Electronic healthcare data are now used routinely to estimate whether patients take their medications as their doctor prescribes, as treatment effectiveness relies on patients’ adherence to medication. Yet, performing the actual computations correctly to turn these data into meaningful values is not quite simple. The algorithms should fit the clinical context of the medication studied, and produce the same results if replicated by a different person. Without valid and reliable methods of computing adherence, there is no way to know if we can trust (and compare) the myriad studies and clinical tools that use these estimates.
Sounds familiar? In the age of big data, finding the signal in the noise is everyone’s problem. Yet, solutions need to be developed for every domain of activity. For adherence to medication there is now AdhereR, the result of a collaboration between Alexandra Dima (psychologist and health sciences researcher, University Claude Bernard Lyon 1 – HESPER, and University of Amsterdam – ASCoR), and Dan Dediu (computational linguist, statistician and software developer, Max Planck Institute for Psycholinguistics, Nijmegen). As a package developed for the statistical environment R, AdhereR is freely available for use and further development. The article introducing the software is now published in PLOS ONE (DOI:10.1371/journal.pone.0174426); it describes the methods implemented and gives practical examples of use. The package can be installed from the R repository (CRAN). Its source code is available on GitHub.