Dementia is a clinical and public health issue of growing importance as life expectancy increases across the planet. Despite advances in both genetic research and clinical management strategies, neither cure nor primary prevention is currently feasible. However, screening for dementia is critical for secondary prevention, i.e., early diagnosis and treatment as well as disability limitation and prevention of complications. Screening is also important for community surveillance and the planning of health and human services. The most appropriate screening approach for a particular clinical or research setting should be selected on the basis of the purpose of screening in that setting. Ideally, all elderly individuals, as well as younger persons with known risk factors, should be routinely screened for dementia. Availability of staff resources, including time and skills, should also be taken into account. Objective cognitive testing appears to be the most logical approach to screening for dementia. However, potential confounding variables and psychometric properties of the instruments should be considered. Ancillary measures such as functional disability scales, self-reported cognitive functioning, and informants'' perceptions may usefully supplement (or supplant) cognitive testing in certain populations. The limitations of screening should be recognized.