ctrdata (version 1.25.0)

dfMergeVariablesRelevel: Merge variables, keeping type where possible, optionally relevel factors

Description

Merge variables in a data frame such as returned by dbGetFieldsIntoDf into a new variable, and optionally also map its values to new levels. See ctrdata-trial-concepts for pre-defined cross-register concepts that are already implemented based on merging fields from different registers and calculating a new field.

Usage

dfMergeVariablesRelevel(df = NULL, colnames = "", levelslist = NULL)

Value

A vector, with the type of the columns to be merged

Arguments

df

A data.frame with the variables (columns) to be merged into one vector.

colnames

A vector of names of columns in `df` that hold the variables to be merged, or a selection of columns as per select.

levelslist

A names list with one slice each for a new value to be used for a vector of old values (optional).

Examples

dbc <- nodbi::src_sqlite(
  dbname = system.file("extdata", "demo.sqlite", package = "ctrdata"),
  collection = "my_trials",
  flags = RSQLite::SQLITE_RO)

df <- dbGetFieldsIntoDf(
  fields = c(
    "ctrname",
    "protocolSection.eligibilityModule.healthyVolunteers",
    "f31_healthy_volunteers",
    "eligibility.healthy_volunteers",
    "participants.participantType",
    paste0(
      "authorizedPartI.trialDetails.trialInformation.",
      "populationOfTrialSubjects.clinicalTrialGroups.name"
    ),
    paste0(
      "authorizedApplication.authorizedPartI.trialDetails.trialInformation.",
      "populationOfTrialSubjects.clinicalTrialGroups.name"
    )
  ),
  con = dbc
)

df$isrctnhealthy <- "Healthy volunteer" == df[, 6]
df$ctis1healthy <- "Healthy volunteers" == df[ ,7]
df$ctis2healthy <- "Healthy volunteers" == df[, 8]

table(
  df$ctrname,
  dfMergeVariablesRelevel(
    df = df,
    colnames = 'matches("healthy")'
))