Help for package rankinma

Type:

Package

Title:

Rank in Network Meta-Analysis

Version:

0.2.2

Date:

2024-04-01

Maintainer:

Enoch Kang <y.enoch.kang@gmail.com>

Description:

A supportive collection of functions for gathering and plotting treatment ranking metrics after network meta-analysis.

License:

GPL (≥ 3)

URL:

https://rankinma.shinyapps.io/rankinma/

Depends:

R (≥ 4.2.0)

Imports:

graphics, grDevices, mvtnorm, netmeta, stats, utils

Suggests:

knitr, rmarkdown, testthat (≥ 3.0.0)

Language:

en-US

Encoding:

UTF-8

RoxygenNote:

7.2.2

Config/testthat/edition:

VignetteBuilder:

knitr

NeedsCompilation:

Packaged:

2024-04-01 13:39:09 UTC; KYN

Author:

Chiehfeng Chen

[aut], Enoch Kang

[aut, cre, trl], Wen-Hsuan Hou

[aut], Jin-Hua Chen

[aut], Yu-Chieh Chuang

[aut], Edwin Chan [aut]

Repository:

CRAN

Date/Publication:

2024-04-01 13:50:02 UTC

rankinma: Rank in Network Meta-Analysis

Description

rankinma is an R package that supports users to easily obtain and visualize various metrics of treatment ranking from network meta-analysis no matter using either frequentist or Bayesian approach. Development of package rankinma is based on R version 4.2.2 (2022-10-31 ucrt). Extra imported packages are as follows:

Details

Current version consists of seven functions, including two functions for data preparation (function GetMetrics and SetMetrics) and five functions for visualization of treatment ranking metrics (i.e. PlotBeads, PlotLine, PlotBar, PlotHeat, and PlotSpie). Probabilities of treatments on each possible rank can be visualized using PlotLine and PlotBar. Due to concise information, PlotBeads is recommended to be used for global metrics of treatment ranking, such as P-score and SUCRA. The other four visualization functions can also generate graphics of the global metrics.

References

Salanti, G., Ades, A. E., & Ioannidis, J. P. (2011). Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial. Journal of clinical epidemiology, 64(2), 163-171.
Chaimani, A., Higgins, J. P., Mavridis, D., Spyridonos, P., & Salanti, G. (2013). Graphical tools for network meta-analysis in STATA. PloS one, 8(10), e76654.
Van Valkenhoef, G., Tervonen, T., Zwinkels, T., De Brock, B., & Hillege, H. (2013). ADDIS: a decision support system for evidence-based medicine. Decision Support Systems, 55(2), 459-475.
Rücker, G., & Schwarzer, G. (2015). Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC medical research methodology, 15(1), 1-9.
Daly, C. H., Mbuagbaw, L., Thabane, L., Straus, S. E., & Hamid, J. S. (2020). Spie charts for quantifying treatment effectiveness and safety in multiple outcome network meta-analysis: a proof-of-concept study. BMC Medical Research Methodology, 20, 1-13.
Balduzzi, S., Rücker, G., Nikolakopoulou, A., Papakonstantinou, T., Salanti, G., Efthimiou, O., & Schwarzer, G. (2023). netmeta: An R package for network meta-analysis using frequentist methods. Journal of Statistical Software, 106, 1-40.

Get treatment ranking metrics from network meta-analysis output

Description

GetMetrics() is a function for gathering metrics of treatment ranking from netmeta output.

Usage

GetMetrics(
  data,
  outcome = NULL,
  prefer = NULL,
  metrics = NULL,
  model = "random",
  simt = 1000,
  rob = NULL
)

Arguments

data

DATA of netmeta output.

outcome

STRING for name of outcome.

prefer

STRING for indicating which direction is beneficial treatment effect in terms of "small" and "large" values in statistic test.

metrics

STRING for metrics of treatment ranking in terms of "SUCRA", "P-score", and "P-best" for the value of surface under the cumulative ranking curve, P-score, and probability of achieving the best treatment.

model

STRING for analysis model in terms of "random" and "common" for random-effects model and common-effect model.

simt

INTEGER for times of simulations to estimate surface under the cumulative ranking curve (SUCRA).

rob

STRING for column name of risk of bias.

Value

GetMetrics() returns a data.frame with three columns, including treatment, metrics of treatment ranking, and outcome name.

References

Rücker, G., & Schwarzer, G. (2015). Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC medical research methodology, 15(1), 1-9.

Salanti, G., Ades, A. E., & Ioannidis, J. P. (2011). Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial. Journal of clinical epidemiology, 64(2), 163-171.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#dataMetrics <- GetMetrics(nma, outcome = "HbA1c", prefer = "small",
#metrics = "SUCRA", model = "random", simt = 1000)

# Get P-score
#dataMetrics <- GetMetrics(nma, outcome = "HbA1c", prefer = "small",
#metrics = "P-score", model = "random", simt = 1000)
## End(Not run)

Illustrate bar chart of treatment ranking metrics

Description

PlotBar() is a function for illustrating bar chart in both separated and accumulative styles.

Usage

PlotBar(data, accum = NULL, merge = NULL, color = NULL, rotateX = NULL)

Arguments

data

DATA of metrics for treatment ranking.

accum

LOGIC value for indicating whether use accumulative probabilities. This parameter is only for probabilities but not global metrics of treatment ranking.

merge

LOGIC value for indicating whether merge bar charts together.

color

LIST of colors for treatments in a network meta-analysis, or CHARACTER of a color for the bar on not accumulated bar chart.

rotateX

NUMERIC value between 0 and 360 for rotating x axis labels of bars.

Value

PlotBar() returns a bar chart.

References

Van Valkenhoef, G., Tervonen, T., Zwinkels, T., De Brock, B., & Hillege, H. (2013). ADDIS: a decision support system for evidence-based medicine. Decision Support Systems, 55(2), 459-475.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#dataMetrics <- GetMetrics(nma, outcome = "HbA1c", prefer = "small", metrics = "SUCRA",
#model = "random", simt = 1000)

# Set data for rankinma
#dataRankinma <- SetMetrics(dataMetrics, tx = tx, outcome = outcome,
#metrics = SUCRA, metrics.name = "SUCRA")

# Illustrate bar plot
#PlotBar(dataRankinma)
## End(Not run)

Illustrate beading plot

Description

PlotBeads() is a function for illustrating beading plot.

Usage

PlotBeads(
  data,
  scaleX = "Numeric",
  txtValue = "Effects",
  color = NULL,
  whichRoB = "None",
  lgcBlind = FALSE,
  szPnt = NULL,
  szFntTtl = NULL,
  szFntTtlX = NULL,
  szFntX = NULL,
  szFntY = NULL,
  szFntTxt = NULL,
  szFntLgnd = NULL,
  rotateTxt = 60
)

Arguments

data

DATA of metrics for treatment ranking.

scaleX

STRING for indicating scale on the x axis.

txtValue

STRING for indicating labels of metrics or effects on each point.

color

LIST of colors for treatments in a network meta-analysis.

whichRoB

STRING for indicating how to display risk of bias for each treatment.

lgcBlind

LOGIC value for indicating whether to display with color-blind friendly.

szPnt

NUMERIC value for indicating point size of ranking metrics.

szFntTtl

NUMERIC value for indicating font size of main title.

szFntTtlX

NUMERIC value for indicating font size of title on X-axis.

szFntX

NUMERIC value for indicating font size of numeric scale on X-axis.

szFntY

NUMERIC value for indicating font size of outcome name(s).

szFntTxt

NUMERIC value for indicating font size of value of each point.

szFntLgnd

NUMERIC value for indicating legend font size.

rotateTxt

NUMERIC value between 0 and 360 for rotating labels of text values of each point.

Value

PlotBeads() returns a beading plot.

Author(s)

Chiehfeng Chen & Enoch Kang

References

Chen, C., Chuang, Y.C., Chan, E., Chen, J.H., Hou, W.H., & Kang, E. (2023). Beading plot: A novel graphics for ranking interventions in network evidence. PREPRINT (Version 1) available at Research Square.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#nma.1 <- GetMetrics(nma, outcome = "HbA1c.random", prefer = "small", metrics = "SUCRA",
#model = "random", simt = 1000)
#nma.2 <- GetMetrics(nma, outcome = "HbA1c.common", prefer = "small", metrics = "SUCRA",
#model = "common", simt = 1000)

# Combine metrics of multiple outcomes
#dataMetrics <- rbind(nma.1, nma.2)

# Set data for rankinma
#dataRankinma <- SetMetrics(dataMetrics, tx = tx, outcome = outcome,
#metrics = SUCRA, metrics.name = "SUCRA")

# Illustrate beading plot
#PlotBeads(data = dataRankinma)
## End(Not run)

Illustrate heat plot for treatment ranking

Description

PlotHeat() is a function for illustrating heat plot.

Usage

PlotHeat(data, sorttx = NULL, rotateX = NULL, szFntY = NULL)

Arguments

data

DATA of metrics for treatment ranking.

sorttx

LOGIC value for indicating whether sort heat plot by treatments.

rotateX

NUMERIC value between 0 and 360 for rotating x axis labels of heat plot.

szFntY

NUMERIC value for indicating font size of outcome name(s).

Value

PlotHeat() returns a heat plot.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#nma.1 <- GetMetrics(nma, outcome = "HbA1c.random", prefer = "small", metrics = "SUCRA",
#model = "random", simt = 1000)
#nma.2 <- GetMetrics(nma, outcome = "HbA1c.common", prefer = "small", metrics = "SUCRA",
#model = "common", simt = 1000)

# Combine metrics of multiple outcomes
#dataMetrics <- rbind(nma.1, nma.2)

# Set data for rankinma
#dataRankinma <- SetMetrics(dataMetrics, tx = tx, outcome = outcome,
#metrics = SUCRA, metrics.name = "SUCRA")

# Illustrate heat plot
#PlotHeat(data = dataRankinma)
## End(Not run)

Illustrate line chart of treatment ranking metrics

Description

PlotLine() is a function for illustrating line chart in both simple and composite styles.

Usage

PlotLine(
  data,
  accum = NULL,
  compo = NULL,
  merge = NULL,
  color = NULL,
  rotateX = NULL
)

Arguments

data

DATA of metrics for treatment ranking.

accum

LOGIC value for indicating whether use accumulative probabilities. This parameter is only for probabilities but not global metrics of treatment ranking.

compo

LOGIC value for indicating whether use composite line chart. This parameter is only for probabilities but not global metrics of treatment ranking.

merge

LOGIC value for indicating whether merge line charts together.

color

LIST of colors for treatments in a network meta-analysis, or CHARACTER of a color for the line on not composite line chart.

rotateX

NUMERIC value between 0 and 360 for rotating x axis labels of line chart.

Value

PlotLine() returns a line chart.

References

Chaimani, A., Higgins, J. P., Mavridis, D., Spyridonos, P., & Salanti, G. (2013). Graphical tools for network meta-analysis in STATA. PloS one, 8(10), e76654.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#dataMetrics <- GetMetrics(nma, outcome = "HbA1c", prefer = "small", metrics = "SUCRA",
#model = "random", simt = 1000)

# Set data for rankinma
#dataRankinma <- SetMetrics(dataMetrics, tx = tx, outcome = outcome,
#metrics = SUCRA, metrics.name = "SUCRA")

# Illustrate bar plot
#PlotLine(dataRankinma)
## End(Not run)

Illustrate beading plot

Description

PlotSpie() is a function for illustrating spie plot.

Usage

PlotSpie(data, color = NULL)

Arguments

data

DATA of metrics for treatment ranking.

color

LIST of colors for outcomes in a network meta-analysis.

Value

PlotSpie() returns a spie plot.

References

Daly, C. H., Mbuagbaw, L., Thabane, L., Straus, S. E., & Hamid, J. S. (2020). Spie charts for quantifying treatment effectiveness and safety in multiple outcome network meta-analysis: a proof-of-concept study. BMC Medical Research Methodology, 20, 1-13.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#nma.1 <- GetMetrics(nma, outcome = "HbA1c.random", prefer = "small", metrics = "SUCRA",
#model = "random", simt = 1000)
#nma.2 <- GetMetrics(nma, outcome = "HbA1c.common", prefer = "small", metrics = "SUCRA",
#model = "common", simt = 1000)

# Combine metrics of multiple outcomes
#dataMetrics <- rbind(nma.1, nma.2)

# Set data for rankinma
#dataRankinma <- SetMetrics(dataMetrics, tx = tx, outcome = outcome,
#metrics = SUCRA, metrics.name = "SUCRA")

# Illustrate beading plot
#PlotSpie(data = dataRankinma)
## End(Not run)

Setup data of treatment ranking metrics for rankinma

Description

SetMetrics() is a function for checking and preparing data set of metrics for further ploting in rankinma.

Usage

SetMetrics(
  data,
  outcome = NULL,
  tx = NULL,
  metrics = NULL,
  metrics.name = NULL,
  trans = 0.8
)

Arguments

data

DATAFRAME of treatment, metrics, and name of outcomes.

outcome

VARIABLE string data for of outcome(s).

tx

VARIABLE with string data for treatments.

metrics

VARIABLE with numeric data for global metrics, but it should be "NULL" when using "Probabilities" as metrics.

metrics.name

STRING for metrics of treatment ranking in terms of "SUCRA","P-score", and "P-best" for the value of surface under the cumulative ranking curve, P-score, and probability of achieving the best treatment.

trans

NUMERIC for indicating transparency of colors of treatments.

Value

SetMetrics() returns a confirmed data.frame of treatment, metrics of treatment ranking, and outcome name.

metrics.name

A string shows type of metrics of treatment ranking.

ls.outcome

Strings list outcomes.

ls.tx

Strings list treatments.

n.outcome

An integer shows numbers of outcomes.

n.tx

An integer shows numbers of treatments.

data

A data frame consists of seven columns of core information among all outcomes.

data.sets

A list shows data frame of core information by each outcome.

ptrn.tx

A data frame shows treatments on each outcome.

ptrn.outcome

A data frame shows outcomes by treatments.

color.txs

A data frame shows color of each treatment.

trans

A numeric value shows transparency for colors of each treatment.

Examples

## Not run:
#library(netmeta)
#data(Senn2013)
#nma <- netmeta(TE, seTE, treat1, treat2,
#studlab, data = Senn2013, sm = "SMD")

# Get SUCRA
#nma.1 <- GetMetrics(nma, outcome = "HbA1c.random", prefer = "small", metrics = "SUCRA",
#model = "random", simt = 1000)
#nma.2 <- GetMetrics(nma, outcome = "HbA1c.common", prefer = "small", metrics = "SUCRA",
#model = "common", simt = 1000)

# Combine metrics of multiple outcomes
#dataMetrics <- rbind(nma.1, nma.2)

# Set data for rankinma
#dataRankinma <- SetMetrics(dataMetrics, tx = tx, outcome = outcome,
#metrics = SUCRA, metrics.name = "SUCRA")
## End(Not run)

rankinma: Rank in Network Meta-Analysis

Description

Details

References

Get treatment ranking metrics from network meta-analysis output

Description

Usage

Arguments

Value

References

See Also

Examples

Illustrate bar chart of treatment ranking metrics

Description

Usage

Arguments

Value

References

See Also

Examples

Illustrate beading plot

Description

Usage

Arguments

Value

Author(s)

References

See Also

Examples

Illustrate heat plot for treatment ranking

Description

Usage

Arguments

Value

See Also

Examples

Illustrate line chart of treatment ranking metrics

Description

Usage

Arguments

Value

References

See Also

Examples

Illustrate beading plot

Description

Usage

Arguments

Value

References

See Also

Examples

Setup data of treatment ranking metrics for rankinma

Description

Usage

Arguments

Value

See Also

Examples