Lectures and seminars SFOepi Seminar Series in Biostatistics: Ian White

Abstract: Often, statisticians and epidemiologists say what analysis we want to do, instead of saying what question we are trying to answer or what quantity we are trying to estimate (the estimand). To remedy this, estimands are increasingly talked about, especially in the pharmaceutical industry. Often, though, specifying the estimand may seem like just another tedious task that we have to do. I hope to show you that thinking well about estimands can help to resolve some difficult questions about how to analyse our data. I will use examples from analysis of longitudinal data, trial design, competing risks and meta-analysis, illustrated by trials in neurodegenerative diseases, liaison psychiatry and tuberculosis.

About the speaker: Professor Ian White is based at the MRC Clinical Trials Unit at University College London (UCL), United Kingdom. Professor White is a medical statistician with an interest in developing new methodology for design and analysis of clinical trials, meta-analysis and observational studies. He has worked for many years in the research area of missing data, where he has published landmark methodological papers of multiple imputation. Through his publications and teaching he has contributed to the widespread use of multiple imputation in medical research, and he is now developing extensions to MI methods for missing-not-at-random (MNAR) data. He is also particularly interested in meta-analysis and network meta-analysis, where he has developed methods for assessing and testing inconsistency.  https://www.mrcctu.ucl.ac.uk/about-us/our-people/ian-white/

 If you have any questions, please contact Marie Jansson at marie.jansson@ki.se