Those of you who follow along regularly know the importance I place on workflow for everything from EHR design to practice optimization. Workflow technology is rapidly maturing and deserves consideration for clinical applications. Clinical decision support is an obvious way to introduce workflow technology into healthcare systems. However, computerization of guidelines has a long history in informatics and has resulted in a number of informatics-specific approaches based on guideline modeling languages such as EON, GLIF, and Asbru. Naturally, I’ve often wondered how workflow management systems (WFMS) would compare to traditional computer-interpretable guideline languages (CIGs) for clinical decision support. These two articles provide a starting point for answering this question.
Workflow pattern and guideline languages
A Pattern-Based Analysis of Clinical Computer-Interpretable Guideline Modeling Languages, by Mulyar, van der Aalst, and Peleg, provides a wonderful analysis of CIGs in comparison to standard workflow control-flow patterns. It reveals that CIGs have evolved quickly and have substantial capabilities. When analyzed according to standard workflow patterns, they hold up very well. Since WFMS designers take standard workflow patterns into account when evolving their products, support for workflow patterns, or the lack thereof, offers an objective means for comparing WFMS products. Fortunately, with a few caveats, standard workflow patterns also allow for comparisons between CIGs and WFMS. This paper is from 2007, and it lays the groundwork for future comparisons between the two technologies. (For more about workflow patterns see the tutorial series Clinical Workflow Analysis Using Workflow Patterns.)
Workflow technology and CDS
The efficacy of using WFMS for CDS is directly addressed by Huser, et al. In Implementation of Workflow Engine Technology to Deliver Basic Clinical Decision Support Functionality, the authors relate their experiences in designing a clinical decision support intervention using an open source workflow engine (HealthFlow). Using this WFMS, they were able to integrate CDS capability with their EHR. The description of their project is quite detailed, and it offers valuable insights into the challenges of using a WFMS. If you are interested in evaluating workflow technology, whether for CDS or another application, this paper is a great starting point.
Both papers are located on the Clinical Workflow page.