Robust person-centered longitudinal data analysis for clinical alcohol research

用于临床酒精研究的稳健的以人为中心的纵向数据分析

• 批准号: 8927513
• 负责人: Douglas L Steinley
• 金额: $ 35.2万
• 依托单位: UNIVERSITY OF MISSOURI-COLUMBIA
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-15 至 2017-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8927513
• 关键词: Address; Aftercare; Alcohols; Area; Attention; Behavior; Benchmarking; Clinical; Clinical Data; Clinical Trials; Cluster Analysis; Communities; Computer software; Conflict (Psychology); Consensus; Controlled Clinical Trials; Data; Data Analyses; Data Set; Derivation procedure; Development; Effectiveness; Epidemiology; Foundations; Growth; Health; Health Status; Individual; Lead; Longitudinal Studies; Measures; Mediator of activation protein; Methodology; Methods; Mind; Modeling; NIH Program Announcements; Outcome; Pattern; Performance; Phenotype; Procedures; Process; Randomized; Randomized Clinical Trials; Recovery; Relative (related person); Research; Research Design; Research Personnel; Services; Solutions; Statistical Methods; Statistical Models; Structure; Techniques; Time; Treatment Effectiveness; Typology; Validation; Variant; alcohol research; alcohol use disorder; base; design; effectiveness clinical trial; graphical user interface; indexing; longitudinal analysis; person centered; programs; simulation; user-friendly

DESCRIPTION (provided by applicant): This application proposes to develop three key components in person-centered longitudinal statistical models: (a) a comprehensive simulation study designed to understand the nuanced performance of a wide range of models used to derive group based trajectories, (b) a cross-validation procedure designed to bolster the confidence (in terms of replication and consistency) of these models when employed in controlled, clinical trials and effectiveness trials, and (c) to promote dissemination of these advanced techniques by developing a user- friendly, freely available software with a graphical-user interface.

描述(申请人提供)：本申请建议在以人为中心的纵向统计模型中开发三个关键组件：(A)全面的模拟研究，旨在了解用于推导基于组的轨迹的各种模型的细微差别性能；(B)交叉验证程序，旨在增强这些模型在受控、临床试验和有效性试验中使用时的信心(在重复性和一致性方面)；以及(C)通过开发具有图形用户界面的用户友好的、免费获得的软件来促进这些先进技术的传播。