Background/Aim: To develop a scoring system to predict bone metastasis after radical resection within 5 years. Patients and Methods: We evaluated the patient records of 1,749 patients, of whom 50 patients developed bone metastasis. Treatment-related factors (age, gender, localization, histology, preoperative carbohydrate antigen 199 level, T-stage, lymph node metastasis (LN) and pulmonary metastasis (PM)) were analyzed. Results: We found three independent risk factors, namely rectal cancer (p=0.038), LN (p=0.006) and metachronous PM (p<0.001). Scoring was conducted by adding zero or one point from each variable and resulted in four groups of 0, 1, 2 or 3 points. Three groups were formed, with 0-1 points vs. 2 points vs. 3 points (1.5% vs. 6.6% and 10.5%, p<0.001). Conclusion: This new score can help clinicians identify patients at risk for continuous monitoring and optimize surveillance to be able to detect and treat bone metastases very early in order to avoid skeletal complications.