OBJECTIVES: To measure the association of health-related quality of life (HRQoL) and pre-baseline relapse among multiple sclerosis (MS) patients in a real-world observational study.  METHODS: The Real-World Betaseron Outcomes Study (ROBUST) is a 12-month, US prospective, observational, open-label, single-arm, multi-center outcomes study of interferon beta-1b given every other day for MS.  Patient outcomes, including relapse history, were reported independently by both patients and physicians via a web-based data capture tool.  A total of 226 patients were registered across 52 sites, and 193 completed the baseline survey.  Analysis of variance methods were used to measure the association of SF-12 Health Survey (SF-12) scores and pre-baseline relapse descriptors (relapse status, frequency, severity of typical and worst relapse). SF-12 results were summarized by Physical Component Score (PCS-12) and Mental Component Score (MCS-12). RESULTS: Both physical and mental scores were lower among patients reporting to be in relapse at baseline (N=78) than among those reporting to be relapse-free at baseline (N=115); however, the difference was statistically significant only on the MCS-12 (43.9 vs. 40.5; p=0.045).  Reduction of both PCS-12 and MCS-12 were strongly associated with increasing frequency of patient reported relapses (p=0.001 for both); there was no association between SF-12 and physician reported frequency of relapses.  PCS-12 was also significantly associated with physician-assessed severity of relapse (p=0.002 for typical; p<0.001 for worst relapse, respectively) as well as patient-assessed severity of relapse (p<0.001 for typical; p<0.001 for worst relapse).  Mental scores, however, were not significantly associated with physician- or patient-reported relapse severity. CONCLUSIONS:   As a generic HRQoL tool, the SF-12 was associated with relapses in an MS population.  In patient-reported, as opposed to neurologist-reported, frequency of relapses was associated with a lower HRQoL.  Patient- and neurologist-assessed severity of relapses was linked to a lower physical QoL. These observations suggest that relapses, both in quantity and severity, have a negative impact on HRQoL as measured by the SF-12.