×

Selection models with monotone weight functions in meta analysis. (English) Zbl 1442.62597

Summary: Publication bias, the fact that studies identified for inclusion in a meta analysis do not represent all studies on the topic of interest, is commonly recognized as a threat to the validity of a meta analysis. One way to explicitly model publication bias is via weighted probability distributions. We adopt the non-parametric approach initially introduced by K. B. G. Dear and C. B. Begg [“An approach for assessing publication bias prior to performing a meta-analysis”, Stat. Sci. 7, No. 2, 237–245 (1992; doi:10.1214/ss/1177011363)] but impose that the weight function \(w\) is monotonely non-increasing as a function of the \(p\)-value. Since in meta analysis one typically only has few studies or “observations,” regularization of the estimation problem seems sensible. In addition, virtually all parametric weight functions proposed so far in the literature are in fact decreasing. We discuss how to estimate a decreasing weight function in the above model and illustrate the new methodology on two well-known examples. Some basic properties of the log-likelihood function and computation of a \(p\)-value quantifying the evidence against the null hypothesis of a constant weight function are indicated. In addition, we provide an approximate selection bias adjusted profile likelihood confidence interval for the treatment effect. The corresponding software and the data sets used to illustrate it are provided as the R package selectMeta (Rufibach, 2011).

MSC:

62P10 Applications of statistics to biology and medical sciences; meta analysis
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] Ardia , D. Mullen , K. 2010 DEoptim: differential evolution optimization in R
[2] Carpenter, copas: An R package for fitting the copas selection model, The R Journal 1 pp 31– (2009)
[3] Copas, What works?: selectivity models and meta-analysis, Journal of the Royal Statistical Society, Series C (Applied Statistics) 162 pp 95– (1999) · Zbl 04549849
[4] Copas, A robust p-value for treatment effect in meta-analysis with publication bias, Statistics in Medicine 27 pp 4267– (2008)
[5] Copas, Reanalysis of epidemiological evidence on lung cancer and passive smoking, British Medical Journal 320 pp 417– (2000)
[6] Copas, A sensitivity analysis for publication bias in systematic reviews, Statistical Methods in Medical Research 10 pp 251– (2001) · Zbl 1121.62593
[7] Dear, An approach for assessing publication bias prior to performing a meta-analysis, Statistical Science 7 pp 237– (1992)
[8] Fan, Discussion of ”On profile likelihood” by Murphy and van der Vaart, Journal of the American Statistical Association 95 pp 468– (2000)
[9] Ghosh, Incorporating monotonicity into the evaluation of a biomarker, Biostatistics 8 pp 402– (2007) · Zbl 1144.62091
[10] Givens, Publication bias in meta-analysis: a Bayesian data-augmentation approach to account for issues exemplified in the passive smoking debate, Statistical Science 12 pp 221– (1997)
[11] Hackshaw, The accumulated evidence on lung cancer and environmental tobacco smoke, British Medical Journal 315 pp 980– (1997)
[12] Hedges, Estimation of effect size under nonrandom sampling: the effects of censoring studies yielding statistically significant mean differences, Journal of Educational Statistics 3 pp 109– (1984)
[13] Hedges, Comment to the article ”Selection models and the file drawer problem”, Statistical Science 3 pp 118– (1988)
[14] Hedges, Modeling publication selection effects in meta-analysis, Statistical Science 7 pp 246– (1992)
[15] Hedges, Statistical Methods for Meta-analysis (1985) · Zbl 0666.62002
[16] Hedges, Publication Bias in Meta-analysis pp 145– (2005)
[17] Hedges, Estimating effect size under publication bias: small sample properties and robustness of a random effects selection model, Journal of Educational and Behavioral Statistics 21 pp 299– (1996)
[18] Iyengar, Selection models and the file drawer problem, Statistical Science 3 pp 109– (1988)
[19] Iyengar, Maximum likelihood estimation for weighted distributions, Statistics and Probability Letters 21 pp 37– (1994) · Zbl 0809.62017
[20] Kelly, Stochastic reduction of loss in estimating normal means by isotonic regression, Annals of Statistics 17 pp 937– (1989) · Zbl 0673.62021
[21] Knuth, Literate programming, Computer Journal 27 pp 97– (1984) · Zbl 0533.68005
[22] Lamport, LaTeX: A Document Preparation System (1994)
[23] Lee, On posterior consistency in selection models, Statistica Sinica 11 pp 827– (2001) · Zbl 0978.62021
[24] Leisch, COMPSTAT 2002 - Proceedings in Computational Statistics pp 575– (2002)
[25] Macaskill, A comparison of methods to detect publication bias in meta-analysis, Statistics in Medicine 20 pp 641– (2001)
[26] Murphy, On profile likelihood, Journal of the American Statistical Association 95 pp 449– (2000) · Zbl 0995.62033
[27] Peng, Caching and distributing statistical analyses in R, Journal of Statistical Software 26 pp 1– (2008)
[28] Preston, Adjusting for publication bias: modelling the selection process, Journal of Evaluation in Clinical Practice 10 pp 313– (2004)
[29] R Development Core Team, R: A Language and Environment for Statistical Computing (2010)
[30] Publication Bias in Meta-analysis. Prevention, Assessment and Adjustments (2005) · Zbl 1231.62209
[31] Rufibach, reporttools: R functions to generate LaTeX tables of descriptive statistics, Journal of Statistical Software, Code Snippets 31 pp 1– (2009)
[32] Rufibach , K. 2011 electmeta: estimation of weight functions in meta analysis · Zbl 1442.62597
[33] Schwarzer , G. 2010 meta: Meta-Analysis with R · Zbl 1333.92002
[34] Silliman, Hierarchical selection models with applications in meta-analysis, Journal of the American Statistical Association 92 pp 926– (1997a) · Zbl 0889.62022
[35] Silliman, Nonparametric classes of weight functions to model publication bias, Biometrika 84 pp 909– (1997b) · Zbl 0892.62014
[36] Storn, Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces, Journal of Global Optimization 11 pp 341– (1997) · Zbl 0888.90135
[37] Sun, Semi-parametric estimates under biased sampling, Statistica Sinica 7 pp 545– (1997) · Zbl 0876.62020
[38] Sutton, Recent developments in meta-analysis, Statistics in Medicine 27 pp 625– (2008)
[39] Sutton, Modelling publication bias in meta-analysis: a review, Statistical Methods in Medical Research 9 pp 421– (2000) · Zbl 1121.62668
[40] Woodroofe, Testing uniformity versus a monotone density, Annals of Statistics 27 pp 338– (1999) · Zbl 0978.62039
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.