 |
|
Data Discrepancies in Meta-analyses That Use Standarized Mean Differences
|
|
| Since this article does not have an abstract, we have provided the first 150 words of the full text and any section headings. |
|
|
|
To the Editor: In their study of data extraction errors in meta-analyses that use standardized mean differences (SMDs), Dr Gøtzsche and colleagues1 included our own meta-analysis (referred to as Edmonds et al, 2004).2 However, there is a discrepancy between the data extracted by the authors and those in our meta-analysis.
In Figure 4 of their article, our SMD is reported as "–0.74 (95% confidence interval [CI], –0.98 to –0.49)," whereas we reported "–0.77 (95% CI, –1.26 to –0.28)."2 This is important because our result is much closer to the SMD recalculated by Gøtzsche et al of –0.80 (95% CI, –1.36 to –0.24). Gøtzsche had contacted us with specific queries about the data that we used. We believe that the remaining differences may be explained by different choices of most relevant outcomes, primary data, and sets of assumptions rather than errors in data extraction.
Regarding Wearden et al,3 Gøtzsche et al . . . [Full Text of this Article]
Hugh McGuire, MSc
h.mcguire@iop.kcl.ac.uk
King's College Institute of Psychiatry
London, England
Melissa Edmonds, PhD
Lambeth Hospital
London, England
Jonathan Price, DPhil, MRCPsych
Department of Psychiatry
University of Oxford
Oxford, England
RELATED LETTER
Data Discrepancies in Meta-analyses That Use Standarized Mean Differences—Reply
Peter C. Gøtzsche and Britta Tendal
JAMA. 2007;298(19):2262-2263.
EXTRACT
| FULL TEXT
RELATED ARTICLE
Data Extraction Errors in Meta-analyses That Use Standardized Mean Differences
Peter C. Gøtzsche, Asbjørn Hróbjartsson, Katja Maric, and Britta Tendal
JAMA. 2007;298(4):430-437.
ABSTRACT
| FULL TEXT
|
