In atorus-research/Tplyr: A Traceability Focused Grammar of Clinical Data Summary

' @section Last Updated By:

' Nathan Kosiba

' @section Last Update Date:

' 2/8/2021

A: Population data can be specified by the user
B: Treatment variable can be manually specified for population data
C: Population data subset can be specified on user specified conditions
D: Header N counts will be specified by combining different subgroups available within the population data
E: Manual groups can be specified by combining different subgroups
F: Analysis data can be specified by the user
G: Analysis data subset can be specified on user specified conditions
H: Treatment variable can be manually specified for analysis data
I: n counts of values within a variable can be produced
J: n counts of values within a group of variables can be produced
K: counts can be produced using a pair of nested variables
L: Total n counts can be added
M: Total row sort value can be specified by the user
N: Missing n count handling can be specified including presentation and denominator handling
O: Missing row sort value can be specified by the user
P: Dummy values can be specified for categories that need to be presented but may not exist within the data
Q: Counts can be produced as n (%)
R: When producing n (%), the denominator can be specified using the analysis data
S: When producing n (%), the denominator can be specified using a particular manually specified subset
T: When producing n (%), the denominator can be specified using the population data
U: When producing n (%), the denominator can be specified using grouping of variables
V: Risk difference including confidence interval can be produced based on specified treatment groupings
W: Risk difference arguments can be passed forward into prop.test using args parameter
X: Risk difference can be calculated over user specified cols arguments
Y: Risk difference can be calculated over nested count layers and by variables
Z: The descriptive statistic of n can be produced based on an input variable
AA: The descriptive statistic of mean can be produced based on an input variable
AB: The descriptive statistic of median can be produced based on an input variable
AC: The descriptive statistic of IQR/Q1/Q3 can be produced based on an input variable
AD: The descriptive statistic of standard deviation can be produced based on an input variable
AE: The descriptive statistic of variance can be produced based on an input variable
AF: The descriptive statistic of min can be produced based on an input variable
AG: The descriptive statistic of max can be produced based on an input variable
AH: The descriptive statistic of missing can be produced based on an input variable
AI: Custom descriptive statistics can be produced based on an input variable and a specified formula
AJ: Descriptive statistics can be performed across discrete values within a grouping variable or a group of grouping variables
AK: Multiple statistics can be presented in one line (i.e. combining Q1, Q3 or Min, Max)
AL: Decimal precision can be specified by the user
AM: Integer length can be specified by the user
AN: Decimal precision can be dynamically created from analysis data
AO: Integer length can be dynamically created from analysis data
AP: Presentation format can be specified by the user including desired non-numeric text
AQ: Strings are built to align per user specification within a display
AR: Descriptive statistic missing values can be set to a user specified string
AS: Shift n counts of values using two variables, a 'from' and a 'to' variable, can be produced
AT: Shift n counts of values within a variable can be produced
AU: Shift n counts of values within a group of variables can be produced
AV: Dummy values for shift counts can be specified for categories that need to be presented but may not exist within the data
AW: Shift counts can be produced as n (%)
AX: For shift counts when producing n (%), the denominator can be specified using the analysis data
AY: For shift counts when producing n (%), the denominator can be specified using a particular manually specified subset
AZ: For shift counts when producing n (%), the denominator can be specified using the population data
BA: For shift counts when producing n (%), the denominator can be specified using a grouping of variables
BB: Row labels can be manually specified by the user
BC: Row labels can be nested to put a subgroup within a parent group
BD: Summaries can be stacked on top of one another
BE: Summaries can be sorted based on manual sorting by presentation specified order
BF: Summaries can be sorted based on count based sorting (either ascending or descending) by a specified treatment group
BG: Summaries can be sorted based on alphabetical sorting based on data values
BH: Summaries can be sorted based on a numeric version of the target variable if available
BI: Summary by variables will be sorted by a numeric variable if available and then by factor
BJ: Nested layers can be sorted independently using different methods
BK: Independent layers can be sorted using different methods and stacked using common sorting variables
BL: Count layer default formats can be set at the table level
BM: Descriptive statistics layer default formats can be set at the table level
BN: Shift layer default formats can be set at the table level
BO: Option for count layer default formats can be specified by the user
BP: Option for descriptive statistics layer default formats can be specified by the user
BQ: Option for shift layer default formats can be specified by the user
BR: Option for a cap on auto precision can be specified by the user
BS: Option for custom descriptive statistics can be specified by the user for use in the table
BT: Option for setting scipen internal option can be specified by the user
BU: Option for setting quantile algorithm choice can be specified by the user
BV: Option for setting IBM Rounding can be specified by the user
BW: Column headers can be added to the output object
BX: Row breaks can be added between sections based on grouping variables
BY: Row labels can be masked in a hierarchical fashion
BZ: A table object is returned in a format that is ready to be cosmetically prepared
CA: Count layers can process a cols argument and separate population data passed from the table level along with normal count layer processing
CB: Count layers can process a cols argument, separate population data, and a defined subset passed from the table level along with normal count layer processing