Towards A Representation for Sharable Guidelines

InterMed (TM) is a joint project of researchers at Harvard Medical School, Stanford University School of Medicine, Columbia University, and the American College of Physicians-American Society of Internal Medicine.

Funded by the National Library of Medicine, the Agency for Healthcare Research and Quality, and the US Army.
 

Introduction:

Clinical practice guidelines and protocols are being applied in diverse areas including policy development, utilization management, education, reference, clinical decision support, conduct of clinical trials, and workflow facilitation. Many parties are engaged in developing guidelines, an arduous task with much redundancy and overlap among the resulting products, but there is little standardization to facilitate sharing or to enable adaptation to local practice settings.

GLIF is a specification for structured representation of guidelines. It was developed by the InterMed Collaboratory in order to facilitate sharing of clinical guidelines (Ohno-Machado, Gennari et al. 1998). The InterMed collaboratory was a joint project of medical informatics laboratories at Harvard (the Decision Systems Group at Brigham and Women's Hospital and Laboratory of Computer Science at Massachusetts General Hospital), Stanford, Columbia, and McGill Universities (Shortliffe, Barnett et al. 1996). That work is being continued under new funding by a subgroup of the InterMed collaborators, including the Decision Systems Group at Harvard, McGill, Columbia, Stanford, and the American College of Physicans-American Society of Internal Medicine. The objective of the GLIF specification is to provide a representation for guidelines that have the following characteristics:

Version 2.0 of GLIF (GLIF2) was published in 1998 (Ohno-Machado, Gennari et al. 1998). That version of GLIF has been the basis for several implementations of guideline-based applications, including one in the Brigham's BICS information system (Zielstorff, Teich et al. 1998). Web-based applications for driving clinical consultations (Boxwala, Greenes et al. 1999), and applications that search for eligible clinical protocols (Ohno-Machado, Wang et al. 1999).