Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Abstract: Missing outcome data in randomised trials are often analysed under a missing at random approach (e.g. using mixed models or multiple imputation), which assumes that any differences between missing values and observed values can be explained by other observed data. This is often a good starting point for analysis, but it is important to consider the possible impact of departures from missing at random. I will discuss various general approaches to this problem, using the central idea of sensitivity parameters. In particular, I will describe a simple mean score method for a single incomplete outcome. For a longitudinal outcome, I will discuss various methods that have been proposed for handling dropout (monotonic missing data) and describe an extension to multiple imputation by chained equations (MICE) for missing-not-at-random imputations of non-monotonic missing data. In particular, I will show that the sensitivity parameters that are required by MICE are related to parameters that are easier to elicit, and I will illustrate the method in a smoking cessation study.


Biography: Professor Ian White moved at the start of 2017 to a new position as Professor of Statistical Methods for Medicine at the Medical Research Council Clinical Trials Unit, University College London. This follows 16 years as a programme leader at the Medical Research Council Biostatistics Unit in Cambridge. He originally studied mathematics at Cambridge University, and his first career was as a teacher of mathematics in The Gambia, Cambridge and London. He obtained his MSc in statistics from University College London, where he subsequently worked in the Department of Epidemiology and Public Health. He was then Senior Lecturer in the Medical Statistics Unit at the London School of Hygiene and Tropical Medicine. He received his PhD by publications in 2011.


His research interests are in statistical methods for the design and analysis of clinical trials, observational studies and meta-analyses. He is particularly interested in developing methods for handling missing data, correcting for departures from randomised treatment, prognostic modelling, and network meta-analysis. He runs courses on various topics and has written a range of Stata software. His web page is at

NDPH Events and Seminars

Richard Doll seminars


Click here to go to the home page of the Richard Doll Seminars in public health and epidemiology.

Upcoming events

More events