Introduction: Acute lymphoblastic leukemia (ALL) is the most common malignant neoplasm in children. Although the prognosis is good, complications caused by chemotherapy are still a large challenge for clinicians. Among them, pancreatitis is one of the most serious complications after the application of asparaginase, which is an important part of treatment. The occurrence of pancreatitis may affect chemotherapy tolerance and prognosis of ALL. We intend to establish a predictive model for the risk of pancreatitis in children with ALL during chemotherapy based on clinical data. Method: Collect clinical data of ALL patients under CCLG-ALL, screen variables that may be related to the occurrence of pancreatitis through lasso regression, divide the total patient into a training set and a validation set in a ratio of 8:2, build a prediction model in the training set, and evaluate prediction ability through area under the receiver operating characteristic curve (AUC) and calibration curve. External validation is done in the validation set. Results: A total of 321 patients with ALL were included in this study, including 58 patients with pancreatitis and 263 patients in the control group. Risk factors related to pancreatitis were elevated total bilirubin, direct bilirubin and induction chemotherapy. C statistic and AUC obtained in the training set of this model was 0.862 and 0.86, and the AUC of this model in the validation set was 0.95. Conclusion: This study constructs a risk prediction model for pancreatitis in children with acute lymphoblastic leukemia receiving chemotherapy, and the results suggest that the prediction ability is good. A nomogram based on this model was developed. Among the risk factors, increased total bilirubin and direct bilirubin may indicate that pancreatitis may be related to biliary obstruction, and induction chemotherapy may indicate that children may have predisposing factors for pancreatitis. Further research may be needed in these two aspects in the future.