Synthesis Analysis of Regression Models with a continuous outcome.pdfVIP

Synthesis Analysis of Regression Models with a Continuous Outcome XH Andrew Zhou azhou@u.washington.edu Department of Biostatistics, University of Washington Biostatistics Unit, Seattle VA Medical Center Outline • Research motivation • A new type of meta-analysis - synthesis analysis • Some extensions Meta-analysis • Meta-analysis is a statistical technique for combining the results of several studies that ask the same or similar research questions. • A typical meta-analysis combines all the research results on one topic into one large study with many participants. • In evidence-based medicine, meta-analysis is used as a quantitative approach to identifying best evidence based medical practices. • It is based on the condition that each individual study reports findings on the same model of the same research question. • Advantages of meta-analysis include an increase in the sample size and the validity of statistical inference. History of meta-analysis • Karl Pearson performed the first meta-analysis in 1904, trying to overcome the problem of reduced statistical power in studies with small sample sizes. • A meta-analysis in medicine was not published until 1955. • Today meta-analysis is widely used in epidemiology and evidence-based medicine. Potentia bias in meta-analysis • Inclusion and exclusion criteria used to select the studies for the meta-analysis. • Methods used to perform the meta-analysis (fixed versus random effects). • Conclusions by the authors regarding the reliability of the results of their meta-analysis. • Broad applicability for the conclusions of a particula