The Clinician's Advisor (CLAD) application is built on the same knowledge and processing substrate as Maryland Virtual Patient (MVP). It is intended to reduce the cognitive load of clinicians caused by the vast amount of information available, and to improve overall patient outcome through providing high-value decision-making assistance.

CLAD will assist clinicians by providing advice (along with its justification), answering questions, providing prognoses, carrying out administrative tasks (e.g., finding out if a given procedure is covered by the patient's insurance company), and so on. CLAD will ultimately be able to receive new information in two ways, through language recorded in the chart and by listening in to the conversation between the doctor and the patient; the first of these is implemented and the second is under development.

For CLAD, we have been enhancing the decision-making capabilities of our agents by (a) incorporating more features - including the preferences of different "stakeholders" - and (b) supporting decision-making under uncertainty. We have selected to use influence diagrams to model the relevant clinical knowledge and reasoning, as they are a convenient graphical representation facilitating the creation of Bayesian networks and/or decision trees that are well-suited to our domain.