Comprehensive Meta-Analysis

Data entry

• Enter data in 100+ formats.
• Enter data using a spreadsheet interface.
• Enter data for each study in its own format.  For example, enter data for one study as number of events and sample size, for another study as odds ratio and confidence interval, and so on.
• Include multiple study designs in the same analysis.  Enter data from studies that used independent groups, paired/matched designs, or crossover trials.  The program will compute the standard errors for each, and allow you to merge the different designs in the same analysis. 
• Include data from prospective and retrospective trials in the same analysis.
• Mix and match continuous, binary, and correlational data within the same analysis.  The program will convert among odds ratios, standardized mean differences, and correlations.

Click here for a list of all supported formats.

Data structure

• Independent subgroups within studies
• Multiple outcomes within studies
• Multiple time-points within outcomes
• Multiple comparisons (Placebo vs A and Placebo vs B) within studies

Computational options

• Computational models for one group:  Fixed effect, random effects
• Computational models for Anova:  Fixed effect, mixed models, fully random effects
• Computational models for meta-regression:  Fixed effect, method of moments, unrestricted maximum likelihood, maximum likelihood, z-distribution or Knapp-Hartung
• Inverse variance, Mantel Haenszel, Peto
• Includes options to use the same computational algorithms as other programs such as RevMan or Stata. 

Data import

• Import data from other programs, including STATA, Revman, Excel, and SPSS.

Reports

• The program will create a two-page text document that reports all the statistics in a format appropriate for publication. Optionally, the program will annotate the report with five pages of text that explain the meaning of each statistic
• The text can be exported to Word or other programs

Prediction Intervals

• The program will compute the prediction interval, which tells us how widely the effect size varies across studies. This is a critically important item of information, and any many publications recommend that this be reported as part of a meta-analysis
• The prediction interval can be added to the forest plot
• The program will create a plot that displays the entire distribution of true effects. This plot can be exported to Word or PowerPoint

Special purpose analyses

• Cumulative analysis.  Runs the analysis k times, adding an additional study with each iteration, to show how the treatment effect changed over time.
• Sensitivity analysis.  Runs the analysis k times, each time removing one study, to show that study’s impact on the combined effect.

High-resolution plots

• Create forest plots with one click.
• Specify the symbols for studies, subgroups, and combined effects.  Make the symbol a constant size or proportional to study weight.
• Modify colors globally or for each element of the plot.  Set templates for formats and color schemes.
• Export directly to Word™ or PowerPoint™ or save as file.

Publication bias

• Funnel plots using precision
• Funnel plots using standard error
• Fail-Safe N
• Orwin’s Fail-Safe N
• Trim and Fill
• Rank correlation test
• Regression test
• Program produces statistics, plots, and a detailed text report for each of the publication bias procedures.

Effect size measures

• For studies that look at treatment effects or other effect sizes the program will compute the  odds ratio, risk ratio, risk difference, standardized mean difference (d), bias-corrected standardized mean difference (Hedges’s g), raw mean difference, correlation, hazard ratio, rate ratio, and more.
• For studies that look at events or means for one group the program will compute the risk, the rate, or the mean.
• For other measures (e.g. regression coefficients) the program will work with generic indices.  This is available both for raw values and for log values.