meta (version 6.5-0)

metabind: Combine and summarize meta-analysis objects

Description

This function can be used to combine meta-analysis objects and is, for example, useful to summarize results of various meta-analysis methods or to generate a forest plot with results of several subgroup analyses.

Usage

metabind(..., name = NULL, pooled = NULL, backtransf = NULL, outclab = NULL)

Value

An object of class c("metabind", "meta") with corresponding generic functions (see meta-object).

Arguments

...

Any number of meta-analysis objects or a single list with meta-analyses.

name

An optional character vector providing descriptive names for the meta-analysis objects.

pooled

A character string or vector indicating whether results of a common effect or random effects model should be considered. Either "common" or "random", can be abbreviated.

backtransf

A logical indicating whether results should be back transformed in printouts and plots. If backtransf=TRUE (default), results for sm="OR" are printed as odds ratios rather than log odds ratios, for example.

outclab

Outcome label for all meta-analyis objects.

Author

Guido Schwarzer

Details

This function can be used to combine any number of meta-analysis objects which is useful, for example, to summarize results of various meta-analysis methods or to generate a forest plot with results of several subgroup analyses (see Examples).

Individual study results are not retained with metabind as the function allows to combine meta-analyses from different data sets (e.g., with randomized or observational studies). This is possible using R function metamerge which can be used to combine results of two meta-analyses of the same dataset.

See Also

metagen, forest.metabind, metamerge

Examples

data(Fleiss1993cont)

# Add some (fictitious) grouping variables:
#
Fleiss1993cont$age <- c(55, 65, 55, 65, 55)
Fleiss1993cont$region <- c("Europe", "Europe", "Asia", "Asia", "Europe")

m1 <- metacont(n.psyc, mean.psyc, sd.psyc, n.cont, mean.cont, sd.cont,
  data = Fleiss1993cont, sm = "SMD")

# Conduct two subgroup analyses
#
mu1 <- update(m1, subgroup = age, subgroup.name = "Age group")
mu2 <- update(m1, subgroup = region, subgroup.name = "Region")

# Combine subgroup meta-analyses and show forest plot with subgroup
# results
#
mb1 <- metabind(mu1, mu2)
mb1
forest(mb1)

# Use various estimation methods for between-study heterogeneity
# variance
#
m1.pm <- update(m1, method.tau = "PM")
m1.dl <- update(m1, method.tau = "DL")
m1.ml <- update(m1, method.tau = "ML")
m1.hs <- update(m1, method.tau = "HS")
m1.sj <- update(m1, method.tau = "SJ")
m1.he <- update(m1, method.tau = "HE")
m1.eb <- update(m1, method.tau = "EB")

# Combine meta-analyses and show results
#
taus <- c("Restricted maximum-likelihood estimator",
  "Paule-Mandel estimator",
  "DerSimonian-Laird estimator",
  "Maximum-likelihood estimator",
  "Hunter-Schmidt estimator",
  "Sidik-Jonkman estimator",
  "Hedges estimator",
  "Empirical Bayes estimator")
#
m1.taus <- metabind(m1, m1.pm, m1.dl, m1.ml, m1.hs, m1.sj, m1.he, m1.eb,
  name = taus, pooled = "random")
m1.taus
forest(m1.taus, print.I2 = FALSE, print.pval.Q = FALSE)