INTRODUCTION

Studies into the prevalence of MS in Switzerland by Bing and collaborators (Bing and Reese 1926) rank among the pioneering works in epidemiology. Since then, two main types of epidemiological method have been used in the study of MS. First, numerous studies attempt to determine prevalence and incidence rates of the disease in different geographical regions as precisely as possible in order to find a pattern of disease distribution which would elucidate its etiological factors. Second, by comparing people affected by MS with healthy individuals in the same population, specific factors become apparent by which MS patients may be distinguished from nonaffected people. Factors are sought by these methods which cause or trigger manifestations of the disease. Such factors may either be related to genetic disposition or be environmental in nature. Theoretically, epidemiological studies should be able to answer the questions raised. However, they are burdened with methodological difficulties because the epidemiologist – in contrast to the experimenter in a laboratory – is not able to determine the experimental conditions under which the studies are carried out. It is therefore of the utmost importance that a rigorous methodology is respected in such studies (Kurtzke 1983b; Nelson et al. 1986; Martyn 1991; Beer and Kesselring 1994).

Precise criteria must first be determined by which the reliability of the clinical diagnosis can be verified. Because no single pathognomic test is available, the probability of the diagnosis must be checked against lists of criteria.