Objective: To determine the long-term efficacy of gabapentin as a treatment of pain after spinal cord injury. Design: All patients with traumatic-onset spinal cord injury treated with gabapentin were identified and followed up using a longitudinal observational design with two contact points (6 and 36 months after the trial) using a semi-structured interview. The first follow-up interview attempted to capture all 31 patients placed on therapeutic trial. The second follow-up interview attempted to capture those reporting a favorable response (n = 14) to the therapeutic trial at the first follow-up. Results: Of the 27 patients contacted at the first follow-up (87% response rate), 6 (22%) discontinued the trial secondary to intolerable side effects; therefore, the pain analgesic effects of gabapentin in these patients could not be determined. Of the remaining 21 patients, 14 (67%) reported a favorable response (i.e., a 2 or greater point reduction on a 0-10 pain-rating scale). The second follow-up interview captured 11 (79% response rate) of the 14 patients reporting a favorable response at the first interview, and 91% (10 of 11 patients) continued to report that gabapentin was an effective analgesic. There was no evidence to suggest dosing difficulties due to tolerance over the 3-year period. Sedation, dizziness, and forgetfulness were the most common side effects. Conclusions: Gabapentin may be an effective treatment of pain after spinal cord injury among those able to tolerate initial and long-term side effects.