Mastering meta-analysis in R: a practical review and recommendations

Mastering meta-analysis in R: a practical review and recommendations

Herney A. García-Perdomo 1, 2 , Daniel A. Nieva-Posso 2

1 Division of Urology/Urooncology, Department of Surgery, Cali, Colombia; 2 UROGIV Research Group. School of Medicine, Universidad del Valle, Cali, Colombia

*Correspondence: Herney A. García-Perdomo. Email: herney.garcia@correounivalle.edu.co

Abstract

Meta‐analysis is a powerful statistical method that synthesizes results from multiple independent studies to generate an overall quantitative estimate of effect sizes. With the growing demand for reproducible and transparent research, R has become a preferred tool for conducting meta‐analyses. This manuscript reviews the fundamental principles of meta‐analysis and demonstrates its practical implementation in R using several packages. We describe how to compute effect sizes, choose appropriate models, assess heterogeneity, and diagnose publication bias. In addition, we explore alternative metaanalytic approaches – including network meta‐analysis, cumulative meta‐analysis, individual participant data meta‐analysis, Bayesian meta‐analysis, and multivariate meta‐analysis – and provide an overview of the R packages that support these methods. The manuscript presents examples, tables, and figures alongside recent references to guide researchers in applying meta‐analytic techniques effectively.

Keywords: Meta‐analysis. R. Publication bias. Network meta‐analysis. Bayesian meta‐analysis.

Contents

The full content will be available shortly. Thank you for your patience!

⇄ Versión en Español
DOI: 10.24875/RUC.25000060
Add to Mendeley
    Descargar PDF
    English versión
    DOI: 10.24875/RUC.25000060

    Editorial times

    The editorial process consists of 6 stages:

    1. Reception of the manuscript (indeterminate, depending on whether the author meets the requirements): its objective is to verify that the manuscript complies with the specifications of these instructions for authors and that the submitted documentation is complete.

    2. Initial editorial review (maximum 5 business days): its objective is to corroborate the relevance, timeliness, originality, and scientific contribution of the manuscript, as well as the methodological and statistical soundness of the study. At this time, it will be submitted to an electronic plagiarism detection system. Derived from this, a rejected opinion may be obtained, or it will be sent for review by peer researchers.

    3. Review by peer researchers (maximum 30 business days): the opinion of at least two experts in the area in question will be obtained, who will evaluate the technical and methodological aspects of the investigation.

    4. Editorial review (maximum 7 business days): its objective is to make a decision based on the opinion of peer reviewers. The opinion can be rejected, major changes, minor changes or accepted. In the case of major or minor changes, it will be submitted again for evaluation by the initial peer reviewers.

    5. Final edition (6 weeks): its objective is the technical and linguistic edition (and translation), layout of galleys, DOI assignment, and correction by the author.

    6. Advanced publication: All manuscripts will be published ahead of print on the journal’s website as soon as they complete the editing process, until they are incorporated into a final issue of the journal.