Cognitive systems at the point of care: The CREDO program.

CREDO is a framework for understanding human expertise and for designing and deploying systems that support cognitive tasks like situation and risk assessment, decision-making, therapy planning and workflow management. The framework has evolved through an extensive program of research on human decision-making and clinical practice. It draws on concepts from cognitive science, and has contributed new results to cognitive theory and understanding of human expertise and knowledge-based AI. These results are exploited in a suite of technologies for designing, implementing and deploying clinical services, early versions of which were reported by Das et al. (1997) [9] and Fox and Das (2000) [26]. A practical outcome of the CREDO program is a technology stack, a key element of which is an agent specification language (PROforma: Sutton and Fox (2003) [55]) which has proved to be a versatile tool for designing point of care applications in many clinical specialties and settings. Since software became available for implementing and deploying PROforma applications many kinds of services have been successfully built and trialed, some of which are in large-scale routine use. This retrospective describes the foundations of the CREDO model, summarizes the main theoretical, technical and clinical contributions, and discusses benefits of the cognitive approach.