Hotelling T^2检验与多元方差分析在通心络对心梗疗效中的应用  

Application of Hotelling T^2 Test and Multivariate Analysis of Variance in the Efficacy of Tonxinluo on Myocardial Infarction

在线阅读下载全文

作  者:何穗智[1] 吴伟康[2] 杨跃进 吴以岭[4] 谭红梅[2] 何甘霖[5] 刘素芳[6] 欧顺云[6] 邓卓燊[6] 曾百乐[6] 

机构地区:[1]中山大学公共卫生学院医学统计与流行病学系,广州510080 [2]中山大学中西医结合研究所 [3]中国医学科学院阜外医院,北京100037 [4]河北医科大学医药研究院,石家庄050035 [5]广州医学院,510182 [6]中山大学公共卫生学院

出  处:《中国医疗前沿》2010年第14期1-2,9,共3页China Healthcare Innovation

基  金:国家重点基础研究研究发展计划项目(编号:2005CB523305)

摘  要:目的探讨通心络疗效分析的统计学方法,为开发中医新药提供实验研究的科学论据。方法采用Hotelling T2检验、多元方差分析,分析处理通心络对心梗干预前后的测量资料。结果多元方差分析结果显示,通心络(TXL)与骨髓间充质干细胞(MSCs)两个因素对LVEF、ESV、Segments、Thickening、Defectarea 5个指标均有统计学意义(p<0.05),且其交互效应具有统计学意义(p<0.05)。通心络与骨髓间充质干细胞对心梗干预的差异不具有统计学意义。结论 Hotelling T2检验与多元方差分析用于检验通心络疗效是可行的,结果满意。提示研究者可充分利用多元分析方法为医学科研服务。Objective To discuss the statistical methods used on analysing the efficacy of Tongxinluo (TXL), which provided the scientific arguments of experimental study for the development of the new Chinese medicine.Methods Using Hotelling T^2 test and multivariate analysis of variance to analyse and treat the measurement data of TXL interfering myocardial infarction (MI) pre and post.Results Results of multivariate analysis of variance showed that, two factors:TXL and Mesenchymal stem cells (MSCs) effect on 5 indices:LVEF, ESV, Segments, Thickening and Defect area, had statistical significance (p=0.000, p=0.000), moreover their interaction effect also had statistical significance (p=0.000).But the difference of TXL and MSCs interfering the MI had no statistical significance.Conclusion Hotelling T^2 test and multivariate analysis of variance used for testing the efficacy of TXL was feasible, with satisfactory results.It prompted that researchers could take full advantage of multi-analysis method to services medical research.

关 键 词:通心络 HOTELLING T^2检验 多元方差分析 

分 类 号:R195.1[医药卫生—卫生统计学] R54[医药卫生—卫生事业管理]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象