Interest in vascular causes for cognitive impairment is increasing, in recognition that such causes are common, and possibly preventable. This has led to attempts to better define vascular dementia and its natural history. Several sets of criteria for the diagnosis of vascular dementia have been proposed. We provide a brief overview of the background to the initiation of a Canadian consensus conference, established by the Consortium of Canadian Centres for Clinical Cognitive Research (C5R) and report the conclusions reached at that conference. To date, no one set of criteria is demonstrably superior to another; we have therefore not endorsed any of the competing sets, nor have we recommended our own. Instead we suggest that empiric studies are required to establish valid criteria. A diagnostic checklist, which combines existing criteria and additional data, is attached for clinicians wishing to participate in such studies.