extractDirectRes: Extract summary results for a direct meta-analysis

In RichardBirnie/mautils: Utilities to facilitate an automated workflow for meta-analysis

Description

Extract summary results for a direct meta-analysis

Usage

extractDirectRes(metaRes, effect, intervention = "Int", comparator = "Con",
  interventionCode = NA, comparatorCode = NA, backtransf = FALSE)

Arguments

metaRes	An object of class c("metagen", "meta") or c("metabin", "meta"). These objects are lists containing the results of direct meta-analysis. See metagen, metabin for a description of exactly what is contained in the list
effect	a character string describing the effect estimate, e.g. 'Rate Ratio', 'Odds Ratio', 'Hazard Ratio'
intervention	A character string describing the name of the intervention. Defaults to 'Int' if not provided
comparator	A character string describing the name of the comparator. Defaults to 'Con' if not provided
backtransf	A logical indicating whether results should be back transformed. If backtransf=TRUE then log odds ratios (or hazard ratios etc) will be converted to odds ratios on plots and print outs.

Details

This function extracts the results of a meta-analysis from a list-type object produced by the meta package and returns them as a data frame. If there is only one study comparing two treatments then no meta-anlaysis is performed but the results of the primary study are included in the output. In this case the fields Model, Tau.sq, method.tau and I.sq in the output will be blank as these fields have no meaning for a single study. This is useful if you want to use these results to perform simple indirect (Bucher) analyses.

If more than one study is available then both fixed effect and random effects results will be returned

Value

A data frame with the following columns:

• Intervention The name of the intervention

• InterventionCode The ID number of the intervention in the current set of analyses. NA if not provided

• Comparator The name of the comparator

• ComparatorCode The ID number of the comparator in the current set of analyses

• Effect The type of effect measure. Takes the value of the effect argument

• Model The type of model. Fixed effect or Random Effects. Blank if there is only one study

• log.TE The treatment effect on log scale, e.g. log OR

• seTE.log The standard error for the log treatment effect

• log.lower, log.upper The upper and lower 95% confidence intervals for the log treatment effect

• z, p The z-value and corresponding p-value for the test of effect

• level The level for the confidence intervals. Defaults to 0.95 for a 95% confidence interval

• TE, lower, upper The treatment effect with lower and upper confidence intervals backtransformed to a linear scale

• Tau.sq The heterogeneity variance

• I.sq, I.sq.lower, I.sq.upper The heterogeneity statistic I-squared with upper and lower confidence intervals.

See Also

metagen, metabin

RichardBirnie/mautils documentation built on July 12, 2019, 8:56 p.m.