Systematic Evaluation of the Effects of Cognitive Remediation across Affective and Non-Affective Psychosis

项目总结 包括精神分裂症[SZ]和双相情感障碍[BD]在内的精神障碍是最昂贵和 世界各地令人衰弱的精神疾病。SZ和BD都与明显的认知缺陷有关， 它们是社区功能不佳、残疾和生活质量下降的最强有力的预测因素之一 (1-3)。认知补救(CR)是为数不多的针对认知症状的干预措施之一 直接去吧。越来越多的证据表明，CR改善了该组织中这些疾病患者的认知能力 级别(例如4-6)。然而，许多关键的知识差距限制了我们发展强大的、个性化的能力 CR干预并为最有可能受益的患者提供途径：首先，CR结果很高 治疗反应的异质性和预测因素尚未确定；第二，潜在的 调节认知和功能增强的神经生物学机制在很大程度上仍未确定。 CR试验具有挑战性，因为干预措施往往漫长而密集，而且确定和保留 精神障碍患者是很困难的。因此，大多数CR研究都没有足够的能力来明确 确定反应的预测因素，通过诊断比较治疗效果，检查CR快速驱动的能力 神经变化，并确定神经变化与认知或功能相关的程度 结果。这些知识差距限制了我们开发强大的、有针对性的CR计划并交付 将它们转给最有可能受益的患者。 该项目的目的有两个：1)识别与认知相关的患者特征 治疗反应包括基线人口统计学、临床和认知变量，以及多变量模型； 2)检查CR在关键信息处理中驱动脑电测量的快速神经变化的能力 系统，以及这种变化与主要认知结果的关联。在过去的八年里，我们的 该小组已经完成了两个独立的随机对照试验，对SZ患者进行了70小时的CR计划 和BD，由NIMH和大脑和行为研究基金会提供资金。这些研究采用了 相同的CR范例、主动控制条件、主要(认知)和次要(临床、功能) 在相同的时间点(基线、中点和治疗后)和脑电方案进行结果测量。 我们将系统地检查整个样本的CR反应的调节因素，包括人口统计学、临床 和认知变量；然后我们将使用主成分分析(PCA)和回归来评估 CR反应的预测因素，将单变量预测因素与我们的多变量模型进行比较，并探索 应答者和非应答者的集群行为。最后，我们将检查CR培训的早期效果 脑电测量的神经生理学，以及神经调节与认知反应的关系。这个项目 通过合并两个相关的数据集来形成一个井，从而填补知识中的关键空白 动力性研究能够回答情感性和非情感性精神病的这些问题。

PROJECT SUMMARY Psychotic disorders including schizophrenia [SZ] and bipolar disorder [BD] are among the most costly and debilitating psychiatric diseases worldwide. Both SZ and BD are associated with marked cognitive deficits, which are among the strongest predictors of poor community functioning, disability, and reduced quality of life (1-3). Cognitive remediation (CR) is one of the few available interventions to target cognitive symptoms directly. Growing evidence indicates that CR improves cognition in patients with these illnesses at the group level (e.g. 4-6). However, a number of critical knowledge gaps limit our ability to develop robust, individualized CR interventions and provide access to patients who are most likely to benefit: first, CR outcomes are highly heterogeneous and predictors of treatment response have not been identified; and second, the underlying neurobiological mechanisms that mediate cognitive and functional enhancement remain largely undetermined. CR trials are challenging, as interventions are often lengthy and intensive, and ascertainment and retention of patients with psychotic disorders is difficult. As a result, the majority of CR studies are underpowered to clearly identify predictors of response, compare treatment effects by diagnosis, examine the ability of CR to drive rapid neural change, and determine the extent to which neural change is associated with cognitive or functional outcomes. These knowledge gaps have limited our ability to develop robust, targeted CR programs and deliver them to patients most likely to benefit. The aims of this project are twofold: 1) identification of patient characteristics associated with cognitive treatment response including baseline demographic, clinical, and cognitive variables, and multivariate models; and 2) examination of CR's ability to drive rapid, EEG-measured neural change in key information processing systems, and the associations of this change with primary cognitive outcomes. Over the last eight years our group has completed two separate randomized controlled trials of a 70-hour CR program in patients with SZ and BD with funding from NIMH and the Brain and Behavior Research Foundation. These studies employed identical CR paradigms, active control conditions, primary (cognitive) and secondary (clinical, functional) outcome measures taken at the same timepoints (baseline, midpoint, and post-treatment), and EEG protocols. We will systematically examine moderator of CR response across the sample including demographic, clinical and cognitive variables; we will then use principal components analysis (PCA) with regression to evaluate factors predictive of CR response, compare univariate predictors with our multivariate models, and explore clustering behavior in responders vs. non-responders. Finally, we will examine early effects of CR training on EEG-measured neurophysiology, and the relationship of neural modulation to cognitive response. This project is uniquely positioned to fill critical gaps in the knowledge by merging two related datasets to form a well- powered study able to answer these questions across affective and non-affective psychosis.