Analysis of the ‘Epidemic' of Multiple Sclerosis in the Faroe Islands

Abstract

The claim has been made that British troops introduced multiple sclerosis (MS), a transmissible disease, into the Faroe Islands during World War II, causing a three-tier epidemic which resulted in the appearance of 32 cases from 1943 until 1973. Assumptions underlying this hypothesis include the belief that the disease was absent from the Faroe Islands before 1940, the view that ascertainment of cases was complete and that Faroese patients who had either been born in Denmark or had been away from the Islands for 3 or more years before the onset of the disease had to be excluded. All the calculations were based on the presumed date of appearance of the first symptoms of MS. We reject the hypotheses of an epidemic and of the transmissibility of MS for several reasons. The most important one is that the date of onset of illness bears no relationship to the probable date of acquisition, which is widely believed to occur between the ages of 5 and 15 years. The criteria for exclusion of cases are arbitrary and instead of the accepted 32 cases, we believe that 42 cases should be counted in all analyses, including at least 2 with onset of illness before 1940. Only 15 of these 42 were in the age range 5–14 years during the British occupation. We cannot accept the statement that the disease was unknown in the Faroes before 1940 and believe that case ascertainment was incomplete in the Islands, which share close geographic, ethnic, and environmental similarities with other North Sea countries of high MS incidence. The theory of transmission is unconvincing and the characteristics of the putative agent unrealistic. The extremely high incidence of disease, which has statistical significance, is based on a very small number of cases in a very small population, and is of very doubtful biological significance.

Analysis of the ‘Epidemic' of Multiple Sclerosis in the Faroe Islands

“…Acceptance of this case supports the hypothesis that an epidemic occurred, rejection of the case supports the null hypothesis that no epidemic occurred. Similar conclusions are reached using the reported date of onset of the cases in data set B [3] in which the maximal cluster is 17 cases in a series of 42 cases. Again acceptance or rejection of case 41 alters the conclusion about the occurrence of an epidemic, dem onstrating the instability and tenuousness of conclusions based on such small samples.…”

“…The distribution of these cases by the reported dates of onset is shown in ta ble II. For clarity during the remainder of this paper the cases accepted by Kurtzke and Hyllested [1,2] will be considered data set A, and the cases accepted by Poser et al [3] data set B.…”

“…This is predicated on the results of immi gration studies [3] which indicate that the incidence rate of MS in the area where the individual lived between the ages of 5 and 14 is more predictive of the risk of acquiring the disease than the incidence in the area of sub sequent residence. The data were analyzed for each of the 10 years between 5 and 14 for each patient who had lived in the Faroes during this period of time; the results can be ade quately summarized, and more easily repre sented, by selecting age 14 as the representa tive age of disease acquisition.…”

“…First, that case finding was com- Number of clinical onset cases 1941 0 1942 0 1943 1 1944 3 1945 3 1946 3 1947 2 1948 2 1949 2 1950 16 0 1951 0 1952 1 1953 1 1954 1 1955 1 1956 1 1957 1 1958 1 1959 2 1960 10 1 1961 1 1962 0 1963 0 1964 0 1965 1 1966 0 1967 0 1968 1 1969 1 1970 5 1 1971 0 1972 0 1973 1 There are serious impedimenta to accept ing the first six assumptions listed above, as discussed in detail in part I of these two papers [3], In part II, we address the seventh assump tion, specifically the statistical analysis of the demographicdetailsregardingtheMSpatients in the Faroes, paying particular attention to the problems of analysis of small samples.…”

Increasing Incidence of Multiple Sclerosis in South Lower Saxony, Germany

“…In contrast, there are small islands where the incidence has decreased [Orkney, Shetlands, 14, 15] and still others where an epidemic with increasing and de creasing incidence figures was observed [Faroer Islands, 16]. The last reported findings were critically analyzed recently in this jour nal [28,29] and it is in fact very difficult to be sure that the methodological problems have been solved adequately. It seems even more difficult to find a common exogenous factor which could be responsible for these observed increases and decreases in inci dence figures.…”