kld_measure

The user specify the (posterior) mean and standard error (or posterior
standard deviation) of two estimated treatment effects, X and Y, that refer
to the same pairwise comparison and are assumed to follow a normal
distribution. The function returns the Kullback-Leibler Divergence (KLD)
measure of 1) approximating X with Y, 2) approximating Y with X, and 3) their
average.

A comprehensive suite of functions to perform and visualise
pairwise and network meta-analysis with aggregate binary or continuous
missing participant outcome data. The package covers core Bayesian one-stage
models implemented in a systematic review with multiple interventions,
including fixed-effect and random-effects network meta-analysis,
meta-regression, evaluation of the consistency assumption via the
node-splitting approach and the unrelated mean effects model (original and
revised model proposed by Spineli, (2022) <doi:10.1177/0272989X211068005>), and
sensitivity analysis (see Spineli et al., (2021) <doi:10.1186/s12916-021-02195-y>).
Missing participant outcome data are addressed in all models of the package
(see Spineli, (2019) <doi:10.1186/s12874-019-0731-y>, Spineli et al., (2019)
<doi:10.1002/sim.8207>, Spineli, (2019) <doi:10.1016/j.jclinepi.2018.09.002>,
and Spineli et al., (2021) <doi:10.1002/jrsm.1478>).
The robustness to primary analysis results can also be investigated using a
novel intuitive index (see Spineli et al., (2021) <doi:10.1177/0962280220983544>).
Methods to evaluate the transitivity assumption using trial dissimilarities
and hierarchical clustering are provided
(see Spineli, (2024) <doi:10.1186/s12874-024-02436-7>, and
Spineli et al., (2025) <doi:10.1002/sim.70068>). A novel index to
facilitate interpretation of local inconsistency is also available
(see Spineli, (2024) <doi:10.1186/s13643-024-02680-4>)
The package also offers a rich, user-friendly visualisation toolkit that aids
in appraising and interpreting the results thoroughly and preparing the
manuscript for journal submission. The visualisation tools comprise the
network plot, forest plots, panel of diagnostic plots, heatmaps on the extent
of missing participant outcome data in the network, league heatmaps on
estimation and prediction, rankograms, Bland-Altman plot, leverage plot,
deviance scatterplot, heatmap of robustness, barplot of Kullback-Leibler
divergence, heatmap of comparison dissimilarities and dendrogram of comparison
clustering. The package also allows the user to export the results to an Excel
file at the working directory.

Loukia Spineli

rnmamod

Bayesian Network Meta-Analysis with Missing Participants

Chrysostomos Kalyvas

Katerina Papadimitropoulou

kld_measure function

<dl><dt>mean_y</dt>
<dd>A real number that refers to the mean of the estimated
treatment effect Y on the scale of the selected effect measure
(in logarithmic scale for relative effect measures).</dd>
<dt>sd_y</dt>
<dd>A positive integer that refers to the posterior standard
deviation or the standard error of the estimated treatment effect Y on the
scale of the selected effect measure (in logarithmic scale for relative
effect measures).</dd>
<dt>mean_x</dt>
<dd>A real number that refers to the mean of the estimated
treatment effect X on the scale of the selected effect measure
(in logarithmic scale for relative effect measures).</dd>
<dt>sd_x</dt>
<dd>A positive integer that refers to the posterior standard
deviation or the standard error of the estimated treatment effect X on the
scale of the selected effect measure (in logarithmic scale for relative
effect measures).</dd></dl>

Arguments

Function for the Kullback-Leibler Divergence of two normally distributed
treatment effects for the same pairwise comparison — kld_measure

<dl>

<dt>mean_y</dt>
<dd>A real number that refers to the mean of the estimated
treatment effect Y on the scale of the selected effect measure
(in logarithmic scale for relative effect measures).</dd>


<dt>sd_y</dt>
<dd>A positive integer that refers to the posterior standard
deviation or the standard error of the estimated treatment effect Y on the
scale of the selected effect measure (in logarithmic scale for relative
effect measures).</dd>


<dt>mean_x</dt>
<dd>A real number that refers to the mean of the estimated
treatment effect X on the scale of the selected effect measure
(in logarithmic scale for relative effect measures).</dd>


<dt>sd_x</dt>
<dd>A positive integer that refers to the posterior standard
deviation or the standard error of the estimated treatment effect X on the
scale of the selected effect measure (in logarithmic scale for relative
effect measures).</dd>

</dl>

kld_sym	The symmetric KLD value as the average of two KLD values .

kld_x_true	The KLD value when approximating X by Y (X is the 'truth').

kld_y_true	The KLD value when approximating Y by X (Y is the 'truth').

kld_measure: Function for the Kullback-Leibler Divergence of two normally distributed treatment effects for the same pairwise comparison

Description

Usage

Value

Arguments

References

See Also