Shared-Parameter Joint Models of Fatigue and Time until Death of Non-Resectable Lung Cancer Patients
AuthorPhillips, David A.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractWith researchers showing greater interest in the relationship between longitudinal and survival outcomes, joint models are being used with greater frequency. Joint models of longitudinal and time to event outcomes offer distinct advantages. First, joint models can reduce bias in estimates of the relationship between surrogate markers and survival endpoints. Second, this class of model can provide sensitivity analysis of longitudinal estimates in the presence of potential missing data when the longitudinal outcome and survival outcome are related. The aim of this paper is to demonstrate the usefulness of this methodology when dealing with potentially related outcomes. Using a data set from a clinical trial aimed at reducing fatigue with physical activity amongst non-resectable lung cancer patients, several joint models and conventional models such as a linear mixed model and Cox proportional hazards model were generated. Both the longitudinal and survival estimates from these models were compared to demonstrate the utility of joint models. Furthermore, the implementation of joint models is discussed as a result of the analysis.
Degree ProgramGraduate College