interval during the first 3 weeks after injury was associated with Motor Index score improvement over a 6 month period. Joint position sense score was the best clinical predictor of outcome (R-square 0.64, p less than 0.0001). Mean SSEPu + t correlated with outcome more closely than the combination of Motor Index score and pinprick sensory score. Mean SSEPu + t in combination with all three clinical indicators produced the strongest correlation with outcome (R-square 0.87, p less than 0.0001). This study confirms the prognostic value of quantitative SSEP analysis for patients with acute spinal cord injuries.