ADaMs with {admiral} and friends

Objectives

By the end of this ADaM section you will have:
- Gained an understanding of how to use admiral, metatools/metacore, and xportr packages to create ADaMs
- Seen code executed to create variables and records common to ADSL and ADVS.
- Seen small code snippets of certain functions with play data.
- Checked in on our fav participant - Barb.
- Completed two short exercises using functions from admiral package.

Play by Play to achieve our Objectives

🕙11:00 - 11:15 Set the Stage: Overview of how we will work, datasets, specs, packages, functions
🕥11:15 - 11:40 Subject-level dataset: ADSL
🕚11:40 - 11:50 Check in on Barb / Exercise
🕦11:50 - 12:20 Basic Data Structure: ADVS
🕚12:20 - 12:30 Check in on Barb / Exercise
🕚12:30 Lunch!

End to End with Barb!!

flowchart LR
  A[Raw] --> B[SDTM]
  B[SDTM] --> C[ADaM]
  C[ADaM] --> D[ARD]
  D[ARD] --> E[TFL]
  
%% Apply the custom class "highlight" to node B
class C highlight;

%% Define the style for the "highlight" class:
classDef highlight fill:#F9B, stroke:#333, stroke-width:2px;

We are going to take a glimpse of where Barb is in ADSL and ADVS

Set the Stage

🕥11:00 - 11:15

Set the Stage

How we will work for the next 90 minutes?
Data (SDTMs, ADSL, and ADVS)
Specs (Excel document with lots of tabs)
pharmaverse packages
- playing nice with tidyverse and base R!!
Functions to make ADaM variables and records.

How we will proceed today?

I will pull out certain functions of the ADSL and ADVS R scripts.
- Let’s take a quick peek at the scripts.
- I will discuss certain functions and function arguments from scripts for deriving variables/records.
- Will pull up full function arguments to briefly highlight other powers. 💡
- When possible will have some play data and mini-runs of the function. 🏸

How we will proceed today continued?

Two exercises set up to help explore two key functions discussed:
- derive_vars_dtm() at end of ADSL.
- derive_param_computed() at end of ADVS.
Brief notes on the code in script and exercises:
- Code in scripts should run out of the box.
- You can run now, but if something doesn’t work let’s fix at the exercises.
- Exercises will need some minor updates.
- 3 folks not presenting can help diagnose issues during exercises.

Datasets

ADSL - Subject Level Dataset

ADSL - Subject Level Dataset, i.e each subject has a record. Focus on adding variables.
We will use SDTM data from the pharmaversesdtm package.
ADSL is available in the pharmaverseadam package

ADVS - Vital Signs Dataset

ADVS - Is a Basic Data Structure (BDS) Dataset. Focus on adding records, but also some variables!
We will use SDTM data vs from the pharmaversesdtm package and ADSL from dataset created by the adsl.R script.
ADVS is available in the pharmaverseadam package

Specs

Spec-details

Gives specifics on how we derive variables in the ADaMs (extreme traceability)
P21-like spec
- P21 is spec/data validation tool widely used in the industry for creating and validating SDTMs and ADaMs.
These specs are not fit for purpose - just a helpful guide
What is in the the Specs:
- Two datasets, dataset labels and variable keys
- Variables - labels, lengths, types, linked to method
- Codelists / Control Terms!!

Spec-peek

Let’s take a quick peek at metadata

The core pharmaverse packages for creating ADaMs

Focus on applying information from our Specs to the ADaMs

Focus on deriving variables and records driven by our Specs

Focus on preparing the dataset as an xpt file to be sent to regulatory agencies.

pharmaverse plays nice!

The ADaM code presented today utilizes all the previously mentioned packages, but also makes use of tidyverse and base R code.
If a function isn’t meeting your needs or looks too cumbersome, then that is okay!
These packages are all modular and play well with others.

admiral

Package	Function	Purpose
admiral	derive_vars_merged()	`left_join` on steroids
admiral	derive_param_computed()	Core function for wrappers used in ADVS
admiral	derive_vars_dy()	Create the relative study day
admiral	derive_vars_dtm()	Take a `---DTC` variable and turn it into a `---DTM` variable
admiral	derive_vars_duration()	Duration between timepoints
admiral	restrict_derivation()	HOF: Execute a derivation on a subset of the input dataset.

Most functions in admiral follow these conventions:

var_ or _vars_ functions add variable(s) to ADaMs
_param_ functions add records to ADaMs
Other not following this convention typically execute more advanced stuff!!

metacore, metatools and xportr

Package	Function	Purpose
metacore	spec_to_metacore()	Creates a “special” object from your specs
metatools	combine_supp()	Join Parent and Supplementary Datasets
metatools	create_var_from_codelist()	Numeric Vars from Specs!
metatools	drop_unspec_vars()	Easily drop variables not in Spec
xportr	xportr_label()	Apply labels from Spec
xportr	xportr_write()	Write out an xpt file

Most functions from these packages are about applying metadata from specs to your dataset.
xportr follows a standard naming convention with xportr_.

`ADSL`

🕥 11:15 - 11:40

Note: Referencing lines in file at each section with a footnote (Lines: ##-##)

The Big Picture of `ADSL`

We have an ADSL script:

Using {pharmaversesdtm} data: dm, suppdm, ex, ae, vs, etc
Uses the ADSL section of our spec file
Creating a single record for each subject with a lot of variables.
306 observations and 54 variables.
Write out an xpt file for regulatory send-off!

Reading in our spec for `ADSL` ¹

adsl_spec <- spec_to_metacore(
  path = "slides/03-ADaM/metadata/posit_specs.xlsx",
  where_sep_sheet = FALSE,
  quiet = TRUE
) %>%
  select_dataset("ADSL")

Takes a P21-like spec and makes it into an object we can easily access.
where_sep_sheet - option to tell if the information for the dataset is in a separate sheet, like in older p21 specs or in a single sheet like newer p21 specs
quiet argument should be set to TRUE while developing your spec and ADaMs.
Same call in ADVS script.

Combine Parent and Supplementary Data ¹

dm_suppdm <- combine_supp(dm, suppdm)

One line of code to join two datasets: parent and supplementary!
Supps usually are a collection non-standard data and linking to parent
Function is from {metatools}

Let’s turn a `--DTC` to a `DTM` or `DT` variable ¹

ex_ext <- ex %>%
  derive_vars_dtm(
    dtc = EXSTDTC,
    new_vars_prefix = "EXST",
    time_imputation = "last",
    flag_imputation = "time"
)

ds_ext <- derive_vars_dt(
  ds,
  dtc = DSSTDTC,
  new_vars_prefix = "DSST"
)

adsl11 <- adsl10 %>%
  derive_vars_dt(
    dtc = DTHDTC,
    new_vars_prefix = "DTH",
    highest_imputation = "M",
    date_imputation = "first"
  )

Example from documentation

mhdt <- tribble(
  ~MHSTDTC,
  "2019-07-18T15:25:40",
  "2019-07-18T15:25",
  "2019-07-18",
  "2019-02",
  "2019",
  "2019---07",
  ""
)

derive_vars_dtm(
 mhdt,
 new_vars_prefix = "AST",
 dtc = MHSTDTC,
 highest_imputation = "M",
 date_imputation = "last",
 time_imputation = "last",
)

# A tibble: 7 × 4
  MHSTDTC               ASTDTM              ASTDTF ASTTMF
  <chr>                 <dttm>              <chr>  <chr> 
1 "2019-07-18T15:25:40" 2019-07-18 15:25:40 <NA>   <NA>  
2 "2019-07-18T15:25"    2019-07-18 15:25:59 <NA>   S     
3 "2019-07-18"          2019-07-18 23:59:59 <NA>   H     
4 "2019-02"             2019-02-28 23:59:59 D      H     
5 "2019"                2019-12-31 23:59:59 M      H     
6 "2019---07"           2019-12-31 23:59:59 M      H     
7 ""                    NA                  <NA>   <NA>

Woah…there is more!!

derive_vars_dtm(
  dataset,
  new_vars_prefix,
  dtc,
  highest_imputation = "h",
  date_imputation = "first",
  time_imputation = "first",
  flag_imputation = "auto",
  min_dates = NULL,
  max_dates = NULL,
  preserve = FALSE,
  ignore_seconds_flag = FALSE
)

A Simple merge ¹

adsl10 <- adsl09 %>%
  derive_vars_merged(
    dataset_add = ds_ext,
    by_vars = exprs(STUDYID, USUBJID),
    new_vars = exprs(RANDDT = DSSTDT),
    filter_add = DSDECOD == "RANDOMIZED",
  )

Filter, arrange, join and rename is a common pattern in ADaMs.
by_vars are going to be the shared variables to connect the two datasets
In admiral, all arguments which can be set to multiple variables, a calculation, or an assignment must be started with rlang::exprs().
Check out Programming Concepts and Conventions

Let’s get a more complicated merge ¹

derive_vars_merged(
  dataset_add = ex_ext,
  by_vars = exprs(STUDYID, USUBJID),
  order = exprs(EXSTDTM, EXSEQ),
  new_vars = exprs(TRTSDTM = EXSTDTM, TRTSTMF = EXSTTMF),
  filter_add = (EXDOSE > 0 |
    (EXDOSE == 0 &
      str_detect(EXTRT, "PLACEBO"))) &
    !is.na(EXSTDTM),
  mode = "first",
)

order - for each by group the first or last observation from the additional dataset is selected with respect to the specified order.
mode - Determines if the first or last observation is selected

Whoa!!! Way more options!!

derive_vars_merged(
  dataset,
  dataset_add,
  by_vars,
  order = NULL,
  new_vars = NULL,
  filter_add = NULL,
  mode = NULL,
  exist_flag = NULL,
  true_value = "Y",
  false_value = NA_character_,
  missing_values = NULL,
  check_type = "warning",
  duplicate_msg = NULL,
  relationship = NULL
)

Let’s derive a Duration Variable ¹

derive_vars_duration(
    new_var = DTHADY,
    start_date = TRTSDT,
    end_date = DTHDT
  )

Example from documentation

# Derive age in years
data <- tribble(
  ~USUBJID, ~BRTHDT,           ~RANDDT,
  "P01",    ymd("1984-09-06"), ymd("2020-02-24"),
  "P02",    ymd("1985-01-01"), NA,
  "P03",    NA,                ymd("2021-03-10"),
  "P04",    NA,                NA
)

derive_vars_duration(data,
  new_var = AAGE,
  new_var_unit = AAGEU,
  start_date = BRTHDT,
  end_date = RANDDT,
  out_unit = "years",
  add_one = FALSE,
  trunc_out = TRUE
)

# A tibble: 4 × 5
  USUBJID BRTHDT     RANDDT      AAGE AAGEU
  <chr>   <date>     <date>     <dbl> <chr>
1 P01     1984-09-06 2020-02-24    35 years
2 P02     1985-01-01 NA            NA <NA> 
3 P03     NA         2021-03-10    NA <NA> 
4 P04     NA         NA            NA <NA>

Ohhh there is more!!

derive_vars_duration(
  dataset,
  new_var,
  new_var_unit = NULL,
  start_date,
  end_date,
  in_unit = "days",
  out_unit = "DAYS",
  floor_in = TRUE,
  add_one = TRUE,
  trunc_out = FALSE,
  type = "duration"
)

Duration versus Interval

The lubridate package calculates two types of spans between two dates: duration and interval. While these calculations are largely the same, when the unit of the time period is month or year the result can be slightly different.
The difference arises from the ambiguity in the length of “1 month” or “1 year”. Months may have 31, 30, 28, or 29 days, and years are 365 days and 366 during leap years. Durations and intervals help solve the ambiguity in these measures.

Let’s apply Control Terms / Code Lists ¹

adsl16 %>%
  create_var_from_codelist(adsl_spec, input_var = AGEGR1, out_var = AGEGR1N) %>%
  create_var_from_codelist(adsl_spec, input_var = RACE, out_var = RACEN) %>%
  create_var_from_codelist(adsl_spec, input_var = RACEGR1, out_var = RACEGR1N) %>%
  create_var_from_codelist(adsl_spec, input_var = REGION1, out_var = REGION1N) %>%
  create_var_from_codelist(adsl_spec, input_var = TRT01P, out_var = TRT01PN) %>%
  create_var_from_codelist(adsl_spec, input_var = TRT01A, out_var = TRT01AN)

Numeric Variables like the above are helpful for sorting options in Tables.

More details

create_var_from_codelist(
  data,
  metacore,
  input_var,
  out_var,
  codelist = NULL,
  decode_to_code = TRUE,
  strict = TRUE
)

codelist - Optional argument to supply a codelist. Must be a data.frame with code and decode columns such as those created by the function metacore::get_control_term. If no codelist is provided the codelist associated with the column supplied to out_var will be used. By default codelist is NULL.

decode_to_code - Direction of the translation. Default value is TRUE, i.e., assumes the input_var is the decode column of the codelist. Set to FALSE if the input_var is the code column of the codelist.

strict - A logical value indicating whether to perform strict checking against the codelist. If TRUE will issue a warning if values in the input_var column are not present in the codelist. If FALSE no warning is issued and values not present in the codelist will likely result in NA results.

Check in on Barb / Exercise

🕚11:40 - 11:50

Check in on Barb in `ADSL`

flowchart LR
  B[SDTM] --> C[ADaM]
  C[ADaM] --> D[ARDs/TFLs]
  
%% Apply the custom class "highlight" to node B
class C highlight;

%% Define the style for the "highlight" class:
classDef highlight fill:#F9B, stroke:#333, stroke-width:2px;

USUBJID: 01-701-1034
SEX: F
AGE: 77
RANDDT: 2014-07-01
TRTSDT: 2014-07-01
TRTSDTM: 2014-07-01 23:59:59
TRTEDT: 2014-12-30
TRTDURD: 183
EOSDT: 2014-12-30
EOSSTT: COMPLETED

Check-in Exercise

Impute the date and time to be last month/day and 23:59:59
Two arguments to update: ????
Check out admiral derive_vars_dtm() documentation

library(tibble)
library(lubridate)
library(admiral)

posit_mh <- tribble(
  ~USUBJID, ~MHSTDTC,
  1,        "2019-07-18T15:25:40",
  1,        "2019-07-18T15:25",
  1,        "2019-07-18",
  2,        "2024-02",
  2,        "2019",
  2,        "2019---07",
  3,        ""
)

derive_vars_dtm(
  dataset = posit_mh,
  new_vars_prefix = "AST",
  dtc = MHSTDTC,
  highest_imputation = "M",
  date_imputation = "????",
  time_imputation = "????"
)

Check-in Exercise Solution

posit_mh <- tribble(
  ~USUBJID, ~MHSTDTC,
  1,        "2019-07-18T15:25:40",
  1,        "2019-07-18T15:25",
  1,        "2019-07-18",
  2,        "2024-02",
  2,        "2019",
  2,        "2019---07",
  3,        ""
)

derive_vars_dtm(
  dataset = posit_mh,
  new_vars_prefix = "AST",
  dtc = MHSTDTC,
  highest_imputation = "M",
  date_imputation = "last",
  time_imputation = "last"
)

# A tibble: 7 × 5
  USUBJID MHSTDTC               ASTDTM              ASTDTF ASTTMF
    <dbl> <chr>                 <dttm>              <chr>  <chr> 
1       1 "2019-07-18T15:25:40" 2019-07-18 15:25:40 <NA>   <NA>  
2       1 "2019-07-18T15:25"    2019-07-18 15:25:59 <NA>   S     
3       1 "2019-07-18"          2019-07-18 23:59:59 <NA>   H     
4       2 "2024-02"             2024-02-29 23:59:59 D      H     
5       2 "2019"                2019-12-31 23:59:59 M      H     
6       2 "2019---07"           2019-12-31 23:59:59 M      H     
7       3 ""                    NA                  <NA>   <NA>

`ADVS`

🕦11:50 - 12:20

Note: Referencing lines in file at each section with a footnote (Lines: ##-##)

The Big Picture of ADVS

We have an ADVS script:
- Using pharmaversesdtm data: vs
- Using the ADVS section of our spec file.
- Creating multiple records for each subject.
- Some ADSL variables included.
- 60K records and 106 variables.
- Lot’s of functions are repeated within the ADVS script used in ADSL script.

Let’s talk about lookup tables ¹

  derive_vars_merged_lookup(
    dataset_add = param_lookup,
    new_vars = exprs(PARAMCD),
    by_vars = exprs(VSTESTCD),
    print_not_mapped = TRUE 
  )

Moving from **TEST/**TESTCD to PARAM/PARAMCD.
More information in the PARAM/PARAMCD - useful for displays.
Can be used for other lookup tables, e.g. Grading/Toxicity.

A Fuller Look!

derive_vars_merged_lookup(
  dataset,
  dataset_add,
  by_vars,
  order = NULL,
  new_vars = NULL,
  mode = NULL,
  filter_add = NULL,
  check_type = "warning",
  duplicate_msg = NULL,
  print_not_mapped = TRUE
)

Let’s add more records for each subject ¹

advs <- advs %>%
  derive_param_computed(
    by_vars = exprs(STUDYID, USUBJID, VISIT, VISITNUM, ADT, ADY, VSTPT, VSTPTNUM),
    parameters = "WEIGHT",
    set_values_to = exprs(
      AVAL = AVAL.WEIGHT / (AVAL.HEIGHT / 100)^2,
      PARAMCD = "BMI",
      PARAM = "Body Mass Index (kg/m^2)",
      AVALU = "kg/m^2"
    ),
    constant_parameters = c("HEIGHT"),
    constant_by_vars = exprs(USUBJID)
)

Only two arguments not used!

derive_param_computed(
  dataset = NULL,
  dataset_add = NULL,
  by_vars,
  parameters,
  set_values_to,
  filter = NULL,
  constant_by_vars = NULL,
  constant_parameters = NULL,
  keep_nas = FALSE
)

filter - The specified condition is applied to the input dataset before deriving the new parameter, i.e., only observations fulfilling the condition are taken into account.
keep_na - If the argument is set to TRUE, observations are added even if some of the values contributing to the computed value are NA

Let’s derive DTYPE summary records ¹

derive_summary_records(
    dataset_add = advs_2, 
    by_vars = exprs(STUDYID, USUBJID, !!!adsl_vars, PARAMCD, AVISITN, AVISIT, ADT, ADY, AVALU),
    filter_add = !is.na(AVAL),
    set_values_to = exprs(
      AVAL = mean(AVAL),
      DTYPE = "AVERAGE"
  )
)

A DTYPE variable indicates a record within a parameter has been imputed or modified, DTYPE will indicate the method used to populate the analysis value, e.g AVERAGE, MAXIMUM, MINIMUM, LOCF, PHANTOM.

Let’s restrict! ¹

restrict_derivation(
  derivation = derive_var_extreme_flag,
  args = params(
    by_vars = exprs(STUDYID, USUBJID, BASETYPE, PARAMCD),
    order = exprs(ADT, VISITNUM, VSSEQ),
    new_var = ABLFL,
    mode = "last",
    true_value = "Y"
  ),
  filter = (!is.na(AVAL) &
    ADT <= TRTSDT & !is.na(BASETYPE) & is.na(DTYPE))
)

Check out Higher Order Functions for more information.
call_derivation() is super handy if you are doing similar calls with same function, but just changing one or two arguments.

Let’s add a BASE variable ¹

advs <- derive_var_base(
  advs,
  by_vars = exprs(STUDYID, USUBJID, PARAMCD, BASETYPE),
  source_var = AVAL,
  new_var = BASE,
  # Below arguments are default values and not necessary to add in our case
  filter = ABLFL == "Y"
)

Used to calculate CHG and PCHG variables

Let’s add an Analysis Sequence Variable ¹

advs <- derive_var_obs_number(
  advs,
  new_var = ASEQ,
  by_vars = exprs(STUDYID, USUBJID),
  order = exprs(PARAMCD, ADT, AVISITN, VISITNUM, ATPTN, DTYPE),
  check_type = "error"
)

ASEQ uniquely indexes records within a subject within an ADaM dataset.
ASEQ is useful for traceability when the dataset is used as input to another ADaM dataset
check_type - If “message”, “warning” or “error” is specified, the specified message is issued if the observations of the input dataset are not unique with respect to the by variables and the order.

Let’s get that ADaM ready for regulatory agencies ¹

advs_prefinal <- advs %>%
  drop_unspec_vars(advs_spec) %>% 
  check_variables(advs_spec, dataset_name = "ADVS") %>% 
  order_cols(advs_spec) %>% 
  sort_by_key(advs_spec)

Functions from {metatools}
Note similar code in ADSL script!

Let’s get that data read for regulatory agencies ¹

advs_final <- advs_prefinal %>%
  xportr_type(advs_spec) %>%
  xportr_length(advs_spec) %>%
  xportr_label(advs_spec) %>%
  xportr_format(advs_spec, domain = "ADVS") %>%
  xportr_df_label(advs_spec, domain = "ADVS") %>%
  xportr_write(file.path(dir, "advs.xpt"), metadata = advs_spec, domain = "ADVS")

Functions from {xportr}
Note similar code in ADSL script!
domain is needed depending on how metacore object is created.

Check in on Barb / Exercise

🕦 12:20 - 12:30

Check in on Barb in ADVS

flowchart LR
  B[SDTM] --> C[ADaM]
  C[ADaM] --> D[ARDs/TFLs]
  
%% Apply the custom class "highlight" to node B
class C highlight;

%% Define the style for the "highlight" class:
classDef highlight fill:#F9B, stroke:#333, stroke-width:2px;

USUBJID: 01-701-1034
PARAM/PARAMCD: MAP / Mean Arterial Pressure (mmHg)
ABLFL/ADT: Y / 2014-07-01
BASE: 94.33333
AVIST/ADT: WEEK 2 / 2014-07-15
AVAL: 115.0
CHG: 20.6666667
PCHG: 21.9081272

Check-in Exercise

We are going to update the function derive_param_computed() for Mean Arterial Pressure.
Formula: \[ \frac{2DIABP + SYSBP}{3} \]

ADVS <- tribble(
  ~USUBJID,      ~PARAMCD, ~PARAM,                            ~AVALU,  ~AVAL, ~VISIT,
  "01-701-1015", "DIABP",  "Diastolic Blood Pressure (mmHg)", "mmHg",    51, "BASELINE",
  "01-701-1015", "SYSBP",  "Systolic Blood Pressure (mmHg)",  "mmHg",   121, "BASELINE",
  "01-701-1028", "DIABP",  "Diastolic Blood Pressure (mmHg)", "mmHg",    79, "BASELINE",
  "01-701-1028", "SYSBP",  "Systolic Blood Pressure (mmHg)",  "mmHg",   130, "BASELINE",
)

derive_param_computed(
  ADVS,
  by_vars = exprs(USUBJID, VISIT),
  parameters = c("????", "????"),
  set_values_to = exprs(
    AVAL = (AVAL.SYSBP + ?? * AVAL.DIABP) / ??,
    PARAMCD = "MAP",
    PARAM = "Mean Arterial Pressure (mmHg)",
    AVALU = "mmHg",
  )
)

Check-in Exercise Solution

ADVS <- tribble(
  ~USUBJID,      ~PARAMCD, ~PARAM,                            ~AVALU,  ~AVAL, ~VISIT,
  "01-701-1015", "DIABP",  "Diastolic Blood Pressure (mmHg)", "mmHg",    51, "BASELINE",
  "01-701-1015", "SYSBP",  "Systolic Blood Pressure (mmHg)",  "mmHg",   121, "BASELINE",
  "01-701-1028", "DIABP",  "Diastolic Blood Pressure (mmHg)", "mmHg",    79, "BASELINE",
  "01-701-1028", "SYSBP",  "Systolic Blood Pressure (mmHg)",  "mmHg",   130, "BASELINE",
) 

derive_param_computed(
  ADVS,
  by_vars = exprs(USUBJID, VISIT),
  parameters = c("SYSBP", "DIABP"),
  set_values_to = exprs(
    AVAL = (AVAL.SYSBP + 2 * AVAL.DIABP) / 3,
    PARAMCD = "MAP",
    PARAM = "Mean Arterial Pressure (mmHg)",
    AVALU = "mmHg",
  )
)

# A tibble: 6 × 6
  USUBJID     PARAMCD PARAM                           AVALU  AVAL VISIT   
  <chr>       <chr>   <chr>                           <chr> <dbl> <chr>   
1 01-701-1015 DIABP   Diastolic Blood Pressure (mmHg) mmHg   51   BASELINE
2 01-701-1015 SYSBP   Systolic Blood Pressure (mmHg)  mmHg  121   BASELINE
3 01-701-1028 DIABP   Diastolic Blood Pressure (mmHg) mmHg   79   BASELINE
4 01-701-1028 SYSBP   Systolic Blood Pressure (mmHg)  mmHg  130   BASELINE
5 01-701-1015 MAP     Mean Arterial Pressure (mmHg)   mmHg   74.3 BASELINE
6 01-701-1028 MAP     Mean Arterial Pressure (mmHg)   mmHg   96   BASELINE

Questions and Resources

admiraldiscovery

pharmaverse examples

pkgdown sites for xportr, metatools, metacore and admiral

Extension packages for admiral: admiralophtha, admiralonco, admiralpeds

Closing Thoughts

Only able to show a small number of pharmaverse packages and functions today, but please delve deeper to explore the full breadth

If you’d like to contribute to pharmaverse, check out options at pharmaverse.org

Finally, be sure to join our community on Slack!

Packages and Session Information

sessionInfo()

R version 4.5.1 (2025-06-13)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 24.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so;  LAPACK version 3.12.0

locale:
 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8       
 [4] LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8   
 [7] LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C          
[10] LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C   

time zone: UTC
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices datasets  utils     methods   base     

other attached packages:
[1] lubridate_1.9.4 admiral_1.3.1   tibble_3.3.0    countdown_0.4.0
[5] glue_1.8.0     

loaded via a namespace (and not attached):
 [1] jsonlite_2.0.0    dplyr_1.1.4       compiler_4.5.1    renv_1.1.5       
 [5] tidyselect_1.2.1  xml2_1.3.8        stringr_1.5.1     tidyr_1.3.1      
 [9] yaml_2.3.10       fastmap_1.2.0     R6_2.6.1          generics_0.1.4   
[13] admiraldev_1.3.1  knitr_1.50        pillar_1.11.0     rlang_1.1.6      
[17] utf8_1.2.6        stringi_1.8.7     roxygen2_7.3.2    xfun_0.52        
[21] timechange_0.3.0  cli_3.6.5         withr_3.0.2       magrittr_2.0.3   
[25] digest_0.6.37     hms_1.1.3         lifecycle_1.0.4   vctrs_0.6.5      
[29] evaluate_1.0.4    rmarkdown_2.29    purrr_1.1.0       tools_4.5.1      
[33] pkgconfig_2.0.3   htmltools_0.5.8.1

ADaMs with {admiral} and friends

Objectives

Play by Play to achieve our Objectives

End to End with Barb!!

Set the Stage

Set the Stage

How we will proceed today?

How we will proceed today?

How we will proceed today continued?

Datasets

ADSL - Subject Level Dataset

ADVS - Vital Signs Dataset

Specs

Spec-details

Spec-peek

The core pharmaverse packages for creating ADaMs

pharmaverse plays nice!

admiral

metacore, metatools and xportr

ADSL

The Big Picture of ADSL

Reading in our spec for ADSL 1

Combine Parent and Supplementary Data 1

Let’s turn a --DTC to a *DTM or *DT variable 1

Example from documentation

Woah…there is more!!

A Simple merge 1

Let’s get a more complicated merge 1

Whoa!!! Way more options!!

Let’s derive a Duration Variable 1

Example from documentation

Ohhh there is more!!

Duration versus Interval

Let’s apply Control Terms / Code Lists 1

More details

Check in on Barb / Exercise

Check in on Barb in ADSL

Check-in Exercise

Check-in Exercise Solution

ADVS

The Big Picture of ADVS

Let’s talk about lookup tables 1

A Fuller Look!

Let’s add more records for each subject 1

Only two arguments not used!

Let’s derive DTYPE summary records 1

Let’s restrict! 1

Let’s add a BASE variable 1

Let’s add an Analysis Sequence Variable 1

Let’s get that ADaM ready for regulatory agencies 1

Let’s get that data read for regulatory agencies 1

Check in on Barb / Exercise

Check in on Barb in ADVS

Check-in Exercise

Check-in Exercise Solution

Questions and Resources

Closing Thoughts

Packages and Session Information

`ADSL`

The Big Picture of `ADSL`

Reading in our spec for `ADSL` ¹

Combine Parent and Supplementary Data ¹

Let’s turn a `--DTC` to a `DTM` or `DT` variable ¹

A Simple merge ¹

Let’s get a more complicated merge ¹

Let’s derive a Duration Variable ¹

Let’s apply Control Terms / Code Lists ¹

Check in on Barb in `ADSL`

`ADVS`

Let’s talk about lookup tables ¹

Let’s add more records for each subject ¹

Let’s derive DTYPE summary records ¹

Let’s restrict! ¹

Let’s add a BASE variable ¹

Let’s add an Analysis Sequence Variable ¹

Let’s get that ADaM ready for regulatory agencies ¹

Let’s get that data read for regulatory agencies ¹