Abstract: Objective: To investigate the risk factors of upper limb dysfunction in postoperative breast cancer patients,construct a risk prediction model of upper limb dysfunction based on the health ecology theory,and provide reference for early identification and prevention of upper limb dysfunction. Methods: A total of 127 patients with postoperative breast cancer were selected by convenience sampling method.Among them,67 patients were included in the case group and 60 patients in the control group according to the Rewe shoulder function score.Binary logistic regression model was used to analyze the influencing factors of postoperative upper limb dysfunction,and receiver operating characteristic curve was established to analyze the predictive efficiency.The "rms" package in R software was used to build the nomogram. Results: There were significant differences in age,education level,pain degree,upper limb edema,upper limb exercise,sleep status,self-efficacy score,kinesophobia score and intimate relationship score among groups with or without upper limb dysfunction (P<0.05 to P<0.01).Logistic regression analysis confirmed that age,upper limb exercise,sleep status,self-efficacy score,kinesophobia score and intimate relationship score were all important risk factors leading to upper limb dysfunction (P<0.01).Through the above meaningful indicators,the nomogram prediction model was constructed,and the model predicted well. Conclusions: Patients with higher age,lower upper limb exercise frequency,worse sleep,lower self-efficacy score,higher kinesophobia score,and lower intimate relationship score after breast cancer surgery are more likely to have postoperative upper limb dysfunction.The model constructed in this study has a good prediction,which can provide reference for risk assessment and intervention of upper limb dysfunction after breast cancer surgery.