NIH Enrollment Tables in HTML"

In codified: Produce Standard/Formalized Demographics Tables

knitr::opts_chunk$set(echo = TRUE)

Install codified Package

First, install the codified package if necessary, and then load it into memory.

# if( !requireNamespace("codified", quietly = TRUE) )
#   remotes::install_github(repo = "OuhscBbmc/codified")

library(codified)

Local Data Source

Establish Datasets

path <- system.file("misc/example-data-1.csv", package = "codified")
col_types <- readr::cols_only(
  record_id = readr::col_integer(),
  name_last = readr::col_character(),
  dob       = readr::col_date(format = ""),
  gender    = readr::col_integer(),
  race      = readr::col_integer(),
  ethnicity = readr::col_integer()
)
ds <- readr::read_csv(path, col_types = col_types) |>
  dplyr::mutate(
    gender     = as.character(gender),
    race       = as.character(race),
    ethnicity  = as.character(ethnicity)
  )
ds |>
  head(10) |>
  knitr::kable(caption = "Observed Dataset (first ten rows)")

ds_lu_gender <- tibble::tribble(
  ~input,   ~displayed            ,
  "0"   ,  "Female"               ,
  "1"   ,  "Male"                 ,
  "U"   ,  "Unknown/Not Reported"
)
knitr::kable(ds_lu_gender, caption = "Gender Mapping")

ds_lu_race <- tibble::tribble(
  ~input , ~displayed                                   ,
  "1"    , "American Indian/Alaska Native"              ,
  "2"    , "Asian"                                      ,
  "3"    , "Native Hawaiian or Other Pacific Islander"  ,
  "4"    , "Black or African American"                  ,
  "5"    , "White"                                      ,
  "M"    , "More than One Race"                         ,
  "6"    , "Unknown or Not Reported"
)
knitr::kable(ds_lu_race, caption = "Race Mapping")

ds_lu_ethnicity <- tibble::tribble(
  ~input,   ~displayed                      ,
  "2"   ,  "Not Hispanic or Latino"         ,
  "1"   ,  "Hispanic or Latino"             ,
  "0"   ,  "Unknown/Not Reported Ethnicity"
)
knitr::kable(ds_lu_ethnicity, caption = "Ethnicity Mapping")

Apply Map to Observed Dataset

ds_summary_long <- codified::table_nih_enrollment(
  d              = ds,
  d_lu_gender    = ds_lu_gender,
  d_lu_race      = ds_lu_race,
  d_lu_ethnicity = ds_lu_ethnicity
)

knitr::kable(ds_summary_long, caption = "Counts of Each Subgroup")

Conform to NIH Cosmetics

codified::table_nih_enrollment_pretty(
  d              = ds,
  d_lu_gender    = ds_lu_gender,
  d_lu_race      = ds_lu_race,
  d_lu_ethnicity = ds_lu_ethnicity
)

REDCap Data Source

A hosted (fake) clinical trial dataset demonstrates how to extract demographic data from REDCap and then present the demographic data in the NIH Inclusion Enrollment Report format.

Establish Datasets

First, install the REDCapR package if necessary, and then load it into memory.

# if( !requireNamespace("REDCapR", quietly = TRUE) )
#   remotes::install_github("OuhscBbmc/REDCapR")

library(REDCapR)

Next, download the data from the REDCap database into the ds_2 data.frame. If you're running the most recent version of REDCapR (available on GitHub), the code will be:

ds_2 <- REDCapR::redcap_read(
  redcap_uri = "https://bbmc.ouhsc.edu/redcap/api/",  # URL of REDCap Server.
  token      = "F304DEC3793FECC3B6DEEFF66302CAD3",    # User-specific token/password.
  guess_type = FALSE                                  # Keep all variables as strings/characters.
)$data

However CRAN policy understandably discourages vignettes from using "Internet resources" so this vignette mimics the code above with this local call. On your own computer, feel free to call that demonstration REDCap project.

ds_2 <-
  readr::read_csv(
    file      = system.file("misc/example-data-2.csv", package = "codified"),
    col_types = readr::cols(.default = readr::col_character())
  )

Conform to NIH Cosmetics

Now, convert these demographic data into a properly formatted NIH enrollment table. Pass the ds_lu_gender, ds_lu_race, and ds_lu_ethnicity metadata, which was defined above. As a reminder, these translate values like 1 to Male and 3 to Native Hawaiian or Other Pacific Islander.

table_nih_enrollment_pretty(
  d               = ds_2,
  d_lu_gender     = ds_lu_gender,
  d_lu_race       = ds_lu_race,
  d_lu_ethnicity  = ds_lu_ethnicity
)

Collapsing Levels

Many observed datasets may collect race with a different set of levels. For instance in ds_3, the American Indian level is separate from the Alaska Native level. In the first and second rows in the metadata below, the two levels are effectively combined into theAmerican Indian/Alaska Native level, so it complies with the format of the NIH Enrollment table.

ds_lu_race_3 <- tibble::tribble(
  ~input                      , ~displayed                                  ,
  "American Indian"           , "American Indian/Alaska Native"             , # Combine w/ Alaska Native
  "Alaska Native"             , "American Indian/Alaska Native"             , # Combine w/ American Indian
  "Asian"                     , "Asian"                                     ,
  "Native Hawaiian"           , "Native Hawaiian or Other Pacific Islander" , # Combine w/ Pacific Islanders
  "Pacific Islander"          , "Native Hawaiian or Other Pacific Islander" , # Combine w/ Hawaiian
  "Black or African American" , "Black or African American"                 ,
  "White"                     , "White"                                     ,
  "More than One Race"        , "More than One Race"                        ,
  "Unknown or Not Reported"   , "Unknown or Not Reported"
)

ds_3 <- tibble::tribble(
  ~subject_id,   ~gender  , ~race            ,  ~ethnicity                 ,
           1L,   "Female" , "American Indian",  "Not Hispanic or Latino"   ,
           2L,   "Male"   , "American Indian",  "Not Hispanic or Latino"   ,
           3L,   "Male"   , "American Indian",  "Not Hispanic or Latino"   ,
           4L,   "Female" , "Alaska Native"  ,  "Not Hispanic or Latino"   ,
           5L,   "Male"   , "Alaska Native"  ,  "Not Hispanic or Latino"   ,
           6L,   "Male"   , "Alaska Native"  ,  "Not Hispanic or Latino"   ,
           7L,   "Male"   , "White"          ,  "Not Hispanic or Latino"   ,
           8L,   "Male"   , "White"          ,  "Not Hispanic or Latino"
)

table_nih_enrollment_pretty(
  d               = ds_3,
  d_lu_race       = ds_lu_race_3
)

Feedback Welcome

Please share your opinions with us by creating an issue at https://github.com/OuhscBbmc/codified/issues.

Upcoming Features

1. Create tables in a fresh docx file with flextable. Issue #5
2. Create tables in a fresh pdf file Issue #5.
3. Populate tables in an existing pdf file with staplr?
4. The 'totals' column & row are calculated. The NIH form does this already, so this is mostly for the purpose of checking. Issue #4.
5. Allow variable names to change (e.g., Gender instead of gender). Issue #3.
6. pkgdown site, like REDCapR.

Group discussion

• Any holes in the functionality or concept?
• What else (e.g., features or explanation) do you think would be helpful?
• Are people able to run this package? Is the vignette clear?
• What other standardized tables/graphs/whatever could this be applied to?
• IRB continuation tables?
• Graphs?

codified documentation built on Aug. 12, 2022, 5:06 p.m.