Predicting Internet-Based Treatment Response for Major Depressive Disorder

多达53%的学生报告在大学期间经历过抑郁，而这些抑郁 发作与学业问题、共病和自杀的频率较高有关。虽然 有有效的治疗选择，大多数人(70%)不寻求服务，甚至 对于那些这样做的人，回复率仍然不高(约40%-50%)。作为提高可访问性的一种手段 在治疗方面，已经开发和测试了基于互联网的抑郁症干预措施。尽管增加了 可获得性，对基于互联网的干预措施的反应仍有很大差异，治疗经常失败 会导致症状的持续和恶化。因此，识别出有很高可能性的个人 对基于互联网的治疗做出反应将是一项重大进步，并解决了一个关键的未得到满足的需求。 近年来，检验治疗效果异质性的有效方法--勾画 哪些个体可能对特定的治疗有反应--已经被开发出来。然而，它们的用途是 确定抑郁症治疗反应的预测因素仍不清楚。为了解决这一未得到满足的需求， 拟议的研究将测试一种新的、经济有效的、可行的预测差异治疗的方法。 基于互联网的认知行为疗法(ICBT)治疗抑郁症后的反应具有代表性 学院样本(波士顿学院和大学联盟，包括7所学校)。委员会成员 该联盟承诺通过严格的在线筛选来筛选所有新生(N=~14,000人) 评估并向抑郁症状水平升高的学生(即轻微或严重)提供iCBT 抑郁)。将采取下列步骤。首先，在研究的初始阶段，来自 波士顿财团将通过在线调查进行抑郁症筛查，他们还将完成 基于网络的神经认知任务，探索抑郁症的关键机制。抑郁的学生会 被邀请注册iCBT，并将根据以下方面的评估开发预测算法 不同的分析单位(即临床特征、神经认知指数)来识别iCBT应答者。 第二，在发展阶段后，将招募独立的大学生样本进行验证 预测算法。这一验证阶段将使用临床指标和神经认知数据。另外， 针对研究领域内关键机制的功能磁共振成像(FMRI)数据 标准(RDoC)将从参与者的子集中获得。神经数据将被整合以确定 他们是否改进了预测算法。第三，跨独立样本的数据将被合并， 这将增加改进我们对急性和持续反应的预测模型的能力。总而言之， 大学生中抑郁的比率令人震惊，只有少数学生利用心理 医疗服务。这项拟议的研究将使我们治疗抑郁症的方法个人化， 最终，这将提高有效性，并更好地为整个大学校园的精神卫生保健提供信息。

As many as 53% of students report experiencing depression during college, and these depressive episodes are associated with a higher frequency of academic problems, comorbidity, and suicide. Although there are effective options for treatment, the majority of individuals (>70%) do not pursue services, and even for those who do, response rates remain modest (~40-50%). As a means of increasing accessibility to treatment, internet-based interventions for depression have been developed and tested. Despite increased availability, response to internet-based interventions continues to vary substantially, and failed treatment often contributes to persistence and worsening of symptoms. Therefore, identifying individuals with a high likelihood of responding to internet-based treatment would represent a major advance and address a critical unmet need. In recent years, promising approaches for testing the heterogeneity of the treatment effects – delineating which individuals are likely to respond to a given treatment – have been developed. However, their use for identifying predictors of treatment response in depression remains unclear. To address this unmet need, the proposed study will test a new, cost-effective, and feasibly-scaled method of predicting differential treatment response following internet-based cognitive behavioral therapy (iCBT) for depression in a large, representative college sample (Boston Consortium of Colleges and Universities which includes 7 schools). Members of the consortium have committed to screening all incoming freshmen (N = ~14,000) through a rigorous online assessment and to offer iCBT to students with elevated levels of depressive symptoms (i.e., minor or major depression). The following steps will be pursued. First, in the initial phase of the study, freshmen students from the Boston Consortium will be screened for depression through an online survey, and they also will complete web-based neurocognitive tasks probing key mechanisms underpinning depression. Depressed students will be invited to enroll in iCBT, and a predictive algorithm will be developed based on assessments across different units of analysis (i.e., clinical characteristics, neurocognitive indices) to identify iCBT responders. Second, after the development phase, an independent sample of college students will be recruited to validate the predictive algorithm. This validation phase will use clinical indicators and neurocognitive data. Additionally, functional magnetic resonance imaging (fMRI) data targeting key mechanisms within the Research Domain Criteria (RDoC) will be acquired from a subset of participants. Neural data will be integrated to determine whether they improve the predictive algorithm. Third, data across independent samples will be combined, which will increase power to refine our predictive model for both acute and sustained response. In summary, there are alarming rates of depression among college students, and only a minority of students utilize mental health services. The proposed research will personalize our approach to depression treatment, which, ultimately, will improve effectiveness and better inform mental health care across college campuses.