The definitive diagnosis of Alzheimer's disease remains histopathological. To date many published studies have been limited by being based on clinical diagnosis alone, which has a significant false positive rate. In a prospective, longitudinal, clinicopathological study which aims to develop an antemortem diagnostic test for Alzheimer's disease, we have achieved 96% (250/260) consent and 97% (59/61) completion rates. We here describe some of the factors we believe are responsible for these high rates and which may be relevant to other studies requiring autopsy to verify clinical diagnoses.