galaxy {xmeta}

Galaxy Plot: A New Visualization Tool of Bivariate Meta-Analysis Studies

Description

A new visualization method that simultaneously presents the effect sizes of bivariate outcomes and their standard errors in a two-dimensional space.

Usage

galaxy(data, y1, s1, y2, s2, scale1, scale2, scale.adj, 
              corr, group, study.label, annotate, xlab, ylab, main, legend.pos)

Arguments

data

dataset with at least 4 columns for the effect sizes of the two outcomes and their standard errors

y1

column name for outcome 1, default is 'y1'

s1

column name for standard error of y1, default is 's1'

y2

column name for outcome 2, default is 'y2'

s2

column name for standard error of y2, default is 's2'

scale1

parameter for the length of the cross hair: the ellipse width is scale1 / s1 * scale.adj

scale2

parameter for the length of the cross hair: the ellipse height is scale2 / s2 * scale.adj

scale.adj

a pre-specified parameter to adjust for scale1 and scale2

corr

column name for within-study correlation

group

column name for study group

study.label

column name for study label

annotate

logical specifying whether study label should be added to the plot, default is FALSE.

xlab

x axis label, default y1

ylab

y axis label, default y2

main

main title

legend.pos

The position of the legend for study groups if group is specified, see legend, default is 'bottomright'.

Details

This function returns the galaxy plot to visualize bivariate meta-analysis data, which faithfully retains the information in two separate funnel plots, while providing useful insights into outcome correlations, between-study heterogeneity and joint asymmetry. Galaxy plot: a new visualization tool of bivariate meta-analysis studies. Funnel plots have been widely used to detect small study effects in the results of univariate meta-analyses. However, there is no existing visualization tool that is the counterpart of the funnel plot in the multivariate setting. We propose a new visualization method, the galaxy plot, which can simultaneously present the effect sizes of bivariate outcomes and their standard errors in a two-dimensional space. The galaxy plot is an intuitive visualization tool that can aid in interpretation of results of multivariate meta-analysis. It preserves all of the information presented by separate funnel plots for each outcome while elucidating more complex features that may only be revealed by examining the joint distribution of the bivariate outcomes.

Author(s)

Chuan Hong, Chongliang Luo, Yong Chen

References

Hong, C., Duan, R., Zeng, L., Hubbard, R., Lumley, T., Riley, R., Chu, H., Kimmel, S., and Chen, Y. (2020) Galaxy Plot: A New Visualization Tool of Bivariate Meta-Analysis Studies, American Journal of Epidemiology, https://doi.org/10.1093/aje/kwz286.

Examples

data(sim_dat)
galaxy(data=sim_dat, scale.adj = 0.9, corr = 'corr', group = 'subgroup', 
        study.label = 'study.id', annotate = TRUE, main = 'galaxy plot')