The number of new strokes diagnosed in the communities of Fargo, North Dakota, and Moorhead, Minnesota, was determined by intensive review of medical records in all hospitals, clinics and nursing homes in the area for the period 1965 to 1966. These two adjacent communities have unusual advantages for such a study because their neurological diagnostic facilities are of excellent quality, the population served is stable, and the communities are isolated from other medical facilities by many miles.
Out of a population of 94,000 about one-third of which was over 35 years of age and almost all of which was white, a total of 408 patients were diagnosed as having suffered a new cerebral thrombosis (154 cases), hemorrhage (66 cases), embolus (15 cases), subarachnoid hemorrhage (30 cases) or unspecified CVA (143 cases). The average annual incidence of these strokes was respectively 83, 35, 8, 16 and 76 per 100,000 population. These rates are similar to the rates reported elsewhere in a United States, a Japanese and an English community in which similar case-finding methods were used. The similarity in incidence of various types of stroke suggested by these data is in contrast to the appreciable differences in frequency suggested by mortality data. The possibility exists, therefore, that geographical differences in stroke rates based on mortality may represent an artifact. More valid measures of frequency and population selectivity of cerebrovascular disease may show that this disease, unlike coronary vascular disease, is actually similar in populations differing widely in environmental and racial characteristics.
Besides the average annual frequency of strokes, the age-specific frequencies were calculated and revealed a marked increase of all types with age over 40 years except for subarachnoid hemorrhage. The latter showed a bimodal age distribution with peaks in the fifth and after the seventh decade. Seasonal incidence and mortality showed a spring and late fall peak. The sex differences in frequency were very small over a wide age range.
Data were also collected on brain stem vascular accidents, ischemic attacks and cerebral arteriosclerosis. An additional 245 patients were listed under these rubrics. These vascular events tend to be underreported and accuracy of diagnosis may be less. Moreover, comparable data from other communities using similar case-ascertainment techniques are lacking. However, if they were included, the average annual incidence of "strokes" would be 347 per 100,000 population.
Additional community-wide studies with special attention to ischemic attacks, brain stem vascular accidents and patients labeled vaguely as cerebral arteriosclerosis would probably yield a more accurate estimate of the true frequency of cerebrovascular accidents. Ideally, all such patients should be examined neurologically and classified according to standardized criteria if data are desired which can be compared among different communities.
