Abstract
The seemingly contradictory literature on the use of melatonin in insomnia is best understood by considering issues of time of administration (day or night), dose range (physiological or pharmacological), subject selection (normals or insomniacs), choice of dependent variable (self-report or electroencephalogram), and comprehensiveness of the reported variables (sleep onset or sleep maintenance). Available data on nighttime administration to normals and insomniacs are reviewed. It is concluded that there is not yet a convincing body of evidence, using generally accepted measures, that melatonin administration improves sleep in insomniacs with noncircadian sleep disturbance.
