Objective To develop a stress hyperglycemia (SH) risk assessment tool for ICU, and to provide reference for prevention of SH in ICU patients. Methods Risk factors of SH in ICU patients were identified through meta-analysis and Logistic regression model was constructed. Simulated data generated by the Rstudio software were used to draw scatter diagram for identifying cut-off values of risk grading and draw the risk assessment diagram.Finally, the risk assessment diagram was validated using data of ICU patients. Results Seven papers were included and 9 factors as age greater than 60 years old and pulmonary disease were introduced in Logistic regression model. The incidence of 36.1% and 53.4% were selected as the cut-off values of low, medium and high risk, and the risk assessment diagram was drawn. The area under the receiver operating characteristic (ROC) curve was 0.776（95%CI: 0.669-0.883）, and the sensitivity and specificity were 0.706 and 0.846 respectively. Conclusion Based on the results of meta-analysis, the SH risk assessment tool for ICU patients has high predictive accuracy and is of significance for early and rapid identification of high-risk patients.