Study design: A 6-month prospective design. Objective: To investigate the relationship between ability of obstacle crossing and falls in independent ambulatory participants with spinal cord injury (SCI). Setting: A tertiary rehabilitation center. Methods: Ninety-four participants were evaluated for their SCI characteristics, ability of walking over small obstacles and functional ability relating to dynamic balance control, lower-extremity motor strength (LEMS) and walking ability. Their fall data were then prospectively monitored every month for 6 months. Results: Twenty-four participants failed in obstacle crossing. However, only eight of the thirty-three participants who fell during the follow-up period were unable to clear the obstacle while walking. The LEMS and functional ability of the participants who failed were significantly poorer than those of individuals who passed an obstacle-crossing test (P ⩽ 0.001). For the falls, significant differences between the groups were found only in age and tactile scores. The findings further indicated that failures on obstacle crossing were not significantly associated with falls (P40.05). Conclusion: Ability of obstacle crossing in a closed/controlled environment clearly correlated with intrinsic causes, whereas a fall in an open environment may be associated with not only intrinsic but also extrinsic causes. Therefore, apart from functional ability, rehabilitation professionals may need to consider extrinsic factors around falls in order to manage risk of injury to the patients.