Objectives/background: Colorectal cancer (CRC) can be prevented by routine colonoscopy. CRC screening in special populations, e.g. spinal cord injury and disorders, presents unique barriers and, potentially, a higher risk of complications. We were concerned about potentially higher risks of complications and sought to determine the safety of colonoscopy. Methods: Retrospective observational design using medical record review for 311 patients who underwent 368 colonoscopies from two large VA SCI centers from 1997-2008. Patient demographics and peri-procedural characteristics, including indication, bowel prep quality, and pathological findings are presented. Descriptive statistics are presented. Results: The population was predominantly male and Caucasian, and 199 (64%) had high-level injuries (T6 or above). Median age at colonoscopy was 61 years (interquartile range 53-69). Just <1/2 of the colonoscopies were diagnostic, usually for evidence of rectal bleeding. Although a majority of colonoscopies were reported as poorly prepped, the proportion that were adequately prepped increased over time (from 3.7 to 61.3%, P = <0.0001). Of the 146 polyps removed, 101 (69%) were adenomas or carcinomas. Ten subjects had 11 complications, none of which required surgical intervention. Conclusions: Although providing quality colonoscopic care in this population is labor intensive, the data suggests that it appears safe and therapeutically beneficial. The results indicate that the risk of screening is outweighed by the likelihood of finding polyps. Recognition of the benefit of colonoscopy in this population may have improved bowel prep and reporting over time. Spinal cord injury providers should continue to offer screening or diagnostic colonoscopy to their patients when indicated, while being aware of the special challenges that they face.