Seriously misleading results using inverse of Freeman-Tukey double arcsine transformation in meta-analysis of single proportions

Guido Schwarzer, Hiam Chemaitelly, Laith Aburaddad, Gerta Rücker

Research output: Contribution to journalArticle

Abstract

Standard generic inverse variance methods for the combination of single proportions are based on transformed proportions using the logit, arcsine, and Freeman-Tukey double arcsine transformations. Generalized linear mixed models are another more elaborate approach. Irrespective of the approach, meta-analysis results are typically back-transformed to the original scale in order to ease interpretation. Whereas the back-transformation of meta-analysis results is straightforward for most transformations, this is not the case for the Freeman-Tukey double arcsine transformation, albeit possible. In this case study with five studies, we demonstrate how seriously misleading the back-transformation of the Freeman-Tukey double arcsine transformation can be. We conclude that this transformation should only be used with special caution for the meta-analysis of single proportions due to potential problems with the back-transformation. Generalized linear mixed models seem to be a promising alternative.

Original languageEnglish
Pages (from-to)476-483
Number of pages8
JournalResearch Synthesis Methods
Volume10
Issue number3
DOIs
Publication statusPublished - 1 Sep 2019

Fingerprint

interpretation

Keywords

  • back-transformation
  • generalized linear mixed model
  • harmonic mean
  • random intercept logistic regression
  • variance stabilization

ASJC Scopus subject areas

  • Education

Cite this

Seriously misleading results using inverse of Freeman-Tukey double arcsine transformation in meta-analysis of single proportions. / Schwarzer, Guido; Chemaitelly, Hiam; Aburaddad, Laith; Rücker, Gerta.

In: Research Synthesis Methods, Vol. 10, No. 3, 01.09.2019, p. 476-483.

Research output: Contribution to journalArticle

@article{e677577a97474c63ae3e4f10c2c9e7c0,
title = "Seriously misleading results using inverse of Freeman-Tukey double arcsine transformation in meta-analysis of single proportions",
abstract = "Standard generic inverse variance methods for the combination of single proportions are based on transformed proportions using the logit, arcsine, and Freeman-Tukey double arcsine transformations. Generalized linear mixed models are another more elaborate approach. Irrespective of the approach, meta-analysis results are typically back-transformed to the original scale in order to ease interpretation. Whereas the back-transformation of meta-analysis results is straightforward for most transformations, this is not the case for the Freeman-Tukey double arcsine transformation, albeit possible. In this case study with five studies, we demonstrate how seriously misleading the back-transformation of the Freeman-Tukey double arcsine transformation can be. We conclude that this transformation should only be used with special caution for the meta-analysis of single proportions due to potential problems with the back-transformation. Generalized linear mixed models seem to be a promising alternative.",
keywords = "back-transformation, generalized linear mixed model, harmonic mean, random intercept logistic regression, variance stabilization",
author = "Guido Schwarzer and Hiam Chemaitelly and Laith Aburaddad and Gerta R{\"u}cker",
year = "2019",
month = "9",
day = "1",
doi = "10.1002/jrsm.1348",
language = "English",
volume = "10",
pages = "476--483",
journal = "Research Synthesis Methods",
issn = "1759-2879",
publisher = "John Wiley and Sons Ltd",
number = "3",

}

TY - JOUR

T1 - Seriously misleading results using inverse of Freeman-Tukey double arcsine transformation in meta-analysis of single proportions

AU - Schwarzer, Guido

AU - Chemaitelly, Hiam

AU - Aburaddad, Laith

AU - Rücker, Gerta

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Standard generic inverse variance methods for the combination of single proportions are based on transformed proportions using the logit, arcsine, and Freeman-Tukey double arcsine transformations. Generalized linear mixed models are another more elaborate approach. Irrespective of the approach, meta-analysis results are typically back-transformed to the original scale in order to ease interpretation. Whereas the back-transformation of meta-analysis results is straightforward for most transformations, this is not the case for the Freeman-Tukey double arcsine transformation, albeit possible. In this case study with five studies, we demonstrate how seriously misleading the back-transformation of the Freeman-Tukey double arcsine transformation can be. We conclude that this transformation should only be used with special caution for the meta-analysis of single proportions due to potential problems with the back-transformation. Generalized linear mixed models seem to be a promising alternative.

AB - Standard generic inverse variance methods for the combination of single proportions are based on transformed proportions using the logit, arcsine, and Freeman-Tukey double arcsine transformations. Generalized linear mixed models are another more elaborate approach. Irrespective of the approach, meta-analysis results are typically back-transformed to the original scale in order to ease interpretation. Whereas the back-transformation of meta-analysis results is straightforward for most transformations, this is not the case for the Freeman-Tukey double arcsine transformation, albeit possible. In this case study with five studies, we demonstrate how seriously misleading the back-transformation of the Freeman-Tukey double arcsine transformation can be. We conclude that this transformation should only be used with special caution for the meta-analysis of single proportions due to potential problems with the back-transformation. Generalized linear mixed models seem to be a promising alternative.

KW - back-transformation

KW - generalized linear mixed model

KW - harmonic mean

KW - random intercept logistic regression

KW - variance stabilization

UR - http://www.scopus.com/inward/record.url?scp=85071898566&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071898566&partnerID=8YFLogxK

U2 - 10.1002/jrsm.1348

DO - 10.1002/jrsm.1348

M3 - Article

VL - 10

SP - 476

EP - 483

JO - Research Synthesis Methods

JF - Research Synthesis Methods

SN - 1759-2879

IS - 3

ER -