Simulated 4-Arm Cluster Randomized Trial (CRT) Data

Description

A simulated cluster randomised trial dataset containing 10 schools and 1,000 pupils. This is a 4-arm trial design with one control group and three intervention groups.

Usage

crt4armSimData

Format

A data frame with 1,000 rows and 7 variables:

pupils

Identifier for each pupil

schools

Identifier for each school

interventions

Treatment assignment coded as 0 for control and 1–3 for intervention groups

pretest

Pre-test scores

gender

Binary gender

ethnicity

A categorical variable with 3 categories

posttest

Post-test scores

Source

Simulated

Add a New Intervention Group to Clustered Randomized Trial (CRT)

Description

This function adds a new intervention group to an existing CRT dataset. It models post-test outcomes using fixed and random effects estimated from the original data and incorporates user-specified effect size and attrition for the new intervention.

Usage

crtAddIntervention(
  originalData,
  ns,
  np,
  es,
  attritionrate,
  outcome,
  interventions,
  schoolsID,
  pupilsID,
  continuous_covariates,
  categorical_covariates
)

Arguments

Details

The function performs the following:

• Fits a linear mixed-effects model (lmer) to the original dataset using provided covariates.

• Applies the specified effect size (es) and generates new post-test scores.

• Simulates attrition by removing post-test scores at random.

Value

A data.frame combining the original and new intervention group, including post-test outcomes simulated for the new intervention based on the estimated mixed model.

See Also

lmer from the lme4 package

Examples

data(crt4armSimData)
new_crt5armData <- crtAddIntervention(originalData = crt4armSimData, ns = 2,
np = 100, es = 0.3, attritionrate = 0.1, outcome = "posttest", interventions = "interventions", 
schoolsID = "schools", pupilsID = "pupils", 
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"))
head(new_crt5armData)

Simulate Clustered Randomized Trial (CRT) Data

Description

This function simulates a multiple intervention arms CRT data. The model includes intervention and pre-test scores as covariates.

Usage

crtDataSimulation(
  ni,
  nstreated,
  np,
  ns,
  sigma,
  ICC,
  B0,
  es,
  seed,
  attritionrates,
  covariates
)

Arguments

Value

A data.frame containing:

pupils

Unique pupil ID

schools

School ID

interventions

Intervention group (0 = control, 1 to ni for interventions)

covariates

Simulated covariates

posttest

Simulated posttest scores (NA if attrited)

Examples

covariates <- list(
 list(name = "pretest", type = "continuous", sd = 1, coefficient = 1.7),
 list(name = "gender", type = "categorical", levels = c("Male", "Female"),
 probs = c(0.3, 0.7), reference = "Male", coefficients = list(B = -0.5)),
 list(name = "ethnicity", type = "categorical", levels = c("White", "Black", "Asian"),
 probs = c(0.3, 0.3, 0.4), reference = "White", coefficients = list(B = 1.02, C = 1.3))
)

crtdata <- crtDataSimulation(ni = 3, ns = 10, np = 100, nstreated = c(2, 3, 2, 3), 
sigma = 1, ICC = 0.1, B0 = 1.45, es = c(0.1, 0.2, 0.5), 
seed = 1234, attritionrates = c(0, 0.1, 0.2, 0.1), covariates = covariates)
head(crtdata)

Internal plotting function for comparison plots

Description

Internal plotting function for comparison plots

Usage

forestPlotMultiArms(
  eefAnalyticsList,
  group,
  Conditional = TRUE,
  ES_Total = TRUE,
  modelNames,
  intlabels = NULL,
  intcolors = NULL,
  maintitle = NULL,
  xlabel = NULL,
  ylabel = NULL,
  vlinecolor = "black"
)

Futility Analysis Across Interventions for CRT, MST, or SRT Designs

Description

This function performs a Bayesian futility analysis for each intervention group compared to control, across cluster randomized trials (CRT), multisite trials (MST) or simple randomized trials (SRT).

Usage

futilityAnalysis(
  method = c("crt", "mst", "srt"),
  data,
  outcome = "posttest",
  interventions = "interventions",
  Random = "schools",
  Nsim = 10000,
  Threshold = 0.05,
  FutThreshold = 0.8,
  continuous_covariates = NULL,
  categorical_covariates = NULL
)

Arguments

Details

The function loops over each intervention, fits the appropriate Bayesian model (crtBayes, mstBayes, srtBayes), extracts the posterior probability, and determines futility based on the specified probability threshold.

Value

A data.frame with columns:

• Intervention: Intervention group identifier.

• Futility: 1 if considered futile (posterior probability < FutThreshold), 0 otherwise.

• ProbES: Bayesian posterior probabilities that the observed effect size is greater than or equal to a pre-specified threshold

See Also

crtBayes, mstBayes, srtBayes functions from the eefAnalytics package

Examples

###Futility analysis of cluster randomized trial###
data(crt4armSimData)
futilityAnalysis(method = "crt", data = crt4armSimData, outcome = "posttest", 
interventions = "interventions", Random = "schools", Nsim = 10000, 
Threshold = 0.05, FutThreshold = 0.8,continuous_covariates = c("pretest"),
categorical_covariates = c("gender", "ethnicity"))

###Futility analysis of multisite trial###
data(mst4armSimData)
futilityAnalysis(method = "mst", data = mst4armSimData, outcome = "posttest",
interventions = "interventions", Random = "schools", Nsim = 10000, 
Threshold = 0.05, FutThreshold = 0.8,continuous_covariates = c("pretest"),
categorical_covariates = c("gender", "ethnicity"))

###Futility analysis of simple randomized trial###
data(srt4armSimData)
futilityAnalysis(method = "srt", data = srt4armSimData, outcome = "posttest",
interventions = "interventions", Nsim = 10000, Threshold = 0.05, FutThreshold = 0.8,
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"))

Simulated 4-Arm Multisite Trial (MST) Data

Description

A simulated multisite trial dataset containing 10 schools and 1,000 pupils. This is a 4-arm trial design with one control group and three intervention groups.

Usage

mst4armSimData

Format

A data frame with 1,000 rows and 7 variables:

pupils

Identifier for each pupil

schools

Identifier for each school

interventions

Treatment assignment coded as 0 for control and 1–3 for intervention groups

pretest

Pre-test scores

gender

Binary gender

ethnicity

A categorical variable with 3 categories

posttest

Post-test scores

Source

Simulated

Add a New Intervention Group to Multisite Trial (MST)

Description

This function adds a new intervention group to an existing Multisite Trial (MST) dataset. It fits a linear mixed-effects model to the original data, then uses its estimates to generate post-test outcomes for the new group, incorporating random intercepts, slopes, and user-defined effect size.

Usage

mstAddIntervention(
  originalData,
  ns,
  np,
  es,
  attritionrate,
  intper,
  outcome,
  interventions,
  schoolsID,
  pupilsID,
  continuous_covariates,
  categorical_covariates
)

Arguments

Details

The function:

• Fits a linear mixed-effects model (lmer) with random slopes and intercepts using existing MST data.

• Simulates new schools and pupils, assigning intervention randomly by specified percentage.

• Simulates attrition by removing post-test scores at random.

Value

A data.frame containing the combined dataset with the newly added intervention group and simulated outcomes.

See Also

lmer from the lme4 package

Examples

data(mst4armSimData)
new_mst5armData <- mstAddIntervention(originalData = mst4armSimData, ns = 2, np = 100, es = 0.3,
intper = 0.5, attritionrate = 0.1, outcome = "posttest", interventions = "interventions",
schoolsID = "schools", pupilsID = "pupils", 
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"))
head(new_mst5armData)

Simulate Multisite Trial (MST) Data

Description

This function simulates a multiple intervention arms Multisite Trial (MST) data. The model includes intervention and pre-test scores as covariates.

Usage

mstDataSimulation(
  ni,
  tpi,
  np,
  ns,
  sigma,
  sigmab0,
  sigmab1,
  B0,
  es,
  seed,
  attritionrates,
  covariates
)

Arguments

Value

A data.frame containing:

pupils

Pupil ID

schools

School ID

interventions

Intervention group assignment (0 = control, 1 to ni = intervention groups)

covariates

Simulated covariates

posttest

Posttest score (NA if attrited)

Examples

covariates <- list(
 list(name = "pretest", type = "continuous", sd = 1, coefficient = 1.7),
 list(name = "gender", type = "categorical", levels = c("Male", "Female"), 
 probs = c(0.3, 0.7), reference = "Male", coefficients = list(B = -0.5)),
 list(name = "ethnicity", type = "categorical", levels = c("White", "Black", "Asian"),
 probs = c(0.3, 0.3, 0.4), reference = "White", coefficients = list(B = 1.02, C = 1.3))
)

mstdata <- mstDataSimulation(ni = 3, ns = 10, np = 100, tpi = c(30, 30, 20, 20),
sigma = 1, sigmab0 = 0.5, sigmab1 = 0.5, B0 = 1.45, 
es = c(0.2, 0.3, 0.1), seed = 1234, attritionrates = c(0.1, 0.1, 0.1, 0), covariates = covariates) 
head(mstdata)

Bayesian or Frequentist Analysis with Forest Plot Comparison for Multi-Arm Trial Designs

Description

This function fits Bayesian or frequentist and producing a forest plot across multiple intervention groups for cluster randomized trials (CRT), multisite trials (MST) or simple randomized trials (SRT).

Usage

multiArmAnalysis(
  method = "crtBayes",
  data,
  outcome = "posttest",
  interventions = "interventions",
  Random = "schools",
  Nsim = 10000,
  Threshold = 0.05,
  FREQoption = "Default",
  nPerm = NULL,
  nBoot = NULL,
  bootType = NULL,
  continuous_covariates = NULL,
  categorical_covariates = NULL,
  maintitle = NULL,
  xlabel = NULL,
  ylabel = NULL,
  vlinecolor = "black",
  intlabels = NULL,
  intcolors = NULL
)

Arguments

Details

This function loops through each intervention, fits the requested statistical model, stores the results, and forest plot visualization for easy comparison. It allows flexible customization for plotting aesthetics.

Value

A ggplot object showing intervention effect sizes and their confidence intervals.

See Also

Functions from the eefAnalytics package: crtBayes, crtFREQ, mstBayes, mstFREQ, srtBayes, srtFREQ

Examples

### Bayesian analysis of cluster randomised trials ###
data(crt4armSimData)
multiArmAnalysis(method = "crtBayes", data = crt4armSimData, outcome = "posttest", 
interventions = "interventions", Random = "schools", Nsim = 10000, Threshold = 0.05, 
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"),
intlabels = c("Intervention A", "Intervention B", "Intervention C"),
maintitle = "Forest plot of comparison of effect sizes", xlabel = "Hedges'g", 
ylabel = "Interventions", vlinecolor = "black")

###MLM analysis of multisite trials with residual bootstrap confidence intervals ###
data(mst4armSimData)
multiArmAnalysis(method = "mstFREQ", data = mst4armSimData, outcome = "posttest",
interventions = "interventions", Random = "schools", nBoot = 1000, bootType="residual",
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"),
intlabels = c("Intervention A", "Intervention B", "Intervention C"),
intcolors = c("Intervention A" = "blue", "Intervention B" = "green", "Intervention C" = "red"),
maintitle = "Forest plot of comparison of effect sizes ", xlabel = "Hedges'g",
ylabel = "Interventions", vlinecolor = "black")

###MLM analysis of multisite trials with permutation p-value###
data(mst4armSimData)
multiArmAnalysis(method = "mstFREQ", data = mst4armSimData, outcome = "posttest", 
interventions = "interventions", Random = "schools", nPerm = 1000, 
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"), 
intlabels = c("Intervention A", "Intervention B", "Intervention C"),
intcolors = c("Intervention A" = "blue", "Intervention B" = "green", "Intervention C" = "red"), 
maintitle = "Forest plot of comparison of effect sizes ",
xlabel = "Hedges'g", ylabel = "Interventions", vlinecolor = "black")


###Bayesian analysis of simple randomised trials###
data(srt4armSimData)
multiArmAnalysis(method = "srtBayes", data = srt4armSimData, outcome = "posttest",
interventions = "interventions", Random = "schools", Nsim = 10000, Threshold = 0.05, 
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"),
intlabels = c("Int A", "Int B", "Int C"),
intcolors = c("Int A" = "#1F77B4", "Int B" = "#2CA02C", "Int C" = "#D62728"),
maintitle = "Forest plot of comparison of effect sizes ", xlabel = "Hedges'g", 
ylabel = "Interventions", vlinecolor = "black")

Plot Posterior Probabilities Across Thresholds for CRT, MST, or SRT Designs

Description

This function generates a Bayesian posterior probability plot across multiple thresholds for each intervention group in a clustered randomized trial (CRT), multisite trial (MST), or simple randomized trial (SRT).

Usage

plotPosteriorProbs(
  method = c("crt", "mst", "srt"),
  data,
  outcome = "posttest",
  interventions = "interventions",
  Random = "schools",
  Nsim = 10000,
  continuous_covariates = NULL,
  categorical_covariates = NULL,
  VerticalLine = NULL,
  VerticalLineColor = "#0000FF",
  HorizontalLine = NULL,
  HorizontalLineColor = "#FF0000",
  threshold_range = c(0, 1),
  maintitle = "Posterior Probabilities Across Thresholds",
  xlabel = "Threshold",
  ylabel = "Posterior Probability",
  intcolors = NULL,
  intlabels = NULL,
  xbreaks = NULL,
  ybreaks = seq(0, 1, by = 0.1)
)

Arguments

Details

The function uses crtBayes(), mstBayes(), or srtBayes() from eefAnalytics package depending on the method.

Value

A ggplot object that displays posterior probabilities across thresholds for each intervention.

See Also

crtBayes, mstBayes, srtBayes functions from the eefAnalytics package

Examples

###Plot Posterior Probabilities of cluster randomized trial###
data(crt4armSimData)
plotPosteriorProbs(method = "crt",data = crt4armSimData, outcome = "posttest",
interventions = "interventions", Random = "schools", Nsim = 10000,
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"),
threshold_range = c(0, 0.1), VerticalLine = 0.05, HorizontalLine = 0.8,
VerticalLineColor= "purple", HorizontalLineColor= "black", 
intlabels = c("Intervention A", "Intervention B", "Intervention C"), 
intcolors = c("Intervention A" = "blue", "Intervention B" = "red", 
"Intervention C" = "green"), maintitle= "Posterior probability plot",
xlabel= "Threshold", ylabel= "Posterior probability",
xbreaks= 0.1, ybreaks= seq(0, 1, by = 0.1))

###Plot Posterior Probabilities of multisite trial###
data(mst4armSimData)
plotPosteriorProbs(method = "ms",data = mst4armSimData, outcome = "posttest", 
interventions = "interventions", Random = "schools", Nsim = 10000,
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"),
threshold_range = c(0, 0.1), VerticalLine = 0.05, HorizontalLine = 0.8, 
VerticalLineColor= "purple", HorizontalLineColor= "black",
intlabels = c("Intervention A", "Intervention B", "Intervention C"), 
intcolors = c("Intervention A" = "blue", "Intervention B" = "red",
"Intervention C" = "green"), maintitle= "Posterior probability plot",
xlabel= "Threshold", ylabel= "Posterior probability",
xbreaks= 0.1, ybreaks= seq(0, 1, by = 0.1))

###Futility analysis of simple randomized trial###
data(srt4armSimData)
plotPosteriorProbs(method = "srt",data = srt4armSimData, outcome = "posttest",
interventions = "interventions", Nsim = 10000, threshold_range = c(0, 0.2),
continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"),
VerticalLine = 0.05, HorizontalLine = 0.8, VerticalLineColor= "purple",
HorizontalLineColor= "black", intlabels = c("Intervention A", "Intervention B",
"Intervention C"), intcolors = c("Intervention A" = "#1F77B4", 
"Intervention B" = "#D62728", "Intervention C" = "#2CA02C"),
maintitle= "Posterior probability plot", xlabel= "Threshold", 
ylabel= "Posterior probability", xbreaks= 0.1, ybreaks= seq(0, 1, by = 0.1))

Run the EPTS Shiny Application

Description

Launches a Shiny application that provides an interactive user interface to run the functions provided by the epts package.

Usage

runEPTS()

Examples

if (interactive()) {
  runEPTS()
}

Simulated 4-Arm Simple Randomized Trial (SRT) Data

Description

A simulated simple randomized trial dataset containing 1,000 individuals. This is a 4-arm trial design with one control group and three intervention groups.

Usage

srt4armSimData

Format

A data frame with 1,000 rows and 6 variables:

ID

Identifier for participants ID

interventions

Treatment assignment coded as 0 for control and 1–3 for intervention groups

pretest

Pre-test scores

gender

Binary gender

ethnicity

A categorical variable with 3 categories

posttest

Post-test scores

Source

Simulated

Add a New Intervention Group to Simple Randomized Trial (SRT) Data

Description

This function adds a new intervention group to an existing SRT dataset by generating new participant-level data.

Usage

srtAddIntervention(
  existing_data,
  np,
  es,
  attritionrate,
  outcome,
  interventions,
  id,
  continuous_covariates,
  categorical_covariates
)

Arguments

Value

A data.frame combining the original dataset with the newly simulated intervention group.

Examples

data(srt4armSimData)
new_srt5armData <- srtAddIntervention(existing_data = srt4armSimData, np = 100, 
es = 0.3, attritionrate = 0.1, outcome = "posttest", interventions = "interventions",
id = "ID", continuous_covariates = c("pretest"), categorical_covariates = c("gender", "ethnicity"))
head(new_srt5armData)

Simulate Simple Randomized Trial (SRT) Data

Description

This function simulates a Simple Randomized Trial (SRT), with multiple intervention arms, pre-test and post-test scores, and individual-level attrition. No clustering or hierarchical structure is assumed.

Usage

srtDataSimulation(ni, tpi, np, sigma, B0, es, seed, attritionrates, covariates)

Arguments

Value

A data.frame containing:

ID

Participant ID

interventions

Intervention assignment (0 = control, 1 to ni = intervention groups)

covariates

Simulated covariates

posttest

Posttest score (NA if participant attrited)

Examples

covariates <- list(
 list(name = "pretest", type = "continuous", sd = 1, coefficient = 1.7),
 list(name = "gender", type = "categorical", levels = c("Male", "Female"),
 probs = c(0.3, 0.7), reference = "Male", coefficients = list(B = -0.5)),
 list(name = "ethnicity", type = "categorical", levels = c("White", "Black", "Asian"),
 probs = c(0.3, 0.3, 0.4), reference = "White", coefficients = list(B = 1.02, C = 1.3))
)

srtdata <- srtDataSimulation(ni = 3, np = 1000, tpi = c(30, 30, 20, 20),
sigma = 1, B0 = 1.45,  es = c(0.2, 0.3, 0.1), seed = 1234,
attritionrates = c(0.1, 0.1, 0.1, 0), covariates = covariates) 
head(srtdata)

Superiority Analysis Across Interventions for CRT, MST, or SRT Designs

Description

This function performs a Bayesian superiority analysis, comparing each intervention against a reference intervention, across cluster randomized trials (CRT), multisite trials (MST) or simple randomized trials (SRT).

Usage

superiorityAnalysis(
  method = c("crt", "mst", "srt"),
  data,
  outcome = "posttest",
  interventions = "interventions",
  Random = "schools",
  Nsim = 10000,
  Threshold = 0.05,
  refintervention = 1,
  SupThreshold = 0.8,
  continuous_covariates = NULL,
  categorical_covariates = NULL
)

Arguments

Details

The effect size is estimated against a reference intervention, which by default is intervention 1 but can be reassigned to any other intervention, including the control (refintervention = 0).

Value

A data.frame with columns:

• Intervention: Intervention group identifier.

• ProbES: Posterior probability of superiority over the reference intervention.

• Superiority: Label indicating "Superior", "Not Superior", or "Reference".

See Also

crtBayes, mstBayes, srtBayes functions from the eefAnalytics package

Examples

###Superiority analysis of cluster randomized trial###
data(crt4armSimData)
superiorityAnalysis(method = "crt", data = crt4armSimData, outcome = "posttest",
interventions = "interventions", Random = "schools", Nsim = 10000, refintervention = 2,
Threshold = 0.05, SupThreshold = 0.8,continuous_covariates = c("pretest"),
categorical_covariates = c("gender", "ethnicity"))

###Superiority analysis of multisite trial###
data(mst4armSimData)
superiorityAnalysis(method = "mst", data = mst4armSimData, outcome = "posttest",
interventions = "interventions", Random = "schools", Nsim = 10000, refintervention = 2,
Threshold = 0.05, SupThreshold = 0.8,continuous_covariates = c("pretest"),
categorical_covariates = c("gender", "ethnicity"))

###Superiority analysis of simple randomized trial###
data(srt4armSimData)
superiorityAnalysis(method = "srt", data = srt4armSimData, outcome = "posttest",
interventions = "interventions", Nsim = 10000, refintervention = 2,
Threshold = 0.05, SupThreshold = 0.8,continuous_covariates = c("pretest"),
categorical_covariates = c("gender", "ethnicity"))