Contents
How do you interpret a meta-analysis?
To interpret a meta-analysis, the reader needs to understand several concepts, including effect size, heterogeneity, the model used to conduct the meta-analysis, and the forest plot, a graphical representation of the meta-analysis. These concepts are discussed below and summarized in TABLE 1.
What are meta analytic methods?
Meta-analytic methods then permit reviewers to quantitatively appraise and synthesize outcomes across studies to obtain information on statistical significance and relevance. Systematic reviews of basic research data have the potential of producing information-rich databases which allow extensive secondary analysis.
How do you conduct meta synthesis?
The method is described in six steps: define the research question and the inclusion criteria, select the studies, assess their quality, extract and present the formal data, analyze the data, and express the synthesis.
When to use a meta regression in a study?
Meta-regression serves to appropriately combine and contrast multiple subsets of studies (e.g. by study design, or by individual measurement procedures) when a single summary measure does not seem correct or sufficient to capture all the clinical or methodological diversity in those subsets.
Why do we use random effects in meta regression?
Second, it is wise to allow for the residual heterogeneity among intervention effects not modelled by the explanatory variables. This gives rise to the term ‘random-effects meta-regression’, since the extra variability is incorporated in the same way as in a random-effects meta-analysis (Thompson 1999).
What is the p value of meta regression?
The P value of each regression coefficient will indicate whether this difference is statistically significant. Meta-regression may be performed using the ‘metareg’ macro available for the Stata statistical package.
What are true variances in a meta-regression?
All studies share a common τ2, i.e., they come from the same super-population of studies [7]. The observed sampling variances σi2 are the “true” variances within each study [4]. The distributional assumptions about the random effects and error terms are correct.