Participatory Design and Artificial Intelligence: Strategies to Improve Health Communication for Diverse Audiences

A major public health challenge is to develop large-scale health communication interventions that are successful with diverse and vulnerable audiences. Participatory design approaches are critical to create communication programs that are relevant to people’s literacy, language, culture, access and functional needs. Further, there are powerful synergies in linking participatory design and artificial intelligence methods. This paper focuses on traditional weaknesses of health communication, and participatory design strategies and models that can be used by developers, researchers and health practitioners. Why health communication matters As annual US health care costs escalate to over $2.3 billion (16% of the Gross Domestic Product), there is increasing pressure to improve health and reduce demands on the health care system (CMS 2010). One half of deaths each year have been attributed to preventable behavioral and social factors (McGinnis & Foege 1993). A US Institute of Medicine (2003) report, provided estimates that early detection screening could reduce mortality rates from various cancers from 25 to 80%. Unfortunately, public health and medical efforts have proven woefully inadequate to reach US population health goals (USDHHS 2000). Rates of obesity and diabetes continue to skyrocket (USDHHS 2008a; USDHHS 2008b). Equally distressing are the ever-widening health disparities that result in a disproportionate burden of disease among minority groups in our society (Kreps 2006). Health communication, “the central social process in the provision of health care delivery and the promotion of public health” (Kreps 1988), has been a primary strategy to improve people’s health. This strategy has historically focused on disseminating evidence-based messages from experts to the public in the hope of motivating them to adopt healthy behaviors and use health care effectively. However, there are many other viable channels for disseminating health information to different audiences. Clearly, there is a major gap between our well-intentioned communication actions and what people actually do. _______________________________ Copyright © 2010, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Health communication problems Although one-way, generic messages from experts to the public (“Exercise 30 minutes a day;” “Eat 5 fruits and vegetables a day”) are based on solid scientific research, they have often shown disappointing outcomes. Health communication models and approaches have overemphasized individual risks and rational decision-making (Neuhauser & Kreps 2003). Research during the past half century is helping us understand that people’s health decisions are greatly influenced by dynamic and complex social contexts. Our view is: “We experts have messages to send, but people have lives to live, and rarely do we bridge that gap.” Newer “socio-ecological” models take into consideration how socio-cultural, institutional and environmental factors at multiple levels affect people’s behaviors (Stokols 2000). Such models also draw on important findings of social epidemiological research showing that people’s sense of control over life conditions may be the best predictor of their health (Syme 1991). Recommendations to do better Radical changes to thinking about health communication have prompted key recommendations to improve our efforts (Neuhauser & Kreps 2010): Strengthen health behavior models Socio-ecological models are providing important guidance to improve health communication through attention to multiple social contexts. However, more work is needed to define how people dynamically interact with health decisions and actions within and among these contexts. The fields of social semiotics, cybernetics, informatics, artificial intelligence and others are making significant contributions to our understanding about how the “receiver” (rather than the expert “sender”) perceives and interacts with health information to take more control over their health. For example, social semiotic models focus on gaining a deep understanding of how people react to messages in their social settings. Social cybernetic frameworks posit how people use information to achieve goals within their social systems. 49 AI and Health Communication — Papers from the AAAI 2011 Spring Symposium (SS-11-01)