The popularity of antidepressants in the treatment of insomnia is not supported by a large amount of convincing data, but rather by opinions and beliefs of the prescribing physicians on the advantages of these agents compared with drugs acting on the benzodiazepine receptor or other drugs used for the treatment of insomnia. The existing data do not allow for clear-cut, evidence-based recommendations concerning the use of antidepressants in insomnia. Our conclusions result from a few short-term studies on single agents, clinical experience and inferences from knowledge on the effect of antidepressants in other indications. At present, prescribing antidepressants for short-term treatment of insomnia can be useful if there is some amount of concomitant depressive symptomology or a history of depression, raising the impression that the present insomnia may be a prodromal sign for a new depressive episode. In all other cases, benzodiazepine receptor agonists, especially the nonbenzodiazepines among them (the so-called 'z drugs') should be the drugs of choice. For long-term treatment, antidepressants are among the pharmacological options, in addition to other groups of psychotropics. Off-label use of antidepressants may be considered for chronic insomnia if there is a concomitant depressive symptomalogy (which is not so pronounced that an antidepressant treatment with adequate higher doses would be required) and if there is no specific indication for one of the other groups of psychotropics (e.g. dementia-related nocturnal agitation, in which case an antipsychotic would be preferred, or circadian problems, in which case melatonin or a melatonin agonist would be favoured). If antidepressants are used to treat insomnia, sedating ones should be preferred over activating agents such as serotonin reuptake inhibitors. In general, drugs lacking strong cholinergic activity should be preferred. Drugs blocking serotonin 5-HT2A or 5-HT2C receptors should be preferred over those whose sedative property is caused by histamine receptor blockade only. The dose should be as low as possible (e.g. as an initial dose: doxepin 25 mg, mirtazapine 15 mg, trazodone 50 mg, trimipramine 25 mg). Regarding the lack of substantial data allowing for evidence-based recommendations, we are facing a clear need for well designed, long-term, comparative studies to further define the role of antidepressants versus other agents in the management of insomnia.