help for metabias                 (Roger Harbord, Ross Harris, Jonathan Sterne)

Updated tests for bias in meta-analysis

metabias varlist [if exp] [in range] [, egger peters harbord begg graph nofit or rr level(#) z(newvar) v(newvar) graph_options ]

by ... : may be used with metabias; see help by.


metabias performs updated regression tests for funnel-plot asymmetry in meta-analysis. The Harbord test regresses Z/sqrt(V) against sqrt(V), where Z is the efficient score and V is Fisher's information (the variance of Z under the null hypothesis). The Peters test regresses the intervention effect estimate on 1/n with weights dh/n, where n is the total sample size, d is the number experiencing the event and h is the number not experiencing the event. These may be calculated for the log-odds or log-risk ratio, from 2x2 tables of binary outcomes.

The Egger test is also implemented, and performs a linear regression of the intervention effect estimates on their standard errors, weighting by 1/(variance of the intervention effect estimate). The latter is recommended for intervention effects measured as mean differences, but can suffer from false-positive test results when analysing odds ratios due to the mathematical association between the log odds ratio and its standard error. For completeness, the Begg test is also implemented; although this is widely accepted to be redundant as it suffers the same statistical problems as the Egger test but has lower power.

varlist should contain either four or two variables. When four variables are given these are assumed to be cell counts for the 2x2 table in the order cases and non-cases for the experimental group followed by cases and non-cases for the control group, i.e., d1 h1 d0 h0, as in metan. When two variables are specified these are assumed to be the effect estimate and its standard error, i.e., theta se_theta; it is recommended that ratio-based effect estimates are log-transformed as for metan.


egger peters harbord begg requests that the original Egger test, the Peters test, or Harbordís modified test be used. Note that there is no default: one test must be chosen, and only one test.

graph displays a Galbraith plot (the standard normal deviate of intervention effect estimate against its precision) for the original Egger test, or a modified Galbraith plot of Z / sqrt(V) vs. sqrt(V) for Harbordís modified test. Note that there is no corresponding Galbraith plot for the Peters test.

nofit suppresses the fitted regression line and confidence interval around the intercept.

or uses odds ratios as the effect estimate of interest (the default)

rr specifies that risk ratios be used rather than odds ratios. Note that this is not available for the Peters test.

level(#) specifies the confidence level, in percent, for confidence intervals of the coefficients. The default is the user-specified default contained in level (which, in turn, is by default 95%).

graph_options are options allowed by twoway_scatter. In particular, options for specifying marker labels may well be useful; see marker_label_options. legend(off) is another possibility.


. metabias d1 h1 d0 h0, or harbord . metabias tdeath tnodeath cdeath cnodeath, or harbord graph mlabel(trial) . metabias eventint noeventint eventcon noeventcon, or peters . metabias theta se_theta, egger

History and note on dialog box

This version of metabias revises and extends the previous package by Thomas Steichen first released as sbe19 in STB 41 and updated through to sbe19.5. We are grateful for Tom's permission to release this version under the same name. The previous program is included in the present package as metabias6, which unlike the revised version has an accompanying dialog box.


Begg CB, Mazumdar M. 1994. Operating characteristics of a rank correlation test for publication bias. Biometrics 50: 1088-1101.

Egger M, Smith GD, Schneider M, Minder C. 1997. Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629-634.

Harbord RM, Egger M, Sterne JA. 2006. A modified test for small-study effects in meta-analyses of controlled trials with binary endpoints. Statistics in Medicine 25: 3443-3457.

Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. 2006. Comparison of two methods to detect publication bias in meta-analysis. JAMA 295: 676-680.

Also see

On-line: help for metabias6, metan (if installed), metafunnel (if installed), confunnel (if installed)


Roger Harbord, Department of Social Medicine, University of Bristol, UK Ross Harris, Department of Social Medicine, University of Bristol, UK Jonathan Sterne, Department of Social Medicine, University of Bristol, UK