Design-Statistical Core

项目总结 设计/统计核心(DSC)将是国家老龄研究所(NIA)不可或缺的组成部分 阿尔茨海默病(AD)/AD相关痴呆(AD/ADRD)医疗系统(HCS)合作实验室。会的 与合作实验室工作组的其他核心和团队合作，开发知识并提供指导 进行严格的嵌入式实用临床试验(EPCT)，以改善患者的护理和健康结果 痴呆症患者(PWD)及其照顾者(CG)。它将作为国家资源提供 合作实验室支持的试点研究、职业发展奖获得者(CDA)的生物统计学专业知识，以及 其他由NIA资助的调查人员正在进行这样的试验。DSC将对设计的各个方面做出贡献， 进行和分析这些项目，并将为设计和开发新的统计方法 在中心内残障人士中进行的ePCT分析。推动这一领域的发展，创造持久的知识 作为资料库，设计/统计核心将在以下领域提供指导：1)ePCT的稳健设计；2) EPCT的统计分析，3)处理丢失数据、污染和其他违规情况的方法，4) 应对与AD/ADRD特别相关的陷阱的战略，5)可复制性和再现性，6) 多样性和多重比较，7)审议代表性不足的少数群体和重要的 分组，包括将性作为生物变量对待，8)记录和传播软件 和分析代码，9)出版和传播，10)培训和指导初级量化 科学家和临床研究人员。设计/统计核心的其他主要功能将提供 向整个AD/ADRD合作实验室提供咨询和服务，特别是向试点研究提供咨询和服务， 职业、CDA接受者和其他NIA资助的调查人员。DSC的工作将由 希瑟·阿洛尔博士，耶鲁医学院医学教授， 耶鲁大学公共卫生学院生物统计学，耶鲁阿尔茨海默病研究主任 中心的数据管理和统计核心。她是一位经验丰富的试用者和该领域的创始人 老年生物统计学。阿洛尔博士将与一个由六个特别执行委员会组成的团队一起参加 在AD/ARDR研究、衰老研究、集群随机试验和 在老年人中开展ePCT。DSC的目标是：目标1：提供生物统计援助和 对合作实验室试点研究、CDA接受者和NIA资助的调查人员规划或实施的指导 目标2：开发新的生物统计方法，并创建和传播 指导文件和产品，与在具有HCS的PWD中设计和实施ePCT有关，目标3：至 与技术/数据核心合作，建立和管理对管理数据的访问和共享(例如， 电子健康记录、索赔数据)，供AD/ADRD合作实验室调查人员使用(例如试点项目 在可重复分析和复制研究方面提供指导。

PROJECT SUMMARY The Design/Statistics Core (DSC) will be an indispensable component of the National Institute of Aging (NIA) Alzheimer's disease (AD)/AD-related dementia (AD/ADRD) Healthcare Systems (HCS) Collaboratory. It will work with other Collaboratory Work Group Cores and Teams to develop knowledge and provide guidance to conduct rigorous embedded pragmatic clinical trials (ePCTs) to improve the care and health outcomes of persons with dementia (PWD) and their caregivers (CGs). It will serve as a national resource providing biostatistical expertise to Collaboratory-supported pilot studies, career development awardees (CDAs), and other NIA-funded investigators conducting such trials. The DSC will contribute to all aspects of the design, conduct, and analysis of these projects, and will develop novel statistical methodology for the design and analysis of ePCTs conducted among PWD within HCS. To advance this field and create a durable knowledge repository, the Design/Statistics Core will provide guidance in areas such as: 1) robust design of ePCTs, 2) statistical analysis of ePCTs, 3) methods for handling missing data, contamination, and other irregularities, 4) strategies for dealing with pitfalls of specific relevance to AD/ADRD, 5) replicability and reproducibility, 6) multiplicity and multiple comparisons, 7) consideration of under-represented minorities and important subgroups, including treatment of sex as a biological variable, 8) documentation and dissemination of software and analytic code, 9) publication and dissemination, and 10) training and mentoring of junior quantitative scientists and clinical investigators. The Design/Statistics Core’s other main function will be to provide consultation and services to the AD/ADRD Collaboratory as a whole and particularly to the pilot studies, career, CDA recipients, and other NIA-funded investigators. The work of the DSC will be enabled by the leadership of Dr. Heather Allore PhD, Professor of Medicine in the Yale School of Medicine, and Professor of Biostatistics at the Yale School of Public Health, and Director of the Yale Alzheimer's Disease Research Center's Data Management and Statistics Core. She is an experienced trialist and founder of the field of Gerontological Biostatistics. Dr. Allore will be joined by a team of six exceptional Executive Committee members who bring varied expertise in AD/ARDR research, aging research, cluster randomized trials, and the conduct of ePCTs in older persons. The DSC’s Aims are: Aim 1: To provide biostatistical assistance and guidance to Collaboratory pilot studies, CDAs recipients and NIA-funded investigators planning or conducting ePCTs in PWD and CG; Aim 2: To develop novel biostatistical approaches, and create and disseminate guidance documents and products relevant to the design and conduct of ePCTs in PWD with HCS, Aim 3: To work with the Technical/Data Core, to establish and manage access to and sharing of administrative data (e.g., electronic health records, claims data) for use by AD/ADRD Collaboratory investigators (e.g. pilot project leaders), in providing guidance on reproducible analyses and replication studies.