Autor 
Pereira, Tiago V.
Patsopoulos, Nikolaos A. Salanti, Georgia Ioannidis, John P. A. 
Instituição  Universidade de São Paulo (USP) Universidade Federal de São Paulo (UNIFESP) Univ Ioannina Harvard Univ Fdn Res & Technol Hellas Tufts Univ Stanford Univ 
Resumo  We describe how an appropriate interpretation of the Qtest depends on its power to detect a given typical amount of betweenstudy variance (tau(2)) as well as prior beliefs on heterogeneity. We illustrate these concepts in an evaluation of 1011 metaanalyses of clinical trials with >= 4 studies and binary outcomes. These concepts can be seen as an application of the Bayes theorem. Across the 1011 metaanalyses, power to detect typical heterogeneity was low in most situations. Thus, usually a nonsignificant Q test did not change perceptibly prior convictions on heterogeneity. Conversely, significant results for the Q test typically augmented considerably the probability of heterogeneity. the posterior probability of heterogeneity depends on what tau(2) we want to detect. With the same approach, one may also estimate the posterior probability for the presence of heterogeneity that is large enough to annul statistically significant summary effects; that is half the average withinstudy variance of the combined studies; and that is able to change the summary effect estimate of the metaanalysis by 20%. the discussed analyses are exploratory, and may depend heavily on prior assumptions when power for the Qtest is low. Statistical heterogeneity in metaanalyses should be cautiously interpreted considering the power to detect a specific tau(2) and prior assumptions about the presence of heterogeneity. Copyright (C) 2010 John Wiley & Sons, Ltd. 
Palavrachave 
heterogeneity
Q test Bayes theorem power analysis metaanalysis 
Idioma  Inglês 
Financiador  WoodWhelan research fellowship for a shortterm visit at the Department of Hygiene and Epidemiology at the University of Ioannina PENED training grant from the European Union Social Fund Greek Ministry of Development, General Secretariat of Research and Technology 
Data de publicação  20100401 
Publicado em  Research Synthesis Methods. Hoboken: WileyBlackwell, v. 1, n. 2, p. 149161, 2010. 
ISSN  17592879 (Sherpa/Romeo, fator de impacto) 
Publicador  WileyBlackwell 
Extensão  149161 
Fonte 

Direito de acesso  Acesso restrito 
Tipo  Artigo 
Web of Science  WOS:000209380400005 
Endereço permanente  http://repositorio.unifesp.br/handle/11600/32400 
Arquivo  Tamanho  Formato  Visualização 

Não existem arquivos associados a este item. 