Plans and situated actions: the problem of human-machine communication
Plans and situated actions: the problem of human-machine communication
Conceptions of the user in computer systems design
The social and interactional dimensions of human-computer interfaces
Rationalizing Medical Work: Decision-Support Techniques and Medical Practices
Rationalizing Medical Work: Decision-Support Techniques and Medical Practices
Flexible guideline-based patient careflow systems
Artificial Intelligence in Medicine
Knowledge Acquisition and Modeling in Clinical Information Systems: A Case Study
EKAW '02 Proceedings of the 13th International Conference on Knowledge Engineering and Knowledge Management. Ontologies and the Semantic Web
Hi-index | 0.00 |
The standard model of clinical work is a fixed sequence of tasks covering diagnosis and treatment of patients. Knowledge based systems have been designed according to this sequence. This ideal typical approach accounts for the relative modest success of knowledge based systems in healthcare practice. In reality however, clinical work is highly contingent, ad-hoc and idiosyncratic and therefore hard to fit into in a formal model. A physician is said to manage complex and diverse patient trajectories. Therefore the concept of a trajectory should not only relate to the course of the disease of a patient, but to all the organizational work during that course as well. We will highlight two aspects of this 'messy' view of clinical work and examine the consequences for the design, implementation and evaluation of knowledge based systems. Articulation refers to the fact that a lot of invisible work is being done in order to complete a visible task of a physician. A physician may see a patient, but before she can do that a lot of work has been done to assure that she actually sees the patient. Localization refers to the fact that clinical work is being adapted to local and situational circumstances. This is not primarily related to the variance in medical work as a result of uncertain knowledge about the true clinical state of a patient, but to the constant negotiating with colleagues, local opportunities and restraints and the possibilities of protocols and technologies. In short, the way how patient trajectories are being shaped by human and nonhuman elements. A knowledge based system that has the potential of adaptability to patient trajectories seems to offer new opportunities. Such an approach would place the user in the centerfold of the design, implementation and evaluation of such systems.