Online Guideline Assist for Intensive Care

© Fraunhofer FIT

In an interdisciplinary endeavour of physicians and computer scientists, the Online Guideline Assist – OLGA project develops a methodology for capturing processes in the clinical domain and for utilizing these process descriptions as guidelines for online assistance. The main objective of OLGA is to translate mainly text-based guidelines into executable workflow fragments that are fully integrated with patient data management systems (PDMS).

In hospitals, increasing implementation of quality management procedures and mounting cost pressures foster the application of best practices according to established standards. Hence, a methodology for capturing the know-how about processes and the use of this know-how for process guidance are essential ingredients of any successful hospital management.

Several knowledge and process management approaches have been proposed, such as clinical pathways, clinical practice guidelines and, lately, Standard Operating Procedures (SOP). However, formal representations of this know-how are scarce and mostly limited to well-structured processes for the treatment of chronic diseases.

Existing systems for guideline support concentrate on well-defined and structured processes, such as oncological chemotherapies and chronic diseases. However, there is little support for complex and dynamic situations, mainly because it is hard to reach a consensus among clinical experts about best practices. Then, from a technological point of view, formalization and automation of guidelines face four major challenges:

  • Diversity of control patterns for structuring activities, i.e. some follow well defined paths, whereas others need to be re-arranged on the fly according to the patient's status;
  • Representation of ad-hoc decisions that are based on personal experience rather than on text-book knowledge;
  • Monitoring of the patient's state of health (in a patient data management system – PDMS);
  • Integration with an operational environment.

A distinguishing feature of the OLGA system is its integration with an operational patient data management system. Olga's workflow features are built as turnkey add-ons to the PDMS, but, to ease acceptance by physicians and nurses, the workflows are presented and animated through means of the PDMS. Moreover, awareness concepts are included in the PDMS to make physicians aware of recommendations for treatment or any significant change in a patient’s condition.