https://github.com/insightsengineering/cardx
R Package to Supplement ARD Functions Found in {cards}
https://github.com/insightsengineering/cardx
r
Last synced: 4 months ago
JSON representation
R Package to Supplement ARD Functions Found in {cards}
- Host: GitHub
- URL: https://github.com/insightsengineering/cardx
- Owner: insightsengineering
- License: other
- Created: 2024-01-15T12:27:13.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2026-02-04T16:15:29.000Z (4 months ago)
- Last Synced: 2026-02-05T03:39:10.184Z (4 months ago)
- Topics: r
- Language: R
- Homepage: https://insightsengineering.github.io/cardx/
- Size: 33.9 MB
- Stars: 20
- Watchers: 5
- Forks: 6
- Open Issues: 19
-
Metadata Files:
- Readme: README.Rmd
- Changelog: NEWS.md
- Contributing: .github/CONTRIBUTING.md
- License: LICENSE
- Code of conduct: .github/CODE_OF_CONDUCT.md
- Security: SECURITY.md
Awesome Lists containing this project
README
---
output: github_document
editor_options:
markdown:
wrap: 72
---
# cardx 
[](https://github.com/insightsengineering/cardx/actions/workflows/R-CMD-check.yaml)
[](https://app.codecov.io/gh/insightsengineering/cardx?branch=main)
[](https://CRAN.R-project.org/package=cardx)
[](https://cran.r-project.org/package=cardx)
The **{cardx}** package is an extension of the {cards} package, providing additional functions to create Analysis Results Data Objects (ARDs) using the **R** programming language.
The {cardx} package exports ARD functions that uses utility functions from {cards} and statistical functions from additional packages (such as, {stats}, {mmrm}, {emmeans}, {car}, {survey}, etc.) to construct summary objects.
Summary objects can be used to:
- **Generate Tables and visualizations for Regulatory Submission**
easily in **R**. Perfect for presenting descriptive statistics,
statistical analyses, regressions, etc. and more.
- **Conduct Quality Control checks on existing Tables** in R.
Storing both the results and test parameters supports the re-use and
verification of data analyses.
## Installation
Install cards from CRAN with:
```r
install.packages("cardx")
```
You can install the development version of cards from [GitHub](https://github.com/) with:
``` r
# install.packages("devtools")
devtools::install_github("insightsengineering/cardx")
```
## Examples
### Example ARD Creation
Example t-test:
```{r}
library(cardx)
cards::ADSL |>
# keep two treatment arms for the t-test calculation
dplyr::filter(ARM %in% c("Placebo", "Xanomeline High Dose")) |>
cardx::ard_stats_t_test(by = ARM, variable = AGE)
```
Note that the returned ARD contains the analysis results in addition to
the function parameters used to calculate the results allowing for
reproducible future analyses and further customization.
### Model Input
Some {cardx} functions accept regression model objects as input:
```{r, eval=FALSE}
lm(AGE ~ ARM, data = cards::ADSL) |>
ard_aod_wald_test()
```
Note that the [Analysis Results Standard](https://www.cdisc.org/standards/foundational/analysis-results-standard) should begin with a data set rather than a model object.
To accomplish this we include model construction helpers.
```{r}
construct_model(
data = cards::ADSL,
formula = reformulate2("ARM", response = "AGE"),
method = "lm"
) |>
ard_aod_wald_test()
```
## Additional Resources
- The best resources are the help documents accompanying each {cardx} function.
- Supporting documentation for both companion packages [{cards}](https://insightsengineering.github.io/cards/) and {[gtsummary](https://www.danieldsjoberg.com/gtsummary/index.html)} will be useful for understanding the ARD workflow and capabilities.
## {cardx} + {renv}
The {cardx} package exports functions to create ARDs based on various statistical methods; methods that are primarily implemented in other packages.
{cardx} does not take a hard dependency on these packages, meaning that these packages are not typically installed when {cardx} is installed from CRAN.
As a result, {renv} will not record these packages in its `lock.file` unless there is a direct reference to the underlying statistical package in your code.
For example, if you pass a regression model to `ard_emmeans_contrast()`, there is no direct reference to the {emmeans} package in your script and {renv} will not record the package.
One can circumvent this issue by including some kind of reference to the package in your code.
Below are are couple of common ways to do so.
```r
library(emmeans)
```
Attaching a package with `library()` is great for its simplicity, but you may not want to attach a package if it's not necessary.
```r
invisible(emmeans::emmeans)
```
You can invisibly print a function from the package.
Printing a function does not have an effect on your environment (which is great), but it is somewhat more difficult to read.
(_This is my preferred method._)