The Vienna Transdanube Aging study "VITA" is a prospective, interdisciplinary cohort-study of all 75-years old inhabitants of the 21. and 22. district of Vienna (n = 1,745), which started in May 2000. The study design is described in this paper for the first time. The main scientific question of the study concerns the prediction of incident dementia in the elderly. The main statistical analysis will compare 8 predictors: episodic memory, verbal fluency, subjective memory complaints, depression, APOE-epsilon4, MAO-B activity in thrombocytes, MRT hippocampal atrophy, and MRT atrophy of the substantia innominata. The whole investigation comprises medical and psychosocial interviews, psychological tests, psychiatric and neurological scales, blood characteristic, genetic factors and cranial magnetic resonance imaging. Various variables will be compared with each other concerning sensitivity and specificity of prediction of cognitive decline. The dependent variable of the intended statistical analysis will be the individual's difference between Mini Mental State Examination scores at the two times of investigation. A high level of participation in geriatric epidemiological studies increases the general applicability of results but recruitment procedures must not ignore the individual's right to privacy and integrity. Using a liberal recruitment procedure as recommended by the local ethics commission the level of participation is between 36.7% and 44.3%.