seqmeans: Summarize sequence-specific descriptive statistics

In SMARTAR: Sequential Multiple Assignment Randomized Trial and Adaptive Randomization

Description

Return a message that lists all the treatment sequence embedded in SMART design and summarizes all the sequence-specific descriptive statistics. It also provide design diagram of SMART and graphs of sequence-specific descriptive statistics (boxplot for continuous primary outcome and barchart for binary primary outcome).

Usage

seqmeans(
  data,
  family = c("gaussian", "binomial")[1],
  plot = "d",
  digits = NULL,
  color = c("yellow", "forestgreen"),
  pch = c(19, 15),
  title = NULL,
  xlab = NULL,
  ylab = NULL,
  xtext = NULL,
  legend = c("Evaluation", "Treatment"),
  reference = TRUE
)

Arguments

data	Input data frame of the sequential randomized trial (SMART) data used for analysis. The data should include the variables of stage-specific treatments (At; t=1,2,3), intermediate evaluation (Ot; t=1,2,3) and final primary outcome (Y), where t represent the number of stages embedded in design. If stage-1 treatment (A1) takes into account the information of baseline evaluation, O1 needed to be include in data, otherwise not.
family	A character string to specify the type of final primary outcome. The default is family=“gaussian”, which refers to the continuous primary outcome. If family=”binomial” then the primary outcome will be treated as binary variable.
plot	A character string to specify the output figure. If plot=”d” then output the design diagram of SMART; If plot=”s” then output boxplot for continuous primary outcome or bar plot for binary primary outcome by sequence. The default is plot=”d”.
digits	An integer indicating the number of decimal places for sequence-specific mean and variance. Default is digits=NULL.
color	Characters indicating the color of boxplot for continuous primary outcome and barplot for binomial primary outcome. The first Default is “yellow", the second default is c("yellow","forestgreen").
pch	Two integer indicating the point shape of design diagram of SMART. Default is c(19,15).
title	An character indicating the title of boxplot, barplot for primary outcome or design diagram of SMART. For primary outcome, the default is "Primary outcome by treatment sequence (O1,A1,O2...)", for design diagram, the default is "Design diagram of SMART".
xlab	Characters to specify the label of the horizontal axis of the output figure.
ylab	Characters to specify the label of the vertical axis of the output figure.
xtext	Characters indicating the text of x axis boxplot.
legend	Characters to specify the legend of the design diagram of SMART. Default is legend=c("Evaluation","Treatment").
reference	Logic argument to add a reference line to the graph of descriptive statistics of the primary outcome. The value of the reference line is equal to the average of all sequence-specific means. If TRUE, add a reference line (mean of outcome) to boxplot of the primary outcome, otherwise do not add reference line. The default is reference=FALSE.

Value

an object of information of all the treatment sequences and sequences-specific descriptive statistics defined in a SMART data

• SEQ: Index of treatment sequences.

• O1: Baseline evaluation outcome.

• A1: Action of stage-1 treatment.

• O2: Intermeidate outcome evaluated at the end of stage 1.

• A2: Action of stage-2 treatment.

• O3: Intermeidate outcome evaluated at the end of stage 2.

• A3: Action of stage-3 treatment.

• N: Number of subjects following a certain treatment sequence.

• MEAN: Sequence-specified sample mean.

• VAR: Sequence-specified sample variance.

References

Thall P., Millikan R. and Sung H.G. (2000), “Evaluating multiple treatment courses in clinical trials,” Statistics in Medicine, 19, 1011-1028

Lavori P. W. and Dawson R. (2000), “A design for testing clinical strategies: biased adaptive within-subject randomization,” Journal of the Royal Statistical Society, Series A, 163, 29-38.

Murphy, S. A. (2005), “An experimental design for the development of adaptive treatment strategies,” Statistics in Medicine, 24, 1455-1481.

Examples

#get descriptive statistics
seqmeans(data=codiacs,family="gaussian",plot="d",digits=2,pch=c(18,14),
xtext=1:4,xlab="SMARTAR design")
seqmeans(data=codiacs, family="gaussian", plot="s",digits=2,
color="lightblue",
title="Primary outcome",ylab="Primary outcome")

SMARTAR documentation built on July 31, 2020, 1:06 a.m.