Objective To investigate the risk of metabolic syndrome（ MS） in patients with schizophrenia， develop and validate its prediction model. Methods The data of 362 schizophrenia patients in Traditional Chinese Medicine Hospital of Meishan from January 2022 to April 2024 were selected for retrospective analysis. The patients were divided into 253 cases in model group and 109 cases in validation group using the "caret" package in the R language software in a ratio of 7∶3 to collect factors that may affect MS. In model group， patients were divided into MS group and non-MS group according to whether they were combined with MS or not， and the general information， disease and treatment-related indexes of patients in the two groups were compared. LASSO regression was used to screen potential variables and then multifactor Logistic regression was performed to screen potential predictors， which was used to establish the multifactor Logistic regression model. Model validation was performed after visualization in a nomogram. Results There were 62 patients（ 24.51%） with MS in model group. The differences in age， duration of disease， history of smoking， history of drinking， family history of MS， 21-item Three-Factor Eating Questionnaire（ TFEQ-21）scores， average daily exercise time， type of medication， and years of medication between patients in MS group and non-MS group were statistically significant（ all P＜0.05）. Multifactorial Logistic regression analysis showed that age［ OR=1.027， 95%CI（ 1.007，1.047）］， duration of disease［ OR=1.946， 95%CI（ 1.405，2.695）］， body mass index［ OR=1.066， 95%CI（ 1.004，1.132）］， family history of MS［ OR=2.514， 95%CI（ 1.197， 5.281）］， TFEQ-21 score［ OR=1.218， 95%CI（ 1.130， 1.312）］， and type of medication［ OR=2.802， 95%CI（ 1.126， 6.970）］ were independent predictors of MS in patients with schizophrenia（ P＜0.05）. In model group， the area under the receiver operating characteristic（ ROC） curve was 0.856［95%CI（ 0.803，0.908）］， with a sensitivity of 80.6% and a specificity of 73.3%. In validation group， the area under the ROC curve was 0.854［ 95%CI （0.799，0.908）］， with a sensitivity of 77.4% and a specificity of 78.5%. The model curves of model group and validation group were basically fitted diagonally to the ideal model curves. Clinical effectiveness analysis showed the highest net benefit in predicting MS in patients with schizophrenia using the model when the predictive probability threshold was 0.05 to 0.95. Conclusions MS in patients with schizophrenia is mainly influenced by age， disease duration， and body mass index， and a nomogram based on these factors can be used to predict the risk of MS in patients with schizophrenia.