中文摘要：

目的 建立高血压患者并发脑卒中的风险评估模型。方法通过Meta分析的方法获得高血压患者并发脑卒中的危险因素及相应的合并风险值，构建Rothman—Keller模型的危险评分表，再通过二项分布函数方法生成一组随机数据集，确定风险评估等级划分界值，并采用实际数据验证。结果共纳入31篇文献，进入Rothman-Keller模型的因素包括收缩压、舒张压、高密度脂蛋白胆固醇、低密度脂蛋白胆固醇、同型半胱氨酸、高尿酸血症、空腹血糖受损／糖尿病、脑卒中家族史、颈动脉斑块、是否规律服药、体育锻炼、吸烟。实际人群发病情况与模型预测结果的ROC曲线下面积及95％CI为0．784（0．718，0．849），P〈0．001。结论本模型的预测效果较好，可以应用于高血压患者并发脑卒中的一级预防，有助于降低高血压患者并发脑卒中的风险。

英文摘要：

Objective To build a risk assessment model of stroke in hypertensive patients. Methods The risk factors and corresponding combined risk values of hypertension patients complicating with stroke were obtained through meta-analysis. The risk score table of Rothman-Keller model was constructed. Then, a group of random data sets was generated through binomial distribution function method, and the dividing value of risk assessment grading was determined. Real data of 208 hypertension patients were used for verification. Results Thirty-one studies were included in the meta-analysis. The factors recruited in the Rothman-Keller model included systolic blood pressure, diastolic blood pressure, high-density lipoprotein cholesterol, low- density lipoprotein cholesterol, homocysteine, hyperuricemia, impaired fasting glucose/diabetes, family history of stroke, carotid plaque, regular drug taking, physical exercise, and smoke. The area under the receiver operating characteristic （ROC） curve of actual data and the forecasted results of the model （95% CI） was 0. 784 and 0. 718 - 0. 849 （P 〈 0. 001 ）. Conclusion The forecasted effects of our model are good, which can be applied to the primary prevention of hypertension patients complicating with stroke. It is conducive to reducing the risks of eomplieated stroke of hypertension patients.