This article describes creating an ADSL
ADaM. Examples are currently presented and tested using DM
, EX
, AE
, LB
and DS
SDTM domains. However, other domains could be used.
Note: All examples assume CDISC SDTM and/or ADaM format as input unless otherwise specified.
APxxSDT
, APxxEDT
, …)TRT0xP
, TRT0xA
)TRTSDT
, TRTEDT
, TRTDURD
)LSTALVDT
)To start, all data frames needed for the creation of ADSL
should be read into the environment. This will be a company specific process. Some of the data frames needed may be DM
, EX
, DS
, AE
, and LB
.
For example purpose, the CDISC Pilot SDTM datasets—which are included in pharmaversesdtm—are used.
library(admiral)
library(dplyr, warn.conflicts = FALSE)
library(pharmaversesdtm)
library(lubridate)
library(stringr)
dm <- pharmaversesdtm::dm
ds <- pharmaversesdtm::ds
ex <- pharmaversesdtm::ex
ae <- pharmaversesdtm::ae
lb <- pharmaversesdtm::lb
dm <- convert_blanks_to_na(dm)
ds <- convert_blanks_to_na(ds)
ex <- convert_blanks_to_na(ex)
ae <- convert_blanks_to_na(ae)
lb <- convert_blanks_to_na(lb)
The DM
domain is used as the basis for ADSL
:
APxxSDT
, APxxEDT
, …)
See the “Visit and Period Variables” vignette for more information.
If the variables are not derived based on a period reference dataset, they may be derived at a later point of the flow. For example, phases like “Treatment Phase” and “Follow up” could be derived based on treatment start and end date.
TRT0xP
, TRT0xA
)
The mapping of the treatment variables is left to the ADaM programmer. An example mapping for a study without periods may be:
For studies with periods see the “Visit and Period Variables” vignette.
TRTSDTM
, TRTEDTM
, TRTDURD
)
The function derive_vars_merged()
can be used to derive the treatment start and end date/times using the ex
domain. A pre-processing step for ex
is required to convert the variable EXSTDTC
and EXSTDTC
to datetime variables and impute missing date or time components. Conversion and imputation is done by derive_vars_dtm()
.
Example calls:
# impute start and end time of exposure to first and last respectively,
# do not impute date
ex_ext <- ex %>%
derive_vars_dtm(
dtc = EXSTDTC,
new_vars_prefix = "EXST"
) %>%
derive_vars_dtm(
dtc = EXENDTC,
new_vars_prefix = "EXEN",
time_imputation = "last"
)
#> The default value of `ignore_seconds_flag` will change to "TRUE" in admiral
#> 1.4.0.
adsl <- adsl %>%
derive_vars_merged(
dataset_add = ex_ext,
filter_add = (EXDOSE > 0 |
(EXDOSE == 0 &
str_detect(EXTRT, "PLACEBO"))) & !is.na(EXSTDTM),
new_vars = exprs(TRTSDTM = EXSTDTM, TRTSTMF = EXSTTMF),
order = exprs(EXSTDTM, EXSEQ),
mode = "first",
by_vars = exprs(STUDYID, USUBJID)
) %>%
derive_vars_merged(
dataset_add = ex_ext,
filter_add = (EXDOSE > 0 |
(EXDOSE == 0 &
str_detect(EXTRT, "PLACEBO"))) & !is.na(EXENDTM),
new_vars = exprs(TRTEDTM = EXENDTM, TRTETMF = EXENTMF),
order = exprs(EXENDTM, EXSEQ),
mode = "last",
by_vars = exprs(STUDYID, USUBJID)
)
This call returns the original data frame with the column TRTSDTM
, TRTSTMF
, TRTEDTM
, and TRTETMF
added. Exposure observations with incomplete date and zero doses of non placebo treatments are ignored. Missing time parts are imputed as first or last for start and end date respectively.
The datetime variables returned can be converted to dates using the derive_vars_dtm_to_dt()
function.
adsl <- adsl %>%
derive_vars_dtm_to_dt(source_vars = exprs(TRTSDTM, TRTEDTM))
Now, that TRTSDT
and TRTEDT
are derived, the function derive_var_trtdurd()
can be used to calculate the Treatment duration (TRTDURD
).
adsl <- adsl %>%
derive_var_trtdurd()
EOSDT
)
The functions derive_vars_dt()
and derive_vars_merged()
can be used to derive a disposition date. First the character disposition date (DS.DSSTDTC
) is converted to a numeric date (DSSTDT
) calling derive_vars_dt()
. The DS
dataset is extended by the DSSTDT
variable because the date is required by other derivations, e.g., RANDDT
as well. Then the relevant disposition date is selected by adjusting the filter_add
argument.
To add the End of Study date (EOSDT
) to the input dataset, a call could be:
# convert character date to numeric date without imputation
ds_ext <- derive_vars_dt(
ds,
dtc = DSSTDTC,
new_vars_prefix = "DSST"
)
adsl <- adsl %>%
derive_vars_merged(
dataset_add = ds_ext,
by_vars = exprs(STUDYID, USUBJID),
new_vars = exprs(EOSDT = DSSTDT),
filter_add = DSCAT == "DISPOSITION EVENT" & DSDECOD != "SCREEN FAILURE"
)
The ds_ext
dataset:
The adsl
dataset:
The derive_vars_dt()
function allows to impute partial dates as well. If imputation is needed and missing days are to be imputed to the first of the month and missing months to the first month of the year, set highest_imputation = "M"
.
EOSSTT
)
The function derive_vars_merged()
can be used to derive the End of Study status (EOSSTT
) based on DSCAT
and DSDECOD
from DS
. The relevant observations are selected by adjusting the filter_add
argument. A function mapping DSDECOD
values to EOSSTT
values can be defined and used in the new_vars
argument. The mapping for the call below is
"COMPLETED"
if DSDECOD == "COMPLETED"
NA_character_
if DSDECOD
is "SCREEN FAILURE"
"DISCONTINUED"
otherwiseExample function format_eosstt()
:
format_eosstt <- function(x) {
case_when(
x %in% c("COMPLETED") ~ "COMPLETED",
x %in% c("SCREEN FAILURE") ~ NA_character_,
TRUE ~ "DISCONTINUED"
)
}
The customized mapping function format_eosstt()
can now be passed to the main function. For subjects without a disposition event the end of study status is set to "ONGOING"
by specifying the missing_values
argument.
adsl <- adsl %>%
derive_vars_merged(
dataset_add = ds,
by_vars = exprs(STUDYID, USUBJID),
filter_add = DSCAT == "DISPOSITION EVENT",
new_vars = exprs(EOSSTT = format_eosstt(DSDECOD)),
missing_values = exprs(EOSSTT = "ONGOING")
)
This call would return the input dataset with the column EOSSTT
added.
If the derivation must be changed, the user can create his/her own function to map DSDECOD
to a suitable EOSSTT
value.
DCSREAS
, DCSREASP
)
The main reason for discontinuation is usually stored in DSDECOD
while DSTERM
provides additional details regarding subject’s discontinuation (e.g., description of "OTHER"
).
The function derive_vars_merged()
can be used to derive a disposition reason (along with the details, if required) at a specific timepoint. The relevant observations are selected by adjusting the filter_add
argument.
To derive the End of Study reason(s) (DCSREAS
and DCSREASP
), the function will map DCSREAS
as DSDECOD
, and DCSREASP
as DSTERM
if DSDECOD
is not "COMPLETED"
, "SCREEN FAILURE"
, or NA
, NA
otherwise.
adsl <- adsl %>%
derive_vars_merged(
dataset_add = ds,
by_vars = exprs(USUBJID),
new_vars = exprs(DCSREAS = DSDECOD, DCSREASP = DSTERM),
filter_add = DSCAT == "DISPOSITION EVENT" &
!(DSDECOD %in% c("SCREEN FAILURE", "COMPLETED", NA))
)
This call would return the input dataset with the column DCSREAS
and DCSREASP
added.
If the derivation must be changed, the user can define that derivation in the filter_add
argument of the function to map DSDECOD
and DSTERM
to a suitable DCSREAS
/DCSREASP
value.
The call below maps DCSREAS
and DCREASP
as follows:
DCSREAS
as DSDECOD
if DSDECOD
is not "COMPLETED"
or NA
, NA
otherwiseDCSREASP
as DSTERM
if DSDECOD
is equal to OTHER
, NA
otherwise
adsl <- adsl %>%
derive_vars_merged(
dataset_add = ds,
by_vars = exprs(USUBJID),
new_vars = exprs(DCSREAS = DSDECOD),
filter_add = DSCAT == "DISPOSITION EVENT" &
DSDECOD %notin% c("SCREEN FAILURE", "COMPLETED", NA)
) %>%
derive_vars_merged(
dataset_add = ds,
by_vars = exprs(USUBJID),
new_vars = exprs(DCSREASP = DSTERM),
filter_add = DSCAT == "DISPOSITION EVENT" & DSDECOD %in% "OTHER"
)
RANDDT
)
The function derive_vars_merged()
can be used to derive randomization date variable. To map Randomization Date (RANDDT
), the call would be:
adsl <- adsl %>%
derive_vars_merged(
dataset_add = ds_ext,
filter_add = DSDECOD == "RANDOMIZED",
by_vars = exprs(STUDYID, USUBJID),
new_vars = exprs(RANDDT = DSSTDT)
)
This call would return the input dataset with the column RANDDT
is added.
DTHDT
)
The function derive_vars_dt()
can be used to derive DTHDT
. This function allows the user to impute the date as well.
Example calls:
adsl <- adsl %>%
derive_vars_dt(
new_vars_prefix = "DTH",
dtc = DTHDTC
)
This call would return the input dataset with the columns DTHDT
added and, by default, the associated date imputation flag (DTHDTF
) populated with the controlled terminology outlined in the ADaM IG for date imputations. If the imputation flag is not required, the user must set the argument flag_imputation
to "none"
.
If imputation is needed and the date is to be imputed to the first day of the month/year the call would be:
adsl <- adsl %>%
derive_vars_dt(
new_vars_prefix = "DTH",
dtc = DTHDTC,
date_imputation = "first"
)
See also Date and Time Imputation.
DTHCAUS
)
The cause of death DTHCAUS
can be derived using the function derive_vars_extreme_event()
.
Since the cause of death could be collected/mapped in different domains (e.g. DS
, AE
, DD
), it is important the user specifies the right source(s) to derive the cause of death from.
For example, if the date of death is collected in the AE form when the AE is Fatal, the cause of death would be set to the preferred term (AEDECOD
) of that Fatal AE, while if the date of death is collected in the DS
form, the cause of death would be set to the disposition term (DSTERM
). To achieve this, the event()
objects within derive_vars_extreme_event()
must be specified and defined such that they fit the study requirement.
An example call to derive_vars_extreme_event()
would be:
adsl <- adsl %>%
derive_vars_extreme_event(
by_vars = exprs(STUDYID, USUBJID),
events = list(
event(
dataset_name = "ae",
condition = AEOUT == "FATAL",
set_values_to = exprs(DTHCAUS = AEDECOD),
),
event(
dataset_name = "ds",
condition = DSDECOD == "DEATH" & grepl("DEATH DUE TO", DSTERM),
set_values_to = exprs(DTHCAUS = DSTERM),
)
),
source_datasets = list(ae = ae, ds = ds),
tmp_event_nr_var = event_nr,
order = exprs(event_nr),
mode = "first",
new_vars = exprs(DTHCAUS)
)
The function also offers the option to add some traceability variables (e.g. DTHDOM
would store the domain where the date of death is collected, and DTHSEQ
would store the xxSEQ
value of that domain). The traceability variables should be added to the event()
calls and included in the new_vars
parameter of derive_vars_extreme_event()
.
adsl <- adsl %>%
select(-DTHCAUS) %>% # remove it before deriving it again
derive_vars_extreme_event(
by_vars = exprs(STUDYID, USUBJID),
events = list(
event(
dataset_name = "ae",
condition = AEOUT == "FATAL",
set_values_to = exprs(DTHCAUS = AEDECOD, DTHDOM = "AE", DTHSEQ = AESEQ),
),
event(
dataset_name = "ds",
condition = DSDECOD == "DEATH" & grepl("DEATH DUE TO", DSTERM),
set_values_to = exprs(DTHCAUS = DSTERM, DTHDOM = "DS", DTHSEQ = DSSEQ),
)
),
source_datasets = list(ae = ae, ds = ds),
tmp_event_nr_var = event_nr,
order = exprs(event_nr),
mode = "first",
new_vars = exprs(DTHCAUS, DTHDOM, DTHSEQ)
)
Following the derivation of DTHCAUS
and related traceability variables, it is then possible to derive grouping variables such as death categories (DTHCGRx
) using standard tidyverse code.
adsl <- adsl %>%
mutate(DTHCGR1 = case_when(
is.na(DTHDOM) ~ NA_character_,
DTHDOM == "AE" ~ "ADVERSE EVENT",
str_detect(DTHCAUS, "(PROGRESSIVE DISEASE|DISEASE RELAPSE)") ~ "PROGRESSIVE DISEASE",
TRUE ~ "OTHER"
))
The function derive_vars_duration()
can be used to derive duration relative to death like the Relative Day of Death (DTHADY
) or the numbers of days from last dose to death (LDDTHELD
).
Example calls:
adsl <- adsl %>%
derive_vars_duration(
new_var = DTHADY,
start_date = TRTSDT,
end_date = DTHDT
)
adsl <- adsl %>%
derive_vars_duration(
new_var = LDDTHELD,
start_date = TRTEDT,
end_date = DTHDT,
add_one = FALSE
)
LSTALVDT
)
Similarly as for the cause of death (DTHCAUS
), the last known alive date (LSTALVDT
) can be derived from multiples sources using derive_vars_extreme_event()
.
An example could be (DTC dates are converted to numeric dates imputing missing day and month to the first):
adsl <- adsl %>%
derive_vars_extreme_event(
by_vars = exprs(STUDYID, USUBJID),
events = list(
event(
dataset_name = "ae",
order = exprs(AESTDTC, AESEQ),
condition = !is.na(AESTDTC),
set_values_to = exprs(
LSTALVDT = convert_dtc_to_dt(AESTDTC, highest_imputation = "M"),
seq = AESEQ
),
),
event(
dataset_name = "ae",
order = exprs(AEENDTC, AESEQ),
condition = !is.na(AEENDTC),
set_values_to = exprs(
LSTALVDT = convert_dtc_to_dt(AEENDTC, highest_imputation = "M"),
seq = AESEQ
),
),
event(
dataset_name = "lb",
order = exprs(LBDTC, LBSEQ),
condition = !is.na(LBDTC),
set_values_to = exprs(
LSTALVDT = convert_dtc_to_dt(LBDTC, highest_imputation = "M"),
seq = LBSEQ
),
),
event(
dataset_name = "adsl",
condition = !is.na(TRTEDT),
set_values_to = exprs(LSTALVDT = TRTEDT, seq = 0),
)
),
source_datasets = list(ae = ae, lb = lb, adsl = adsl),
tmp_event_nr_var = event_nr,
order = exprs(LSTALVDT, seq, event_nr),
mode = "last",
new_vars = exprs(LSTALVDT)
)
Traceability variables can be added by specifying the variables in the set_values_to
parameter of the event()
function.
adsl <- adsl %>%
select(-LSTALVDT) %>% # created in the previous call
derive_vars_extreme_event(
by_vars = exprs(STUDYID, USUBJID),
events = list(
event(
dataset_name = "ae",
order = exprs(AESTDTC, AESEQ),
condition = !is.na(AESTDTC),
set_values_to = exprs(
LSTALVDT = convert_dtc_to_dt(AESTDTC, highest_imputation = "M"),
LALVSEQ = AESEQ,
LALVDOM = "AE",
LALVVAR = "AESTDTC"
),
),
event(
dataset_name = "ae",
order = exprs(AEENDTC, AESEQ),
condition = !is.na(AEENDTC),
set_values_to = exprs(
LSTALVDT = convert_dtc_to_dt(AEENDTC, highest_imputation = "M"),
LALVSEQ = AESEQ,
LALVDOM = "AE",
LALVVAR = "AEENDTC"
),
),
event(
dataset_name = "lb",
order = exprs(LBDTC, LBSEQ),
condition = !is.na(LBDTC),
set_values_to = exprs(
LSTALVDT = convert_dtc_to_dt(LBDTC, highest_imputation = "M"),
LALVSEQ = LBSEQ,
LALVDOM = "LB",
LALVVAR = "LBDTC"
),
),
event(
dataset_name = "adsl",
condition = !is.na(TRTEDT),
set_values_to = exprs(LSTALVDT = TRTEDT, LALVSEQ = NA_integer_, LALVDOM = "ADSL", LALVVAR = "TRTEDTM"),
)
),
source_datasets = list(ae = ae, lb = lb, adsl = adsl),
tmp_event_nr_var = event_nr,
order = exprs(LSTALVDT, LALVSEQ, event_nr),
mode = "last",
new_vars = exprs(LSTALVDT, LALVSEQ, LALVDOM, LALVVAR)
)
AGEGR1
or REGION1
)
Numeric and categorical variables (AGE
, RACE
, COUNTRY
, etc.) may need to be grouped to perform the required analysis. admiral provides the derive_vars_cat()
function to create such groups. This function is especially useful if more than one variable needs to be created for each condition, e.g., AGEGR1
and AGEGR1N
.
Additionally, one needs to be careful when considering the order of the conditions in the lookup table. The category is assigned based on the first match. That means catch-all conditions must come after specific conditions, e.g. !is.na(COUNTRY)
must come after COUNTRY %in% c("CAN", "USA")
.
# create lookup tables
agegr1_lookup <- exprs(
~condition, ~AGEGR1,
AGE < 18, "<18",
between(AGE, 18, 64), "18-64",
AGE > 64, ">64",
is.na(AGE), "Missing"
)
region1_lookup <- exprs(
~condition, ~REGION1,
COUNTRY %in% c("CAN", "USA"), "North America",
!is.na(COUNTRY), "Rest of the World",
is.na(COUNTRY), "Missing"
)
adsl <- adsl %>%
derive_vars_cat(
definition = agegr1_lookup
) %>%
derive_vars_cat(
definition = region1_lookup
)
Alternatively, you can also solve this task with custom functions:
format_agegr1 <- function(var_input) {
case_when(
var_input < 18 ~ "<18",
between(var_input, 18, 64) ~ "18-64",
var_input > 64 ~ ">64",
TRUE ~ "Missing"
)
}
format_region1 <- function(var_input) {
case_when(
var_input %in% c("CAN", "USA") ~ "North America",
!is.na(var_input) ~ "Rest of the World",
TRUE ~ "Missing"
)
}
adsl %>%
mutate(
AGEGR1 = format_agegr1(AGE),
REGION1 = format_region1(COUNTRY)
)
SAFFL
)
Since the populations flags are mainly company/study specific no dedicated functions are provided, but in most cases they can easily be derived using derive_var_merged_exist_flag
.
An example of an implementation could be:
adsl <- adsl %>%
derive_var_merged_exist_flag(
dataset_add = ex,
by_vars = exprs(STUDYID, USUBJID),
new_var = SAFFL,
condition = (EXDOSE > 0 | (EXDOSE == 0 & str_detect(EXTRT, "PLACEBO")))
)
The users can add specific code to cover their need for the analysis.
The following functions are helpful for many ADSL derivations:
derive_vars_merged()
- Merge Variables from a Dataset to the Input Datasetderive_var_merged_exist_flag()
- Merge an Existence Flagderive_var_merged_summary()
- Merge Summary VariablesSee also Generic Functions.
Adding labels and attributes for SAS transport files is supported by the following packages:
metacore: establish a common foundation for the use of metadata within an R session.
metatools: enable the use of metacore objects. Metatools can be used to build datasets or enhance columns in existing datasets as well as checking datasets against the metadata.
xportr: functionality to associate all metadata information to a local R data frame, perform data set level validation checks and convert into a transport v5 file(xpt).
NOTE: All these packages are in the experimental phase, but the vision is to have them associated with an End to End pipeline under the umbrella of the pharmaverse. An example of applying metadata and perform associated checks can be found at the pharmaverse E2E example.