Automating individualized coaching and authentic role-play practice for brief intervention training.

OBJECTIVES: Brief intervention helps to reduce alcohol abuse, but there is a need for accessible, cost-effective training of clinicians. This study evaluated STAR Workshop , a web-based training system that automates efficacious techniques for individualized coaching and authentic role-play practice. METHODS: We compared STAR Workshop to a web-based, self-guided e-book and a no-treatment control, for training the Engage for Change (E4C) brief intervention protocol. Subjects were medical and nursing students. Brief written skill probes tested subjects' performance of individual protocol steps, in different clinical scenarios, at three test times: pre-training, post-training, and post-delay (two weeks). Subjects also did live phone interviews with a standardized patient, post-delay. RESULTS: STAR subjects performed significantly better than both other groups. They showed significantly greater improvement from pre-training probes to post-training and post-delay probes. They scored significantly higher on post-delay phone interviews. CONCLUSION: STAR Workshop appears to be an accessible, cost-effective approach for training students to use the E4C protocol for brief intervention in alcohol abuse. It may also be useful for training other clinical interviewing protocols.

Evaluation of a medical diagnosis system using simulator test scenarios.

This paper describes an informal but systematic method for how to test and verify a knowledge-based system in a large open-ended medical target domain. The system used is Guardian, an intelligent system for monitoring and diagnosis of post-cardiac surgery patients in an intensive-care unit. The knowledge base is tested and verified by running the system on a series of realistic test scenarios, both with an embedded simulator and with an external simulation system. The same scenarios are presented to human test subjects, making it possible to compare and analyze the performance of the knowledge-based system with that of human physicians. The use of simulators instead of clinical data also means that it is possible to test crucial scenarios which occur seldom in medical practice. Our results show that a system like Guardian might indeed be useful in medical care.

The patient advocate: a cooperative agent to support patient-centered needs and demands.

Knowledge-based monitoring and therapy planning systems were mainly built for the convenience of health care providers. They neglected the consumers of health care, namely, the patients. Our approach is concentrated on the individual patients' demands and needs. We are designing, building, and demonstrating a cooperative agent to support patients' management of their own health-related behavior on a day-to-day basis at home. Clinical treatment protocols are represented in an intention-based time-oriented representation language to overcome the drawbacks of vague or ill-structured problem definitions (e.g., missing functional dependencies). These representations are used to guide the patients, to provide necessary explanations, and to observe and critique whether the patients obey the instructions of the health-care providers. We will present a prototype which supports women with gestational diabetes mellitus.

Guaranteeing real-time response with limited resources.

Unanticipated problems detected by patient-monitoring systems may sometimes require real-time response in order to provide high-quality care and avoid catastrophic outcomes. In this paper, we present an approach for guaranteeing a response to such events by a monitoring agent even in situations where we have limited problem-solving resources. We show that an action-based hierarchy can accomplish this goal. We also analyze the performance of this hierarchy under varying resource availability and discuss decision-theoretic approaches to enable us to best structure such a hierarchy. We also describe an implementation of these ideas, called ReAct, in the BB1 architecture. All the ideas are illustrated with examples from the surgical intensive care unit (SICU).