Purpose. Exercise tests represent an important clinical tool to evaluate cardio-respiratory fitness and to predict future adverse cardiovascular events. However, use of such tests in patients with rheumatoid arthritis (RA) is relatively uncommon despite well-established evidence that low exercise capacity and high CVD mortality are features of this disease. Therefore, this study examined the validity and reliability of a sub-maximal step test for use in RA patients. Methods. Thirty patients (24 females) (mean ± SD age 53 ± 10 years) performed a sub-maximal step test on two occasions to estimate the criterion measure of cardio-respiratory fitness ([Formula: see text]). A further maximal cycling test provided a direct fitness measurement ([Formula: see text]). Pearson correlation coefficient, intraclass correlation coefficient (ICC), Bland and Altman plots, and 95% limits of agreement (LOA) were used to determine the validity and reliability of the sub-maximal test. Results. Estimated [Formula: see text] correlated well with directly measured [Formula: see text] (r = 0.79, LoA ±5.7 mL·kg(-1) ·min(-1)). Test-retest reproducibility for estimated [Formula: see text] was excellent (ICC = 0.97, LoA ±2.2 mL·kg(-1) ·min(-1)). Conclusion. The sub-maximal step test studied here represents a valid and reproducible method to estimate cardio-respiratory fitness in RA patients. This test may be useful for the assessment and management of CVD risk in a clinical setting.