rifttable

rifttable: Results Tables to Bridge the Rift Between Epidemiologists and Their Data

https://github.com/stopsack/rifttable

Keywords

Repository

rifttable: Results Tables to Bridge the Rift Between Epidemiologists and Their Data

README.Rmd

---
output: 
  github_document:
    df_print: kable
---



```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
```

# Automated, Reproducible Generation of Results Tables: Bridging the Rift Between Epidemiologists and Their Data


[![R-CMD-check](https://github.com/stopsack/rifttable/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/stopsack/rifttable/actions/workflows/R-CMD-check.yaml)
[![Codecov test coverage](https://codecov.io/gh/stopsack/rifttable/graph/badge.svg)](https://app.codecov.io/gh/stopsack/rifttable)


Tables are the key format in which epidemiologists present their results. Many results tables in applied studies merely show point estimates and confidence intervals, or even p-values, from regression models: a “growing rift between epidemiologists and their data” ([Rothman 2017](https://doi.org/10.1007/s10654-017-0314-3)). “Actual,” descriptive data, such as counts stratified by exposure and a main confounder or effect modifier, are often lacking. 

**rifttable** creates presentation-ready results tables for epidemiologists in an automated, reproducible fashion. The user provides the final analytical dataset and specifies the design of the table, with rows and/or columns defined by exposure(s), effect modifier(s), and estimands as desired, allowing to show descriptors and inferential estimates in one table -- bridging the rift between epidemiologists and their data, one table at a time.


## Installation

The rifttable package can be installed from CRAN:

```{r, eval = FALSE}
install.packages("rifttable")
```

Development versions can be installed from 
[GitHub](https://stopsack.github.io/rifttable/) using:

```{r, eval = FALSE}
remotes::install_github("stopsack/rifttable")
```

The latter installation procedure requires the remotes package, obtainable via `install.packages("remotes")`.


## Example

```{r example}
library(rifttable)

example_design <- tibble::tribble(
  ~label,                       ~type,                   ~stratum,
  "Overall",                    "",                      "",
  "– Deaths/N",                 "outcomes/total",        c("Low", "High"),
  "– Risk",                     "risk",                  c("Low", "High"),
  "– Risk ratio (95% CI)",      "rr",                    c("Low", "High"),
  "– Risk difference (95% CI)", "rd",                    c("Low", "High"),
  "",                           "",                      "",
  "Low hormone receptor",       "",                      "",
  "– Deaths/N (Risk)",          "outcomes/total (risk)", "Low",
  "– Risk difference (95% CI)", "rd",                    "Low",
  "High hormone receptor",      "",                      "",
  "– Deaths/N (Risk)",          "outcomes/total (risk)", "High",
  "– Risk difference (95% CI)", "rd",                    "High"
) |> 
  dplyr::mutate(
    exposure = "stage",
    outcome = "death",
    effect_modifier = "receptor"
  )

rifttable(
  design = example_design,
  data = risks::breastcancer
)
```

For more examples, see the [Get Started vignette](https://stopsack.github.io/rifttable/articles/rifttable.html).

Owner

GitHub Events

Packages