Abstract：Objective Construction Cox proportional hazards model to predict the probability of patients dropping out in MMT, and identification dropout risk factors in time, implementation of individual interventions, improve MMT maintenance treatment rate. Methods Collecting the relevant information of patients in MMT of Futian and Nanshan from 2009 to 2014, and then dividing the samples into training and testing samples, the training sample was used to construct the Cox model, and the testing sample was used to evaluate the reliability and validity of the model. RESULT From the constructed Cox model, we selected 8 variables to predict the patients dropping out, and the model’s sensitivity is 82.24%, specificity is 80.76% and the consistent rate is 81.76%, indicating that the model predictions and actual dropout situation have higher consistency. CONCLUSION The Cox proportional hazards model can be used to predict the patient’s dropout in MMT.
牛姬飞, 杨长贵, 康春香, 严燚, 司徒潮满, 舒彬, 姜世强. 基于Cox模型的美沙酮维持治疗患者脱失预测研究[J]. 实用预防医学, 2016, 23(5): 529-531.
Niu Ji-fei, Yang Changgui, Kang Chunxiang, Yanyi, Situ Chaoman, Shu Bin, Jiang shiqiang. Research on Patients dropping out prediction based on Cox Model in Community-based Methadone Maintenance Treatment. , 2016, 23(5): 529-531.
2016, Vol. 23 
