ADSL_Subject_Level_Analysis_Data

# Load necessary libraries
library(dplyr)
library(lubridate)

# Demographic Variables
dm1 <- sdtm.dm %>%
  mutate(
    AETHNIC = ifelse(is.na(ETHNIC), "NOT COLLECTED", ETHNIC),
    ARACE = case_when(
      RACE == "WHITE" ~ "W",
      RACE == "BLACK OR AFRICAN AMERICAN" ~ "B",
      RACE == "NATIVE HAWAIIAN OR OTHER PACIFIC ISLANDERS" ~ "HP",
      RACE == "ASIAN" ~ "A",
      RACE == "AMERICAN INDIAN OR ALASKA AMERICAN" ~ "AA",
      TRUE ~ NA_character_
    ),
    TRTSDT = as.Date(substr(RFSTDTC, 1, 10), format = "%Y-%m-%d"),
    TRTSTM = hms::as_hms(substr(RFSTDTC, 12, 16)),
    TRTEDT = as.Date(substr(RFENDTC, 1, 10), format = "%Y-%m-%d"),
    TRTETM = hms::as_hms(substr(RFENDTC, 12, 16)),
    TRT01P = ifelse(ARMCD == "DRUG A", "TRTA", NA_character_),
    TRT01A = ifelse(ACTARMCD == "DRUG A", "TRTA", NA_character_),
    SAFFL = ifelse(!is.na(RFSTDTC), "Y", "N"),
    FASFL = ifelse(!is.na(RFSTDTC), "Y", "N")
  )

# Merge dm1 and sdtm.ds to get DSDECOD and DSSTDTC
dm2 <- left_join(dm1, sdtm.ds %>% select(USUBJID, DSDECOD, DSSTDTC), by = "USUBJID") %>%
  mutate(
    EOSDT = as.Date(substr(DSSTDTC, 1, 10), format = "%Y-%m-%d"),
    EOSSTT = ifelse(DSDECOD == "COMPLETED", "COMPLETED", "DISCONTINUED")
  )

# Sort and remove duplicates from sdtm.EX
ex <- SDTM.EX %>%
  distinct(USUBJID, EXSTDTC, EXENDTC, .keep_all = TRUE) %>%
  arrange(USUBJID, EXSTDTC, EXENDTC)

# Split the data into ex1 and ex2 based on USUBJID
ex1 <- ex %>%
  group_by(USUBJID) %>%
  filter(row_number() == 1) %>%
  ungroup()

ex2 <- ex %>%
  group_by(USUBJID) %>%
  filter(row_number() == n()) %>%
  ungroup()

# Merge dm3 with ex1 and ex2
dm4 <- left_join(dm2, ex1 %>% select(USUBJID, EXSTDTC), by = "USUBJID")
dm5 <- left_join(dm4, ex2 %>% select(USUBJID, EXENDTC), by = "USUBJID")

# Final dataset processing
Final <- dm5 %>%
  mutate(
    TRT01SDTM = as_datetime(EXSTDTC),
    TRT01SDT = as.Date(substr(EXSTDTC, 1, 10), format = "%Y-%m-%d"),
    TRT01EDTM = as_datetime(EXENDTC),
    TRT01EDT = as.Date(substr(EXENDTC, 1, 10), format = "%Y-%m-%d")
  )

# Create ADAM.ADSL dataset with specified labels and lengths
ADSL <- Final %>%
  select(
    STUDYID, USUBJID, SUBJID, SITEID, BRTHDTC, AGE, AGEU, SEX, RACE, ARACE,
    ETHNIC, AETHNIC, SAFFL, FASFL, ARM, ARMCD, ACTARMCD, ACTARM, TRT01P, TRT01A,
    RFSTDTC, RFENDTC, TRTSDT, TRTSTM, TRT01SDTM, TRT01SDT, TRTEDT, TRTETM,
    TRT01EDTM, TRT01EDT, EOSSTT, EOSDT, COUNTRY
  )

# Save the data to ADAM.ADSL (assuming you want to write to disk)
write.csv(ADSL, "E://users/tiany/ADSL.csv", row.names = FALSE)