Basic aspects of meta-analysis. Part 2

Meta-analysis combines the results of several scientific studies to obtain a summary quantitative estimation of effect size in order to compare the results of several studies and to identify patterns among them and possible sources of biases. Since the research teams, patients, research protocol, clinical guidelines and time intervals are often different between original scientific studies all these sources of difference can influence the results of each study, causing statistical heterogeneity. Meta-analyses are the highest level of credibility within evidence-based medicine, as they allow us to take into account the influence of many confounding factors and publication biases on the true effect size. Understanding the possible sources of erroneous conclusions in papers will allow researchers to analyze the results of such studies correctly and properly plan their own experiments. In this article the reader will be introduced to methods for identifying and quantifying hidden heterogeneity, such as subgroup analysis and meta-regression. In addition, the reader will learn how to calculate effect size in studies, estimate publication bias mathematically and graphically, and incorporate this estimate into the overall average effect size estimate in meta-analyses.

1. Suvorov A.Y., Latushkina I.V., Gulyaeva K.А., et al. Basic aspects of meta-analysis. Part 1. Sechenov Medical Journal. 2023; 14(1): 4–14. https://doi.org/10.47093/2218-7332.2023.14.1.4-14

2. Borenstein M., Higgins J.P.T. Meta-Analysis and Subgroups. Prev Sci. 2013 Apr; 14(2): 134–143. https://doi.org/10.1007/s11121013-0377-7. PMID: 23479191

3. Berkey C.S., Hoaglin D.C., Mosteller F., Colditz G.A. A randomeffects regression model for meta-analysis. Stat Med. 1995 Feb 28; 14(4): 395–411. https://doi.org/10.1002/sim.4780140406. PMID: 7746979

4. Hedges L.V. Modeling publication selection effects in metaanalysis. J Phys A Math Theor. 1992; 7(2): 246–255. https://doi.org/10.1088/1751-8113/44/8/085201

5. Dear K.B.G., Begg C.B. An approach for assessing publication bias prior to performing a meta-analysis. Stat Sci. 1992; 7(2): 237–245. https://doi.org/10.1214/ss/1177011363

6. Sutton A.J., Song F., Gilbody S.M., Abrams K.R. Modelling publication bias in meta-analysis: a review. Stat Methods Med Res. 2000 Oct; 9(5): 421–445. https://doi.org/10.1177/096228020000900503. PMID: 11191259

7. Sterne J.A.C., Egger M. Funnel plots for detecting bias in meta-analysis: guidelines on choice of axis. J Clin Epidemiol. 2001 Oct; 54(10): 1046–1055. https://doi.org/10.1016/S08954356(01)00377-8. PMID: 11576817

8. Egger M., Smith G.D., Schneider M., Minder C. Bias in metaanalysis detected by a simple, graphical test. BMJ. 1997 Sep 13; 315(7109): 629–634. https://doi.org/10.1136/bmj.315.7109.629. PMID: 9310563; PMCID: PMC2127453

9. Begg C.B., Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994 Dec; 50(4): 1088–1101. https://doi.org/10.2307/2533446. PMID: 7786990

10. Harbord R.M., Egger M., Sterne J.A.C. A modifi ed test for smallstudy effects in meta-analyses of controlled trials with binary endpoints. Stat Med. 2006 Oct 30; 25(20): 3443–3457. https://doi.org/10.1002/sim.2380. PMID: 16345038

11. Macaskill P., Walter S.D., Irwig L. A comparison of methods to detect publication bias in meta-analysis. Stat Med. 2001 Feb 28; 20(4): 641–654. https://doi.org/10.1002/sim.698. PMID: 11223905.

12. Lin L., Chu H. Quantifying publication bias in meta-analysis. Biometrics. 2018 Sep; 74(3): 785–794. https://doi.org/10.1111/biom.12817. Epub 2017 Nov 15. PMID: 29141096; PMCID: PMC5953768

13. Bakbergenuly I., Hoaglin D.C., Kulinskaya E. Estimation in meta-analyses of mean difference and standardized mean difference. Stat Med. 2020 Jan 30; 39(2): 171–191. https://doi.org/10.1002/sim.8422. Epub 2019 Nov 11. PMID: 31709582; PMCID: PMC6916299

14. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Second Edition. New Jersey: Lawrence Erlbaum Associates; 1988. 579 p. ISBN 0-8058-0283-5

15. Friedrich J.O., Adhikari N.K.J., Beyene J. The ratio of means method as an alternative to mean differences for analyzing continuous outcome variables in meta-analysis: A simulation study. BMC Med Res Methodol. 2008; 8(1): 32. https://doi.org/10.1186/1471-2288-8-32