中文摘要：

目的：既往对于阿司匹林与氯吡格雷联合治疗预防脑卒中再发效果是否优于阿司匹林单独用药的研究结果不一致,本研究利用北京市城镇职工医疗保险数据库的资料,在大样本人群数据的基础上比较联合用药和单独用药对缺血性脑卒中患者再入院的影响。方法：采用回顾性队列研究的设计方法,从北京市城镇职工医疗保险数据库中提取主诊断为缺血性脑卒中的患者。患者的首条入院记录作为本研究的基线,根据患者的基线用药情况分为阿司匹林单独用药组、阿司匹林和氯吡格雷联合用药组。随访患者用药后是否因为主要结局事件再次入院,主要结局事件包括：（1）缺血性脑卒中复发;（2）脑梗死的出血性转化;（3）心肌梗死;（4）消化道出血。采用KaplanMeier方法比较两组之间的生存情况,并用Log-Rank检验生存曲线的差异。为控制混杂因素对基线的影响,对患者的基线数据采用倾向评分进行1∶1配对,并采用Cox比例风险模型计算风险比（hazard ratio,HR）。结果：从2010年1月至2013年9月纳入研究的患者共计27 695人,其中联合用药组4 047人,单独用药组23 648人。由于患者的基线特征不均衡可比,所以用倾向评分进行1∶1配比,配比后两组各有4 046人。调整了一般人口学特征如年龄、性别、民族及伴随疾病情况和合并用药情况后,两组的生存曲线差异没有统计学意义（P=0.06）,组间的主要结局事件的HR值为0.91（0.82-1.01,P=0.07）,差异没有统计学意义。协变量中性别HR=1.36（1.20-1.55,P〈0.05）,伴随糖尿病HR=1.36（1.20-1.54,P〈0.05）、血脂异常HR=1.13（1.00-1.27,P=0.05）、心脏病HR=1.39（1.22-1.58,P〈0.05）差异有统计学意义,合并使用其他抗血小板药物HR=1.05（0.95-1.17,P〉0.05）不增加再入院风险。结论：联合使用阿司匹林和氯吡格雷预防患者再次入院的效果与单独使用阿司匹林的效果差异

英文摘要：

Objective： To see the influence of different antiplatelet therapies on stroke patients ＇readmission by performing a deep data-mining into Beijing Healthcare Insuring Database,based on a large sample size. Methods： Aretrospective cohort study,was adopted to extract patients primarily diagnosed as ischemic stroke from healthcare database. The first hospital records were considered as the patient＇s baseline in this study,who were divided into MAPT（ aspirin） and DAPT（ aspirin and clopidogrel） according to the patient＇s baseline medications. A follow-up was conducted to see whether the patients would have rehospitalization record because of major result events after medication. The major result events,included：（ 1） recurrence of ischemic stroke;（ 2） hemorrhagic transformation of ischemic stroke;（ 3） myocardial infarction;（ 4） the digestive hemorrhage. The Kaplan-Meier figure was used to compare the survival situations between these two groups,the log-rank test was used to test the difference of the survival curve,and 1 ∶ 1 propensity score matching was calculated from the patients＇ baseline data. Cox proportional hazards model was used to calculate the hazard ratio（ HR）. Results： A total of27 695 patients From January 2010 to September 2013 were included,4 047 with DAPT,and 23 648 with MAPT. Because the baseline characteristics of the patients was disequilibrium,so we used 1 ∶ 1 propensity score matching,after which,the number of the two groups was 4 046 each. Adjusted for the general demographic characteristics such as age,sex,nationality,complication and drug combination,no statistical significance was observed between the survival curves of the two groups（ P = 0. 06）. HR value of major result events between the groups was 0. 91（ 0. 82- 1. 01,P = 0. 07）,which was not statistically significant. The covariate gender HR = 1. 36（ 1. 20- 1. 55,P〈0. 05）,accompanied by diabetes HR =1. 36（ 1. 20- 1. 54,P〈0. 05）,dyslipidemia HR = 1. 13（ 1. 00- 1. 27,P = 1.