Objective To construct a risk prediction model of the postoperative delirium for patients with acute type A aortic dissection（ATAAD）,and to verify the prediction effect of the model. Methods A retrospective study was conducted among 245 adult patients undergoing aortic dissection surgery in the cardiac surgery department of a tertiary A hospital in Wenzhou from October 2020 to April 2021. According to the occurrence of postoperative delirium,the patients were divided into a delirium group （n=51） and a non-delirium group （n=194）,and the data of the 2 groups were compared. The risk prediction scoring model was constructed based on the multivariate logistic regression analysis of ATAAD postoperative delirium risk factors. The Hosmer-Lemeshow test was used to determine the model’s goodness of fit of the model,and the receiver operator characteristic curve （ROC） was used to test the model differentiation. The risk prediction model was validated by the prospective study with inclusion of 51 ATAAD patients admitted to the cardiac surgery department from July,2020 to June,2021. Results There were 6 factors included in the risk prediction model of patients with postoperative delirium,namely age over than 55.50 years（1 point）,male（1 point）,preoperative blood lactic acid value more than 1.85 mmol/L（1 point）,hypothermia cycle arrest time more than 36.50 minutes（1 point）,length of stay in CCU（cardiovascular intensive care unit）more than 8.50 days（2 points）,and other postoperative complications（1 point）. The Hosmer-Lemeshow test of the prediction scoring model showed that when P=0.393,the area under the ROC was 0.926. When the patient’s score was ≤4,the incidence of postoperative delirium in patients with ATAAD was≤15.63%;while it was≥5,the incidence was ≥50.00%. The sensitivity,specificity and accuracy of the risk stratification model were 76.47%,93.30% and 89.80%,respectively. The result of model verification shows that when the area under the ROC was 0.896,the Hosmer-Lemeshow test P=0.359. The sensitivity,specificity and accuracy of the risk stratification model were 75.00%,92.31% and 88.24%,respectively. Conclusion This risk prediction model provides an effective prediction effect,and it is simple and convenient to use in clinical application,which could be a reference for medical staff to take preventive treatment and nursing care.