目的应用分类树模型构建缺血性脑卒中发病风险的预测模型,并评价其应用价值。方法采用1:1配比病例对照研究设计,选择深圳市2所综合性医院的309名缺血性脑卒中患者为病例组,同时选择按年龄、性别匹配的健康者作为对照;采用卡方自动交互检测(CHAID)法建立缺血性脑卒中发病风险的预测模型,采用错分概率Risk值、索引图及受试者工作特征曲线(ROC)评价模型的应用价值。结果所建立的分类树模型共包括4层,共19个结点,共筛检出6个解释变量;其中最为重要的预测因素为体育锻炼和高血压病史。模型错分概率Risk值为0.207,利用预测概率绘制的ROC曲线下面积为0.789,与0.5比较,差异有统计学意义(P=0.001),模型拟合的效果较好。结论分类树模型不仅能有效地拟合缺血性脑卒中发病风险的预测模型,还可以有效地筛检变量间的交互作用效应。 Objective To establish a prediction model of risk of ischemic stroke by using classification tree model and evaluate its application value. Methods A 1: 1 case-control study was designed. 309 ischemic stroke patients in two general hospitals in Shenzhen were selected as the case group, and the healthy adults who were matched by age and gender were selected as the control. The interactive prediction (CHAID) method was used to establish the prediction model of the risk of ischemic stroke. The risk value, index chart and receiver operating characteristic curve (ROC) were used to evaluate the value of the model. Results The classification tree model consisted of 4 layers and 19 nodes, of which 6 explanatory variables were selected. The most important predictors were physical exercise and history of hypertension. The risk of model misclassification was 0.207, and the area under the ROC curve using prediction probability was 0.789. Compared with 0.5, the difference was statistically significant (P = 0.001). The model fitting effect was better. Conclusion Classification tree model can not only fit the prediction model of ischemic stroke risk effectively, but also effectively screen the interaction effects between variables.