Clinical diagnostic decision support systems and diagnostic process definitions

Diagnosis

Miller and Geissbuhler (n.d.) define diagnosis as "the placing of an interpretive, higher level label on a set of raw, more primitive observations" thereby treating diagnosis as a single event and not a chain of events or a process. This definition has not been widely accepted due to the practical nature of most diagnosis especially clinical diagnosis. For purposes of trying to explain this term various models have been developed by healthcare informaticians, on such model is the Clinical Diagnostic Decision Support System (CDDSS) (Coiera, 2003). This model has played a very significant part in the clinical knowledge management technologies by supporting the clinical process. A CDDSS is a computer-based algorithm that is used by clinicians in the diagnostic process and helps them in a few component steps. This model incorporates active knowledge systems which generate case specific advice through a few data of a patient and presents a further variation of the definition of diagnosis. The typical function of this system is the integration of medical knowledge base, patient data and an inference engine that generates case specific advice.

The CDDSS is based on the fact that no diagnostic problem can be defined in an absolute sense and it has to be understood that the origin of the problem to be solved is the clinician-user's mind. Thus diagnosis, as defined by CDDSS, is a task that each CDDSS functionality can perform (or assist users in performing), implying that it is a series of events. Since this system was trying to redefine what had already existed, its development was not very easy. A number of assumptions that had a strong influence on the development of CDDSS and have been commonly referred to as myths, these assumptions included the following. "Diagnosis is the dominant decision-making issue in medicine," "Clinicians will use knowledge-based systems if the programs can be shown to function at the level of experts," and "Clinicians will use stand-alone decision-support tools." (Shortliffe, 1986). These myths which were held by clinicians played a part in delaying the success of the CDDSS development but they are being overtaken by time and are fading away. Initially, CDDSS was viewed as a tool that was to replace rather than supplement the diagnostic capabilities of the clinician-users. However, this has changed among the experienced users. Some naive users have also expected too much from the system with some even expecting the system to behave in a similar manner to them (experts) (Shortliffe, 2006). These misunderstandings have led to the system being opposed for a long time.