基于数据挖掘对单纯性肥胖中医临床用药规律研究.pdfVIP

基于数据挖掘对单纯性肥胖中医临床用药规律研究

姚婷;赵鋆

【摘要】目的收集有关单纯性肥胖的中医临床治疗研究,将其处方和用药拆解为各

个独立事物体,并寻找各个事物的出现频次、事物间的联系和规律.筛选出最佳关联

规则,指导今后的临床试验研究.方法(1)对维普、万方、知网和中国生物医药数据

库建库至2015年12月收录的有关中医药治疗单纯性肥胖的临床试验进行文献

检索;(2)整理文献,将文献中出现的处方和用药描述进行整理;(3)采用Excel和Spss

Moldeler数据挖掘软件,通过Apriori计算方法,求得各关联规则的支持度和置信度

并排序.结果共收集文献总数1070篇,最后纳入文献共83篇,收集处方109首,药

物156味.药对分析组最佳关联规则为茯苓→陈皮(S=27.52%C=90.00%).药组分

析组最佳关联规则是茯苓→陈皮、白术(S=18.35%C=100.00%).组方分析组最佳

关联规则是陈皮→半夏、白术、茯苓(S=14.68%C=87.50%).症状与药物之间最佳

关联规则是形体肥胖→陈皮、白术、茯苓(S=23.53%C=91.67%).结论结合药物

与药物、症状与药物的关联规律,认为以利水渗湿药、消食药、泻下药治疗单纯性

肥胖病脾虚痰阻证居多.%ObjectiveTheresearchesontheclinicaltreatment

ofsimpleobesityinChineseMedicinewascollectedandseparatedthe

prescription,medicationandsymptomsintoIndependentthings.Thenthe

frequencyofoccurrenceofeachthingandrelationsandrulesbetween

thingswerefoundout.Thebestassociationrulesandguidesforfuture

clinicaltrialswerescreenedout.Methods(1)SearchingoutClinicaltrial

literatureonsimpleobesitytreatedbytraditionalChinesemedicinein

VIP,Wanfangdata,CNKIandChinesebiomedicaldatabasefromwhen

theywereestablishedtoDecember2015.(2)Theprescriptionanddrug

descriptionintheliteratureweresortedout.(3)ExcelandSpssMoldeler

wasusedtoobtainandsortthesupportandconfidenceoftheassociation

rulesthroughAprioricalculationmethod.Results1070articleswere

collected,83articlesmettherequirement.109prescriptionsand156

medicationswerecollected.ThebestAssociationrulesofthecouplet

medicinesanalysisgroupwasPoriacocos→driedtangerineororangepeel

(S=27.52%C=90.00%).ThebestAssociationrulesofthedruggroup

analysisgroupwasPoriacocos→driedtangerineororangepeel,bighead

atractylodesrhizome(S=18.35%C=100.00%).ThebestAssociationrulesof

thecompositionalanalysisofrecipegroupwasdriedtangerineororange

peel→Pinelliaternata,bigheadatractylodesrhizome,Poriacocos(