The Open Medical Informatics Journal

2010, 4 : 245-254
Published online 2010 December 14. DOI: 10.2174/1874431101004010245
Publisher ID: TOMINFOJ-4-245

RESEARCH ARTICLE
System-Agnostic Clinical Decision Support Services: Benefits and Challenges for Scalable Decision Support

Kensaku Kawamoto, * , Guilherme Del Fiol , Charles Orton and David F Lobach
Division of Clinical Informatics, Department of Community and Family Medicine, Box 2914, Duke University Medical Center, Durham, NC 27710, USA

* Address correspondence to this author at the Division of Clinical Informatics, Department of Community and Family Medicine, Duke Institute for Genome Sciences and Policy, Box 2914, Duke University Medical Center, Durham, NC 27710, USA; Tel: (919) 684-2340; Fax: (919) 684-8675; E-mail: kawam001@mc.duke.edu

ABSTRACT

System-agnostic clinical decision support (CDS) services provide patient evaluation capabilities that are independent of specific CDS systems and system implementation contexts. While such system-agnostic CDS services hold great potential for facilitating the widespread implementation of CDS systems, little has been described regarding the benefits and challenges of their use. In this manuscript, the authors address this need by describing potential benefits and challenges of using a system-agnostic CDS service. This analysis is based on the authors’ formal assessments of, and practical experiences with, various approaches to developing, implementing, and maintaining CDS capabilities. In particular, the analysis draws on the authors’ experience developing and leveraging a system-agnostic CDS Web service known as SEBASTIAN. A primary potential benefit of using a system-agnostic CDS service is the relative ease and flexibility with which the service can be leveraged to implement CDS capabilities across applications and care settings. Other important potential benefits include facilitation of centralized knowledge management and knowledge sharing; the potential to support multiple underlying knowledge representations and knowledge resources through a common service interface; improved simplicity and componentization; easier testing and validation; and the enabling of distributed CDS system development. Conversely, important potential challenges include the increased effort required to develop knowledge resources capable of being used in many contexts and the critical need to standardize the service interface. Despite these challenges, our experiences to date indicate that the benefits of using a system-agnostic CDS service generally outweigh the challenges of using this approach to implementing and maintaining CDS systems.

Keywords:

Clinical decision support, decision support service, Web service, SEBASTIAN, service-oriented architecture.