Melatonin, secreted nocturnally by the pineal gland, is an endogenous sleep regulator. Impaired melatonin production and complaints on poor quality of sleep are common among the elderly. Non-restorative sleep (perceived poor quality of sleep) and subsequently poor daytime functioning are increasingly recognized as a leading syndrome in the diagnostic and therapeutic process of insomnia complaints. The effects of 3-weeks prolonged-release melatonin 2 mg (PR-melatonin) versus placebo treatment were assessed in a multi-center randomized placebo-controlled study in 170 primary insomnia outpatients aged > or =55 years. Improvements in quality of sleep (QOS) the night before and morning alertness (BFW) were assessed using the Leeds Sleep Evaluation Questionnaire and changes in sleep quality (QON) reported on five categorical unit scales. Rebound insomnia and withdrawal effects following discontinuation were also evaluated. PR-melatonin significantly improved QOS (-22.5 versus -16.5 mm, P = 0.047), QON (0.89 versus 0.46 units; P = 0.003) and BFW (-15.7 versus -6.8 mm; P = 0.002) compared with placebo. The improvements in QOS and BFW were strongly correlated (Rval = 0.77, P < 0.001) suggesting a beneficial treatment effect on the restorative value of sleep. These results were confirmed in a subgroup of patients with a greater symptom severity. There was no evidence of rebound insomnia or withdrawal effects following treatment discontinuation. The incidence of adverse events was low and most side-effects were judged to be of minor severity. PR-melatonin is the first drug shown to significantly improve quality of sleep and morning alertness in primary insomnia patients aged 55 years and older-suggesting more restorative sleep, and without withdrawal symptoms upon discontinuation.