comorbidity

comorbidity: An R package for computing comorbidity scores - Published in JOSS (2018)

https://github.com/ellessenne/comorbidity

Keywords

Repository

An R package for computing comorbidity scores.

README.Rmd

---
output: github_document
editor_options: 
  chunk_output_type: console
---



```{r, echo = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE
)
options(width = 100)
```

# The {comorbidity} Package: Computing Comorbidity Scores in R 

Last updated: `r Sys.time()`


[![R-CMD-check](https://github.com/ellessenne/comorbidity/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/ellessenne/comorbidity/actions/workflows/R-CMD-check.yaml)
[![Codecov test coverage](https://codecov.io/gh/ellessenne/comorbidity/branch/master/graph/badge.svg)](https://app.codecov.io/gh/ellessenne/comorbidity?branch=master)
[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/comorbidity)](https://cran.r-project.org/package=comorbidity)
[![CRAN_Logs_Badge](http://cranlogs.r-pkg.org/badges/comorbidity)](https://cran.r-project.org/package=comorbidity)
[![CRAN_Logs_Badge_Total](http://cranlogs.r-pkg.org/badges/grand-total/comorbidity)](https://cran.r-project.org/package=comorbidity)
[![JOSS DOI](http://joss.theoj.org/papers/10.21105/joss.00648/status.svg)](https://doi.org/10.21105/joss.00648)
[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://makeapullrequest.com/)


`comorbidity` is an R package for computing comorbidity scores such as the weighted Charlson score and the Elixhauser comorbidity score; both ICD-10 and ICD-9 coding systems are supported.

## Installation

`comorbidity` is on CRAN. You can install it as usual with:

```{r cran-installation, eval = FALSE}
install.packages("comorbidity")
```

Alternatively, you can install the development version from GitHub with:

```{r gh-installation, eval = FALSE}
# install.packages("remotes")
remotes::install_github("ellessenne/comorbidity")
```

## Simulating ICD-10 codes

The `comorbidity` packages includes a function named `sample_diag()` that allows simulating ICD diagnostic codes in a straightforward way. For instance, we could simulate ICD-10 codes:

```{r simulate-data}
# load the comorbidity package
library(comorbidity)
# set a seed for reproducibility
set.seed(1)
# simulate 50 ICD-10 codes for 5 individuals
x <- data.frame(
  id = sample(1:5, size = 50, replace = TRUE),
  code = sample_diag(n = 50)
)
x <- x[order(x$id, x$code), ]
print(head(x, n = 15), row.names = FALSE)
```

It is also possible to simulate from two different versions of the ICD-10 coding system. The default is to simulate ICD-10 codes from the 2011 version:

```{r simulate-data-2011}
set.seed(1)
x1 <- data.frame(
  id = sample(1:3, size = 30, replace = TRUE),
  code = sample_diag(n = 30)
)
set.seed(1)
x2 <- data.frame(
  id = sample(1:3, size = 30, replace = TRUE),
  code = sample_diag(n = 30, version = "ICD10_2011")
)
# should return TRUE
all.equal(x1, x2)
```

Alternatively, you could use the 2009 version:

```{r simulate-data-2009}
set.seed(1)
x1 <- data.frame(
  id = sample(1:3, size = 30, replace = TRUE),
  code = sample_diag(n = 30, version = "ICD10_2009")
)
set.seed(1)
x2 <- data.frame(
  id = sample(1:3, size = 30, replace = TRUE),
  code = sample_diag(n = 30, version = "ICD10_2011")
)
# should not return TRUE
all.equal(x1, x2)
```

## Simulating ICD-9 codes

ICD-9 codes can be easily simulated too:

```{r simulate-data-icd9}
set.seed(2)
x9 <- data.frame(
  id = sample(1:3, size = 30, replace = TRUE),
  code = sample_diag(n = 30, version = "ICD9_2015")
)
x9 <- x9[order(x9$id, x9$code), ]
print(head(x9, n = 15), row.names = FALSE)
```

## Computing comorbidity scores

The main function of the `comorbidity` package is named `comorbidity()`, and it can be used to compute any supported comorbidity score; scores can be specified by setting the `score` argument, which is required. 

Say we have 3 individuals with a total of 30 ICD-10 diagnostic codes:

```{r simulate-data-cs}
set.seed(1)
x <- data.frame(
  id = sample(1:3, size = 30, replace = TRUE),
  code = sample_diag(n = 30)
)
```

We could compute the Charlson comorbidity domains:

```{r charlson}
charlson <- comorbidity(x = x, id = "id", code = "code", map = "charlson_icd10_quan", assign0 = FALSE)
charlson
```

We set the `assign0` argument to `FALSE` to not apply a hierarchy of comorbidity codes, as described in `?comorbidity::comorbidity`.

Alternatively, we could compute the Elixhauser score:

```{r elixhauser}
elixhauser <- comorbidity(x = x, id = "id", code = "code", map = "elixhauser_icd10_quan", assign0 = FALSE)
elixhauser
```

Weighted an unweighted comorbidity scores can be obtained using the `score()` function:

```{r score}
unw_cci <- score(charlson, weights = NULL, assign0 = FALSE)
unw_cci

quan_cci <- score(charlson, weights = "quan", assign0 = FALSE)
quan_cci

all.equal(unw_cci, quan_cci)
```

Code for the Elixhauser score is omitted, but works analogously.

Conversely, say we have 5 individuals with a total of 100 ICD-9 diagnostic codes:

```{r simulate-data-cs-9}
set.seed(3)
x <- data.frame(
  id = sample(1:5, size = 100, replace = TRUE),
  code = sample_diag(n = 100, version = "ICD9_2015")
)
```

The Charlson and Elixhauser comorbidity codes can be easily computed once again:

```{r charlson-9}
charlson9 <- comorbidity(x = x, id = "id", code = "code", map = "charlson_icd9_quan", assign0 = FALSE)
charlson9
```

```{r elixhauser-9}
elixhauser9 <- comorbidity(x = x, id = "id", code = "code", map = "elixhauser_icd9_quan", assign0 = FALSE)
elixhauser9
```

Scores:

```{r score-9}
unw_eci <- score(elixhauser9, weights = NULL, assign0 = FALSE)
vw_eci <- score(elixhauser9, weights = "vw", assign0 = FALSE)
all.equal(unw_eci, vw_eci)
```

## Citation

If you find `comorbidity` useful, please cite it in your publications:

```{r citation}
citation("comorbidity")
```

## References

More details on which comorbidity mapping and scoring algorithm are available within the package can be found in the two accompanying vignettes, which can be accessed on CRAN or directly from your R session:

```r
vignette("A-introduction", package = "comorbidity")
vignette("B-comorbidity-scores", package = "comorbidity")
```

The list of available algorithms can be printed interactively using the `available_algorithms()` function:

```{r}
available_algorithms()
```

## Copyright

The icon for the hex sticker was made by Freepik from .

Owner

JOSS Publication

comorbidity: An R package for computing comorbidity scores

Published

March 30, 2018

DOI

10.21105/joss.00648

Volume 3, Issue 23, Page 648

Authors

Alessandro Gasparini
Biostatistics Research Group, Department of Health Sciences, University of Leicester

Editor

Thomas J. Leeper

Tags

comorbidity scores administrative data epidemiology biostatistics ICD

View PDF Review Thread Software Archive

Papers & Mentions

GitHub Events

Committers

Last synced: 7 months ago

Issues and Pull Requests

Last synced: 6 months ago

Packages