Welcome to visit Zhongnan Medical Journal Press Series journal website!
Zhongnan Medical Journal Press Series of Journals Center for Evidence-Based and Translational Medicine of Wuhan University

The title
  • The title
  • The keywords
  • The author
Submit
Home Articles Vol 36,2023 No.8 Detail

The misuse of the I2 test in the evaluation of heterogeneity of Meta-analysis

Published on Aug. 30, 2023Total Views: 1125 times Total Downloads: 427 times Download Mobile

Author: Shi-Qi WANG Qing-Qing JIANG Shen HUANG Yu-Lin XIE Shi-Yi CAO

Affiliation: School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China

Keywords: Meta-analysis I2 test I2 statistic Heterogeneity

DOI: 10.12173/j.issn.1004-4337.202304137

Reference: Wang SQ, Jiang QQ, Huang S, Xie YL, Cao SY. The misuse of the I2 test in the evaluation of heterogeneity of Meta-analysis[J]. Journal of Mathematical Medicine, 2023, 36(8): 561-564. DOI: 10.12173/j.issn.1004-4337.202304137[Article in Chinese]

  • Abstract
  • Full-text
  • References
Abstract

The heterogeneity of Meta-analysis refers to the differences among the individual studies included in the Meta-analysis. Selecting appropriate methods to identify and measure heterogeneity is an important step to evaluate the reliability of Meta-analysis results. The I-squared (I2) test is the commonly used heterogeneity test of Meta-analysis, but according to the definition of I2 statistic, I2 test cannot truly show the heterogeneity, and the current evaluation methods of heterogeneity of Meta-analysis are seriously misused. This study aimed to systematically review the commonly used heterogeneity evaluation methods of Meta-analysis, introduce the misuse of I2 statistic in the test of heterogeneity of Meta-analysis through two examples, and propose statistical indicators that can truly show the heterogeneity of Meta-analysis, in order to provide reference for improving the reliability and normativity of Meta-analysis researches.

Full-text
Please download the PDF version to read the full text: download
References

1.陈维, 赵守盈, 罗杰, 等. 元分析中三种统计异质性估计方法的比较[J]. 西南师范大学学报(自然科学版), 2015, 40(4): 112-116. [Chen W, Zhao SY, Luo J, et al. Comparison of three estimators of statistical heterogeneity in meta-analysis[J]. Journal of Southwest China Normal University(Natural Science Edition), 2015, 40(4): 112-116.] DOI: 10.13718/j.cnki.xsxb.2015.04.022.

2.Borenstein M. In a meta-analysis, the I-squared statistic does not tell us how much the effect size varies[J]. J Clin Epidemiol, 2022(152): 281-284. DOI: 10.1016/j.jclinepi.2022.10.003.

3.Migliavaca CB, Stein C, Colpani V, et al. Meta-analysis of prevalence: I2 statistic and how to deal with heterogeneity[J]. Res Synth Methods, 2022, 13(3): 363-367. DOI: 10.1002/jrsm.1547.

4.王若琦, 秦超英. Meta分析中异质性检验方法的改进[J].科学技术与工程, 2012, 12(10): 2256-2259. [Wang RQ, Qin CY. The improvement of testing methods for heterogeneity in meta-analysis[J]. Science Technology and Engineering, 2012, 12(10): 2256-2259.] DOI: 10.3969/j.issn.1671-1815.2012.10.002.

5.Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis[J]. Stat Med, 2002, 21(11): 1539-1558. DOI: 10.1002/sim.1186.

6.Sorita A, Ahmed A, Starr SR, et al. Off-hour presentation and outcomes in patients with acute ischemic stroke: a systematic review and meta-analysis[J]. Eur J Intern Med, 2014, 25(4): 394-400. DOI: 10.1016/j.ejim.2014.03.012.

7.Naghshi S, Aune D, Beyene J, et al. Dietary intake and biomarkers of alpha linolenic acid and risk of all cause, cardiovascular, and cancer mortality: systematic review and dose-response meta-analysis of cohort studies[J]. BMJ, 2021(375): n2213. DOI: 10.1136/bmj.n2213.

8.Borenstein M, Higgins JP, Hedges LV, et al. Basics of meta-analysis: I2 is not an absolute measure of heterogeneity[J]. Res Synth Methods, 2017, 8(1): 5-18. DOI: 10.1002/jrsm.1230.

9.Borenstein M. Research note: in a meta-analysis, the I2 index does not tell us how much the effect size varies across studies[J]. J Physiother, 2020, 66(2): 135-139. DOI: 10.1016/j.jphys.2020.02.011.

Popular papers
Last 6 months