[Calendar Icon] Events Calendar

Dr. Andrew Leroux
Assistant Professor
Department for Biostatistics and Informatics
University of Colorado School of Public Health

Circadian Patterns of Objectively Measured Physical Activity and Mortality Modelled Using Additive Functional Cox Regression 

Modelling the association between functional covariates and time-to-event outcomes is a relatively new area of methodologic research. In this presentation, we introduce the additive functional Cox model which extends the linear functional proportional hazards model by allowing the association between the functional covariate and log hazard to vary non-linearly in both the functional domain and the value of the functional covariate. We apply this method to study the association between physical activity profiles measured via wearable accelerometers and risk of all-cause mortality. Applications to three datasets are presented: The National Health and Nutrition Examination Survey (NHANES) 2003-2006, NHANES 2011-2014, and the UK Biobank. Implementation using high quality, fast, and open source software and future research directions are discussed.