Mentoring and research in applying digital technologies to prevent chronic Illnesses

 描述(申请人提供)：对于这个K24职业中期研究员奖以患者为中心的研究(POR)奖项，我的目标是：1)通过应用新兴的数字技术和社交媒体，成为2型糖尿病和心血管疾病初级预防领域的国际领导者；2)利用机器学习在测试创新、成本效益高的生活方式干预措施(体力活动)方面进行高影响力的研究；以及3)提供HIH优质的导师和培训环境，支持在心血管和糖尿病预防领域进行高POR影响的研究。这一K24奖项将使我不仅在护理学院，而且在加州大学旧金山分校担任更广泛和更正式的导师角色。这项新研究的总体目标是开发决策规则算法和量化模型，以优化对未来体力活动干预依从性的实时预测，并在干预试验中对这些算法和模型进行初步测试。具体目标是：1)确定基线因素在多大程度上可以预测自我监测的依从性(使用移动应用程序和佩戴先进的加速计)；2)确定基于APP的早期自我监测和评估(例如，先进的加速计和日记的使用，以及最初几周的步数)可以在多大程度上得到优化，以对未来的治疗失败进行实时预测；3)使用开发的实时预测模型(在目标2中开发)与“常规体力活动”干预相比较，评估“体力活动加”干预措施的可行性，以了解三个月期间适度和高强度体力活动的总每日步数和持续时间的变化。为了调整干预强度，我们将评估研究访问和实时反馈的频率，以最大限度地减少治疗失败和最大限度地坚持治疗。POR的初级调查人员将数字移动技术作为其研究工具的一部分，并产生了大量连续的数据。然而，在这一研究领域只有几个培训机会；特别是在严格和充分利用收集的丰富数据的方法方面。因此，拟议的K24申请也将满足这些受训者跨学科的迫切需求。

 DESCRIPTION (provided by applicant): For this K24 midcareer investigator award in patient-oriented research (POR) award, my goals are: 1) to become an international leader in the primary prevention of type 2 diabetes and cardiovascular diseases by applying emerging digital technologies and social media; 2) to conduct high-impact research in testing innovative, cost-effective lifestyle interventions (physical activity) using machine leaning; and 3) to provide a hih quality mentorship and training environment that supports the conduct of high POR impact research in cardiovascular and diabetes prevention. This K24 award will allow me to take on a broader and more formal mentorship role not only in the School of Nursing but also throughout the University of California, San Francisco campus. The overall goals of the new research in this application are to develop decision rule algorithms and quantitative models to optimize real-time predictions of future physical activity intervention adherences and to pilot test these algorithms and models in an intervention trial. The specific aims are: 1) to determine the extent to which baseline factors can predict adherence in self-monitoring (using mobile apps and wearing advanced accelerometers); 2) to determine the extent to which early app-based self-monitoring and evaluation (e.g., advanced accelerometers and diary use, and step counts in the first few weeks) can be optimized to make real-time predictions of future treatment failures; 3) to assess the feasibility of the "Physical Activity Plus" intervention using developed real-time prediction models (developed in Aim 2), compared to a "regular physical activity" intervention, for changes in total daily steps and duration of moderate and intensive physical activity over a three-month period. To tailor intervention intensity, we will evaluate the frequency of study visits and real-time feedback to minimize treatment failures and maximize adherence. Junior investigators in the POR have adapted digital mobile technologies as part of their research tools and have generated large volumes of continuous data. However, only a few training opportunities are available in this research area; in particular, for methods to rigorously and fully utilize the richness of the data collected. Therefore, the proposed K24 application will also meet these mentees' urgent needs across disciplines.