Sero-epidemiological studies have demonstrated association between multiple sclerosis (MS) and prior Epstein–Barr virus (EBV) infection and it has been hypothesised that peripheral EBV reactivation may drive inflammation. Recent investigation has shown significant differences in median serum levels of EBV nuclear antigen-1 (EBNA-1) IgG between disease subgroups and correlation with disease activity reflected by number of Gd-enhancing lesions and T2 lesion volume. These data have led to hopes that EBNA-1 may be an accessible and effective marker of disease activity. We examined the applicability of these findings in clinical practice, assessing 100 subjects (25 PPMS, 25 RRMS, 25 RRMS in relapse, 25 controls) for serum EBNA-1 using both the Liaison quantitative chemiluminescent assay and Biotest ELISA. We showed no difference in levels between disease subgroups and no correlation with phenotypic characteristics including age at onset (r=−0.17, p=0.16), disease duration (r=0.03, p=0.78), EDSS (r=0.03, p=0.78) or MSSS (r=0.02, p=0.9). There was a poor correlation between the two test methods used (intraclass correlation coefficient 0.67; 0.56–0.78) suggesting potential problems with test interpretation. We have been unable to determine a clinical value for serum EBNA-1 levels in MS or to confirm reported association with disease course and clinical disease activity, although further investigation with larger sample sizes and parallel MR data may be of value to clarify this issue further.