摘要
目的建立原发性高血压发病概率预测模型并对模型预测效果进行评价,为防治高血压发病提供依据。方法调查285例原发性高血压患者和260例对照病例,采用logistic回归分析的方法研究影响高血压发病的危险因素,建立发病概率预测模型并利用ROC曲线进行评价。结果影响高血压发病的危险因素包括家族史、BMI、文化程度、饮酒、蔬菜水果摄入、饮食偏咸、吃动物内脏、体育锻炼程度、血压差等,ROC曲线下面积为0.930,模型具有较好的预测价值。结论建立了合理的logistic回归概率预测模型,模型评价效果较好,能较为准确地预测高血压发病概率。
[Objective]To establish predictive model of incidence probability of primary hypertension patients,to evaluate predictive effect of this model,then provide evidence for its prevention and treatment.[Methods]Retrospective investigation was performed on risk factors of285 cases of primary hypertension and 260 cases of control patients,then the predictive model of incidence probability was established,with ROC method evaluating predictive model.[Results]The main risk factors influencing primary hypertension incidence included family history,BMI,educational background,drinking,intake of vegetables and fruits,salty food,eating internal organs of animals,degree of physical exercise and differential blood pressure.The area of ROC curve was 0.930,which indicated that the model had good predictive.[Conclusion]A rational predictive model was established by logistic regression method.It is good for predicting the incidence probability of primary hypertension.
出处
《职业与健康》
CAS
2011年第18期2051-2053,共3页
Occupation and Health
关键词
原发性高血压
危险因素
发病概率
预测模型
Primary hypertension
Risk factors
Incidence probability
Predictive model
