Please note: this event has passed
Abstract: In conventional prediction models, predictors are typically measured at a single fixed time point such as at baseline or the most recent follow-up. Dynamic prediction has emerged as a more appealing prediction technique that takes account of longitudinal history of biomarkers for making predictions. In this talk I will present results from a simulation study comparing the prediction performance of two well-known approaches for dynamic prediction, namely joint modelling and landmarking approaches. We assessed the performance of both approaches in terms of extended definitions of discrimination and calibration, namely dynamic area under the receiver operating characteristic curve (dynAUC) and expected prediction error (PE). We used bootstrap simulation to be as impartial as possible to both methods as landmarking is a pragmatic approach which does not specify a statistical model for the longitudinal markers, and therefore any comparison based on model-based data simulation may potentially be more advantageous to joint modelling approach. The simulation was based on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) data with repeat Mini-Mental State Examination (MMSE) scores as the longitudinal biomarker and time-to-Alzheimer’s disease (AD) as the survival outcome. The results show that joint modelling approach has performed better than the landmarking approach in terms of both discrimination and calibration, although the margin of gain in performance by using joint models over landmarking was relatively small indicating that landmarking approach was close enough, despite not having a precise statistical model characterising the evolution of the longitudinal markers.
Biography: Dr Mizanur Khondoker is an Associate Professor in Medical Statistics at the Norwich Medical School, University of East Anglia. He is an applied statistician with a core interest in dementia research focusing on identifying early biomarkers, risk prediction and prevention of dementia. His expertise includes machine learning, analysis of high dimensional data, particularly those arising from electronic health records, and modern bioinformatics and genomics technologies, dynamic predictions using joint models and landmark approach with particular focus on risk prediction for dementia, modelling of neuropsychological test scores measuring cognitive decline, latent class analysis and growth mixture models.