BACKGROUND: Six-minute step test (6MST) has been used to assess functional capacity in chronic conditions; however, its reproducibility in the COPD population has not been evaluated. Our study objective was to evaluate 6MST reproducibility. METHODS: The test was performed in a single 20-cm height step, and subjects were instructed to step up and down (cadence-free) for 6 min. Subjects underwent three tests. The first and second were controlled by the same assessor with a 30-min interval. The third test was controlled by a different assessor one week later. For intrarater comparison, the first and second performances of the test were used, and for inter-assessor comparison, the better performance of the first two tests was compared with the third test. RE-SULTS: Excellent intra-rater and inter-rater relative reproducibility was observed (intraclass correlation coefficient > 0.8), and there was no statistical difference (repeated measures of analysis of variance) among the performances of the three tests. Intra-rater error values were acceptable (mean error of 5.7 steps and limits of agreement between ؊7 and 18 steps). Inter-rater error values were not acceptable (mean error of 4.4 steps and limits of agreement between ؊20 and 29 steps. CONCLUSIONS: 6MST proved to be reproducible in the COPD population when performed by the same assessor.