Enhancing alcohol media literacy (AML) is a critical strategy to attenuate negative effects of alcohol-related media messages. We aimed to systematically review previous studies on AML in children and adolescents. Through database and hand searches, we identified four non-experimental and 13 experimental studies. From the 17 studies, we extracted data on theoretical framework, AML measures, intervention contents and main findings. Our review revealed the following. First, nine and four studies used a single theory and at least two theories, respectively. The most frequently used theory was the Message Interpretation Process model. Second, 13 studies assessed AML using participant self-reporting or evaluators' ratings of participant performance, but no study reported validities of AML measures. Eight studies assessed tobacco or general media literacy as well as AML. Third, all experimental studies used three to five media literacy core concepts proposed by the Center for Media Literacy in AML interventions. Finally, all non-experimental studies found strong cross-sectional associations between higher AML and improved drinking outcomes; all experimental studies found strong intervention effects, such as an increase in AML or improvement in drinking outcomes. In addition, intervention effects differed across participants' gender and drinking experiences. Given our review, health professionals should implement strategies to enhance children and adolescents' AML. Future research should investigate (i) causal inferences in non-experimental studies, (ii) long-term intervention effects in experimental studies and (iii) differences in intervention effects across participants' characteristics, while using solid theoretical frameworks and multidimensional AML concepts. Lay Summary Currently, alcohol consumption and great exposure to alcohol-related media messages in youth are significant public health concerns. After systematically reviewing non-experimental studies on alcohol media literacy (AML) in children and adolescents, we found that high AML levels were associated with improved drinking-related characteristics: a decrease in positive alcohol expectancies, intention to drink and risk of alcohol consumption. Additionally, we found that AML education targeting children and adolescents was effective in enhancing AML and improving the aforementioned drinking-related characteristics. Given our findings, we strongly recommend providing AML education to children and adolescents in order to attenuate negative effects of alcohol-related media messages on drinking-related characteristics.