RT Journal Article SR Electronic T1 Survival and cause of death in multiple sclerosis: a prospective population-based study JF Journal of Neurology, Neurosurgery & Psychiatry JO J Neurol Neurosurg Psychiatry FD BMJ Publishing Group Ltd SP 1016 OP 1021 DO 10.1136/jnnp.2007.127332 VO 79 IS 9 A1 C Hirst A1 R Swingler A1 D A S Compston A1 Y Ben-Shlomo A1 N P Robertson YR 2008 UL http://jnnp.bmj.com/content/79/9/1016.abstract AB Background: Detailed studies of mortality in multiple sclerosis (MS) are limited. Studying death certificates in a prospective cohort of patients known to have MS is of value in establishing mortality data and can also provide important information on the accuracy and use of death certificates for epidemiological studies. Methods: A population-based survey performed in South Wales in 1985 identified 441 patients. Cases were flagged with the Office of Population Censuses and Surveys and death certificates collected prospectively for more than 20 years. Results: Median observed survival time was 38.0 years from symptom onset. Mean age at death was 65.3 for women and 65.2 years for men. Mean age at death in patients dying from MS-related causes was 62.5 and 69.3 years (p<0.001) for unrelated deaths. Those dying of MS-related causes had a younger age at disease onset (32.5) compared with those dying of unrelated causes (36.8 years) (p = 0.01). Cause of death was related to MS in 57.9% and unrelated in 42.1% of individuals. In 27% of patients, “MS” was absent from the death certificate. The most common cause of death was respiratory disease (47.5%). The standardised mortality ratio was 2.79 (95% CI 2.44 to 3.18) so that MS patients were almost three times more likely to die prematurely relative to the general population. Conclusions: These results confirm a continuing trend of premature death in patients with MS. Relying on data derived from death certificates will underestimate disease prevalence. Differences were identified between those dying from MS-related causes and those dying from other causes.