OBJECTIVE: A potential issue with the Walking Impairment Questionnaire (WIQ) is that it is relatively complex. We estimated the number of errors made by patients when self-completing the WIQ, and assessed the benefit of correcting missing, duplicate or paradoxical (i.e., reported lower difficulty for a higher-intensity task) answers.DESIGN: Prospective non-interventional study. MATERIALS: All consecutive new patients with claudication over a 3-month period. METHODS: The WIQ was self-completed before patients performed a constant-load treadmill walking test (maximised to 750 m). MAIN OUTCOME MEASURE: We analysed the coefficient of determination of the linear relationship between overall WIQ score (mean of the available subscales when at least two subscales are available) and treadmill maximal walking distance (MWD), before and after correction of errors. RESULTS: We studied 73 patients. Thirty-seven questionnaires had to be corrected for one or more errors. The coefficient of determination between the overall WIQ score and MWD was R(2) = 0.391 (n = 56) and R(2) = 0.426 (n = 73) before and after correction, respectively. CONCLUSION: Supervision of self-completed WIQs detects errors in almost half of the questionnaires, resulting in a missing overall WIQ score in 23% of cases among uncorrected questionnaires. The overall WIQ score correlates only moderately with MWD, even after correction.