The MRC Cognitive Function and Ageing Study (CFAS) is a prospective longitudinal study of a population-based cohort of elderly people in six UK sites, evaluated using psychometric instruments and questionnaires to elucidate physical and mental health. Data from the core study includes prevalence and incidence rates for dementia and longitudinal measures of cognitive decline together with data on genetic risk factors for dementia. A neuropathology study runs in collaboration with the core study based on premortem counselling of individual respondents or carers. Analysis of pathological data from the first 209 accumulated brain donations showed that both Alzheimer-type pathologies (ATP) and vascular pathologies (including congophilic amyloid angiopathy (CAA)) were common in both demented and non-demented respondents. Although many cases fulfil conventional diagnostic criteria for the pathological diagnosis of Alzheimer disease, the data differ from those published from conventional studies of hospital or memory clinic cohorts. In particular, there are individuals whose total burden of pathology is inappropriately high or low compared with their clinical dementia status, even when all pathologies are considered in a multivariable model of dementia risk factors (25% of respondents misdiagnosed from pathology findings). Vascular pathology is so common that few dementia cases lack a mixed component of both ATP and vascular lesions (pure AD cases, 21%). More recently, the study has examined white matter pathology in this cohort as a potential manifestation of small-vessel disease (SVD) in the ageing brain. Using an MRI strategy to image formalin-fixed brain slices, the study shows that white matter lesions (WMLs) are common (94% overall frequency) and are an independent risk factor for dementia using multivariable analysis.