Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Data extraction errors in meta-analyses that use standardized mean differences.

Peter C Gøtzsche1, Asbjørn Hróbjartsson, Katja Maric

  • 1Nordic Cochrane Centre, Rigshospitalet, Copenhagen Ø, Denmark. pcg@cochrane.dk

JAMA
|July 27, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Factors influencing the trustworthiness of non-randomized studies of interventions: a survey of international experts.

BMC medicine·2026
Same author

SPIRIT 2025 statement: updated guideline for protocols of randomised trials.

Lancet (London, England)·2026
Same author

Unpacking the black box - a questionnaire study of how medical journals identify and manage conflicts of interest of editors and peer reviewers.

Journal of clinical epidemiology·2026
Same author

Author adherence and experiences with PRISMA-P 2015: a cross-sectional study.

Journal of clinical epidemiology·2026
Same author

Outcome switching in cohort studies of interventions: meta-epidemiological study.

BMJ (Clinical research ed.)·2026
Same author

Active placebo control interventions in pharmacological trials: a scoping review of the development process and a terminological analysis.

Journal of clinical epidemiology·2026

Many meta-analyses using standardized mean differences (SMDs) contain errors, impacting study reliability. Researchers and readers should exercise caution with these findings due to potential data extraction inaccuracies.

Area of Science:

  • Biostatistics
  • Medical Research Methodology

Background:

  • Meta-analyses synthesizing data from diverse scales require transformation to a uniform scale, such as the standardized mean difference (SMD).
  • The reliability of meta-analyses employing SMDs has not been thoroughly investigated.

Purpose of the Study:

  • To assess the accuracy of standardized mean differences (SMDs) reported in published meta-analyses.
  • To identify the prevalence and nature of errors in SMD calculations within meta-analyses.

Main Methods:

  • Systematic review of meta-analyses published in 2004 that reported results as SMDs.
  • Random selection of two trials from each meta-analysis for independent SMD calculation using Hedges adjusted g.
  • Comparison of independently calculated SMDs with author-reported values, with a discrepancy threshold of 0.1.

Related Experiment Videos

Main Results:

  • Errors in SMD calculations or confidence intervals were found in 37% of meta-analyses for at least one of two selected trials.
  • In 63% of meta-analyses, errors were detected in at least one of the two examined trials.
  • Subsequent in-depth review of 10 meta-analyses with significant discrepancies revealed errors in 70%, including one retraction and two cases where statistical significance changed.

Conclusions:

  • A high proportion of meta-analyses utilizing SMDs contain errors, primarily due to data extraction inaccuracies.
  • These errors can potentially alter or reverse study findings, underscoring the need for critical appraisal.
  • Researchers, reviewers, and policymakers should approach meta-analyses reporting SMDs with significant caution.