Advanced Statistical Methods to Model and Adjust for Bias in Meta-Analysis

The following methods are implemented to evaluate how sensitive the results of a meta-analysis are to potential bias in meta-analysis and to support Schwarzer et al. (2015) , Chapter 5 'Small-Study Effects in Meta-Analysis': - Copas selection model described in Copas & Shi (2001) ; - limit meta-analysis by Rücker et al. (2011) ; - upper bound for outcome reporting bias by Copas & Jackson (2004) ; - imputation methods for missing binary data by Gamble & Hollis (2005) and Higgins et al. (2008) .


*** metasens, version 0.3-2, 2017-12-06 ***

** Major changes **

  • Version of R package meta must be larger or equal 4.9-0

  • For limit meta-analysis, R_b measure of between-study heterogeneity added (Crippa et al., 2016, DOI:10.1002/sim.6980)

  • P-values can be printed in scientific notation

  • P-values equal to 0 are actually printed as "0" instead of "< 0.0001"

  • Thousands separator can be used in printouts and forest plots for large numbers

** User-visible changes **

  • print.summary.limitmeta():
    • new argument print.Rb to specify if heterogeneity measure should be shown in output

** Internal changes **

  • Code customisations due to changes in R package meta, version 4.9-0

*** metasens, version 0.3-1, 2016-10-15 ***

** User-visible changes **

  • plot.copas():
    • default label on x-axis in two bottom plots should read '... largest se' instead of '... largest sd'

2016-02-16, version 0.3-0

Version of R package meta must be larger or equal 4.0-0

Function copas:

  • new argument 'sign.rsb' (which has been available in function summary.copas since version 0.6-3 of R package copas)
  • use internal chklevel function from R package meta to check significance level

Functions plot.copas, print.copas and summary.copas:

  • consider default value for argument 'sign.rsb' from copas object
  • use internal chklevel function from R package meta to check significance level

Function print.copas:

  • bug fix such that other values for sign.rsb than 0.1 are considered
  • use internal chklevel function from R package meta to check significance level

Function limitmeta:

  • argument 'sm' removed (not necessary)
  • set z- and p-value of adjusted effect equal to NA for metaprop objects

Function print.summary.limitmeta:

  • bug fix such that no error occurs for metaprop objects

Help pages updated accordingly

2014-12-06, version 0.2-0

Argument 'backtransf' available in R package meta during creation of an R object is incorporated in R package metasens.

R function copas, limitmeta, funnel.limitmeta:

  • new argument 'backtransf'

R function orbbound:

  • new argument 'backtransf'
  • new argument 'left' to chose whether selection bias is expected on the left or right side of the funnel plot

R function forest.orbbound, print.copas, print.limitmeta, print.orbbound, print.summary.copas, print.summary.limitmeta:

  • new argument 'backtransf' which replaces argument 'logscale'

Several help pages updated to reflect changes in version 0.3-0

2014-06-24, version 0.1-0

First version of R package metasens published on CRAN (replacement of R package copas).

Changes compared to version 0.8-0 of R package copas.

New functions:

  • limitmeta (limit meta-analysis)
  • print.limitmeta, summary.limitmeta, print.summary.limitmeta (print and summary functions)
  • funnel.limitmeta (funnel plot for limit meta-analysis object)
  • radialregression (regression based on radial plot, used internally)

Functions print.copas and print.summary.copas:

  • new argument 'logscale' (print results for relative effect measures on log scale)

New dataset nsaids

New help pages added and update of existing help pages

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.5-0 by Guido Schwarzer, a month ago

Browse source code at

Authors: Guido Schwarzer [cre, aut] , James R. Carpenter [aut] , Gerta Rücker [aut]

Documentation:   PDF Manual  

Task views: Clinical Trial Design, Monitoring, and Analysis, Meta-Analysis, Missing Data

GPL (>= 2) license

Depends on meta

See at CRAN