Combining data sources to identify effect moderation for personalized mental health treatment

项目摘要 确定“什么对谁有效”是各种疾病预防和治疗的关键目标 领域，包括精神健康。通过了解哪些人从哪些项目中获益最多 我们有可能将稀缺的资源引导到那些受益最大的人手中， 以及减少在找到一种治疗方法之前尝试多种治疗的个人的“搅动” 这对他们来说很管用。确定效果调节剂--与治疗大小有关的因素 效果--对提供治疗和预防干预措施至关重要，但这样做 使用标准研究设计非常困难。随机试验，治疗的黄金标准 估计平均影响，通常不足以检测到适度。大规模非政府组织 实验研究可能会提供另一种检查效果缓和的方法，但可能会受到影响。 从困惑中解脱。需要新的方法来最好地利用可用的数据来学习如何 个性化心理健康治疗。这项工作将综合、扩展和应用方法 当有多项研究可用时，确定效果调节剂，特别关注 心理健康研究中的复杂性。这些方法将得到广泛应用，并将在 一个估计精神分裂症药物治疗效果的例子，使用11个数据 来自杜克大学卫生系统的随机对照试验和非实验数据 电子健康记录。该工作将：1)扩展针对以下场景的缓和方法 多个随机实验，2)开发使用组合数据集的数据的方法 通过实验和非实验设计来确定效果缓和，以及3) 将这些方法传播给心理健康研究人员。通过开发方法来充分利用 实验数据和非实验数据的优势这项工作有可能取得进展 走向个性化心理健康，从而改善我们预防和治疗心理健康的方式 人口中的挑战。

Project Summary Determining “what works for whom” is a key goal in prevention and treatment across a variety of areas, including mental health. By understanding which individuals benefit most from which treatments we have the possibility of directing scarce resources to those who will most benefit, and of reducing the “churn” of individuals attempting multiple treatments before finding the one that works for them. Identifying effect moderators—factors that relate to the size of treatment effects--is crucial for delivery of treatment and prevention interventions, but doing so is incredibly difficult using standard study designs. Randomized trials, the gold standard for estimating average effects, are typically under-powered to detect moderation. Large-scale non- experimental studies may provide another way to examine effect moderation, but can suffer from confounding. New methods are needed to best harness the data available to learn how to personalize mental health treatments. This work will synthesize, extend, and apply methods for identifying effect moderators when multiple studies are available, with a particular focus on the complexities in mental health research. The methods will apply broadly and will be illustrated in an example estimating the effects of medication treatment for schizophrenia, using data from 11 randomized controlled trials and non-experimental data from the Duke University Health System electronic health record. The work will: 1) Extend moderation methods for scenarios with multiple randomized experiments, 2) Develop methods for using data from combined datasets with both experimental and non-experimental designs to identify effect moderation, and 3) Disseminate the methods to mental health researchers. By developing methods to take full advantage of both experimental and non-experimental data this work has the potential to move towards personalized mental health, thus improving how we prevent and treat mental health challenges in the population.