In the United States, roughly 2/3 of all hypnotic prescriptions go to chronic users, who have taken hypnotics for an average of 5 years or more. Two large prospective epidemiological studies have shown that reported hypnotic use, especially use 30 times per month, is associated with an excess hazard of death. Indeed, use of hypnotics 30 times per month is associated with a similar mortality hazard to smoking 1-2 packs of cigarettes per day. Moreover, the hypnotic user's wish to improve daytime function is usually unfulfilled. The preponderance of evidence is that hypnotics impair performance, cognition and memory, increase the risk of automobile accidents and falls and promote unfavourable changes in personality. Due to tolerance, the sleep-promoting effects of hypnotics appear to be lost with chronic use. With long-term use, there is little controlled evidence that hypnotics produce benefits of any sort. More study of long-term hypnotic effects by public agencies is needed, but available evidence weighs strongly against long-term prescribing.