Article type
Abstract
Background:
Network meta-analysis (NMA) addresses the challenges of selecting from multiple treatments by offering comprehensive insights into their effectiveness and safety. Endorsed by articles and supported by the World Health Organization, NMA employs treatment-ranking metrics like the likelihood of being the best treatment (P-best), the surface under the cumulative ranking curve (SUCRA), and P-score, aiding decision-making. However, interpreting vast data from NMA can be complex. Graphics play a vital role in clarifying NMA results.
Objectives:
To introduce a visually intuitive graphic to simplify treatment rankings and facilitate decision-making amid the intricacies of NMA
Methods:
This work introduced the beading plot, a variant of the number line plot, to comprehensively overview multiple outcomes in network evidence synthesis. This visualization tool utilizes a scale of 0 to 1 for global metrics such as SUCRA and P-score and P-best. Employing the rankinma R package accessible on the Comprehensive R Archive Network (https://cran.r-project.org/web/packages/rankinma/index.html), the beading plot synthesizes rankings from various outcomes in NMAs, demonstrated using data from a published NMA by Wang et al (2022) examining exercise's prophylactic effects on trismus in patients with nasopharyngeal cancer, involving 11 randomized controlled trials with 805 cases.
Results:
Before the beading plot, basic plots displayed probabilities of each exercise module on every possible rank in short-term mouth opening analysis. Line and bar charts depicted probabilities and cumulative probabilities, albeit with increased treatment numbers leading to unclear patterns. Besides, heat and spie plots were also performed for global metrics across treatments. The beading plot, delineating 4 outcomes, employed a probability line plot, highlighting superior strategies for preventing trismus or reducing mouth opening distance compared with usual care.
Conclusions:
The beading plot offers a practical means to visualize treatment rankings across multiple outcomes simultaneously, including key metrics like SUCRA, P-score, and P-best. Its probability-like number line format presents information in a reader-friendly manner, aiding decision-making by condensing abundant data from network evidence. By providing a comprehensive overview, the beading plot has the potential to empower clinicians and patients in identifying optimal treatments amid complex comparisons. Nonetheless, its usage should be tempered by consideration of the overall evidence certainty.
Network meta-analysis (NMA) addresses the challenges of selecting from multiple treatments by offering comprehensive insights into their effectiveness and safety. Endorsed by articles and supported by the World Health Organization, NMA employs treatment-ranking metrics like the likelihood of being the best treatment (P-best), the surface under the cumulative ranking curve (SUCRA), and P-score, aiding decision-making. However, interpreting vast data from NMA can be complex. Graphics play a vital role in clarifying NMA results.
Objectives:
To introduce a visually intuitive graphic to simplify treatment rankings and facilitate decision-making amid the intricacies of NMA
Methods:
This work introduced the beading plot, a variant of the number line plot, to comprehensively overview multiple outcomes in network evidence synthesis. This visualization tool utilizes a scale of 0 to 1 for global metrics such as SUCRA and P-score and P-best. Employing the rankinma R package accessible on the Comprehensive R Archive Network (https://cran.r-project.org/web/packages/rankinma/index.html), the beading plot synthesizes rankings from various outcomes in NMAs, demonstrated using data from a published NMA by Wang et al (2022) examining exercise's prophylactic effects on trismus in patients with nasopharyngeal cancer, involving 11 randomized controlled trials with 805 cases.
Results:
Before the beading plot, basic plots displayed probabilities of each exercise module on every possible rank in short-term mouth opening analysis. Line and bar charts depicted probabilities and cumulative probabilities, albeit with increased treatment numbers leading to unclear patterns. Besides, heat and spie plots were also performed for global metrics across treatments. The beading plot, delineating 4 outcomes, employed a probability line plot, highlighting superior strategies for preventing trismus or reducing mouth opening distance compared with usual care.
Conclusions:
The beading plot offers a practical means to visualize treatment rankings across multiple outcomes simultaneously, including key metrics like SUCRA, P-score, and P-best. Its probability-like number line format presents information in a reader-friendly manner, aiding decision-making by condensing abundant data from network evidence. By providing a comprehensive overview, the beading plot has the potential to empower clinicians and patients in identifying optimal treatments amid complex comparisons. Nonetheless, its usage should be tempered by consideration of the overall evidence certainty.