Skip to contents

Extract Common ADSL Variables Across Analysis Domains

extract_common_adsl_variables.Rd

This function identifies variables from the ADSL (Subject-Level Analysis Dataset) that are also present in other analysis domains. It provides comprehensive information about variable reuse and consistency across study datasets, which is essential for data validation and understanding variable provenance in clinical studies.

Usage

extract_common_adsl_variables(
  variable_info_df,
  adsl_dataset_name = "ADSL",
  include_sas_info = TRUE,
  min_domains_count = 1,
  sort_by = "domains_count"
)

Arguments

variable_info_df

A data.frame containing variable information extracted from define.xml. Expected columns include: OID, Name, SASFieldName, DataType, Length, Description, Origin, CodelistOID, Dataset, and others.

adsl_dataset_name

Character string specifying the name of the subject-level dataset. Default is "ADSL".

include_sas_info

Logical indicating whether to include SAS-specific information in the output. Default is TRUE.

min_domains_count

Integer specifying minimum number of domains a variable must appear in to be included in results. Default is 1.

sort_by

Character string specifying how to sort results. Options are "domains_count" (default), "variable_name", or "data_type".

Value

A data.frame with the following columns:

Variable

Variable name from ADSL

SASFieldName

SAS field name (if include_sas_info = TRUE)

DataType

Data type (text, integer, float)

Length

Variable length specification

Description

Variable description from define.xml

Origin

Variable origin (Predecessor, Derived, Assigned)

CodelistOID

Reference to codelist if applicable

DomainsCount

Number of domains containing this variable

DomainsFound

Comma-separated list of domain names

SASLength

SAS length specification (if include_sas_info = TRUE)

Details

The function performs the following steps:

  1. Validates input parameters and data structure

  2. Extracts variables from the specified ADSL dataset

  3. Identifies matching variables in other analysis domains

  4. Counts domain occurrences for each common variable

  5. Formats and sorts results according to specified criteria

Common ADSL variables typically include:

  • Study identifiers (STUDYID, USUBJID, SUBJID)

  • Demographics (AGE, SEX, RACE, etc.)

  • Treatment assignments (ARM, TRT01P, TRT01A, etc.)

  • Study dates (TRTSDT, TRTEDT, etc.)

  • Baseline characteristics

See also

extract_variable_info_from_define for extracting variable information from define.xml

Author

Clinical Data Science Team

Examples

if (FALSE) { # \dontrun{
# Using the included define.xml file (recommended)
define_path <- system.file("define.xml", package = "adrgOS")
if (file.exists(define_path)) {
  variable_info <- extract_variable_info_from_define(define_path)
  common_vars <- extract_common_adsl_variables(variable_info)
  print(common_vars)

  # Filter by minimum domain count
  high_freq_vars <- extract_common_adsl_variables(
    variable_info,
    min_domains_count = 3
  )
  print(high_freq_vars)
}

# Alternative: Using custom sample data
sample_data <- data.frame(
  Name = c("STUDYID", "USUBJID", "AGE", "SEX", "STUDYID", "USUBJID", "AVAL",
           "STUDYID", "USUBJID", "PARAMCD"),
  Dataset = c("ADSL", "ADSL", "ADSL", "ADSL", "ADAE", "ADAE", "ADAE",
              "ADLBC", "ADLBC", "ADLBC"),
  DataType = c("text", "text", "integer", "text", "text", "text", "float",
               "text", "text", "text"),
  Length = c("12", "20", "3", "1", "12", "20", "8", "12", "20", "8"),
  Description = c("Study Identifier", "Unique Subject ID", "Age", "Sex",
                  "Study Identifier", "Unique Subject ID", "Analysis Value",
                  "Study Identifier", "Unique Subject ID", "Parameter Code"),
  Origin = c("Assigned", "Assigned", "Collected", "Collected",
             "Assigned", "Assigned", "Derived", "Assigned", "Assigned", "Derived"),
  stringsAsFactors = FALSE
)

# Basic usage
result <- extract_common_adsl_variables(sample_data)
print(result)

# Filter by minimum domain count (now works with improved sample data)
result_filtered <- extract_common_adsl_variables(
  sample_data,
  min_domains_count = 2
)
print(result_filtered)

# Sort by variable name without SAS info
result_sorted <- extract_common_adsl_variables(
  sample_data,
  include_sas_info = FALSE,
  sort_by = "variable_name"
)
print(result_sorted)
} # }