SummaryBackground and objectives Exercise capacity is impaired at a younger age in CKD patients than in the general population. This study examined the reliability of the Duke Activity Status Index (DASI) questionnaire as a measure of exercise capacity in medically stable adults with stage 3-4 CKD (estimated GFR [eGFR], 15-59 ml/min per 1.73 m 2 ).Design, setting, participants, & measurements Peak oxygen uptake ( _ VO 2 peak), estimated from DASI responses and cardiopulmonary exercise test measurements, was obtained at baseline and 6 months in a derivation sample (n=23) and once in a validation sample (n=20). Bland-Altman analysis and linear mixed models were used to estimate bias, concordance correlation coefficients, and intraclass correlation coefficients as the proportion of the variance due to participant (intertest reliability) and method (test-retest reliability).Results The two samples were homogeneous with respect to age (mean 60614 years), eGFR (35.5615 ml/min per 1.73 m 2 ), male sex (53%), and diabetes (56%). Mean measured _ VO 2 peak was 16.564 ml/kg per min. The DASI questionnaire overestimated _ VO 2 peak by 4.3 ml/kg per min. Intertest reliability was 53% when eGFR was $35 ml/min per 1.73 m 2 (median) and 60% when eGFR was ,35 ml/min per 1.73 m 2 (P,0.01). Test-retest reliability was 81% when eGFR was ,35 ml/min per 1.73 m 2 and 71% when eGFR was $35 ml/min per 1.73 m 2 (P,0.01).Conclusions The DASI questionnaire may be a reliable measure of exercise capacity in CKD patients, especially when eGFR is ,35 ml/min per 1.73 m 2 . Larger prospective studies are needed to determine its prognostic value.