Introduction: Dementia is a neurodegenerative disease with insidious onset and progressive progression, of which the most common type is Alzheimer's disease (AD). Lithium, a trace element in the body, has neuroprotective properties. However, whether lithium can treat dementia or AD remains a highly controversial topic. Therefore we conducted a meta-analysis. Methods: A systematic literature review was conducted in PubMed, Embase, and Web of Science. Comparison of the effects of lithium on Alzheimer's disease or dementia in terms of use, duration, and dosage, and meta-analysis to test whether lithium therapy is beneficial in ameliorating the onset of dementia or Alzheimer's disease. Sensitivity analyses were performed using a stepwise exclusion method. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of included studies. We determined the relative risk (RR) between patient groups using a random effects model. Results: A total of seven studies were included. The forest plot results showed that taking lithium therapy reduced the risk of Alzheimer's disease (RR 0.59, 95% CI: 0.44-0.78), and is also protective in reducing the risk of dementia (RR 0.66, 95% CI: 0.56-0.77). The duration of lithium therapy was able to affect the dementia incidence (RR 0.70, 95% CI: 0.55-0.88); however, it is unclear how this effect might manifest in AD. It's also uncertain how many prescriptions for lithium treatment lower the chance of dementia development. Conclusion: The duration of treatment and the usage of lithium therapy seem to lower the risk of AD and postpone the onset of dementia.