International Workshop on Dynamic Modeling of Health Behavior Change and Maintenance: Moving the Field Forward

与健康相关的不良行为和习惯造成了约40%的可预防死亡和大部分慢性病负担。然而，我们目前对健康相关行为的理解主要是基于对人类行为的静态快照，而不是对不断变化的生物、社会和个人环境状态做出反应的持续的、动态的行为反馈循环。通过新兴技术，包括可穿戴和可部署的传感器和移动电话，丰富的连续数据流正变得越来越可用。这些数据与主要来自工程领域的复杂建模技术相结合，可以提供前所未有的机会，在背景下和实时地理解行为。然而，为了利用这些机会，行为/健康科学家和数据建模师之间以及跨复杂数据建模技术的不同学派之间需要专门的合作。拟议的讲习班侧重于为新兴技术日益提供的时间密集、背景丰富和个性化的数据建立模型的新概念。这次研讨会的主要目标是1)融合不同计算建模方法的子学科的技术，2)促进共享词汇/本体的开发，以促进建模者、行为和健康科学家之间的交流，以及3)将动态行为理论的严谨性提高到下一个水平，走向因果和预测性动态模型。21世纪数据建模和健康行为研究的挑战是朝着能够捕捉行为状态和相关影响因素的复杂和快速变化的行为的计算、动态建模的方向发展。这些新模型将为及时、适应性干预(JITAI)铺平道路，这种干预在人们最容易接受和最有可能受益的时刻提供背景反馈。为了实现这一目标，这里提出的研讨会将汇集各种类型的建模专家，例如(但不限于)系统动力学、社会网络、基于代理的建模、机器学习和贝叶斯推理，与行为科学家和卫生保健专业人员合作，将这一努力推向新的水平。拟议的研讨会将遵循数字健康干预措施开发和评估方法国际研讨会，该研讨会将于2015年9月10日至11日在英国伦敦举行，由苏珊·米奇博士和杰里米·怀亚特博士领导，由英国医学研究理事会(MRC)赞助。这里提出的研讨会将健康行为和移动干预领域的国际领导者与大数据建模领域的领导者聚集在一起。这次研讨会将促进不同的“大数据”模型师和行为学家之间前所未有的合作，以开发新的范式来建模时间密集的、有背景的行为数据，以指导未来跨健康和健康领域的联合技术援助。

Poor health-related behaviors and habits are responsible for approximately 40% of preventable deaths and the majority of the chronic disease burden. However, our current understanding of health-related behavior is largely based on static snapshots of human behavior, rather than ongoing, dynamic feedback loops of behavior in response to ever-changing biological, social and personal environmental states. Rich streams of continuous data are becoming increasingly available through emerging technologies, including wearable and deployable sensors and mobile phones. This data, combined with sophisticated modeling techniques emanating primarily from engineering fields, can provide unprecedented opportunities to understand behavior in context and in real time. However, to take advantage of these opportunities, dedicated collaborations between behavioral/health scientists and data modelers, as well as across different schools of complex data modeling techniques, are required. The proposed workshop focuses on developing new concepts for modeling the temporally dense, contextually rich and personalized data increasingly afforded by emerging technologies. The major goals of this proposed workshop are to 1) amalgamate techniques from sub-disciplines across different computational modeling approaches, 2) facilitate development of a shared vocabulary/ontology to facilitate communication between modelers, behavioral and health scientists and 3) advance the rigor of dynamic behavioral theories to the next level towards causal and predictive dynamic models.The challenge to 21st century data modeling and health behavior research is to move toward computational, dynamic modeling of behavior that can capture complex and rapid changes in behavioral state and related influencing factors. These new models will pave the way for Just-In-Time, Adaptive Interventions (JITAI) that provide feedback in context, in the moment, when people are most receptive and most likely to benefit. To accomplish this, the workshop proposed here will bring together experts in various types of modeling, for example (but not limited) to systems dynamics, social networks, agent-based modeling, machine learning, and Bayesian inference to work together with behavioral scientists and health care professionals to move this endeavor to the next level. The proposed workshop will follow the International Workshop on Methodologies for Developing and Evaluating Digital Health Interventions, which will be held in London UK, 10-11 September 2015, led by Dr. Susan Michie and Dr. Jeremy Wyatt, sponsored by the Medical Research Council (MRC), UK. The workshop proposed here brings the international leaders in health behavior and mobile interventions together with leaders in big data modeling. This workshop will facilitate an unprecedented collaboration between different streams of "big data" modelers and behaviorists to develop new paradigms for modeling temporally dense, contextualized behavioral data that can guide future JITIAs across health and wellness domains.