经倾向指数匹配后的肝癌患者疗效评价  

Effect evaluation of hepatocellular carcinoma patients after propensity score matching

作　　者：王素珍[1] 孟维静[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,261053

出　　处：《郑州大学学报（医学版）》2012年第6期769-772,共4页Journal of Zhengzhou University(Medical Sciences)

基　　金：国家自然科学基金资助项目81141112;山东省自然基金资助项目ZR2009CM117

摘　　要：目的:将肝癌患者分为介入治疗(TACE)和介入加放疗(TCR)2组,用倾向指数匹配法均衡组间的协变量,评价2种治疗方式的效果。方法:采用logistic回归模型计算肝癌患者的倾向指数,然后按照倾向指数进行组间卡钳匹配,对匹配后的数据进行生存分析。结果:匹配前2组的生存率差异无统计学意义,Log-Rank检验的χ2=1.792,P=0.181;匹配后差异有统计学意义,Log-Rank检验的χ2=4.026,P=0.045。结论:采用倾向指数匹配法能有效降低混杂偏倚,对肝癌患者的治疗效果做出正确评价。Aim：To evaluate the effects of transcatheter arterial chemoembolization（TACE）combined with or without radiation therapy on two groups of hepatocellular carcinoma（HCC） patients after balancing the covariates by propensity score.Methods：Each patient＇s propensity score from the model was calculated and made caliper matching according to the propensity score.And then make survival analysis for the matched data.Results：There were no significant differences in the survival rate between the 2 groups before matching,with χ2 of 1.792 and P of 0.181 from Log-Rank test.But the differences in the survival rate were statistically significant between the 2 groups after matching,with χ2 of 4.026 and P of 0.045 from Log-Rank test.Conclusion：Propensity score matching method can effectively reduce the confounding bias of non-randomized clinical observational data,helping us evaluate the curative effect of HCC patients correctly.

关 键 词：倾向指数 匹配法 肝癌 生存分析 

分 类 号：R181.2[医药卫生—流行病学]