The aim of this proposed session is to showcase the application and development of statistical methods that address challenges and biases when observational data are used to help solve problems relating to the statistical study of pprecision medicine. The session consists of four speakers, and cover the topics of missing data, model mis-specification, measurement error, and irregular observation times, all in the context of estimation of adaptive treatment strategies using techniques drawing on the causal inference literature. 

Speakers

The speakers in this session represent a diverse group of outstanding researchers. All speakers are presently employed at Canadian universities, however they represent three different maternal languages (English, French, and Finnish), three different countries of origin (Canada, England, and Finland), different sexual orientations (two members of the LGBTQ+ community), different genders (one woman), and different career stages: from very early career (PhD 2021) to a full professor. The session is organized and chaired by a senior female researcher.

Dr. Janie Coulombe is an Assistant Professor in the Department of Mathematics and Statistics at the Université de Montréal. She has graduated with a PhD in Biostatistics from McGill University (2021). Her research focuses on causal inference under irregular observation times and on the special features of observational data. After graduating with her PhD, Dr. Coulombe pursued a postdoctoral fellowship at McGill University working in collaboration with Dr. Erica Moodie (McGill University) and Dr. Susan Shortreed (University of Washington, Kaiser Permanente Washington). In 2022, she took up the position of Assistant Professor at the Université de Montréal.

Dr. Denis Talbot is a Professor of Biostatistics at the Department of Social and Preventive Medicine of Laval University’s School of Medicine and a regular investigator at the University Health Centre of the Québec – Laval University Research Center. He holds a research career award from the Fonds de recherche du Québec – Santé. His research interests concern various topics in causal inference including model selection and personalized medicine.

Dr. Michael Wallace is an Associate Professor at the University of Waterloo. Their research focuses on precision medicine and dynamic treatment regimes, developing new methodology to identify treatment decision rules based on patient-level data. More broadly, they study the effects of assumption violations in causal inference problems, with a particular focus on measurement error.

Dr. Olli Saarela is an Associate Professor at Dalla Lana School of Public Health, University of Toronto. Before joining U of T, he completed his PhD at University of Helsinki, Finland, followed by a postdoctoral fellowship at McGill University. He is a biostatistician with methodological research interests in causal inference, Bayesian inference and survival analysis. 

Organizer

Dr. EricaMoodie is a Professor of Biostatistics and a Canada Research Chair (Tier 1) in Statistical Methods for Precision Medicine. She obtained her MPhil in Epidemiology in 2001 from the University of Cambridge and a PhD in Biostatistics in 2006 from the University of Washington, before joining the faculty at McGill. Her main research interests are in causal inference and longitudinal data with a focus on precision medicine. She is the 2020 recipient of the CRM-SSC Prize in Statistics and an Elected Member of the International Statistical Institute. Dr Moodie serves as an Associate Editor of Biometrics and a Statistical Editor of Journal of Infectious Diseases. She holds a chercheur de merite career award from the Fonds de recherche du Quebec-Sante.

Tentative talk titles 

Janie Coulombe - An individualized treatment rule for the choice of antidepressant drug with multiply imputed longitudinal data

Denis Talbot - Double robust estimation of partially adaptive treatment strategies for hormonal therapy treatments for breast cancer

Michael Wallace - Precision medicine with imprecise measurements: Exploring measurement error in dynamic treatment regimes

Olli Saarela - Model-based estimation of multistage dynamic treatment regimes based on irregularly observed data

Causal inference attempts to uncover the structure of the data and eliminate all non-causative explanations for an observed association. The goal of most, if not all, statistical inference is to uncover causal relationships, but it is not in general possible to conclude causality from standard statistical inference procedures, merely that the observed association between two variables is not due to chance. The need for causal inference procedures is apparent in many fields, but is perhaps most pressing in the field of health research, where quantifying the efficacy of new therapies, or uncovering the etiology of diseases, is often rendered complicated due to difficulties inherent in observational studies. Even in experimental studies, partial compliance to treatment regimens can compromise a well-designed experiment. The complexity of models, and corresponding inference procedures, is heightened in the context of longitudinal studies, where time-dependent confounding may be present.

The statistical study of adaptive treatment strategies, also called dynamic treatment regimes, is an area that has grown to prominence over the last two decades. While sequentially randomized trials can provide high quality evidence for the efficacy of tailored sequences of treatments, the sample sizes needed to estimate heterogeneous effects that form the basis of treatment personalization are such that non-experimental data sources are commonly used to estimate adaptive treatment strategies. Thus, methods in causal inference have formed the basis for most estimation approaches in this area. 

The aim of this proposed session is to showcase the application and development of statistical methods that address challenges and biases in observational data to help solve problems relating to personalization of care. The session would consist of four speakers, and cover the topics of missing data, model mis-specification, measurement error, and irregular observation times.

Organiser: Prof. Erica EM Moodie 
Chair: Haoyu Wu

Speaker: Michael Wallace 

Speaker: Olli Saarela 

Speaker: Dr Janie Coulombe 

Speaker: Prof. Denis Talbot