Bayesian random effects meta‐analysis of trials with binary outcomes: methods for the absolute risk difference and relative risk scales

When conducting a meta‐analysis of clinical trials with binary outcomes, a normal approximation for the summary treatment effect measure in each trial is inappropriate in the common situation where some of the trials in the meta‐analysis are small, or the observed risks are close to 0 or 1. This problem can be avoided by making direct use of the binomial distribution within trials. A fully Bayesian method has already been developed for random effects meta‐analysis on the log‐odds scale using the BUGS implementation of Gibbs sampling. In this paper we demonstrate how this method can be extended to perform analyses on both the absolute and relative risk scales. Within each approach we exemplify how trial‐level covariates, including underlying risk, can be considered. Data from 46 trials of the effect of single‐dose ibuprofen on post‐operative pain are analysed and the results contrasted with those derived from classical and Bayesian summary statistic methods. The clinical interpretation of the odds ratio scale is not straightforward. The advantages and flexibility of a fully Bayesian approach to meta‐analysis of binary outcome data, considered on an absolute risk or relative risk scale, are now available. Copyright © 2002 John Wiley & Sons, Ltd.

[1]  D G Altman,et al.  Confidence intervals for the number needed to treat , 1998, BMJ.

[2]  A Whitehead,et al.  Borrowing strength from external trials in a meta-analysis. , 1996, Statistics in medicine.

[3]  K A L'Abbé,et al.  Meta-analysis in clinical research. , 1987, Annals of internal medicine.

[4]  A. Raftery,et al.  How Many Iterations in the Gibbs Sampler , 1991 .

[5]  A. Maynard.  Evidence-based medicine: an incomplete method for informing treatment choices , 1997, The Lancet.

[6]  J B Carlin,et al.  Tutorial in biostatistics. Meta-analysis: formulating, evaluating, combining, and reporting by S-L. T. Normand, Statistics in Medicine, 18, 321-359 (1999) , 2000, Statistics in medicine.

[7]  M. Daniels.  A prior for the variance in hierarchical models , 1999 .

[8]  S D Walter,et al.  Variation in baseline risk as an explanation of heterogeneity in meta-analysis. , 1997, Statistics in medicine.

[9]  N. Laird,et al.  Meta-analysis in clinical trials. , 1986, Controlled clinical trials.

[10]  H. V. van Houwelingen,et al.  Investigating underlying risk as a source of heterogeneity in meta-analysis. , 1999, Statistics in medicine.

[11]  J. Hutton,et al.  Number needed to treat: properties and problems , 2000 .

[12]  M. McIntosh,et al.  The population risk as an explanatory variable in research synthesis of clinical trials. , 1996, Statistics in medicine.

[13]  Douglas G Altman,et al.  The relation between treatment benefit and underlying risk in meta-analysis , 1996, BMJ.

[14]  J. Fleiss.  Review papers : The statistical basis of meta-analysis , 1993 .

[15]  J. Carlin.  Meta-analysis for 2 x 2 tables: a Bayesian approach. , 1992, Statistics in medicine.

[16]  C H Schmid,et al.  An empirical study of the effect of the control rate as a predictor of treatment efficacy in meta-analysis of clinical trials. , 1998, Statistics in medicine.

[17]  Walter R. Gilks,et al.  A Language and Program for Complex Bayesian Modelling , 1994 .

[18]  S G Thompson,et al.  Investigating underlying risk as a source of heterogeneity in meta-analysis. , 1997, Statistics in medicine.

[19]  Christopher H Schmid,et al.  Summing up evidence: one answer is not always enough , 1998, The Lancet.

[20]  H Kragt,et al.  Importance of trends in the interpretation of an overall odds ratio in the meta-analysis of clinical trials. , 1992, Statistics in medicine.

[21]  H C Van Houwelingen,et al.  A bivariate approach to meta-analysis. , 1993, Statistics in medicine.

[22]  David J. Spiegelhalter,et al.  Bayesian Approaches to Randomized Trials , 1994, Bayesian Biostatistics.

[23]  S G Thompson,et al.  Systematic Review: Why sources of heterogeneity in meta-analysis should be investigated , 1994, BMJ.

[24]  A Whitehead,et al.  A general parametric approach to the meta-analysis of randomized clinical trials. , 1991, Statistics in medicine.

[25]  D. Sackett,et al.  The number needed to treat: a clinically useful measure of treatment effect , 1995, BMJ.

[26]  D J Spiegelhalter,et al.  Bayesian approaches to random-effects meta-analysis: a comparative study. , 1995, Statistics in medicine.

[27]  B J Biggerstaff,et al.  Incorporating variability in estimates of heterogeneity in the random effects model in meta-analysis. , 1997, Statistics in medicine.

[28]  F. Song,et al.  Cholesterol lowering and mortality: the importance of considering initial level of risk. , 1993, BMJ.

[29]  C S Berkey,et al.  A random-effects regression model for meta-analysis. , 1995, Statistics in medicine.

[30]  R. Kass,et al.  Reference Bayesian Methods for Generalized Linear Mixed Models , 2000 .

[31]  D. Gavaghan,et al.  Size is everything – large amounts of information are needed to overcome random effects in estimating direction and magnitude of treatment effects , 1998, Pain.

[32]  H. Morgenstern.  Uses of ecologic analysis in epidemiologic research. , 1982, American journal of public health.

[33]  Christopher H. Schmid,et al.  Exploring Heterogeneity in Randomized Trials Via Meta-Analysis , 1999 .

[34]  M. Egger,et al.  Who benefits from medical interventions? , 1994, BMJ.

[35]  S Greenland,et al.  Quantitative methods in the review of epidemiologic literature. , 1987, Epidemiologic reviews.

[36]  I Olkin,et al.  Heterogeneity and statistical significance in meta-analysis: an empirical study of 125 meta-analyses. , 2000, Statistics in medicine.