Improving Eligibility Prescreening for Alzheimer's Disease and Related Dementias Clinical Trials with Natural Language Processing

利用自然语言处理改善阿尔茨海默病和相关痴呆症临床试验的资格预筛

• 批准号: 10396839
• 负责人: Betina Ross Saldua Idnay
• 金额: $ 2.08万
• 依托单位: COLUMBIA UNIVERSITY HEALTH SCIENCES
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-01-01 至 2022-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10396839
• 关键词: Improving; Eligibility; Prescreening; Alzheimer; Disease

The increasing prevalence of Alzheimer’s disease and related dementias (ADRD) presents major financial and care delivery challenges to the United States (US) healthcare system. There are currently an estimated 5.8 million Americans age 65 and older living with ADRD, with a projected increase to 13.8 million by 2050. Deaths associated with ADRD increased by 146.2% from 2000 to 2018, making it the sixth-leading cause of death in the US and the only disease in the top 10 causes of death that cannot be prevented, cured, or even slowed. Further complicating treatment advances is that on average, ADRD disease-modifying treatment development requires 13 years, with a failure rate of new therapies of more than 99%. A major bottleneck which contributes to this high failure rate is eligibility prescreening, which involves costly, time-consuming, and inefficient manual review of complex clinical data sources by clinical research staff. Natural language processing (NLP), an informatics approach used to extract relevant data from a variety of structured and unstructured data types, may improve eligibility prescreening for ADRD clinical trials. NLP has been used to identify potentially eligible patients in other disease-specific clinical trials that resulted to prompting research teams when appropriate research is available for specific patients, yet this has not been utilized in ADRD clinical trials. The proposed study will evaluate the clinical research staff’s technology adoption of an NLP-driven eligibility prescreening tool for ADRD clinical trials. Criteria2Query (C2Q), a novel open source NLP-driven eligibility prescreening tool, was developed to translate free-text eligibility criteria to standards-based cohort definition queries. In the proposed mixed-methods study, clinical research staff will participate in usability testing, and accuracy and efficiency evaluation of ADRD clinical trial eligibility prescreening using C2Q for patients seen in an Aging and Dementia clinical practice. Guided by the adapted Fit between Individuals, Task, and Technology Framework, the specific aims are to: (1) examine the usability of an NLP-driven tool for ADRD clinical trial eligibility prescreening, and (2) assess the efficiency and accuracy of eligibility prescreening done by clinical research staff using an NLP-driven tool for ADRD clinical trials. The proposed aims are consistent with the priorities of the Agency for Healthcare Research and Quality (AHRQ) in supporting research to increase accessibility and affordability of health care by examining innovative market approaches to care delivery. Findings from the proposed study have the potential to produce key findings for successful ADRD clinical research recruitment to accelerate knowledge discovery and development of a disease-modifying treatment for ADRD. Lastly, R36 dissertation award will provide a valuable opportunity for a pre-doctoral student to build a strong foundation for her long-term goal of becoming an independent nurse scientist with expertise in leveraging informatics in clinical research for the aging population.

阿尔茨海默氏病和相关痴呆症 (ADRD) 的患病率不断上升，给经济带来了重大影响 以及美国医疗保健系统面临的护理服务挑战。目前估计有 580 万 65 岁及以上的美国人患有 ADRD，预计到 2050 年将增加到 1380 万。 从 2000 年到 2018 年，与 ADRD 相关的死亡人数增加了 146.2%，成为导致 ADRD 的第六大原因。 是美国十大死亡原因中唯一无法预防、治愈甚至无法预防的疾病 减慢了。使治疗进展进一步复杂化的是，平均而言，ADRD 疾病缓解治疗 开发需要13年，新疗法的失败率超过99%。主要瓶颈 导致这种高失败率的原因是资格预筛选，这涉及成本高昂、耗时且 临床研究人员对复杂临床数据源的手动审查效率低下。自然语言 处理（NLP），一种信息学方法，用于从各种结构化和 非结构化数据类型，可以改善 ADRD 临床试验的资格预筛选。 NLP 已被用于 在其他疾病特异性临床试验中识别潜在合格的患者，从而促进研究 当针对特定患者进行适当的研究时，团队需要进行研究，但这尚未在 ADRD 中使用 临床试验。 拟议的研究将评估临床研究人员对 NLP 驱动的技术的采用情况 ADRD 临床试验资格预筛选工具。 Criteria2Query (C2Q)，一种新颖的开源 NLP 驱动的 资格预筛选工具，旨在将自由文本资格标准转化为基于标准的群体 定义查询。在拟议的混合方法研究中，临床研究人员将参与可用性 使用 C2Q 进行 ADRD 临床试验资格预筛选的测试、准确性和效率评估 在老龄化和痴呆症临床实践中看到的患者。以个体、任务之间适应的适应性为指导， 和技术框架，具体目标是：（1）检查 NLP 驱动的 ADRD 工具的可用性 临床试验资格预筛选，以及 (2) 评估资格预筛选的效率和准确性 由临床研究人员使用 NLP 驱动的工具进行 ADRD 临床试验。拟议的目标是一致的 与医疗保健研究和质量局 (AHRQ) 的优先事项相一致，支持研究 通过研究创新的市场护理方法来提高医疗保健的可及性和可负担性 送货。拟议研究的结果有可能为 ADRD 的成功提供关键发现 招募临床研究以加速疾病缓解的知识发现和开发 ADRD 的治疗。最后，R36 论文奖将为博士生提供宝贵的机会 学生为她成为一名独立护士科学家的长期目标奠定坚实的基础 在老龄化人口临床研究中利用信息学的专业知识。