Zopiclone is a non-benzodiazepine hypnotic which was first reviewed in Drugs in 1986. At that time, zopiclone had shown hypnotic efficacy superior to that of placebo, but had not been extensively compared with benzodiazepine hypnotics in patients with insomnia. A much larger body of clinical data is now available, allowing a more detailed evaluation than was previously possible. Together with results from earlier studies, subsequent clinical trials have shown that zopiclone is generally at least as effective as the benzodiazepines (regardless of duration of action) in the treatment of insomnia, although comparisons between zopiclone and flurazepam have produced inconsistent results. Tolerance to the effects of zopiclone was not seen in short term clinical trials (< or = 4 weeks); data from longer term studies are conflicting and the potential for tolerance during long term zopiclone treatment is therefore unclear. Zopiclone has a relatively low propensity to cause residual clinical effects (such as difficulty in waking or reduced morning concentration). Rebound of insomnia to a level below that at baseline can occur after withdrawal of zopiclone, but, on the basis of data from short term studies, does not appear to be common. Data from prescription-event monitoring suggest that zopiclone does not have a high dependence potential (at least in those who are not regular drug abusers/addicts). Zopiclone is well tolerated in both the elderly and younger patients with insomnia. A bitter aftertaste is usually the most common adverse event, but is relatively infrequent (3.6% in the largest available postmarketing study). Thus, zopiclone is now firmly established as an effective and well tolerated hypnotic agent. Although the available data on rebound insomnia and dependence liability are encouraging, potential differences between zopiclone and the benzodiazepines in these respects may have little clinical relevance in the context of short term intermittent use of hypnotics, as it currently recommended. A low propensity for rebound insomnia and dependence might prove valuable during long term hypnotic therapy (which, although not recommended, is a clinical reality). However, the risk-benefit profile of zopiclone in this context remains unknown. Nevertheless, zopiclone is clearly a suitable alternative to the benzodiazepines for the short term treatment of insomnia.