EUVIMEDEUVIMED entdecken
Metaanalyse

Sensitivity analysis for publication bias in meta-analysis of sparse data based on exact likelihood

Taojun Hu iD, Yi Zhou iD, Satoshi Hattori iD

Biometrics · 01.07.2024 · 80 (3)

DOI: 10.1093/biomtc/ujae092 ↗

Kurz gefasst

Metaanalyse in der Zeitschrift Biometrics, online erschienen am 10.09.2024. Verfasst von Taojun Hu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Statistics, Graduate School of Medicine, Osaka University, Osaka, 565-0871, Japan. Hauptthemen laut MEDLINE: Computer Simulation, Meta-Analysis as Topic, Publication Bias. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 28 Literaturangaben. Gefördert von: Grant-in-Aid for Challenging Exploratory Research, Ministry of Education, Science, Sports and Technology.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Meta-analysis is a powerful tool to synthesize findings from multiple studies. The normal–normal random-effects model is widely used to account for between-study heterogeneity. However, meta-analyses of sparse data, which may arise when the event rate is low for binary or count outcomes, pose a challenge to the normal–normal random-effects model in the accuracy and stability in inference since the normal approximation in the within-study model may not be good. To reduce bias arising from data sparsity, the generalized linear mixed model can be used by replacing the approximate normal within-study model with an exact model. Publication bias is one of the most serious threats in meta-analysis. Several quantitative sensitivity analysis methods for evaluating the potential impacts of selective publication are available for the normal–normal random-effects model. We propose a sensitivity analysis method by extending the likelihood-based sensitivity analysis with the $t$-statistic selection function of Copas to several generalized linear mixed-effects models. Through applications of our proposed method to several real-world meta-analyses and simulation studies, the proposed method was proven to outperform the likelihood-based sensitivity analysis based on the normal–normal model. The proposed method would give useful guidance to address publication bias in the meta-analysis of sparse data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Computer SimulationMeta-Analysis as TopicPublication BiasHumansData Interpretation, StatisticalLikelihood FunctionsLinear ModelsSensitivity and Specificity

Stichwörter der Autor:innen

generalized linear mixed modelmeta-analysis of sparse datapublication biasresearch synthesisselection functionsensitivity analysis

Autor:innen und Institutionen

  • Department of Biomedical Statistics, Graduate School of Medicine, Osaka University, Osaka, 565-0871, Japan.
  • Department of Biostatistics, School of Public Health, Peking University, Beijing, 100191, China.
  • Beijing International Center for Mathematical Research, Peking University, Beijing, 100871, China.
  • Integrated Frontier Research for Medical Science Division, Institute for Open and Transdisciplinary Research Initiatives (OTRI), Osaka University, Osaka, 565-0871, Japan.

ORCID-Profile: Taojun Hu, Yi Zhou, Satoshi Hattori

Förderung

  • Grant-in-Aid for Challenging Exploratory Research
  • Ministry of Education, Science, Sports and Technology

Zitieren

APA

Hu, T., Zhou, Y., & Hattori, S. (2024). Sensitivity analysis for publication bias in meta-analysis of sparse data based on exact likelihood. Biometrics, 80(3). https://doi.org/10.1093/biomtc/ujae092

Vancouver

Hu T, Zhou Y, Hattori S. Sensitivity analysis for publication bias in meta-analysis of sparse data based on exact likelihood. Biometrics. 2024;80(3). doi:10.1093/biomtc/ujae092

BibTeX
@article{Hu2024,
  author = {Hu, Taojun and Zhou, Yi and Hattori, Satoshi},
  title = {Sensitivity analysis for publication bias in meta-analysis of sparse data based on exact likelihood},
  journal = {Biometrics},
  year = {2024},
  volume = {80},
  number = {3},
  doi = {10.1093/biomtc/ujae092}
}
Kostenlos starten →