Optimizing digital health technologies to improve therapeutic skill use and acquisition

项目摘要 暴饮暴食（即，在一段不连续的时间内吃了大量的食物，并伴有失落感 控制进食）是包括神经性贪食症（BN）和暴食症在内的几种进食障碍的关键症状 进食障碍（BED）虽然认知行为疗法（CBT）可以是一种有效的治疗方法， 在暴饮暴食的情况下，40-50%的BED患者和近70%的BN患者未能达到缓解。一 越来越多的研究表明，许多患者可能无法从CBT中获益的一个关键原因是次优 技能获得率（即，成功地执行在治疗中学到的技能的能力）和利用（即， 患者练习或使用治疗技能的频率）。技能的使用和获取不佳可能 在某些患者亚群中特别高，例如那些经历自我调节缺陷的患者。 这项研究表明，可以改善技能使用和获得的治疗增强（特别是 对于那些需要额外支持才能在CBT中取得成功的人来说）可能有很大的潜力来提高结果。 NIMH最近发现，数字健康技术（DHTs）具有很大的潜力，可以“促进 会议之间的技能实践/收购”，并已选择这作为一个高优先级的研究计划（不- MH-18-031）。DHT可以通过几种途径来提高技能的使用和获取，其中之一是 使用微干预（即，在人们日常生活中向他们提供简短的数字干预措施）。 微干预的复杂程度可以从简单的自动提醒到练习 先进的即时适应性干预（JITAI）系统，使用机器学习或 其他先进的算法，在需要的特定时刻提供个性化的干预措施。最近 我们团队的试点工作支持JITAI的实用性，作为提高技能利用率和获取的一种方式， 作为治疗增强，但没有将JITAI与更简单的自动提醒微 干预措施。此外，我们的试点工作还发现，频繁监测技能的使用本身就是一个问题。 这是一种非常有效的方法来鼓励技能练习。这些结果表明， JITAI可能不是所有人都需要从DHT增强中受益。 开发最有效的DHT的能力需要使用更大规模的临床试验，以帮助识别 哪些数字组件（以及复杂度）在提高技能使用和获取方面最有效， 好的临床效果。我们建议使用2 × 3全因子设计，其中264名BN或BED患者 被分配到六种治疗条件之一，即，代表自我监控的每一种排列 复杂性（技能监测开启与技能监测关闭）和微干预复杂性（无微干预） 干预与自动提醒消息与JITAIs）作为CBT的增强。结果 成分分析建立了未来的工作，以评估一个优化的DHT只包含有效成分 （可以预期其具有上级目标参与、功效、效率和可传播性）。

Project Summary Binge eating (i.e., eating a large amount of food within a discrete time period accompanied by a sense of loss of control over eating) is a key symptom of several eating disorders including bulimia nervosa (BN) and binge eating disorder (BED). While cognitive behavioral therapy (CBT) can be an effective treatment approach for binge eating, 40-50% of patients with BED and nearly 70% of patients with BN fail to achieve remission. A growing body of research suggests that a key reason many patients may fail to benefit from CBT is suboptimal rates of skill acquisition (i.e., the ability to successfully perform a skill learned in treatment) and utilization (i.e., the frequency with which a patient practices or employs therapeutic skills). Poor skill use and acquisition may be particularly high among certain subsets of patients such as those who experience deficits in self-regulation. This research suggests that treatment augmentations that could improve skill use and acquisition (particularly for those who need additional support to succeed in CBT) could have high potential to enhance outcomes. The NIMH has recently identified that digital health technologies (DHTs) have high potential to “promote between-session skill practice/acquisition” and have selected this as a high priority research initiative (NOT- MH-18-031). DHTs may be able to improve skill use and acquisition via several pathways, one of which is the use of micro-interventions (i.e., short digital interventions delivered to people as they go about their daily lives). Micro-interventions can range in complexity from something as simple as an automated reminder to practice a therapeutic skill to advanced just-in-time adaptive intervention (JITAI) systems that use machine learning or other advanced algorithms to deliver personally tailored interventions in specific moments of need. Recent pilot work from our team supports the utility of JITAIs as a way to improve skill utilization and acquisition when used as a treatment augmentation but did not compare JITAIs to more simple automated reminder micro- interventions. Additionally, our pilot work also found that frequent monitoring of skill use was in and of itself a surprisingly effective method for encouraging skill practice. These results suggest that the added complexity of JITAIs may not be necessary for all individuals to experience benefit from a DHT augmentation. The ability to develop maximally effective DHTs requires the use of a larger clinical trial that can help to identify which digital components (and at which complexity) are most effective at improving skill use and acquisition as well clinical outcomes. We propose to use a 2 x 3 full factorial design in which 264 individuals with BN or BED are assigned to one of six treatment conditions, i.e., representing each permutation of self-monitoring complexity (Skills-Monitoring On vs. Skills-Monitoring Off) and micro-intervention complexity (No Micro- Interventions vs. Automated Reminder Messages vs. JITAIs) as an augmentation to CBT. Results of the component analysis set up future work to evaluate an optimized DHT containing only effective components (which can be expected to have superior target engagement, efficacy, efficiency and disseminability).