Electrical Engineering and Computer Science
Mathematical Sciences
University of Wisconsin-Milwaukee
Milwaukee, WI 53201
The PEAS (Patient Education and Activation System) project aims to prepare people to take a more active role in health care decisions. The project is investigating strategies for helping people identify their health care concerns, learn what actions they can take on their own, and, if necessary, be able to verbalize their concerns to health care professionals. These strategies combine a multi-modal computer interface (including typed text and mouse-inputs) with intelligent tutoring and intelligent discourse processing. As PEAS interacts with a patient, it will vary the content and pace of the interaction and suggest relevant learning activities.
PEAS can help improve the effectiveness of health care providers, by helping bridge the communication gap that exists between patients and their doctors. Modern medical care offers doctors a wide variety of options for responding to a patient's illness or injury. Evaluating these options requires that the patient take a more active role in decision-making, because the correct decision often depends on subjective factors such as the patient's life style or environment, her tolerance for risk, or her desired quality of life. However, for patients to give a reliable account of their preferences they must be able to understand the implications of their choices. Computers can help prepare people for decision-making, by giving people access to medical information outside the doctor's office and by adapting the presentation of the information to suit the individual's interests or expertise.
With the collaboration of the Department of Medicine at the University of Wisconsin Medical School, Milwaukee Area Clinical Campus (Sinai Samaritan Medical Center), the PEAS project group has been developing a coordinated set of computer programs for educating medical patients. The system present medical information to users interactively and in understandable terms, tailoring the interaction to the information that the user provides about her medical history and concerns. The system will also direct users to tools that allow them to explore the effects of different health care choices. For example, one tool allows patients to explore their preferences for different medical outcomes. Another tool allows users to consider the health risks and benefits associated with possible nutritional strategies, given their lifestyle and medical history.
LEAF (Layman
Education and Activation Form) addresses the problem of presenting
medical information to patients incrementally and selectively on the basis
of their reported history and interests. This project aims to create a
more informed and ``activated patient'' by extending the normal activity
of filling in a medical history form to include educational activities
that help patients understand the terminology of the form and suggest topics
that they might want to discuss with their doctor. For example, if the
patient has indicated that they have a preventable medical condition, the
system will offer information about prevention and treatment strategies.
It will also suggest questions that she might ask her doctor. Unlike a
brochure, the presentation of this information is incremental, and interactive,
allowing the system to monitor the patient's attention and level of understanding
and adapt the interaction appropriately. LEAF will also filter out irrelevant
parts of the form, which, because they are irrelevant, are most likely
to contain terminology that is unfamiliar or confusing.
PETIL (Preference Ellicitation Tool using Inductive Learning) addresses the problem of assessing patient preferences for different treatment options based on their subjective feelings, PETIL allows a patient to explore her preferences for different health outcomes and then see how those preferences might affect similar medical decisions. The tool presents a set of scenarios involving risky outcomes, along with questions that the user must answer. The system uses the answers to construct a model of the patient and shows her how the system would predict her response for similar scenarios. In future work the tool will include a component that analyzes the model for properties such as risk aversion or inconsistency, and direct the user to appropriate educational resources.
NEST (Nutrition Education System) addresses the problem of educating the general public about life skills, such as good nutrition and budgeting. Chronic and preventable illnesses consume an ever increasing amount of the U.S. health care dollar (76% percent of every dollar spent on direct medical care goes for chronic diseases (JAMA, November 1996); bad habits and unhealthy lifestyle cause sixty-five percent of cancer deaths (Cancer Causes and Control, November 1996)). However, social service organizations cannot keep up with the growing need for education programs. Computers can help bridge the gap. NEST uses an underlying statistical model along with an interactive, multi-modal interface to implement a proven educational approach: identify particular concerns of the student and then identify specific activities that are likely to reduce the effects of those problems.
The PEAS project has two working groups: The medical group is headed
by Dr. Robert McNutt, MD, chair of the Department of Medicine. They are
addressing the problems of providing medical content and designing clinical
evaluations. The computer group is headed by Susan McRoy, Assistant Professor
of Computer Science at the University of Wisconsin Milwaukee. This group
is designing and implementing appropriate tools as well as addressing the
problems of how to collect, analyze, and present the relevant information
efficiently and effectively.