In 1997, the results from the Third National Acute Spinal Cord Injury Study (NASCIS 3) were published. We undertook the present study to determine the treatment protocols for patients with spinal cord injuries in Colorado and assess whether there were any barriers to the administration of corticosteroids. Cross-sectional. In May 1999, surveys were mailed to every trauma facility medical director and emergency medical system physician advisor in the state. Physicians were asked to provide information about their facilities' or agencies' current practice(s) for administering steroids to patients with spinal cord injuries. They were also asked about their opinion on whether the data on corticosteroid treatment for spinal cord injury support its use. Ninety-eight percent (39 out of 41) of the medical directors who responded and treat patients with spinal cord injuries said that their facilities do administer steroids to those patients. Fourteen percent reported following the NASCIS 3 protocol; 75%, the NASCIS 2 protocol. About half of the medical directors were either uncertain or did not believe that the data regarding the corticosteroid treatment for spinal cord injury supported its use. The majority of physician advisors responded that they do not authorize the administration of corticosteroids to patients with spinal cord injuries in the field, primarily because of short transport times. Our study demonstrated relatively poor compliance with the NASCIS 3 protocol, but good compliance with the NASCIS 2 protocol. There was skepticism about the efficacy of corticosteroid treatment among some Colorado physicians that treat patients with spinal cord injuries acutely; however, this does not completely explain the findings.