Alzheimer's disease: Preliminary study of spatial distribution at birth place

Alzheimer's disease (AD) is a neurodegenerative disorder which is characterized by a progressive loss of memory and the alteration of cognitive functions. At least three chromosomal segments have been associated with early-onset AD in genetic linkage studies. These results argue for a certain degree of heterogeneity in the genetic origin of some forms of AD, although environmental risk factors cannot be ruled out in late-onset AD. In this preliminary study, we analyzed the geographical distribution of the birth places of a sample of 235 AD cases born in a defined region of Quebec (Canada), between 1895 and 1935. We wished to test the hypothesis that risk factors acting at, or around birth place and time play a role in the etiology of AD. The field of study was divided into rural and urban areas. A reference population of live births was used to compute a measure of odds ratio (OR). The OR results showed a statistically significant excess of AD cases in the rural area as compared to the reference population. When stratified for sex, the OR results showed a global excess of female AD cases in both the rural and the urban areas. For men, only the urban area presented a statistically significant deficit. We also analyzed the structures of the genealogical kinships of the rural and urban sub-groups. Although AD cases from the rural sub-group were more closely related to each other than those from the urban one, removal of the kin pairs from the OR analysis seemed to have little effect on the rural/urban distribution of cases. Therefore, the OR results would not appear to be due primarily to a difference in the kinship structures of the two sub-groups. This could mean that some risk factors for AD afflict women more strongly than men, the effect being different depending on the urban or rural origin. However, potential biases such as a higher rate of report for women, differential migration between birth places or a differential mortality ratio between sexes could produce spurious results in the direction of what we have observed in this preliminary study.