Epidemiology of multiple sclerosis in the counties of Northumberland and Durham

Considerable variations in the scope and the methods of ascertainment employed in different epidemiolo-gical studies, to say nothing of divergent diagnostic criteria, have seriously jeopardized the value ofmany such observations in estimating the true prevalence of multiple sclerosis in various parts of the world. Acheson has pointed out that studies based on mortality statistics (Limburg, 1950; Acheson, 1961) are subject to misreporting and under-reporting which reduce their accuracy and comparability. Studies of hospital records (Chiavacci, Hoff, and Polvan, 1950), augmented on occasion by inter-views of practising physicians and examination of death certificates (Ipsen, 1950), represent further attempts to obtain more accurate data. Investigations of special population groups such as U.S. veterans (Acheson and Bachrach, 1960; Acheson, Bachrach, and Wright, 1960), for which comprehensive medical records are available, have afforded opportunities for more circumscribed studies. However, the personally conducted population survey, drawing on every available source to compile a comprehensive register of patients, has been recognized as a superior method. Essential to the survey is the establishment of firm diagnostic criteria, which not only lend reliability to the investigation but also permit the subsequent employment of data in authentic comparative studies. One such survey in the field of multiple sclerosis was published more than 30 years ago (Allison, 1931) and important contributions on similar lines have been made b