Developing an Adaptive Intervention for Suicidal Adolescents Following Inpatient Hospitalization: A Pilot SMART

摘要 这份为期4年的K23指导职业发展奖申请书提出了一项重点研究计划 和培训，以促进申请人过渡到独立的职业生涯，以患者为中心的研究 专门针对有自杀风险的青少年进行适应性和技术强化干预。上下文： 尽管青少年自杀具有公共卫生意义，但对自杀青少年缺乏有效的干预措施。 特别是，迫切需要对住院精神病的青少年进行有效的干预。 有反复自杀行为和相关危机的风险很高。因为有自杀倾向的年轻人 不同水平的出院后自杀风险和对干预措施的反应不同，优化 出院后的结果(防止自杀危机复发)需要一种个性化的方法。适应性 干预(Ais)可以使治疗的类型、强度和时机个体化，非常适合于 解决这种异质性。此外，鉴于移动通信的普遍存在和年轻人的偏好， (短信)，整合技术以加强对自杀青年的干预可能会加强 他们的影响。研究策略：为自杀技术增强的人工智能的发展提供信息 青少年，申请人建议进行序贯、多任务、随机试验(SMART)试点 进行激励性访谈(MI)--增强安全计划干预(MI-SafeCope)。建立在 申请人的试点研究，目标1将侧重于最终确定干预组件(MI增强的安全计划 住院时提供；出院后每日短信助推器；出院后电话助推器 呼叫)、实施协议和保真度评估工具，为智能飞行员做准备。具体来说， 申请人将通过基于以下内容的迭代精炼过程来开发文本消息助推器组件 根据专家和青少年焦点小组的反馈(N=20)。目标2将专注于进行SMART 飞行员(N=80)与自杀青少年住院患者(13-17岁)一起完善和证明可接受性和 研究程序和MI-SafeCope组件及其顺序的可行性，以指导实施 全尺寸SMART(R01提案)。全面的智能反过来又将导致构建 在接受急诊护理后自杀的青少年。培训计划：促进申请者成为一名 独立临床科学家开发预防青少年自杀的干预措施，具体培训目标为 为：(1)获得干预科学方面的专业知识，重点是开发、实施和评估 有自杀危险的青少年；(2)培养整合技术的技能，以加强和提供干预措施 这些年轻人；和(3)接受分析战略方面的培训，以便进行聪明的研究和分析方法，从而 告知人工智能系统的发展。这些目标将通过以下领域专家的密切指导来实现 这些内容领域；通过精心挑选的教学活动和出席会议；通过应用研究 经验；以及通过手稿和赠款提案(例如，全功率SMART的R01)准备。

ABSTRACT This 4-year K23 Mentored Career Development Award application proposes a program of focused research and training to facilitate the applicant's transition to an independent career in patient-oriented research with a specialization in adaptive and technology-enhanced interventions for youth at risk for suicide. CONTEXT: Despite the public health significance of youth suicide, efficacious interventions for suicidal teens are lacking. In particular, a critical need exists for effective interventions for psychiatrically hospitalized adolescents who are at a high risk for repeated suicidal behavior and related crises. Because suicidal youth are highly heterogeneous in terms of varying levels of post-discharge suicide risk and response to intervention, optimizing post-discharge outcomes (preventing relapse of suicidal crises) requires a personalized approach. Adaptive interventions (AIs), which can individualize the type, intensity, and timing of treatment, are ideally suited for addressing this heterogeneity. In addition, given the ubiquity and youth preferences for mobile communication (text messaging), integrating technology to augment interventions for suicidal youth can potentially strengthen their impact. RESEARCH STRATEGY: To inform the development of a technology-augmented AI for suicidal teens, the applicant proposes to conduct a Sequential, Multiple Assignment, Randomized Trial (SMART) pilot of a Motivational Interview (MI)-enhanced safety planning intervention (MI-SafeCope). Building on the applicant's pilot research, Aim 1 will focus on finalizing the intervention components (MI-enhanced safety plan delivered at hospitalization; post-discharge daily text message boosters; post-discharge telephone booster call), implementation protocol, and fidelity assessment tools in preparation for the SMART pilot. Specifically, the applicant will develop the text message booster component through a process of iterative refinement based on feedback from experts and adolescent focus groups (N=20). Aim 2 will focus on conducting the SMART pilot (N=80) with suicidal adolescent inpatients (ages 13-17) to refine and demonstrate acceptability and feasibility of study procedures and MI-SafeCope components and their sequence to guide the implementation of a full-scale SMART (R01 proposal). The full-scale SMART will, in turn, lead to the construction of the AI for suicidal teens following acute care. TRAINING PLAN: To facilitate the applicant's career goal of becoming an independent clinical scientist developing interventions to prevent youth suicide, the specific training goals are to: (1) gain expertise in intervention science with a focus on developing, implementing, and evaluating AIs for youth at risk for suicide; (2) develop skills in integrating technology to augment and deliver interventions for these youth; and (3) obtain training in analytic strategies for SMART studies and analytic approaches that can inform the development of AIs. These goals will be accomplished through close mentorship from experts in these content areas; via carefully selected didactic activities and conference attendance; via applied research experiences; and via manuscript and grant proposal (i.e. an R01 for a fully-powered SMART) preparation.