February 17, 2021, 10:00-11:30 MST – Online
Distinguished Visitor Lecture by Ruwanthi Kolamunnage-Dona
Online event hosted by the University of Calgary and CANSSI
Longitudinal study designs are widely used in clinical research, however, they are often analysed by simple statistical methods, which do not fully exploit the information in the resulting data. For example, patients with worse conditions may utilise health care more, or may have worse biomarker values recorded over a shorter period of time than those patients with milder conditions. In observational studies, biomarkers are measured at irregular follow-up visit times, and in randomised controlled trials, participant dropout is common during the intended follow-up, which are often correlated with patient’s prognosis. Joint modelling is a modern statistical method that has the potential to reduce biases and uncertainties due to informative participant follow-up in longitudinal studies. It combines longitudinal biomarkers and clinical event-times together (simultaneously) into a single model through latent associations. The model can also be used to investigate the association between the longitudinal outcomes and important clinical events. In this talk, our work on the methodology of joint modelling, and its advances for competing risks and multiple longitudinal outcomes will be discussed with real applications in health research.
Ruwanthi Kolamunnage-Dona is an Associate Professor in Biostatistics from the Department of Health Data Science at the University of Liverpool. Her research interests are in statistical analysis methods for most efficient use of the data available, and translating this knowledge into clinical applications. She actively involves development of methodologies for joint analysis of longitudinal and survival data with applications in health research, to integrate biomarkers for clinical-decision making and drug discovery. She has published over 50 articles, many as the Lead Methodologist of clinical studies or proof of concepts for joint modelling in the highest impact scientific journals. Over 20% of these Publications are in the top 10% most cited worldwide. She is the leading statistician or the principal investigator for many research grants, including a Medical Research Council funded methodological grant to develop statistical modelling and prediction for multidimensional data. She is leading the Joint Modelling Research Collaboration, which is at the forefront of joint modelling research, and internationally renowned for their research outputs including innovative software and training workshops. She has been a Fellow of the Royal Statistical Society for the past 8 years. She is an Associate Editor for the Research Methods in Medicine & Health Sciences journal, and a Statistical Editor for the Lancet journal of Child & Adolescent Health.