Efficacy of Precision Text Messaging to Increase Physical Activity in Insufficiently-Active Young Adults

项目概要 体力活动和体重管理是理想心血管健康的关键组成部分，但许多年轻人 成年人活动不足，体重增长迅速（平均每年 0.5-1.0 公斤）。体力活动是 减轻体重增加的行之有效的策略，因此迫切需要有效的干预措施来减少体重增加 通过鼓励人们多动、少坐和减少体重增加来降低心血管风险。移动和可穿戴设备 技术在年轻人中几乎无处不在，并提供了获取有关生活的密集数字数据的途径 人和他们的环境。这项技术还提供了一种对即时干预进行干预的方法。 时间基础，以便每个人都能在正确的时间收到正确的干预内容。受到愿景的启发 精确的行为干预，我们应用了控制系统的系统识别工具 工程开发计算模型和算法，优化特定于人的文本剂量 促进身体活动的信息。在一系列初步研究中，我们 (a) 确定了天气指数 与身体活动始终相关，(b) 证明了长期佩戴活动追踪器的可行性 年轻人对短信的接受程度，(c) 丰富了我们的物理计算模型 使用瞬时天气数据来表征对消息的天气分级响应的活动，以及（d） 设计了一个控制器，该控制器使用计算模型中的特定于个人的参数和最近的数据 行为和预测的天气条件，以优化短信发送的选择和时间安排。这个 控制器是精确自适应干预消息传递（Precision AIM）干预的基础。在这个 应用程序中，我们提出了一项单中心临床试验来回答以下问题：“Precision AIM 是否会增加 活动不足的年轻人的体力活动和体重增加的减少比随机分配的要多 来自同一消息库或未分配的活动跟踪器的干预消息（随机 AIM） 干预消息（无 AIM）？”该项目的具体目标是（1）评估 Precision 的功效 AIM 与随机 AIM 和无 AIM 相比，在增加体力活动和减少体重增加方面 活动不足的年轻人，以及（2）确定对以下活动反应较多的参与者的特征 精确 AIM 优于随机 AIM 或无 AIM。我们提出了一项三臂随机对照试验来实现 这些目标。活动不足的年轻人将收到一个消费者活动追踪器，并被随机分配 分配给三个组之一：精确 AIM、随机 AIM 或无 AIM。干预期将持续 12 个月，18 个月时进行维护评估。主要结果是平均每日步数 基于越来越多的证据表明，增加这种易于解释且广泛使用的指标会减少 成年期心血管风险（在基线、3、6、12 和 18 个月时评估）。如果成功的话，底层 来自可穿戴设备的密集数据提供的算法分析将提供一种新的、可扩展的方法 使用消费者可穿戴设备对年轻人心血管风险进行初级和初级预防。

PROJECT SUMMARY Physical activity and weight management are key components of ideal cardiovascular health but many young adults are insufficiently active and gaining weight rapidly (0.5-1.0 kg/year on average). Physical activity is a proven strategy for attenuating weight gain so there is a pressing need for effective interventions that reduce cardiovascular risk by motivating people to move more, sit less, and reduce weight gain. Mobile and wearable technologies are nearly ubiquitous among young adults and provide access to dense digital data about the person and their environment. This technology also provides a means of delivering interventions on a just-in- time basis so each person can receive the right intervention content at the right time. Inspired by the vision of precision behavioral interventions, we have applied system identification tools from control systems engineering to develop computational models and algorithms that optimize person-specific dosing of text messages to promote physical activity. In a series of preliminary studies, we (a) identified weather indices consistently linked with physical activity, (b) demonstrated the feasibility of long-term activity tracker wear and the acceptability of text messages among young adults, (c) enriched our computational model of physical activity using momentary weather data to characterize weather-graded responses to messages, and (d) designed a controller that uses person-specific parameters from the computational model and data on recent behavior and forecasted weather conditions to optimize the selection and timing of text message delivery. This controller is the basis for the Precision Adaptive Intervention Messaging (Precision AIM) intervention. In this application, we propose a single-site clinical trial to answer question, “Does Precision AIM increase insufficiently-active young adults’ physical activity and reduce weight gain more than randomly-assigned intervention messages (Random AIM) from the same message library or an activity tracker with no assigned intervention messages (No AIM)?” The specific aims of this project are (1) to evaluate the efficacy of Precision AIM compared to Random AIM and No AIM for increasing physical activity and reducing weight gain in insufficiently-active young adults, and (2) to identify characteristics of participants who respond more to Precision AIM than Random AIM or No AIM. We propose a three-arm randomized controlled trial to achieve these aims. Insufficiently-active young adults will receive a consumer activity tracker and be randomly assigned to one of three groups: Precision AIM, Random AIM, or No AIM. The intervention period will last for 12 months with a maintenance assessment at 18 months. The primary outcome is average daily step counts based on accumulating evidence that increasing this readily-interpretable and widely-available metric reduces cardiovascular risk in adulthood (assessed at baseline, 3, 6, 12, and 18 months). If successful, the underlying algorithmic analyses informed by dense data from wearable devices will provide a new and scalable approach for using consumer wearables in primordial and primary prevention of cardiovascular risk in young adults.