I am writing to share some work I have done on the conversion of textual dose instructions to Fast Healthcare Interoperability Resources standard structured data, as suggested by NHS Digital, by using Microsoft Excel.
I anticipate that this would be of particular utility to my colleagues in primary care, when dealing with the reports generated from the general practice clinical systems.
While the statement, ‘Take two every six hours’ can be interpreted by a human, unless it is presented as structured data to a computer, the valuable information the statement contains cannot be interpreted programmatically.
If the computer can discern that the dose quantity has a value of two, the timing repeat frequency is once, the timing repeat period is six and the timing repeat period unit is an hour, then the possibility of additional analysis of prescription data becomes possible. For example, the computer can now calculate that eight dose units can be taken per day (2 x 1 x (6/24) = 8).
Use case one: given the dose instructions and a product formulation description, it becomes possible to evaluate the strength of the preparation and how much of it is prescribed in total per day. By additionally evaluating the formulation’s drug name — where appropriate — a conversion factor can be applied, and a morphine daily equivalent (MDE) dose can be calculated.
Practice reports seeking patients taking MDE>120mg output to Microsoft Excel are typically calculated manually in an onerous process. Programmatically calculating MDE for a large data set can help quickly identify patients with prescriptions that exceed 120mg MDE by ranking the data. A pivot table could be utilised to find the total sum MDE for patients taking multiple opioids.
Use case two: given the dose instructions and the quantity supplied, it is possible to calculate the duration of the prescription. This finds application in reports where a review of antibiotic prescription duration compared with National Institute for Health and Care Excellence recommendations is the aim.
I have created some tools to illustrate these use cases. I have published the code and a Microsoft Excel add-in file to GitHub. I have also included some other functions that may be of use. They are described in the readme file of the GitHub repository. I would welcome fellow readers to use and improve the tools using GitHub’s functionality.
Robert Brownsmith MRPharmS, medicines optimisation pharmacist, Hampshire and Isle of Wight Integrated Care Board
