hetmeta (version 0.1.0)

hetmeta: Deriving Measures Of Heterogeneity

Description

The "hetmeta" implements the most common measures of heterogenity in meta-analysis.

Usage

hetmeta(model)

Arguments

model

an object of class "rma.uni".

Value

The hetmeta function returns an object of class "hetmeta" as described in hetmetaObject.

Details

The "hetmeta" function calculates estimates for several heterogeneity measures in meta-analysis based on a meta-analytic model of class rma.uni (see metafor-package for more details).

Specifically, the measures derived in the function are the $R_b$, $I^2$, and $R_I$. To complement those measures, the Dersimonian-Laird $Q$ test is presented, together with the coefficient of variation of the pooled estimate $CV_b$, coefficient of variation of the within-study variances, and the typical within-variance terms as defined in the $I^2$ and $R_I$. See references for more details.

References

Crippa A, Khudyakov P, Wang M, Orsini N, Spiegelman D. A new measure of between-studies heterogeneity in meta-analysis. 2016. Stat. Med. In Press.

Takkouche B, Khudyakov P, Costa-Bouzas J, Spiegelman D. Confidence Intervals for Heterogeneity Measures in Meta-analysis. Am. J. Epidemiol. 2013:kwt060.

Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat. Med. 2002; 21(11):1539-1558.

Takkouche B, Cadarso-Suarez C, Spiegelman D. Evaluation of old and new tests of heterogeneity in epidemiologic meta-analysis. Am. J. Epi- demiol. 1999; 150(2):206-215.

See Also

hetmeta-package, metafor

Examples

## load data
dat <- get(data(dat.gibson2002))

## random-effects model analysis of the standardized mean differences
dat <- escalc(measure = "SMD", m1i = m1i, sd1i = sd1i, n1i = n1i, m2i = m2i,
              sd2i = sd2i, n2i = n2i, data = dat)
res <- rma(yi, vi, data = dat, method = "REML")

## heterogeneity measures
hetmeta(res)


## load BCG vaccine data
data(dat.bcg)

## random-effects model of log relative risks
dat <- escalc(measure="RR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg)
res <- rma(yi, vi, data=dat)

## heterogeneity measures
hetmeta(res)