Multiple sclerosis (MS) has important effects on quality of life but it is unknown how cognitive impairment affects the ability to assess or report this. Our objective was to determine whether cognitive impairment negatively affects the construct validity and the reliability of the Multiple Sclerosis Quality of Life Inventory (MSQLI). A neuropsychological test battery and the Multiple Sclerosis Functional Composite (MSFC) were administered to a sample of 136 patients referred for cognitive testing by their neurologists. Age, sex, education and ethnicity-adjusted T scores were calculated for each cognitive variable. Cognitive impairment was defined as any T score less than the fifth percentile. The MSQLI was administered prior to neuropsychological testing and readministered one to four weeks later: Correlations between the MSFC and the SF-36 were determined and compared between the cognitively impaired and unimpaired groups as the main test of construct validity. Test-retest and internal consistency reliability of each of the scales were compared for the impaired and unimpaired groups. Seventy-six (56%) patients were cognitively impaired. Construct validity and internal consistency reliability did not differ between the cognitively impaired and unimpaired groups. Test retest reliability was lower for the bladder and vision scales in the impaired group, but remained acceptable for the bladder scale (r > 0.7). Cognitive impairment, a common MS manifestation, does not appear to reduce the reliability or validity of the MSQLI as a patient self-report measure of health status and quality of life.