Background Effective screening instruments are needed for the detection of alcohol use disorders (AUDs) in the elderly, in view of the significant physical, psychological and social problems associated with this phenomenon. Methods This paper provides details on the different self-report alcohol screening instruments that have been studied in the elderly, describing both the instruments themselves and their effectiveness as screening instruments for AUDs in different elderly populations. Results The vast majority of studies reviewed were carried out in the US, and a high proportion of these were carried out in Veterans Administrations institutions, thus limiting the generalizability of results. The CAGE was the most widely studied screening instrument, followed by the MAST or variations of the MAST, the AUDIT and variations of the AUDIT, and other screening instruments. Sensitivity and specificity of these instruments varied widely, depending on the prevalence of AUDs in the population being studied, the clinical characteristics of the population and the type of AUD being detected. The CAGE performed poorly in psychiatric populations but a newer instrument, the AUDIT-5, has had promising results to date. No studies focussed on elderly people with cognitive impairment, and there is a need for research in this area. Conclusions Ease of use, patient acceptability, sensitivity and specificity must all be considered when selecting a self-report alcohol screening instrument for use in the elderly. Furthermore, the prevalence of AUDs in the population and the clinical characteristics of that population must also be taken into account. Copyright (C) 2004 John Wiley Sons, Ltd.
