Prescriptive Grammar for Clinical Prescribing Workflow

Introduction

The increasingly high cost of healthcare has cast a spotlight on hospital inefficiencies (Born & Levinson, 2017; MacMillan, Slessarev, & Etchells, 2016), and a search for solutions to allow for the efficient use of hospital resources. Inefficiencies in healthcare settings can lead to a variety of problems including patient safety concerns, wasted resources, and long wait times. To combat these problems, a renewed interest in researching clinical workflow has taken hold. A workflow is defined as an ordered procedure of tasks followed by an individual in an environment (Brixey, Robinson, Turley, & Zhang, 2010). Examining a prescription with this definition, the researchers can observe that a medication prescription is a workflow. A prescription is a set of instructions for in this case a patient and a pharmacist to follow. When analyzing current clinical prescribing workflows in place, there can be observed that there are many existing problems and routes to optimize current routines. Current clinical prescribing workflows involve a variety of paper and digital solutions fractured by being spread across a variety of programs and documents. These environments are still using a variety of paper and electronic methods for storing patient information despite the clear benefits of moving to entirely computerized implementations. Research into the effects of Health Information Technology (HIT) on the current workflow shows that much of the time is spent on redundant data entry tasks (MacMillan, Slessarev, & Etchells, 2016).

This research attempts to model a clinical prescribing workflow to a textual Domain Specific Language (DSL). The researcher’s Parser Combinator Grammar (PCG) will use direct and indirect mapping of medical concepts to the researcher's syntax as shown in Figure 1. The grammar, along with the authors Integrated Development Environment (IDE) will allow for the creation of prescriptions while maintaining the Electronic Health Record (EHR). The researcher's solution will look to implement itself in the existing workflow as in Figure 2.

Figure 1.

Mapping of medication concepts

Figure 2.

Prescribing workflow