dispo var functions

NAMESPACE

@@ -10,6 +10,7 @@ export(convert_dtc_to_dt)
 export(convert_dtc_to_dtm)
 export(derive_aage)
 export(derive_baseline)
+export(derive_disposition_dt)
 export(derive_duration)
 export(derive_extreme_flag)
 export(derive_merged_vars)
@@ -36,7 +37,6 @@ export(filter_extreme)
 export(has_unique_records)
 export(impute_dtc)
 export(initialize)
-export(is_character)
 export(is_date)
 export(is_timeunit)
 export(is_valid_date_entry)

R/assertions.R

@@ -100,8 +100,9 @@ assert_has_only_one_baseline_record <- function(dataset, by) { # nolint
 #' @examples
 #' data(ex)
 #' has_unique_records(ex,
-#'                    by_vars = exprs(USUBJID) ,
-#'                    order = exprs(desc(EXENDTC)))
+#'   by_vars = exprs(USUBJID),
+#'   order = exprs(desc(EXENDTC))
+#' )
 has_unique_records <- function(dataset,
                                by_vars = NULL,
                                order = NULL,
@@ -150,9 +151,11 @@ has_unique_records <- function(dataset,
       # create message
       tbl <- capture.output(print(duplicates))
       if (missing(message)) {
-        message <- paste0("Dataset contains multiple records with respect to ",
-                          paste(all_vars_msg, collapse = ", "),
-                          ".")
+        message <- paste0(
+          "Dataset contains multiple records with respect to ",
+          paste(all_vars_msg, collapse = ", "),
+          "."
+        )
       }
       err_msg <- paste0(
         message,
@@ -169,7 +172,7 @@ has_unique_records <- function(dataset,
     }
     TRUE
   }
-  else{
+  else {
     FALSE
   }
 }
@@ -202,16 +205,19 @@ has_unique_records <- function(dataset,
 #' @examples
 #' data(ex)
 #' assert_has_unique_records(ex,
-#'                           by_vars = exprs(USUBJID) ,
-#'                           order = exprs(desc(EXENDTC)))
+#'   by_vars = exprs(USUBJID),
+#'   order = exprs(desc(EXENDTC))
+#' )
 assert_has_unique_records <- function(dataset,
                                       by_vars = NULL,
                                       order = NULL,
                                       message) {
-  has_unique_records(dataset = dataset,
-                     by_vars = by_vars,
-                     order = order,
-                     message_type = "error")
+  has_unique_records(
+    dataset = dataset,
+    by_vars = by_vars,
+    order = order,
+    message_type = "error"
+  )
 }
 
 #' Is Date/Date-time?
@@ -467,34 +473,6 @@ on_failure(is_valid_month) <- function(call, env) {
 }
 
 
-#' Is the object a character?
-#'
-#' Checks if a character vector was specified
-#'
-#' @param arg The argument to check
-#'
-#' @author Samia Kabi
-#'
-#' @return `TRUE` if the argument is a character, `FALSE` otherwise
-#'
-#' @export
-#'
-#' @examples
-#' date <- "2020-02-03"
-#' assertthat::assert_that(is_character(date))
-is_character <- function(arg) {
-  is.character(arg)
-}
-on_failure(is_character) <- function(call, env) {
-  paste0(
-    "Argument ",
-    deparse(call$arg),
-    " = ",
-    eval(call$arg, envir = env),
-    " is not a character."
-  )
-}
-
 is_named_exprs <- function(arg) {
   is.list(arg) && all(map_lgl(arg, is.language)) && all(names(arg) != "")
 }

R/derive_date_vars.R

@@ -203,7 +203,7 @@ impute_dtc <- function(dtc,
 #' convert_dtc_to_dt("2019-07")
 convert_dtc_to_dt <- function(dtc) {
   # Check dtc is character
-  assert_that(is_character(dtc))
+  assert_that(is.character(dtc))
   dt <- case_when(
     nchar(dtc) >= 10 & is_valid_dtc(dtc) ~ ymd(substr(dtc, 1, 10)),
     TRUE ~ ymd(NA)
@@ -237,7 +237,7 @@ convert_dtc_to_dt <- function(dtc) {
 #' convert_dtc_to_dtm("2019-07-18")
 convert_dtc_to_dtm <- function(dtc) {
   # Check dtc is character
-  assert_that(is_character(dtc))
+  assert_that(is.character(dtc))
   # note T00:00:00 is not printed in dataframe
   dtm <- case_when(
     nchar(dtc) == 19 & is_valid_dtc(dtc) ~ ymd_hms(dtc),
@@ -272,7 +272,7 @@ convert_dtc_to_dtm <- function(dtc) {
 #' compute_dtf(dtc = "2019", dt = as.Date("2019-07-18"))
 compute_dtf <- function(dtc, dt) {
   # Check dtc is character
-  assert_that(is_character(dtc))
+  assert_that(is.character(dtc))
   # check dt is a date
   assert_that(is_date(dt))
 
@@ -316,7 +316,7 @@ compute_dtf <- function(dtc, dt) {
 #' compute_tmf(dtc = "2019-07-18", dtm = as.POSIXct("2019-07-18"))
 compute_tmf <- function(dtc, dtm) {
   # Check dtc is character
-  assert_that(is_character(dtc))
+  assert_that(is.character(dtc))
   # check dt is a date
   assert_that(is_date(dtm))
 

R/derive_disposition_dt.R

@@ -0,0 +1,90 @@
+#' Derive a disposition date
+#'
+#' Derive a disposition status date from the the relevant records in the disposition domain.
+#'
+#' @param dataset Input dataset
+#'
+#' @param dataset_ds Datasets containing the disposition information (e.g.: ds)
+#'
+#' The variable specified in dtc parameter must be in dataset_ds
+#'
+#' @param new_var Name of the disposition date variable
+#'
+#' a variable name is expected
+#'
+#' @param dtc The character date used to derive/impute the disposition date
+#'
+#'   A character date is expected in a format like yyyy-mm-dd or yyyy-mm-ddThh:mm:ss.
+#'   If the year part is not recorded (missing date), no imputation is performed.
+#'
+#' @param filter_ds Filter condition for the disposition data.
+#'
+#'   Filter used to select the relevant disposition data.
+#'
+#'   Permitted Values: logical expression
+#'
+#' @param date_imputation The value to impute the day/month when a datepart is missing.
+#'
+#'   If NULL: no date imputation is performed and partial dates are returned as missing.
+#'
+#'   Otherwise, a character value is expected, either as a
+#'   - format with day and month specified as 'dd-mm': e.g. '15-06' for the 15th of June
+#'   - or as a keyword: 'FIRST', 'MID', 'LAST' to impute to the first/mid/last day/month
+#'
+#'   Default is NULL
+#'
+#' @return the input dataset with the disposition date (new_var) added
+#'
+#' @keywords adsl timing
+#'
+#' @author Samia Kabi
+#'
+#' @export
+#'
+#' @examples
+#' data("dm")
+#' data("ds")
+#' derive_disposition_dt(
+#'   dataset = dm,
+#'   dataset_ds = ds,
+#'   new_var = FRVDT,
+#'   dtc = DSSTDTC,
+#'   filter = expr(DSCAT == "OTHER EVENT" & DSDECOD == "FINAL RETRIEVAL VISIT")
+#' )
+derive_disposition_dt <- function(dataset,
+                                  dataset_ds,
+                                  new_var,
+                                  dtc,
+                                  filter_ds,
+                                  date_imputation = NULL) {
+  # Checks
+  warn_if_vars_exist(dataset, deparse(substitute(new_var)))
+  assert_that(is.data.frame(dataset_ds))
+  assert_has_variables(dataset_ds, deparse(substitute(dtc)))
+
+  # Process the disposition data
+  prefix <- sub("\\DT.*", "", deparse(substitute(new_var)))
+  newvar <- paste0(prefix, "DT")
+  ds_subset <- dataset_ds %>%
+    filter(!!filter_ds) %>%
+    mutate(datedtc___ = !!enquo(dtc)) %>%
+    derive_vars_dt(
+      new_vars_prefix = prefix,
+      dtc = datedtc___,
+      date_imputation = date_imputation,
+      flag_imputation = FALSE
+    ) %>%
+    select(STUDYID, USUBJID, !!enquo(new_var) := !!sym(newvar))
+
+  # Expect 1 record per subject - issue a warning otherwise
+  has_unique_records(
+    dataset = ds_subset,
+    by_vars = "USUBJID",
+    message_type = "warning",
+    message = "the filter used for DS results in several records per patient - please check"
+  )
+
+  # add the new dispo date to the input dataset
+  dataset %>%
+    left_join(ds_subset, by = c("STUDYID", "USUBJID"))
+}

R/globals.R

@@ -23,13 +23,15 @@ globalVariables(c(
   "ONTRTFL",
   "PARAMCD",
   "RANDDT",
+  "STUDYID",
   "TRTDURD",
   "TRTEDT",
   "TRTEDTM",
   "TRTSDT",
   "TRTSDTM",
   "USUBJID",
   "Dataset",
+  "datedtc___",
   "Derivation",
   "Derivation / Comment",
   "desc",

inst/example_scripts/ad_adsl.R

@@ -2,7 +2,7 @@
 #
 # Label: Subject Level Analysis Dataset
 #
-# Input: dm, ex
+# Input: dm, ex, ds
 #
 
 library(dplyr)
@@ -12,10 +12,13 @@ library(admiral)
 # read in predecessor variables from dm
 data("dm")
 data("ex")
+data("ds")
 
 # derive treatment variables (TRT01P, TRT01A)
 adsl <- dm %>%
-  mutate(TRT01P = ARMCD, TRT01A = ACTARMCD) %>%
+  mutate(
+    TRT01P = ARMCD, TRT01A = ARMCD
+  ) %>%
 
   # derive treatment start date (TRTSDTM, TRTSDT)
   derive_var_trtsdtm(dataset_ex = ex) %>%
@@ -28,6 +31,21 @@ adsl <- dm %>%
   # derive treatment duration (TRTDURD)
   derive_var_trtdurd() %>%
 
+  # Disposition date - option 1
+  derive_disposition_dt(
+    dataset_ds = ds,
+    new_var = EOSDT1,
+    dtc = DSSTDTC,
+    filter = expr(DSCAT == "DISPOSITION EVENT")
+  ) %>%
+  derive_disposition_dt(
+    dataset_ds = ds,
+    new_var = FRVDT,
+    dtc = DSSTDTC,
+    filter = expr(DSCAT == "OTHER EVENT" & DSDECOD == "FINAL RETRIEVAL VISIT")
+  ) %>%
+
+  # Option 2
   # derive study completion/discontinuation variables
   derive_merged_vars(
     dataset_add = ds,
@@ -36,6 +54,7 @@ adsl <- dm %>%
       EOSDT = convert_dtc_to_dt(impute_dtc(DSSTDTC, date_imputation = "FIRST")),
       EOSSTT = if_else(DSDECOD == "COMPLETED", "COMPLETED", "DISCONTINUED"),
       DCSREAS = if_else(DSDECOD != "COMPLETED", DSDECOD, "")
-    ))
+    )
+  )
 
 save(adsl, file = "data/adsl.rda", compress = TRUE)