Overview

A number of methods to support the use guidelines in clinical decision support have been or are being developed by the medical informatics community. The SAGE project takes a model-based approach where we develop a formal model that specify the entities and relationship that constitute a computer-interpretable guideline. The model acts as a template for encoding instances of guidelines. Through the model, the medical knowledge in a guideline is formalized so that electronic applications can apply it to generate patient-specific recommendations for clinical decisions and actions.

The SAGE Guideline Model is designed to

  1. use standardized components that are sufficiently granule and precise, to allow interoperability of guideline execution elements with the standard services provided within vendor clinical information systems.
  2. encode guideline knowledge needed to provide situation-specific decision support, and to maintain linked explanatory resource information for the end-user
  3. include organizational knowledge to capture workflow information and resources needed to provide decision-support in enterprise setting
  4. to be well structured, so that the bulk of guideline knowledge can be encoded, through a guideline workbench, by clinicians with basic understanding of the guideline model.
  5. synthesize prior guideline modeling work for encoding guideline knowledge needed to provide situation-specific decision support, and to maintain linked explanatory resource information for the end-user

In these pages, we describe

the approach we have taken to formulate SAGE Guideline Model,

the major components of the model, and

the tool that we use to represent the model.

Details of the SAGE Guideline Model can be found in the SAGE Guideline Model document [pdf]. Additional references are in Link and Reference sections.

With our Guideline Model, we can

  • Represent arbitrarily complex medical knowledge
  • Describe event timing and scheduling within clinical workflows
  • Provide decision logic for medication choices, dosing, contraindications
  • Specify flowsheets as workflow enhancers (nursing, medical, RT, PT, etc.)
  • Describe pro-active monitoring of time sensitive clinical events
  • Customize evidence-based care to level of individual patient
  • Specify alerts and their escalation
  • Provide precise specifications for where in the clinical workflow guidance should be activated
  • Be aware of theoretical workflows and real active workflows; and can provide triggers for connecting them.
  • Specify opportunities for decision support woken up via workflows: fits perfectly with ForeSight modes of operation.