Abstract

Background: Network meta-analysis is developed to compare all available treatments; therefore it enriches evidence for clinical decision-making, offering insights into treatment effectiveness and safety when faced with multiple options. However, the complexity and numerous treatment comparisons in network meta-analysis can challenge healthcare providers and patients. The purpose of this study aimed to introduce a graphic design to present complex rankings of multiple interventions comprehensively. Methods: Our team members developed a “beading plot” to summary probability of achieving the best treatment (P-best) and global metrics including surface under the cumulative ranking curve (SUCRA) and P-score. Implemented via the “rankinma” R package, this tool summarizes rankings across diverse outcomes in network meta-analyses, and the package received an official release on the Comprehensive R Archive Network (CRAN). It includes the `PlotBead()` function for generating beading plots, which represent treatment rankings among various outcomes. Results: Beading plot has been designed based on number line plot, which effectively displays collective metrics for each treatment across various outcomes. Order on the -axis is derived from ranking metrics like P-best, SUCRA, and P-score. Continuous lines represent outcomes, and color-coded beads signify treatments. Conclusion: The beading plot is a valuable graphic that intuitively displays treatment rankings across diverse outcomes, enhancing reader-friendliness and aiding decision-making in complex network evidence scenarios. While empowering clinicians and patients to identify optimal treatments, it should be used cautiously, alongside an assessment of the overall evidence certainty.

Original languageEnglish
Article number235
JournalBMC Medical Research Methodology
Volume24
Issue number1
DOIs
Publication statusPublished - Dec 2024

Keywords

  • Decision-making
  • Network meta-analysis
  • Statistic plot
  • Treatment ranking

ASJC Scopus subject areas

  • Epidemiology
  • Health Informatics

Fingerprint

Dive into the research topics of 'Beading plot: a novel graphics for ranking interventions in network evidence'. Together they form a unique fingerprint.

Cite this