36 PERVASIVE computing Published by the IEEE Computer Society ■ 1536-1268/07/$25.00 © 2007 IEEE Distributed Healthcare: Simultaneous Assessment of Multiple Individuals A s treatments for life-threatening ill- nesses improve, life expectancy increases along with the proportion of healthcare dollars supporting chronic care. Combine this with the growing number of aging baby boomers (who are most at risk for these chronic diseases), and we see a greater demand for healthcare alterna- tives. Issues at stake include rising costs and the importance of quality of life to the chronically ill and the elderly. Most people approach healthcare by reacting to triggered problems: when we get sick, we typ- ically wait until the symptoms start interfering with our daily life, and then we visit a clinic. At this point, particularly for pop- ulations at risk such as the elderly and the chronically ill, the treatment can often be riskier and much more expan- sive than if the problem had been dealt with earlier. Con- versely, a proactive approach to healthcare would in many cases result in more effective and much less expensive treatments, by predicting or detecting conditions earlier. Both early detection and health maintenance require a health-delivery system that can moni- tor health status. Early detection of physical and mental changes requires sensitive and frequent measurement of physiological and behavioral data. Physiological monitoring of an individual’s physical condition usually involves checking changes in heart rate, blood pressure, blood glu- cose levels, and day-to-day weight. Increasing evi- dence shows that we can also use daily behaviors, such as sleep patterns, walking speed, and move- ments in and outside the home, to monitor phys- ical health and even mental condition. However, gathering behavioral data requires an intensity of monitoring that’s difficult and expensive to achieve in a clinic environment. As we show in our work, which permits unobtrusive continu- ous behavioral monitoring of individuals in their home, pervasive computing technology offers a practical and economically feasible way to make frequent assessments. Behavioral monitoring in the home Our research’s basic premise is that we can observe and assess many informative behaviors in a person’s normal life just as a skilled clinician can assess a patient’s state from his posture, gait, and demeanor. For example, a condition such as arthritis causes physical discomfort, which can be reflected in an individual’s overt behaviors. In addition to the obvious physical connection, increasing evi- dence shows that we can relate observable every- day behaviors to neurological states. Studies show that variability in mobility measures, such as Employing pervasive computing technologies can help enable continuous patient monitoring and assessment in various settings outside of hospitals, lowering healthcare costs and allowing earlier detection of problems. HEALTHCARE Tamara L. Hayes, Misha Pavel, Nicole Larimer, Ishan A. Tsay, and John Nutt Oregon Health and Science University Andre Gustavo Adami University of Caxias do Sul