JOINT MODELING OF MULTISTATE AND NONPARAMETRIC MULTIVARIATE LONGITUDINAL DATA

• 成果类型: Article
• 署名作者: You, Lu; Salami, Falastin; Torn, Carina; Lernmark, Ake; Tamura, Roy
• 署名单位: State University System of Florida; University of South Florida; Lund University
• 刊物名称: ANNALS OF APPLIED STATISTICS
• ISSN/ISSBN: 1932-6157
• DOI: 10.1214/24-AOAS1889
• 发表日期: 2024
• 页码: 2444-2461
• 关键词: monte carlo methods; survival; event; time; autoantibodies; likelihood

摘要

It is oftentimes the case in studies of disease progression that subjects can move into one of several disease states of interest. Multistate models are an indispensable tool to analyze data from such studies. The Environmental Determinants of Diabetes in the Young (TEDDY) is an observational study of at-risk children from birth to onset of type-1 diabetes (T1D) up through the age of 15. A joint model for simultaneous inference of multistate and multivariate nonparametric longitudinal data is proposed to analyze data and answer the research questions brought up in the study. The proposed method allows us to make statistical inferences, test hypotheses, and make predictions about future state occupation in the TEDDY study. The performance of the proposed method is evaluated by simulation studies. The proposed method is applied to the motivating example to demonstrate the capabilities of the method.