Various melatonin supplementations have been developed to improve health outcomes in various clinical conditions. Thus, we sought to evaluate and summarize the effect of melatonin treatments in clinical settings for health outcomes. We searched PubMed/Medline, Embase, and Cochrane Library from inception to 4 February 2021. We included meta-analyses of randomized controlled trials investigating the melatonin intervention for any health outcome. Based on the different effect sizes of each meta-analysis, we calculated random models' standardized mean differences or risk ratios. We observed robust evidence supported by statistical significance with non-considerable heterogeneity between studies for sleep-related problems, cancer, surgical patients, and pregnant women. Patients with sleep disorder, sleep onset latency (SMD 0.33, 95% CI: 0.10 - 0.56, P < 0.01) were significantly improved whereas no clear evidence was shown with sleep efficiency (1.10, 95% CI: -0.26 to 2.45). The first analgesic requirement time (SMD 5.81, 95% CI: 2.57-9.05, P < 0.001) of surgical patients was distinctly improved. Female patients under artificial reproductive technologies had significant increase in the top-quality embryos (SMD 0.53, 95% CI: 0.27 - 0.79, P < 0.001), but no statistically clear evidence was found in the live birth rate (SMD 1.20, 95% CI: 0.83 - 1.72). Survival at one year (RR 1.90, 95% CI: 1.28 - 2.83, P < 0.005) significantly increased with cancer patients. Research on melatonin interventions to treat clinical symptoms and sleep problems among diverse health conditions was identified and provided considerable evidence. Future well-designed randomized clinical trials of high quality and subgroup quantitative analyses are essential.