The purpose of this study was to establish the test-retest reliability of a modified 20 m shuttle walk/run test of aerobic capacity for adults with traumatic brain injuries (TBI). A convenience sample of 18 TBI patients (16 males, two females) between 19 and 58 years of age, was tested using an externally paced, progressive, maximal shuttle walk/run, on two separate occasions, within a 1-week period. The test involving walking or running a 20 m shuttle course while maintaining the pace determined by signals from a prerecorded audiotape. The initial slow walking pace (2.4 km/h) was increased gradually, each minute, until the patient could not continue. Statistical analyses revealed excellent reliability for the number of levels completed (ICC = 0.976), total walk/run test time (S) (ICC = 0.983) and maximal heart rate attained during the final level (bpm) (ICC = 0.964). Although small but significant increases were noted for the number of levels completed during the second test, the modified 20 m shuttle walk/run seems to be a reliable field test which may be useful for assessing the aerobic capacity of brain-injured adults.