Background: Community concerns about a potential excess of multiple sclerosis (MS) prompted this study. Objective: To determine the period prevalence of MS in a community bordering a closed oil refinery and a control community. Methods: Cases seen by a neurologist during 1998 to 2001 were obtained from area neurologists and hospital discharge data. Population data were obtained from the year 2000 US Census. Patient data were abstracted by a trained abstractor onto a standardized report form. A consulting neurologist reviewed the form and made a final diagnosis using the Poser criteria plus the category of presumed. Age- adjusted prevalence rates and rates of agreement were calculated. Results: The direct age- adjusted period prevalence for both sexes and all races for the entire study area was 113 per 100,000 (95% CI = 93 to 136). For white subjects only, the prevalence was 123 per 100,000 (95% CI = 102 to 147). With use of an indirect method of age adjustment, the number of observed cases in the community bordering the refinery was similar to the number of cases expected (standardized morbidity ratio = 130.8, 95% CI = 62.3 to 199.3), based on rates from the comparison area. The agreement between the treating neurologist (for definite plus probable cases) and the consulting neurologist (for definite plus probable plus presumed cases) was good (κ = 0.5733). Conclusions: The prevalence of multiple sclerosis (MS) for this area was generally consistent with prevalence estimates calculated in previous studies in other areas. No significant excess was seen in the exposed area. Background: Community concerns about a potential excess of multiple sclerosis (MS) prompted this study. Objective: To determine the period prevalence of MS in a community bordering a closed oil refinery and a control community. Methods: Cases seen by a neurologist during 1998 to 2001 were obtained from area neurologists and hospital discharge data. Population data were obtained from the year 2000 US Census. Patient data were abstracted by a trained abstractor onto a standardized report form. A consulting neurologist reviewed the form and made a final diagnosis using the Poser criteria plus the category of presumed. Age-adjusted prevalence rates and rates of agreement were calculated. Results: The direct age-adjusted period prevalence for both sexes and all races for the entire study area was 113 per 100,000 (95% CI = 93 to 136). For white subjects only, the prevalence was 123 per 100,000 (95% CI = 102 to 147). With use of an indirect method of age adjustment, the number of observed c ases in the community bordering the refinery was similar to the number of cases expected (standardized morbidity ratio = 130.8, 95% CI = 62.3 to 199.3), based on rates from the comparison area. The agreement between the treating neurologist (for definite plus probable cases) and the consulting neurologist (for definite plus probable plus presumed cases) was good (κ = 0.5733). Conclusions: The prevalence of multiple sclerosis (MS) for this area was generally consistent with prevalence estimates estimates in previous studies in other areas. No significant excess was seen in the exposed area.