The current review aims to examine melatonin therapy for both sleep disturbances and cognitive function in dementia. We searched all randomized controlled trials published in Medline, Embase, the Cochrane Library, China National Knowledge Infrastructure, the Cochrane Dementia and Cognitive Improvement Group’s Specialized Register, and Clinical Trials.gov. The grading of recommendations assessment, development and evaluation framework was used to assess the quality of evidence. Seven studies were included (n = 520). Treated participants showed prolonged total sleep time (TST) by 24.36 minutes (P = .02). Sleep efficacy (SE) was marginally improved (P = .07). This effect was stronger under a longer intervention period lasting more than 4 weeks (P = .02). Conversely, cognitive function did not change significantly. Additionally, there was no report of severe adverse events. Given the current studies, we conclude that melatonin therapy may be effective in improving SE and prolonging TST in patients with dementia; however, there is no evidence that this improvement impacts cognitive function.