Title: Nested Data Summary, Adverse Events and REDCap
Version: 1.0.2
Description: Tools and code snippets for summarizing nested data, adverse events and REDCap study information.
License: MIT + file LICENSE
Depends: R (≥ 4.2)
Imports: afex, coda, cowplot, dplyr, forcats, ggh4x, ggplot2, lifecycle, MCMCglmm, modeest, openxlsx, plyr, purrr, reportRmd, rlang, rstatix, stats, stringi, stringr, tibble, utils
Suggests: testthat (≥ 3.0.0)
Config/testthat/edition: 3
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.2.3
Collate: 'data.R' 'globals.R' 'as_numeric_parse.R' 'covsum_nested.R' 'dsmb_ccru.R' 'nice_mcmcglmm.R' 'nice_mcmcglmm_icc.R' 'ae_timeline_plot.R' 'caterpillar_plot.R' 'redcap_data_out.R'
NeedsCompilation: no
Packaged: 2025-02-03 05:26:49 UTC; t109810uhn
Author: Lisa Avery ORCID iD [aut], Zeynep Baskurt [aut], Osvaldo Espin-Garcia [aut], Katrina Hueniken [aut], Katherine Lajkosz ORCID iD [aut], Xuan Li [aut], Tyler Pittman ORCID iD [cre, aut], Anna Santiago ORCID iD [aut], Jessica Weiss [aut], Wei Xu [aut]
Maintainer: Tyler Pittman <tyler.pittman@uhn.ca>
Repository: CRAN
Date/Publication: 2025-02-03 08:00:10 UTC

Simulated adverse events for patients receiving two study agents.

Description

Simulated adverse events for patients receiving two study agents.

Usage

ae

Format

A data frame with 394 rows and 9 variables:

Subject

Patient ID

ae_detail

Adverse event detail, also known as lowest level term

ae_category

Adverse event category, also known as system organ class

CTCAE5_LLT_NM

Common Terminology Criteria for Adverse Events (CTCAE) version 5

AE_VERBATIM_TRM_TXT

Adverse event verbatim text entered by clinical registered nurse, for "Other, specify"

AE_SEV_GD

Adverse event severity grade, scale from 1 to 5

AE_ONSET_DT_INT

Adverse event onset date

CTC_AE_ATTR_SCALE

Attribution scale of adverse event to first study agent

CTC_AE_ATTR_SCALE_1

Attribution scale of adverse event to second study agent

Outputs related adverse event timeline plots including just system organ class (AE category), or system organ class and lowest level term (AE detail). This function can fit up to 5 different attributions. Modify width, height and scale parameters in ggsave() to customize fit for large plot.

Description

Outputs related adverse event timeline plots including just system organ class (AE category), or system organ class and lowest level term (AE detail). This function can fit up to 5 different attributions. Modify width, height and scale parameters in ggsave() to customize fit for large plot.

Usage

ae_timeline_plot(
  subjID,
  subjID_ineligText = NULL,
  baseline_datasets,
  ae_dataset,
  ae_attribVars,
  ae_attribVarsName = NULL,
  ae_attribVarText = NULL,
  startDtVars,
  ae_detailVar,
  ae_categoryVar,
  ae_severityVar,
  ae_onsetDtVar,
  time_unit = c("day", "week", "month", "year"),
  include_ae_detail = TRUE,
  legendPerSpace = NULL,
  fonts = NULL,
  fontColours = NULL,
  panelColours = NULL,
  attribColours = NULL,
  attribSymbols = NULL,
  columnWidths = NULL
)

Arguments

subjID

key identifier field for participant ID in data sets

subjID_ineligText

character text that denotes participant IDs to exclude. For example, c("New Subject") (if provided)

baseline_datasets

list of data frames that contain baseline participant characteristics. For example, list(enrollment_DF,demography_DF,ineligibility_DF)

ae_dataset

data frame that contains subject AEs

ae_attribVars

field(s) that denotes attribution to intervention under study. For example, c("CTC_AE_ATTR_SCALE","CTC_AE_ATTR_SCALE_1") (if provided)

ae_attribVarsName

character text that denotes name of interventions under study. For example, c("Drug 1", "Drug 2") (if provided)

ae_attribVarText

character text that denotes related attribution. For example c("Definite", "Probable", "Possible") (if provided)

startDtVars

field(s) that denotes participant start date (i.e. 10MAY2021). For example, it could be enrollment date or screening date. If more than one field given (unique names are required), each field is assumed to be specific start date for attribution in corresponding field order

ae_detailVar

field that denotes participant AE detail (lowest level term)

ae_categoryVar

field that denotes participant AE category (system organ class)

ae_severityVar

field that denotes participant AE severity grade (numeric)

ae_onsetDtVar

field that denotes participant AE onset date

time_unit

character text that denotes time unit for desired timeline. For example, could be one of c("day","week","month","year") (if provided)

include_ae_detail

boolean that denotes if AE detail should be included in timeline plot. Default is True

legendPerSpace

parameter at denotes proportion of vertical image space dedicated to legend at bottom. Default is 0.05 for AE detail and 0.1 for AE Category

fonts

character text that denotes font for AE category, AE detail, axis, legend and plot labels (if provided)

fontColours

character text that denotes system font colours for AE category and AE detail (if provided)

panelColours

character text that denotes panel background colours for AE category, AE detail and plot area (if provided)

attribColours

character text that denotes colours for attributions, supports up to 10 distinct colours (if provided)

attribSymbols

text that denotes median plot symbols for attributions, supports up to 10 distinct symbols (if provided)

columnWidths

text that denotes character columns widths for AE category and AE detail columns (if provided)

Value

ggplot object of AE timeline plot

Examples

data("drug1_admin", "drug2_admin", "ae");
p <- ae_timeline_plot(subjID="Subject",subjID_ineligText=c("01","11"),
                      baseline_datasets=list(drug1_admin, drug2_admin),
                      ae_dataset=ae,
                      ae_attribVars=c("CTC_AE_ATTR_SCALE","CTC_AE_ATTR_SCALE_1"),
                      ae_attribVarsName=c("Drug 1","Drug 2"),
                      ae_attribVarText=c("Definite", "Probable", "Possible"),
                      startDtVars=c("TX1_DATE_INT","TX2_DATE_INT"),
                      ae_detailVar="ae_detail",
                      ae_categoryVar="ae_category",ae_severityVar="AE_SEV_GD",
                      ae_onsetDtVar="AE_ONSET_DT_INT",time_unit="month",
                      include_ae_detail=FALSE,
                      fonts=c("Calibri","Albany AMT","Gadugi","French Script MT","Forte"),
                      fontColours=c("#FFE135"),
                      panelColours=c("#E52B50",NA,"#FFE4C4"),
                      attribColours=c("#9AB973","#01796F","#FFA343","#CC7722"),   
                      attribSymbols=c(7,8,5,6),
                      columnWidths=c(23))

Modification of the as.numeric function that prints entries that fail to parse as a message

Description

Modification of the as.numeric function that prints entries that fail to parse as a message

Usage

as_numeric_parse(x)

Arguments

x

string or vector to coerce to numeric

Value

No return value, called for side effects

Examples

z <- as_numeric_parse(c(1:5, "String1",6:10,"String2"))
z

Caterpillar plot. Useful for plotting random effects from hierarchical models, such as MCMCglmm::MCMCglmm() object, that have binary outcome.

Description

Caterpillar plot. Useful for plotting random effects from hierarchical models, such as MCMCglmm::MCMCglmm() object, that have binary outcome.

Usage

caterpillar_plot(
  subjID,
  subjLabel = NULL,
  remove.text.subjID = FALSE,
  mcmcglmm_object,
  orig_dataset,
  binaryOutcomeVar,
  prob = NULL,
  title = NULL,
  no.title = FALSE,
  subtitle = NULL,
  ncol = NULL,
  fonts = NULL,
  columnTextWidth = NULL,
  break.label.summary = FALSE
)

Arguments

subjID

key identifier field for participant ID in data sets

subjLabel

text label field in dataset to replace key identifier field for participant ID with in plot (if provided)

remove.text.subjID

boolean indicating if non-numeric text should be removed from subjID in plot label. Note that this can only be used if there are non-duplicate participant IDs when non-numeric text is removed. Default is FALSE (if provided)

mcmcglmm_object

MCMCglmm model output

orig_dataset

data frame supplied to MCMCglmm function

binaryOutcomeVar

name of binary variable (0,1) that denotes outcome in MCMCglmm model

prob

probability for highest posterior density interval, similar to a confidence interval. Default is 0.95 (if provided)

title

title of the plot. Overrides default title (if provided)

no.title

boolean that denotes if title should be outputted in plot. Default is TRUE (if provided)

subtitle

subtitle of the plot. Overrides default subtitle (if provided)

ncol

number of columns in plot. Default is 2 (if provided)

fonts

character text that denotes font for title, subtitle, category labels, x-axis plot labels (if provided)

columnTextWidth

numeric that denotes character width for label text before breaking to start new line. Default is 20 characters (if provided)

break.label.summary

boolean to indicate if new line should start in label before (n, event) summary. Default is FALSE

Value

ggplot object of caterpillar plot

Examples

data("ae");

ae$G3Plus <- 0;
ae$G3Plus[ae$AE_SEV_GD %in% c("3", "4", "5")] <- 1;
ae$Drug_1_Attribution <- 0;
ae$Drug_1_Attribution[ae$CTC_AE_ATTR_SCALE %in% c("Definite", "Probable", "Possible")] <- 1;
ae$Drug_2_Attribution <- 0;
ae$Drug_2_Attribution[ae$CTC_AE_ATTR_SCALE_1 %in% c("Definite", "Probable", "Possible")] <- 1;

prior2RE <- list(R = list(V = diag(1), fix = 1),
  G=list(G1=list(V=1, nu=0.02), G2=list(V=1, nu=0.02)));
  
model1 <- MCMCglmm::MCMCglmm(G3Plus ~ Drug_1_Attribution + Drug_2_Attribution, 
  random=~Subject + ae_category, family="categorical", data=ae, saveX=TRUE, 
  verbose=FALSE, burnin=2000, nitt=10000, thin=10, pr=TRUE, prior=prior2RE);
  
p <- caterpillar_plot(subjID = "Subject",
  mcmcglmm_object = model1,
  prob = 0.99,
  orig_dataset = ae,
  binaryOutcomeVar = "G3Plus")
  
p <- caterpillar_plot(subjID = "ae_category",
  mcmcglmm_object = model1,
  prob = 0.95,
  orig_dataset = ae,
  remove.text.subjID = FALSE,
  ncol = 4,
  binaryOutcomeVar = "G3Plus",
  subtitle = "System organ class (n, event)",
  title = "Odds Ratio for G3+ Severity with 95% Highest Posterior Density Interval",
  fonts = c("Arial", "Arial", "Arial", "Arial"),
  break.label.summary = TRUE)

Nested version of reportRmd covsum()

Description

Nested version of reportRmd covsum()

Usage

covsum_nested(
  data,
  covs,
  maincov = NULL,
  id = NULL,
  digits = 1,
  numobs = NULL,
  markup = TRUE,
  sanitize = TRUE,
  nicenames = FALSE,
  IQR = FALSE,
  all.stats = FALSE,
  pvalue = TRUE,
  effSize = TRUE,
  show.tests = TRUE,
  nCores = NULL,
  nested.test = NULL,
  nsim = NULL,
  excludeLevels = NULL,
  dropLevels = TRUE,
  full = TRUE,
  digits.cat = 0,
  testcont = c("rank-sum test", "ANOVA"),
  testcat = c("Chi-squared", "Fisher"),
  include_missing = FALSE,
  percentage = c("column", "row")
)

Arguments

data

dataframe containing data

covs

character vector with the names of columns to include in table

maincov

covariate to stratify table by

id

covariates to nest summary by

digits

number of digits for summarizing mean data, does not affect p-values

numobs

named list overriding the number of people you expect to have the covariate

markup

boolean indicating if you want latex markup

sanitize

boolean indicating if you want to sanitize all strings to not break LaTeX

nicenames

boolean indicating if you want to replace . and _ in strings with a space

IQR

boolean indicating if you want to display the inter quantile range (Q1,Q3) as opposed to (min,max) in the summary for continuous variables

all.stats

boolean indicating if all summary statistics (Q1,Q3 + min,max on a separate line) should be displayed. Overrides IQR.

pvalue

boolean indicating if you want p-values included in the table

effSize

boolean indicating if you want effect sizes included in the table. Can only be obtained if pvalue is also requested.

show.tests

boolean indicating if the type of statistical used should be shown in a column beside the p-values. Ignored if pvalue=FALSE.

nCores

if > 1, specifies number of cores to use for parallel processing for calculating the nested p-value (default: 1).

nested.test

specifies test used for calculating nested p-value from afex::mixed function. Either parametric bootstrap method or likelihood ratio test method (default: "LRT"). Parametric bootstrap takes longer.

nsim

specifies number of simulations to use for calculating nested p-value with parametric bootstrap method used for nested.test (default: 1000).

excludeLevels

a named list of covariate levels to exclude from statistical tests in the form list(varname =c('level1','level2')). These levels will be excluded from association tests, but not the table. This can be useful for levels where there is a logical skip (i.e. not missing, but not presented). Ignored if pvalue=FALSE.

dropLevels

logical, indicating if empty factor levels be dropped from the output, default is TRUE.

full

boolean indicating if you want the full sample included in the table, ignored if maincov is NULL

digits.cat

number of digits for the proportions when summarizing categorical data (default: 0)

testcont

test of choice for continuous variables,one of rank-sum (default) or ANOVA

testcat

test of choice for categorical variables,one of Chi-squared (default) or Fisher

include_missing

Option to include NA values of maincov. NAs will not be included in statistical tests

percentage

choice of how percentages are presented, one of column (default) or row

See Also

fisher.test,chisq.test, wilcox.test,kruskal.test, and anova

Simulated demography for patients.

Description

Simulated demography for patients.

Usage

demography

Format

A data frame with 12 rows and 2 variables:

Subject

Patient ID

GENDER_CODE

Patient gender

Simulated study agent 1 for patients.

Description

Simulated study agent 1 for patients.

Usage

drug1_admin

Format

A data frame with 12 rows and 2 variables:

Subject

Patient ID

TX1_DATE_INT

Study agent 1 start date of patient on study

Simulated study agent 2 for patients.

Description

Simulated study agent 2 for patients.

Usage

drug2_admin

Format

A data frame with 12 rows and 2 variables:

Subject

Patient ID

TX2_DATE_INT

Study agent 2 start date of patient on study

Outputs the three DSMB-CCRU AE summary tables in Excel format per UHN template

Description

Outputs the three DSMB-CCRU AE summary tables in Excel format per UHN template

Usage

dsmb_ccru(
  protocol,
  setwd,
  title,
  comp = NULL,
  pi,
  presDate,
  cutDate,
  boundDate = NULL,
  subjID,
  subjID_ineligText = NULL,
  baseline_datasets,
  ae_dataset,
  ineligVar = NULL,
  ineligVarText = NULL,
  genderVar,
  enrolDtVar,
  ae_detailVar,
  ae_categoryVar,
  ae_severityVar,
  ae_onsetDtVar,
  ae_detailOtherText = NULL,
  ae_detailOtherVar = NULL,
  ae_verbatimVar = NULL,
  numSubj = NULL,
  fileNameUnderscore = TRUE
)

Arguments

protocol

study protocol name (uppercase, no spaces permitted)

setwd

directory to write Excel summary files to

title

full character vector with name of study

comp

baseline comparison group. For example, cohort (if provided)

pi

character vector name of study principal investigator

presDate

presentation date (i.e. 17NOV2023) for DSMB

cutDate

recent cutoff date for AEs (i.e. 31AUG2023)

boundDate

lower bound cutoff date for AEs (if provided)

subjID

key identifier field for participant ID in data sets

subjID_ineligText

character text that denotes participant IDs to exclude, for example, c("New Subject") (if provided)

baseline_datasets

list of data frames that contain baseline participant characteristics, for example, list(enrollment_DF,demography_DF,ineligibility_DF)

ae_dataset

data frame that contains subject AEs

ineligVar

field that denotes participant ineligibility (if provided)

ineligVarText

character text that denotes participant ineligibility, for example, c("Yes", "Y") (if provided)

genderVar

field that denotes participant gender

enrolDtVar

field that denotes participant enrollment date (i.e. 10MAY2021)

ae_detailVar

field that denotes participant AE detail (lowest level term)

ae_categoryVar

field that denotes participant AE category (system organ class)

ae_severityVar

field that denotes participant AE severity grade (numeric)

ae_onsetDtVar

field that denotes participant AE onset date

ae_detailOtherText

character text that denotes referencing verbatim AE field, for example, c("Other, specify", "OTHER") (if provided)

ae_detailOtherVar

field that denotes participant AE detail other (if provided)

ae_verbatimVar

field that denotes participant AE detail verbatim (if provided)

numSubj

vector to override value for number of participants in summary (if provided)

fileNameUnderscore

boolean that denotes if spaces should be underscore in filename

Value

three Excel files containing DSMB-CCRU AE summary tables

Examples

data("enrollment", "demography", "ineligibility", "ae");
dsmb_ccru(protocol="EXAMPLE_STUDY",setwd="./man/tables/",
  title="Phase X Study to Evaluate Treatments A-D",
  comp="COHORT",pi="Dr. Principal Investigator",
  presDate="30OCT2020",cutDate="31AUG2020",
  boundDate=NULL,subjID="Subject",subjID_ineligText=c("New Subject","Test"),
  baseline_datasets=list(enrollment,demography,ineligibility),
  ae_dataset=ae,ineligVar="INELIGIBILITY_STATUS",ineligVarText=c("Yes","Y"),
  genderVar="GENDER_CODE",enrolDtVar="ENROL_DATE_INT",ae_detailVar="ae_detail",
  ae_categoryVar="ae_category",ae_severityVar="AE_SEV_GD",
  ae_onsetDtVar="AE_ONSET_DT_INT",ae_detailOtherText="Other, specify",
  ae_detailOtherVar="CTCAE5_LLT_NM",ae_verbatimVar="AE_VERBATIM_TRM_TXT",
  numSubj=c(2,4,5,6))

Enrollment data Simulated enrollment for patients.

Description

Enrollment data

Simulated enrollment for patients.

Usage

enrollment

Format

A data frame with 12 rows and 3 variables:

Subject

Patient ID

COHORT

Study cohort for patient

ENROL_DATE_INT

Enrollment date of patient to study

Simulated ineligibility for patients.

Description

Simulated ineligibility for patients.

Usage

ineligibility

Format

A data frame with 11 rows and 2 variables:

Subject

Patient ID

INELIGIBILITY_STATUS

Recorded ineligibility status of patient to study

Nice table of model output from MCMCglmm::MCMCglmm()

Description

Nice table of model output from MCMCglmm::MCMCglmm()

Usage

nice_mcmcglmm(mcmcglmm_object, dataset)

Arguments

mcmcglmm_object

returned output from MCMCglmm()

dataset

dataframe containing data

Value

grouped_df

Examples

## Not run: 
data(ae)

ae$AE_SEV_GD <- as.numeric(ae$AE_SEV_GD);
ae$Drug_1_Attribution <- "No";
ae$Drug_1_Attribution[ae$CTC_AE_ATTR_SCALE %in% c("Definite", "Probable", "Possible")] <- "Yes";
ae$Drug_1_Attribution <- as.factor(ae$Drug_1_Attribution);
ae$Drug_2_Attribution <- "No";
ae$Drug_2_Attribution[ae$CTC_AE_ATTR_SCALE_1 %in% c("Definite", "Probable", "Possible")] <- "Yes";
ae$Drug_2_Attribution <- as.factor(ae$Drug_2_Attribution);

prior2RE <- list(R = list(V = diag(1), fix = 1), G=list(G1=list(V=1, nu=0.02), 
          G2=list(V=1, nu=0.02)));

model1 <- MCMCglmm::MCMCglmm(Drug_1_Attribution ~ AE_SEV_GD + Drug_2_Attribution, 
          random=~ae_detail + Subject, family="categorical", data=ae, saveX=TRUE, 
          verbose=FALSE, burnin=2000, nitt=10000, thin=10, pr=TRUE, prior=prior2RE);

mcmcglmm_mva <- nice_mcmcglmm(model1, ae);

## End(Not run)

Nice table of intraclass correlation coefficients from MCMCglmm::MCMCglmm() model output

Description

Nice table of intraclass correlation coefficients from MCMCglmm::MCMCglmm() model output

Usage

nice_mcmcglmm_icc(mcmcglmm_object, prob = NULL, decimals = NULL)

Arguments

mcmcglmm_object

returned output from MCMCglmm()

prob

probability for highest posterior density interval, similar to a confidence interval. Default is 0.95 (if provided)

decimals

number of decimal places to use in estimates

Value

grouped_df

Examples

## Not run: 
data(ae)
ae$AE_SEV_GD <- as.numeric(ae$AE_SEV_GD);
ae$Drug_1_Attribution <- 0;
ae$Drug_1_Attribution[ae$CTC_AE_ATTR_SCALE %in% c("Definite", "Probable", "Possible")] <- 1;
ae$Drug_2_Attribution <- 0;
ae$Drug_2_Attribution[ae$CTC_AE_ATTR_SCALE_1 %in% c("Definite", "Probable", "Possible")] <- 1;
prior2RE <- list(R = list(V = diag(1), fix = 1), G=list(G1=list(V=1, nu=0.02), 
   G2=list(V=1, nu=0.02)));
model1 <- MCMCglmm::MCMCglmm(Drug_1_Attribution ~ AE_SEV_GD + Drug_2_Attribution, 
   random=~ae_detail + Subject, family="categorical", data=ae, saveX=TRUE, 
   verbose=FALSE, burnin=2000, nitt=10000, thin=10, pr=TRUE, prior=prior2RE);
mcmcglmm_icc <- nice_mcmcglmm_icc(model1);

## End(Not run)

Combines exported REDCap raw and label .csv files together with data dictionary. Tranforms the exported data into Excel sheets by survey instrument with one row per participant

Description

Combines exported REDCap raw and label .csv files together with data dictionary. Tranforms the exported data into Excel sheets by survey instrument with one row per participant

Usage

redcap_data_out(
  protocol,
  pullDate = NULL,
  subjID,
  subjID_ineligText = NULL,
  subjID_eligPattern = NULL,
  varFilter = NULL,
  varFilter_eligPattern = NULL,
  setWD_files,
  setWD_dataDict = NULL,
  outDir
)

Arguments

protocol

study protocol name (i.e. Example_Study)

pullDate

date of data pull. For example, 2024_01_02 (if provided)

subjID

key identifier field(s) for participant ID in data sets

subjID_ineligText

character text that denotes participant IDs to exclude using first key identifier field. For example, c("New Subject") (if provided)

subjID_eligPattern

character text that denotes pattern for participant IDs to include using first key identifier field. For example, c("^Site_A") (if provided)

varFilter

field to use for filtering data (if provided)

varFilter_eligPattern

character text that denotes pattern for filter variable to include. For example, c("^Arm_A") (if provided)

setWD_files

directory where the both raw and label REDCap export .csv files are stored, following the convention for file names of 1_DATA.csv, 1_DATA_LABELs.csv, 2_DATA.csv, 2_DATA_LABELs.csv, etc

setWD_dataDict

directory where the REDCap .csv data dictionary is stored. Make sure that file is saved as CSV UTF-8 comma delimited. Must contain "ictionary" in file name (if provided)

outDir

output directory where the Excel files are saved

Value

two Excel files, one containing variable names and labels and the other containing REDCap survey instrument data by sheet

Examples

## Not run: 
redcap_data_out(protocol="Example_Study",pullDate="2024_01_03",

## End(Not run)

Outputs a nested version of reportRmd::rm_covsum()

Description

Outputs a nested version of reportRmd::rm_covsum()

Usage

rm_covsum_nested(
  data,
  covs,
  maincov = NULL,
  id = NULL,
  caption = NULL,
  tableOnly = FALSE,
  covTitle = "",
  digits = 1,
  digits.cat = 0,
  nicenames = FALSE,
  IQR = FALSE,
  all.stats = FALSE,
  pvalue = TRUE,
  effSize = TRUE,
  p.adjust = "none",
  unformattedp = FALSE,
  show.tests = TRUE,
  just.nested.pvalue = FALSE,
  nCores = NULL,
  nested.test = NULL,
  nsim = NULL,
  testcont = c("rank-sum test", "ANOVA"),
  testcat = c("Chi-squared", "Fisher"),
  full = TRUE,
  include_missing = FALSE,
  percentage = c("column", "row"),
  dropLevels = TRUE,
  excludeLevels = NULL,
  numobs = NULL,
  markup = TRUE,
  sanitize = TRUE,
  chunk_label
)

Arguments

data

dataframe containing data

covs

character vector with the names of columns to include in table

maincov

covariate to stratify table by

id

covariates to nest summary by

caption

character containing table caption (default is no caption)

tableOnly

Logical, if TRUE then a dataframe is returned, otherwise a formatted printed object is returned (default).

covTitle

character with the names of the covariate (predictor) column. The default is to leave this empty for output or, for table only output to use the column name 'Covariate'.

digits

number of digits for summarizing mean data

digits.cat

number of digits for the proportions when summarizing categorical data (default: 0)

nicenames

boolean indicating if you want to replace . and _ in strings with a space

IQR

boolean indicating if you want to display the inter quantile range (Q1,Q3) as opposed to (min,max) in the summary for continuous variables

all.stats

boolean indicating if all summary statistics (Q1,Q3 + min,max on a separate line) should be displayed. Overrides IQR.

pvalue

boolean indicating if you want p-values included in the table

effSize

boolean indicating if you want effect sizes included in the table. Can only be obtained if pvalue is also requested.

p.adjust

p-adjustments to be performed

unformattedp

boolean indicating if you would like the p-value to be returned unformatted (ie not rounded or prefixed with '<'). Best used with tableOnly = T and outTable function.

show.tests

boolean indicating if the type of statistical used should be shown in a column beside the p-values. Ignored if pvalue=FALSE.

just.nested.pvalue

boolean indicating if the just the nested p-value should be shown in a column, and not unnested p-value, unnested statistical tests and effect size. Overrides effSize and show.tests arguments.

nCores

number of cores to use for parallel processing if calculating the nested p-value (if provided).

nested.test

specifies test used for calculating nested p-value from afex::mixed function. Either parametric bootstrap method or likelihood ratio test method (default: "LRT"). Parametric bootstrap takes longer.

nsim

specifies number of simulations to use for calculating nested p-value with parametric bootstrap method used for nested.test (default: 1000).

testcont

test of choice for continuous variables,one of rank-sum (default) or ANOVA

testcat

test of choice for categorical variables,one of Chi-squared (default) or Fisher

full

boolean indicating if you want the full sample included in the table, ignored if maincov is NULL

include_missing

Option to include NA values of maincov. NAs will not be included in statistical tests

percentage

choice of how percentages are presented, one of column (default) or row

dropLevels

logical, indicating if empty factor levels be dropped from the output, default is TRUE.

excludeLevels

a named list of covariate levels to exclude from statistical tests in the form list(varname =c('level1','level2')). These levels will be excluded from association tests, but not the table. This can be useful for levels where there is a logical skip (ie not missing, but not presented). Ignored if pvalue=FALSE.

numobs

named list overriding the number of people you expect to have the covariate

markup

boolean indicating if you want latex markup

sanitize

boolean indicating if you want to sanitize all strings to not break LaTeX

chunk_label

only used if output is to Word to allow cross-referencing

Value

A character vector of the table source code, unless tableOnly=TRUE in which case a data frame is returned

See Also

fisher.test, chisq.test, wilcox.test, kruskal.test, and anova

Examples

## Not run: 
# Example 1
data(ae)
rm_covsum_nested(data = ae, id = c("ae_detail", "Subject"), covs = c("AE_SEV_GD", 
"AE_ONSET_DT_INT"), maincov = "CTC_AE_ATTR_SCALE")

# Example 2: set variable labels and other options, save output with markup
data("ae")
lbls <- data.frame(c1=c('AE_SEV_GD','AE_ONSET_DT_INT'),
    c2=c('Adverse event severity grade','Adverse event onset date'))
ae$AE_SEV_GD <- as.numeric(ae$AE_SEV_GD)
ae <- reportRmd::set_labels(ae, lbls)
output_tab <- rm_covsum_nested(data = ae, id = c("ae_detail", "Subject"), 
    covs = c("AE_SEV_GD", "AE_ONSET_DT_INT"), maincov = "CTC_AE_ATTR_SCALE", 
    testcat = "Fisher", percentage = c("col"), show.tests = FALSE, pvalue = TRUE, 
    effSize = FALSE, full = TRUE, IQR = FALSE, nicenames = TRUE, sanitize = TRUE, 
    markup = TRUE, include_missing = TRUE, just.nested.pvalue = TRUE, 
    tableOnly = TRUE)
cat(reportRmd::outTable(tab=output_tab))
cat(reportRmd::outTable(output_tab, format="html"), file = paste("./man/tables/", 
    "output_tab.html", sep=""))
cat(reportRmd::outTable(output_tab, format="latex"), file = paste("./man/tables/", 
    "output_tab.tex", sep=""))

## End(Not run)