Background Mild cognitive impairment (MCI) is a cognitive state falling between normal aging and dementia. The relation between alcohol intake and risk of MCI as well as progression to dementia in people with MCI (PDM) remained unclear. Objective To synthesize available evidence and clarify the relation between alcohol intake and risk of MCI as well as PDM. Method We searched electronic databases consisting of PubMed, EMBASE, Cochrane Library, and China Biology Medicine disc (CBM) from inception to October 1, 2019. Prospective studies reporting at least three levels of alcohol exposure were included. Categorical meta-analysis was used for quantitative synthesis of the relation between light, moderate and heavy alcohol intake with risk of MCI and PDM. Restricted cubic spline and fixed-effects dose-response models were used for dose-response analysis. Result Six cohort studies including 4244 individuals were finally included. We observed an unstable linear relation between alcohol intake (drinks/week) and risk of MCI (P linear = 0.0396). It suggested that a one-drink increment per week of alcohol intake was associated with an increased risk of 3.8% for MCI (RR, 1.038; 95% CI 1.002-1.075). Heavy alcohol intake (> 14 drinks/week) was associated with higher risk of PDM (RR = 1.76; 95% CI 1.10-2.82). And we found a nonlinear relation between alcohol intake and risk of PDM. Drinking more than 16 drinks/week (P nonlinear = 0.0038, HR = 1.42; 95% CI 1.00-2.02), or 27.5 g/day (P nonlinear = 0.0047, HR = 1.46; 95% CI 1.00-2.11) would elevate the risk of PDM. Conclusion There was a nonlinear dose-response relation between alcohol intake and risk of PDM. Excessive alcohol intake would elevate the risk of PDM.