Objective To infer the relationship between bisphenol A （BPA） exposure and adverse health effects in humans， the human epidemiological data of BPA was evaluated by the weight of evidence approach.Methods Based on the weight of evidence approach， the human epidemiological research articles on BPA published between 2013 and 2020 were systematically searched， screened and classified. The selected articles were evaluated following the guidelines of evaluating the relevance and reliability of toxicological data developed by the China National Center for Food Safety Risk Assessment. The hazard assessment was performed based on different endpoints.Results A total of 188 papers were obtained screening from 134 897 records， of which 68 papers were evaluated as “A/B” or “B” in terms of relevance， and the rest were evaluated as “C”. Only the studies with the relevance evaluation grade of “A/B” or “B” were evaluated for reliability， of which 6 studies were evaluated as “high” reliability and 62 studies were evaluated as “medium” reliability. The endpoints of toxicological effects should include metabolic， reproductive and developmental， neurological， immunological and cardiovascular effects. However， the effect indicators across studies lacked consistency due to study population， exposure duration and indicator design.Conclusion Based on the available evidence of human epidemiological data， the relationship between BPA exposure and adverse health effects in humans is not yet clear.