Advanced Development and Utilization of Assembled Aging Trajectory Files from Multiple Datasets

项目摘要 该项目的目标是创建一个独特而全面的衰老轨迹数据研究库， 任务集、相关资源和分析方法，可用于回答 老龄化和相关科学。具体而言，通过协调和合并多个数据集，该项目将产生- 获取所需的数据基础设施，以了解随着时间的推移，在护理环境，老年综合征，物理， cal功能，以及多个层面（患者、提供者、社区、医疗保健系统和 社会）和跨多个领域（生物，行为，社会文化和物理/建筑环境） 包括慢性疾病和急性疾病史，如COVID-19，暴露于空气污染，邻居， 社会经济和卫生保健系统因素（目标1）。将为用户制定分析战略- 定义的队列及其倾向评分匹配的对照，例如，患有慢性病的老年人 包括阿尔茨海默氏病和相关痴呆（ADRD）、糖尿病、心力衰竭、终末期再狭窄、 终末疾病、转移性癌症和HIV。国家的最先进的分析方法被用来确定模式的ag- 老年人在治疗期间经历的变化轨迹（护理环境、老年综合征、身体功能） 生命的最后几年及其与共同风险因素和远端结局的相关性（目标2）。从装配- 目标1中流血的轨迹文件，通过对齐原始索引时间（如年龄截止点）来导出队列 和诊断时间（例如，ADRD、中风、慢性肾病）。基于模型的方法和MA- chine学习算法，然后用来发现多层次的和潜在的互动预测的预测， 列（例如，独立生活受益人的功能迅速下降）和具体结果（例如，呼吸 诊断为COVID-19的医疗保险受益人中的呼吸机使用情况）（目标3）。独特的资源是 然后共享以传播资源，包括数据集、文档、源代码和方法（Aim 4）。在这个项目的最后，研究基础设施之间的关系，共同风险 因素和老化轨迹将随时可以使用和复制，使调查人员具有前所未有的能力， 解决衰老科学的新挑战。这将使研究人员能够了解潜在的过程， 与整个护理环境中的可逆残疾期相关的系统，以及可能 用于支持功能恢复和减少老年综合征，包括认知能力下降， 目的是减少繁重的护理过渡，并保持功能独立性。这个项目 还将创造所需的资源和方法，以评估创新和干预措施的影响， 在患者、提供者、社区、医疗保健系统和社会/政策层面实施，以改善护理 老年人的生活质量和结果。

PROJECT SUMMARY This goal of this project is to create a unique and comprehensive research repository of aging trajectory da- tasets, related resources, and analytic methods that can be used to answer new and important questions in aging and related sciences. Specifically, by harmonizing and merging multiple data sets this project will gener- ate the data infrastructure needed to understand change over time in care settings, geriatric syndromes, physi- cal functioning, and shared risk factors at multiple levels (patient, provider, community, healthcare system, and society) and across multiple domains (biological, behavioral, sociocultural, and physical/built environments) including chronic conditions and history of acute illness such as COVID-19, exposure to air pollution, neighbor- hood socioeconomic, and health care system factors (Aim 1). Analytic strategies will be developed for user- defined cohorts and their propensity score-matched controls, e.g., older adults who were living with chronic conditions including Alzheimer's disease and related dementias (ADRD), diabetes, heart failure, end-stage re- nal disease, metastatic cancer, and HIV. State-of-the-art analytic methods are used to identify patterns of ag- ing trajectories (care setting, geriatric syndromes, physical functioning) experienced by older adults during the final years of life and their association with shared risk factors and distal outcomes (Aim 2). From the assem- bled trajectory file in Aim 1, cohorts are derived by aligning an originating index time such as age cutoff point and time at diagnosis (e.g., ADRD, stroke, chronic kidney disease). Both a model-based approach and ma- chine learning algorithms are then used to discover multilevel and potentially interactive predictors of trajecto- ries (e.g., rapid functional decline in independent living beneficiaries) and specific outcomes (e.g., respiratory ventilator usage among Medicare beneficiaries diagnosed with COVID-19) (Aim 3). The unique resources are then shared to disseminate resources including datasets, documentation, source code, and methodology (Aim 4). At the end of this project, the research infrastructure to investigate the relationship between shared risk factors and aging trajectories will be ready to use and replicate, giving investigators unprecedented ability to solve new challenges in aging science. This will allow researchers to understand the underlying processes and systems associated with reversible periods of disability across care settings, and interventions that may be used to support recovery of function and reduction of geriatric syndromes including cognitive decline, for the purpose of reducing burdensome care transitions, and maintenance of functional independence. This project will also create the resources and methods needed to evaluate the impact of innovations and interventions im- plemented at the patient, provider, community, healthcare system, and society/policy levels to improve care quality and outcomes for older adults.