A prospective study was designed to document course and outcome. Two hundred fifty-three patients with 274 spinal injuries were reviewed at the time of injury and discharge from hospital, as well as at 1, 2, and 5 years postinjury. Thirty-eight of these patients were identified who had been misdiagnosed at the initial assessment. Fracture location, cause of injury, neurologic deficit, and delay in diagnosis were all documented: 22.9% of cervical injuries, and 4.9% of the thoracolumbar injuries had a delayed diagnosis ranging from less than 1 day to 36 days. The causes of delayed diagnosis were: 1) failure to take X-rays, 2) fractures missed on X-ray, and 3) failure of patients to seek medical attention. Associated factors such as intoxication of the patient, multiple injuries, level of consciousness, or two levels of spinal injury contributed to the delayed diagnosis of these injuries. Certain "at-risk" populations for missed spinal injuries have been identified. In spite of delays in diagnosis, progression of an established neurologic deficit did not appear to occur in our study. However, the development of secondary deficits was significant in the delayed diagnosis group.