Objective To translate the Knowledge, Attitude and Behavior in Medication Errors Questionnaire (KAB-MEQ) into Chinese and evaluate its reliability and validity. Methods We obtained permission of KAB-MEQ from the original authors. In pursuant to the Brislin′s model, KAB-MEQ was translated, and back-translated, with the translated versions being reviewed (by the original authors), culturally adapted, and pretested to form the Chinese version KAB-MEQ. Then the questionnaire was administered to 328 ICU nurses using convenience sampling method, in a bid to evaluate the reliability and validity of the Chinese version. Results The Chinese version KAB-MEQ had 3 dimensions and 20 items, with a cumulative variance contribution rate of 73.689%. After modification of the model in confirmatory factor analysis, the model fitted well (χ2/df=1.702, RMSEA=0.057, IFI=0.972, TLI=0.966, CFI=0.972). The item-level content validity index (CVI) of the questionnaire ranged from 0.889 to 1.000, and the content validity index at the questionnaire level was 0.978. The Cronbach′s α coefficient of the questionnaire was 0.925, the split-half reliability coefficient was 0.818, and the test-retest reliability coefficient was 0.862. Conclusion The Chinese version KAB-MEQ has good reliability and validity, and can be used as a valid tool to assess ICU nurses′ knowledge, attitude and behavior about intravenous medication errors.