〔Abstract〕 Objective To construct nomogram to predict the risk of bone metastasis in patients with non-small cell lung cancer(NSCLC). Methods The clinical data of NSCLC patients diagnosed in the hospital were retrospectively analyzed, including the occurrence of bone metastasis, age, gender, pathological type, smoking status, PS score, TN stage, metastasis of other sites before bone metastasis, carcinoembryonic antigen(CEA) level, alpha fetoprotein(AFP) level, serum calcium(Ca2+), serum phosphorus(P), alkaline phosphatase(ALP) level, which were determined by univariate and multivariate logistic regression analysis. Receiver operating characteristic curve(ROC) and decision curve analysis were used, DCA was used to verify the accuracy and clinical benefit of the model, and nomogram was used to visualize the model. Results Area under the ROC curve(AUC) showed that in the modeling group(n=138) and the validation group(n=92), the AUC value predicted by combined indicators(age, gender, pathological type, CEA, ALP)(modeling group=0.792, validation group=0.629) was higher than that predicted by single indicator. Conclusion The prediction model constructed in this study has good effect and can provide reference for clinical screening of high-risk patients with bone metastasis of NSCLC.