No.35 诊断性试验的Meta分析.ppt

这一节我们通过一个例子，学习整个诊断Meta的过程。内容超多，请先上好厕所。non-infectious origin in critically ill patients and addressThe extracted data were general anddetailedthe heterogeneity of patients and the affect of individualmethodology characteristics,characteristics of the studycovariates.population(adults or children).setting(emergencydepartment,general ward,or intensive care unit).Methodsadmission category(surgical or medical).severity ofSearch strategy and selection criteriaillness(sepsis,severe sepsis,or septic shock).and detailsWe systematically searched Medline(via PubMed).of the procalcitonin assays and cutoffs used.Embase(via OvidSP).ISI Web of Knowledge.theingestigator also recorded the number of trueCochrane Library.Scopus,BioMed Central,and ScienceIpositives and negatives.We contacted theDirect for studies that assessed the accuracy of pro-corresponding authors if further information wascalcitonin for the diagnosis of sepsis.needed.If no response was received after sending aOur medical subject heading terms(for Medline).reminder,the study was excluded.Methodsi部分中首先交代了检索的数据库，包括：Medline、Embase、ISI、Cochrane Library等等。OR proceeding paper OR poster presentation ORbias)of the 14 individual quality-related items asmeta-analysisOR case report).We searched thecovariates in a bivariate random-effects model to testdatabases between inception and Feb 21.2012.We alsothem as possible sources of variation and bias.searched the reference list of each primary studyidentified and of previous systematic reviews.Statistical analysisStudies were included if they assessed the accuracy ofWe tabulated true positives,false negatives,false posi.procalcitonin for differentiation between critically illtives,and true negatives in patients with sepsis andpatients with sepsis from those who have a systemicsystemic inflammatory response syndrome,stratified byinflammatory response syndrome without infection.study.We used the numbers to calculate sensitivity and