covar_contribution_plot {rnmamod} | R Documentation |
Visualising study percentage contributions against a covariate
Description
A scatter plot of the study percentage contributions against the values of a continuous study-level covariate for the treatment effects of comparisons referring to the basic parameters, functional parameters or both. Contributions on the estimated regression coefficients are also presented. Study percentage contributions are based on the proposed methodology of Donegan and colleagues (2018).
Usage
covar_contribution_plot(
contr_res,
comparisons = "basic",
drug_names,
upper_limit = 100,
name_x_axis = NULL,
axis_title_size = 14,
axis_text_size = 14,
strip_text_size = 14,
subtitle_size = 14,
label_size = 4,
seq_by = 0.1,
percentage = FALSE
)
Arguments
contr_res |
An object of S3 class |
comparisons |
Character string indicating the type of comparisons to
plot, with possible values: |
drug_names |
A vector of labels with the name of the interventions in
the order they appear in the argument |
upper_limit |
A positive number to define the upper bound of range of percentage values for the y-axis. The default argument is 100. |
name_x_axis |
Text for the x axis title through the |
axis_title_size |
A positive integer for the font size of x axis title.
|
axis_text_size |
A positive integer for the font size of axis text (both
axes). |
strip_text_size |
A positive integer for the font size of strip text in
facets. |
subtitle_size |
A positive integer for the font size of subtitle.
|
label_size |
A positive integer for the font size of labels appearing on
each data point. |
seq_by |
A positive integer for the sequence of values in the x-axis.
|
percentage |
Logical with values |
Details
A panel of scatter plots is returned on the study percentage contributions to
the treatment effects (and also regression coefficients, if relevant) against
a continuous covariate for each comparison defined by the argument
comparisons
; namely, only those referring to the basic or functional
parameters or all possible pairwise comparisons. Blue and red points indicate
the studies investigating the corresponding comparisons directly and
indirectly, respectively. Each point displays the number of the corresponding
study in the dataset.
If interest also lies on the study percentage contributions to the regression
coefficients, the regression coefficients can be determined to be common
across the comparisons, independent or exchangeable and this assumption is
specified in the study_perc_contrib
function.
Value
If interest lies only on the study percentage contributions to the summary treatment effects of all possible pairwise comparisons, the function returns one plot named 'plot_treat'. If interest lies also on the study percentage contributions to the regression coefficient(s), the function returns also the plot named 'plot_reg'.
Author(s)
Loukia M. Spineli
References
Donegan S, Dias S, Tudur-Smith C, Marinho V, Welton NJ. Graphs of study contributions and covariate distributions for network meta-regression. Res Synth Methods 2018;9(2):243–60. doi: 10.1002/jrsm.1292
See Also
Examples
## Not run:
data("nma.fluoride.donegan2018")
# Get study contributions to random-effects network meta-regression
# results under the assumption of independent treatment-by-covariate
# interaction
res <- study_perc_contrib(study_name = nma.fluoride.donegan2018$study,
base_t = nma.fluoride.donegan2018$t1,
exp_t = nma.fluoride.donegan2018$t2,
ref_t = 1,
obs_se = nma.fluoride.donegan2018$SE,
obs_cov = nma.fluoride.donegan2018$Cov,
covar = nma.fluoride.donegan2018$year,
covar_assum = "independent",
model = "RE",
tau = sqrt(0.03))
# Covariate-contribution plot on the basic parameters only
covar_contribution_plot(contr_res = res,
comparisons = "basic",
drug_names = c("NT", "PL", "DE", "RI", "GE", "VA"),
upper_limit = 15,
name_x_axis = "Randomisation year",
seq_by = 10)
## End(Not run)