基于倾向指数匹配法的肝癌病人疗效评价  

Propensity score matching-based therapeutic effect evaluation in patients with hepatocellular carcinoma

作　　者：王素珍[1] 孟维静[1] 安洪庆[1] 赵晓蒙[1] 吕军城[1] 石福艳[1] 

机构地区：[1]潍坊医学院公共卫生学院卫生统计学教研室,山东潍坊261053

出　　处：《南方医科大学学报》2012年第9期1234-1237,共4页Journal of Southern Medical University

基　　金：国家自然科学基金(81141112);山东省自然科学基金(ZR2009CM117)~~

摘　　要：目的根据临床观测所得数据,采用倾向指数匹配法均衡介入治疗(TACE)和介入联合射频消融术治疗(TACE+RFA)两组之间的混杂因素,评价肝癌病人的疗效。方法以分组变量为因变量,以协变量为自变量建立Logistic回归模型并计算肝癌病人的倾向指数,用倾向指数卡钳匹配法均衡组间的协变量,对匹配后的数据进行生存分析。结果匹配前两组间不均衡的协变量在匹配以后达到了均衡。匹配之前,TACE组和TACE+RFA组的1年生存率分别为52.07%、59.08%,2年生存率分别为32.24%、36.43%,3年生存率分别为16.54%、19.39%;两组中位生存期分别为1.20和1.40年;Logrank检验结果表明两组生存曲线间无统计学差异(P>0.05);匹配之后,TACE组和TACE+RFA组的1年生存率分别为54.39%、62.28%,2年生存率分别为23.15%、40.08%,3年生存率分别为10.20%、18.52%,两组中位生存时间分别为1.10和1.50年;Logrank检验结果表明两组生存曲线之间差别有统计学意义(P=0.0033),说明联合治疗组的生存率高于TACE组。结论采用倾向指数匹配法能有效降低混杂偏倚,对肝癌病人的治疗效果做出正确评价。Objective To evaluate the therapeutic effects of transcatheter arterial chemoembolization （TACE） with or without radiofrequency ablation （RFA） in hepatocellular carcinoma （HCC） patients based on propensity score matching. Methods A logistic regression model was established with the treatment assignment as the dependent variable and the covariates as the independent variables. For each HCC patient, the propensity score was calculated from the model for caliper matching, and a survival analysis of the matched data were carried out. Results The covariates between the groups were balanced after caliper matching based on the propensity scores. Before matching, the one-, two-, and three-year survival rates of TACE and TACE＋ RFA were 52.07% and 59.08%, 32.24% and 36.43%, and 316.54% and 19.39%, with the median survival time of 1.20 and 1.40 years, respectively, showing no significant differences in the overall survival rate between the two groups. After matching, the 1-year, 2-year, and 3-year survival rates of TACE and TACE＋RFA groups were 54.39% and 62.28%, 23.15% and 40.08%, and 10.20% and 18.52%, with the median survival time of 1.10 years and 1.50 years, respectively, showing significant differences in the overall survival between the two groups. The survival rate in TACE＋RFA group was higher than that of TACE only group. Conclusion Propensity score matching can effectively reduce the confounding bias of non-randomized clinical observational data for a more accurate evaluation of the therapeutic effect in HCC patients.

关 键 词：倾向指数 匹配法 肝癌 生存分析 

分 类 号：R735.7[医药卫生—肿瘤]