Calculate the Care Density or Fragmented Care Density Given a Patient-Sharing Network.
CareDensity
Author: Robin Denz
Description
CareDensity
is a small R-Package which can be used to calculate the care density index and the fragmented care density.
Installation
A developmental version of this package may be installed from github using the remotes
R-Package:
library(remotes)
remotes::install_github("RobinDenz1/CareDensity")
Bug Reports and Feature Requests
If you encounter any bugs or have any specific feature requests, please file an Issue.
Examples
Suppose you have the following patient-provider data:
data <- data.frame(PatID=c("1", "1", "1", "2", "2", "3", "3", "4", "5"),
ArztID=c("A", "C", "D", "A", "D", "A", "D", "D", "C"))
This is essentially a description of the following bipartite network:

The simple care density can then be calculated like this:
care_density(data)
Citation
Use citation("CareDensity")
to get the relevant citation information.
License
© 2024 Robin Denz
The contents of this repository are distributed under the GNU General Public License. You can find the full text of this License in this github repository. Alternatively, see http://www.gnu.org/licenses/.