Aging & Mental Health 2000; 4(2): 101±118 Is Alzheimer’s disease preventable? A review of two decades of epidemiological research C. J. GILLEARD Psychology Department, Spring®eld University Hospital,Tooting, London, UK Abstract This paper reviews the epidemiological literature on Alzheimer’s disease. Increasing age emerges as one of the most powerful risk factors for the development of dementia/Alzheimer’s disease across all populations studied. Examination of age-independent risk factors suggests that there are few environmental or clinical factors that seem to be related consistently to increased risk of dementia/Alzheimer’s disease. In contrast,familial/genetic risk factors do seem to in¯u- ence the incidence of dementia/Alzheimer’s disease. Age, ethnicity and gender all have been found to moderate the effects of both genetic and environmentalrisk factors.The overall weight of evidence suggests that the primary influence of both risk and protective factors is to modify rather than determinethe expression of dementia during the course of the adult lifespan. Introduction Dementia is central to many people’s fears of old age. It is feared particularly because it robs people of their independence, their autonomy and their dignity. It is the major determinant of functional dependence in old age. People who suffer from dementia are six times more likely to need help with the activities of daily living than the rest of the older adult population(Aguero-Torres et al., 1998). Care for people rendered functionally dependent by dementia represents a major public health expenditure. The risk of someone suffering from dementia being placed in a nursing home is 20 times that of their age peers (Rockwood et al., 1996). In England and Wales, the cost of nursing and residential home care associated with dementia has been estimated at £1.4 billion (Gray & Fenn, 1994), while the health and social care costs of caring for someone with dementia in the community are 30 times higher than those for non-demented older people (Souetre et al., 1999). For these three powerful reasonsÐreducing the personal indignities that accompany old age, relieving the burden on families and the community associ- ated with functional dependency in old age and removing a signi®cant element in the costs associated with community and institutional forms of careÐthe prevention of dementia is a matter of enormous economic, social and personal signi®cance. The present review considers the evidence base that might support the goal of the primary prevention of Alzhe- imer’s disease. Cognitive failure, Alzheimer’s disease and dementia One of the ®rst issues to consider is whether the target of primary prevention should be general or speci®c. Should the aim be to prevent cognitive decline in old age (general) or prevent the occurrence of particular disease process(es) (speci®c)? Cognitive impairment, dementia and Alzheimer’s disease are terms that re¯ect a common themeÐthe age related loss of mental function. They demonstrate an inclusive, non-re¯exive relationship to each other,such that all instances of Alzheimer’s disease are instances of dementia, and all instances of dementia exemplify later life cognitive impairment but not all forms of later life cognitive impairment are instances of dementia and not all instances of dementia are embodied in Alzheimer’s disease. It is still not clear how many instances of later life cognitive impair- ment arise from factors unrelated to Alzheimer’s disease. This uncertainty arises for several reasons, not least because of the indeterminacy that surrounds the very concept of `Alzheimer’s disease’. Alzheimer’s disease is thought of as a particular neurodegenerative disorder characterized by speci®c pathology in the brain. Characteristic features include senile plaques (extracellular deposits in the cortical tissue containing a core of abnormal û-amyloid protein surrounded by degenerating neurites) and neuro®brillary tangles (bundles of intracellular twisted neuro®lament proteins).These changes, we are often reminded,`are not part of the normal ageing process’ (Cayton, Graham & Warner, 1997, p. 6). Dementia is Correspondence to: Chris J. Gilleard, Psychology Department, Spring®eld University Hospital, London SW17 7DJ, UK. Received for publication 8th April 1999. Accepted 30th August 1999. ISSN 1360±7863 print/ISSN 1364±6915 online/00/020101±18 ½Taylor & Francis Ltd