Depression is a widespread and debilitating illness that affects an individual's ability to carry out their usual activities. It is therefore likely to represent a considerable economic and social burden to society. The prevalence of depression in the elderly has been estimated at around 10%, although it has been suggested that the level of undetected and untreated depression may be especially high in this group. The direct and indirect costs associated with depression in the elderly are therefore likely to be considerable. The application of economic analysis to drug therapy in the treatment of depression has so far been limited. There are no economic studies of the cost effectiveness of antidepressant drug therapy that concentrate exclusively on the treatment of the elderly. Available evidence across all groups suggests that selective serotonin reuptake inhibitors may be more cost effective than tricyclic antidepressants. This is primarily attributable to the lower incidence of adverse effects with the former. Whilst these results may be replicated in the elderly, there are specific characteristics of the elderly population which suggest that it may not be possible to generalise results. These include adverse effects, the costs of detection, the costs of noncompliance, and indirect costs and benefits. Therefore, further research and careful economic evaluation of the treatment of depression in the elderly is necessary to take account of these characteristics.