Sensitivity Analysis of Publication Bias in Meta-analysis : A Bayesian Approach

Due to the selection process in academic publication, all meta-analysis of published literature is more or less affected by the so-called publication bias and tends to overestimate the effect of interest. Statistically, publication bias in meta-analysis is a selection bias which results from a non-random sampling from the population of unpublished studies. Several authors proposed methods of modelling publication bias using a selection model approach, which considers a joint modelling of the weight function representing the publication probability of each study and a regression of the outcome of interest. Copas (1999) showed that in this approach some of the model parameters are not estimable and a sensitivity analysis should be conducted. In implementing the Copas’s sensitivity analysis of publication bias, a practical difficulty arises in determining the range of sensitivity parameters appropriately. We propose in this article a Bayesian hierarchical model which extends Copas’s selectivity model and incorporates the experts’ opinions as a prior distribution of sensitivity parameters. We illustrate this approach with an example of the passive smoking and lung cancer meta-analysis.

[1]  J. Copas,et al.  Inference for Non‐random Samples , 1997 .

[2]  M. A. Best.  Bayesian Approaches to Clinical Trials and Health‐Care Evaluation , 2005 .

[3]  Jonathan A C Sterne,et al.  Systematic reviews in health care: Investigating and dealing with publication and other biases in meta-analysis. , 2001, BMJ.

[4]  D. Rubin,et al.  Inference from Iterative Simulation Using Multiple Sequences , 1992 .

[5]  D. Spiegelhalter,et al.  Bayesian Approaches to Clinical Trials and Health-Care Evaluation: Spiegelhalter/Clinical Trials and Health-Care Evaluation , 2004 .

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

[7]  J. Copas,et al.  Reanalysis of epidemiological evidence on lung cancer and passive smoking , 2000, BMJ : British Medical Journal.

[8]  N J Wald,et al.  The accumulated evidence on lung cancer and environmental tobacco smoke , 1997, BMJ.

[9]  D. Rennie,et al.  Publication bias and public health policy on environmental tobacco smoke. , 1994, JAMA.

[10]  J Q Shi,et al.  A sensitivity analysis for publication bias in systematic reviews , 2001, Statistical methods in medical research.

[11]  W. Gilks,et al.  Adaptive Rejection Sampling for Gibbs Sampling , 1992 .

[12]  W. Gilks,et al.  Adaptive Rejection Metropolis Sampling Within Gibbs Sampling , 1995 .

[13]  Law,et al.  Passive smoking and lung cancer: increased risk is not disputed , 2000 .

[14]  J. Copas.  What works?: selectivity models and meta‐analysis , 1999 .

[15]  Larry V. Hedges,et al.  Estimation of effect size under nonrandom sampling , 1984 .

[16]  P. Schmidt,et al.  Limited-Dependent and Qualitative Variables in Econometrics. , 1984 .

[17]  Jonathan A C Sterne,et al.  Investigating and Dealing with Publication and Other Biases , 2008 .

[18]  J. Copas,et al.  Meta-analysis, funnel plots and sensitivity analysis. , 2000, Biostatistics.

[19]  S. Glantz.  Lung cancer and passive smoking. Nothing new was said. , 2000, BMJ.