Improving Outcomes in Depression in Primary Care in a Low Resource Setting

改善资源匮乏的初级保健中抑郁症的治疗效果

• 批准号: 10624403
• 负责人: STEVEN DENNIS HOLLON
• 金额: $ 115.28万
• 依托单位: HARVARD MEDICAL SCHOOL
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-05-18 至 2026-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10624403
• 关键词: Academia; Adherence; Advocate; Antidepressive Agents; Behavioral; Caring; Characteristics; Clinical; Collaborations; Cost Analysis; Cost Effectiveness Analysis; Detection; Disease remission; Effectiveness; European ancestry; Exclusion Criteria; Fluoxetine; Genotype; Goals; Health system; India; Individual; Intervention; Knowledge; Machine Learning; Masks; Measures; Mediation; Mediator; Mental Depression; Mental Health; Mental Health Services; National Institute of Mental Health; Nepal; Neurocognitive; Outcome; Participant; Patient-Focused Outcomes; Patients; Pharmaceutical Preparations; Policies; Precision therapeutics; Prediction of Response to Therapy; Primary Care; Probability; Psychotherapy; Quality-Adjusted Life Years; Random Allocation; Randomized; Recommendation; Recovery; Research; Research Infrastructure; Resource-limited setting; Sample Size; Sampling; Selective Serotonin Reuptake Inhibitor; Specialist; Specificity; Testing; Treatment outcome; Work; World Health Organization; burden of illness; cost effective; cost effectiveness; effectiveness evaluation; effectiveness outcome; ethnic diversity; functional outcomes; genetic predictors; genome wide association study; health care delivery; improved outcome; individual patient; individual variation; insight; optimal treatments; personalized medicine; polygenic risk score; precision medicine; primary care patient; primary care setting; programs; psychogenetics; psychologic; relative cost; relative effectiveness; response; routine care; socioeconomics; treatment optimization; treatment trial

Project Summary Abstract Depression is the leading mental health related contributor to the Global Burden of Disease. We have shown in previous studies that generic antidepressant medications (ADMs) and brief psychological interventions such as our culturally adapted version of behavioral activation, the Healthy Activity Program (HAP), are effective in achieving remission in primary care patients. However, not everyone responds to either intervention and similar aggregate outcomes can mask considerable individual variability in response. The goal of our proposed research is to see if we can enhance treatment outcomes for patients with moderate to severe depression by personalizing allocation to one of the two treatments; additionally, we aim to identify those patients who are unlikely to respond to either treatment and should be referred to specialist care. To achieve this, we will use machine learning to develop a precision treatment rule (PTR) based on a wide range of baseline moderators which are feasible to assess in routine care settings. Our primary hypothesis is that those patients randomized by chance to their optimal intervention as indicated by the PTR will be more likely to remit and recover than those who are not. Moreover, we hypothesize that using our PTR to select the optimal treatment for each individual patient will prove to be more cost-effective than leaving things to chance. We also plan to explore secondary questions, such as whether we can enhance our mediation tests by including the PTR in interactions with our purported mediators (moderated mediation). Further, we plan to explore whether we can enhance the prescriptive utility of our PTR via genotyping our sample and calculating polygenic risk scores based on very large sample Genome-Wide Association studies. We will test these hypotheses in a controlled trial in primary care settings in India where we have a record of conducting depression treatment trials for two decades. We plan to randomize 1500 individuals to either HAP or ADM and generate our PTR on the first 1000 patients randomized and then test it on the remaining 500 patients. This will be the first test of whether precision medicine can be used to enhance depression treatment outcomes through prediction of differential response to the two treatments recommended by the WHO for depression in primary care. Concurrently, we also should be able to identify baseline predictors of nonresponse to either of these two treatments, so as to identify patients who should be referred to specialist care at the outset. Our findings have the potential to make significant contributions to the prospect of optimizing the treatment of depression in primary care not just in India but also in primary care settings worldwide, and thus support the practice-related goals of NIMH RFA- MH-18-701.

项目概要 摘要 抑郁症是造成全球疾病负担的主要心理健康相关因素。我们已经展示在 之前的研究表明，一般抗抑郁药物（ADM）和简短的心理干预措施，例如 我们的行为激活文化适应版本，即健康活动计划 (HAP)，可以有效地 在初级保健患者中实现缓解。然而，并不是每个人都对干预和干预做出反应 类似的总体结果可能会掩盖反应中相当大的个体差异。我们提出的目标 研究的目的是看看我们是否可以通过以下方式提高中度至重度抑郁症患者的治疗效果： 个性化分配两种治疗方法中的一种；此外，我们的目标是确定那些患有以下疾病的患者： 这两种治疗都不太可能有反应，应转诊至专科护理。为了实现这一点，我们将使用 机器学习基于广泛的基线调节器制定精准治疗规则（PTR） 在常规护理环境中进行评估是可行的。我们的主要假设是这些患者随机 与 PTR 所示的偶然机会相比，他们的最佳干预更有可能缓解和恢复 那些不是的人。此外，我们假设使用我们的 PTR 为每个患者选择最佳治疗方法 事实证明，与个别患者合作比听天由命更具成本效益。我们也计划探索 次要问题，例如我们是否可以通过将 PTR 纳入其中来增强我们的中介测试 与我们所谓的调解员的互动（有调节的调解）。此外，我们计划探索是否可以 通过对样本进行基因分型并计算多基因风险评分，增强 PTR 的规范性效用 基于非常大样本的全基因组关联研究。我们将在受控的环境中测试这些假设 在印度初级保健机构进行的试验，我们有对两名患者进行抑郁症治疗试验的记录 几十年。我们计划将 1500 人随机分配到 HAP 或 ADM，并在前 1000 人上生成我们的 PTR 患者被随机分组​​，然后对剩下的 500 名患者进行测试。这将是第一次测试是否 精准医学可通过预测差异来提高抑郁症的治疗效果 对世界卫生组织推荐的初级保健抑郁症的两种治疗方法的反应。同时，我们 还应该能够确定对这两种治疗方法无反应的基线预测因素，以便 从一开始就确定应转诊至专科护理的患者。我们的发现有可能使 对优化初级保健抑郁症治疗的前景做出了重大贡献，而不仅仅是在 印度以及全世界的初级保健机构，因此支持 NIMH RFA 的实践相关目标 MH-18-701。