blandr: a Bland-Altman Method Comparison package for R

blandr carries out Bland Altman analyses (also known as a Tukey mean-difference plot) as described by JM Bland and DG Altman in 1986.

This package was started in 2015 as existing Bland-Altman R functions did not calculate confidence intervals: my belief is that when drawing Bland-Altman plots, confidence intervals should be considered the gold-standard. blandr fulfils this need!

The benefits

Citations

You can find the citation information through the usual R citation commands:

citation("blandr")
#> 
#> To cite blandr in publications, please use:
#> 
#>   Datta, D. (2017). blandr: a Bland-Altman Method Comparison
#>   package for R. Zenodo. DOI:10.5281/zenodo.824514
#>   https://github.com/deepankardatta/blandr
#> 
#> A BibTeX entry for LaTeX users is
#> 
#>   @Manual{,
#>     title = {blandr: a Bland-Altman Method Comparison package for R},
#>     author = {Deepankar Datta},
#>     doi = {10.5281/zenodo.824514},
#>     year = {2017},
#>     url = {https://github.com/deepankardatta/blandr},
#>   }

The DOI will refer to all versions of blandr. If you need to cite specific releases DOIs, the full versioning information can be found at Zenodo (https://zenodo.org/record/824515), with the full source code at the blandr GitHub page (https://github.com/deepankardatta/blandr/).

Installation

You can install blandr from github with:

install.packages("devtools")
devtools::install_github("deepankardatta/blandr")

Currently the code is hosted at https://github.com/deepankardatta/blandr/ - in future I hope it will be available on CRAN for easy integration into R

Example

This is a basic example which shows you how to solve a common problem:

library(blandr)
load(file="Data/bland.altman.PEFR.1986.rda")
blandr.display ( bland.altman.PEFR.1986$WrightFirst , bland.altman.PEFR.1986$MiniWrightFirst , sig.level=0.95 )
blandr.draw( bland.altman.PEFR.1986$WrightFirst , bland.altman.PEFR.1986$MiniWrightFirst )

Why release a version <1.0?

I am believer that if the function is good enough to work, I should just publish, see if others can improve it and just iterate slowly to get to a version 1.0. What counts as a 1.0? Well I’d like other people to help validate it, and add a few more functions, to a point there should not be much more to add to this.

From what I’ve read this is what’s called a “Minimum Viable Product” (https://en.wikipedia.org/wiki/Minimum_viable_product).

Why the name?

Thinking of a unique name was difficult - and it wasn’t worth spending/wasting time on this initially. Curiously it was going to be called BlandAltmanEdinburgh (as I was in Edinburgh at the time I made this) until I considerd uploading this to GitHub and CRAN.

Hadley Wickham has an excellent set of documentation on creating packages. The one on naming (http://r-pkgs.had.co.nz/package.html) is worth a read. Reflecting on it a lot of the naming issues are to prevent collisions with similarly named packages, so using blandaltman in the name might have cause problems. Mirroring the pre-existing plyr and knitr I thought I’d just add a “r” to “bland”: yes it doesn’t involve Altman’s name, but I couldn’t think of anything better.

If you can think of a better name please let me know!

Further information

Further information can be found in the function help files in the package, as well as in the vignettes.

Future improvements

Whilst this package is good enough for use, there is the scope for iterative improvements.

Future works include:

Help wanted!

Comments on the code, suggestions for improvement, verification tests, validation studies, and even code contributions would be gratefully accepted. Email me at deepankardatta@nhs.net and I’ll try and get back to you as soon as possible. Please do bear in mind this is a side-project, and I can be otherwise busy with a lot of other work.