Objectives: To explore the influence factors of hypertension and new way to predict disease with longitudinal physical examination data. Methods: 943 examiners were extracted as the research sample from five units in six years and were divided into two groups according to hypertension. Generalized Estimating Equations (GEE) model was used to select risk factors of hypertension, set the two response variables as whether diagnosed with hypertension or not. Results: There are differences between 5 tested units on hypertension. Systolic blood pressure, high density lipoprotein cholesterol, hemoglobin concentration change are related with the occurrence of hypertension. Conclusions: Systolic blood pressure, hemoglobin concentration is a risk factor for hypertension. High density lipoprotein cholesterol is the protection factor in hypertension. The establishment of continuous physical examination data change model can provide help for the forecast of disease.%目的：利用连续体检资料的变化，探讨相关疾病的影响因素，为疾病预测提供新思路。方法：抽取5个单位经过6年体检的943人作为研究样本，以是否患有高血压分为两组。运用重复测量广义估计方程（GEE模型）对选择的样本进行模型构建，将二项反应变量设定为是否诊断为高血压，筛选出高血压的危险因素。结果：5个受检单位之间对高血压的影响存在差异。收缩压、高密度脂蛋白胆固醇、血红蛋白浓度的变化对高血压的发生具有统计学意义。结论：收缩压、血红蛋白浓度是高血压的危险因素，高密度脂蛋白胆固醇是高血压的保护因素；通过建立连续体检资料变化的模型能对疾病的预测提供帮助。
展开▼
