PSI Scientific One Day Meeting: Missing Data in Clinical Trials - Past, Present & Future

Add to:

At this meeting we will present the evolution of missing data approaches, looking at how they have been handled in the past, the current established missing data approaches and the impact of the new ICH E9 R1 addendum on the handling of missing data.

Date: Tuesday 28th April 2020
Time: 09:30 - 16:30 (UK Time)
Location: GSK House, Brentford, TW8 9GS
Speakers: David Brown (MHRA), David Wright (AstraZeneca), Simon Day (CTCT Ltd), Michael O’Kelly (IQVIA), Daniel Bratton (GSK), James Roger (Ind) and Mouna Akacha (Novartis).

Incomplete datasets due to missing data is an issue that has been, and will be, around for a long time. At this meeting we will present the evolution of missing data approaches, looking at how they have been handled in the past, the current established missing data approaches and the impact of the new ICH E9 R1 addendum on the handling of missing data, focussing in particular on the treatment policy estimand.

PSI Member: £40+VAT 
Non-Member: £135+VAT (price includes PSI membership for the remainder of 2020)
Please click here to register.

Please click here to view the agenda for this event. 

 Speaker  Biography  Abstract
David Brown cropped

David Brown

David Brown has been a statistical assessor at the MHRA for over 20 years and is currently the acting head of the statistics & pharmacokinetics unit. He was the main author on the CHMP guideline on the choice of the non-inferiority margin. He was on the drafting groups for all the CPMP points to consider statistical guidelines released in the wake of ICH E9 in the early 2000s, which included the points to consider on missing data. He played a key role in the development of the CHMP guideline on the investigation of bioequivalence, and has been involved in more recent CHMP guidance documents, both statistical and non-statistical, via the biostatistics working party. He has also been a member of the PK and scientific advice working parties at the EMA.

The history and future of missing data handling in medicines regulation.

Based on 20+ years of experience in medicines regulation, this talk will give an eye-witness account of the evolution of missing data handling from the days of ignoring it completely, through LOCF, MMRM and other acronyms up to the emergence of the estimand, which some may consider to also be a four letter word!

David Wright cropped
David Wright


David Wright is the Head of Statistical Innovation at AstraZeneca. He has worked at AZ for 3 years across all therapeutic areas of interest for the company. Previously David worked for the MHRA for 16 years where he was the Chair of the Biostatistics Working Party and led the revision of the CHMP guideline on missing data in confirmatory clinical trials.This work sparked his interest in Estimands particularly on how to use the proposed framework in practice. David is a member of the EFPIA/EFSPI Estimand Implementation Working Group.


Aligning how subjects with missing data due to study discontinuation are handled in the primary analysis with the primary Estimand.

The ICH E9 Addendum states that “methods to address the problem presented by missing data can be selected to align with the estimand”. What does this mean in practice? Should this approach apply generally in confirmatory clinical trials? This talk will explore this and give examples in different therapeutic areas of how the Estimand framework might be aligned with how missing data are handled in the primary analysis.

Simon Day cropped 

Simon Day
Clinical Trials Consulting & Training Ltd.

Simon has spent 30 years working in clinical trials, mostly in the pharmaceutical industry but also including five years at the UK and European regulatory agencies. He now works as a statistical and regulatory consultant to pharmaceutical and biotechnology companies around the world. He is a former president of the International Society for Clinical Biostatistics. He is joint editor of Statistics in Medicine, and on the editorial board of Translational Sciences of Rare Diseases. In 2012 he was elected a Fellow of the Society for Clinical Trials. He has published widely in statistical and medical journals, is author of one book “Dictionary for Clinical Trials” and is joint editor of the “Textbook of Clinical Trials”, both published by Wiley.

Missing data handling in non‐inferiority and equivalence trials.

I think we have fallen fowl in the past of taking methods and approaches developed for clinical trials (implicitly, but not always explicitly) intended to show superiority, and copied them across to equivalence and non-inferiority trials without considering any special caveats. Sample size re-estimation is an obvious case in point. Regarding problems of missing data we have perhaps been a bit more cautious, readily understanding that ‘conservative’ approaches (for superiority trials) may typically be anti-conservative (for equivalence and non-inferiority trials).

I will report on a literature review of 109 papers published between May 2015 and April 2016. Apparently, eight of these studies had no missing data for the primary outcome (lucky them). Of the unlucky ones (or the honest ones), 50% reported the primary analysis based on complete cases, and 28% reported single imputation approaches for handling missing data. Only 32% reported conducting analyses of both intention‐to‐treat and per‐protocol populations. Only 11% conducted any sensitivity analyses to test assumptions with respect to missing data.

Not great, really, is it?

Michael OKelly cropped
Michael O’Kelly



Michael O’Kelly has worked as a statistician in the pharmaceutical industry for 27 years. He has been involved in many areas of biostatistics, and has a special interest in statistical modelling and simulation. He co-authored a Best Practice proposal for projects involving Modelling and Simulation, which was adopted by the PSI board in 2017. With colleagues Michael O’Kelly has developed new methods for missing data that are now widely used in clinical trials. His book authored with Bohdana Ratitch, “Clinical trials with missing data: a guide for practitioners”, was published in 2014 by Wiley. He has given courses on missing data for PSI and at many conferences and for many pharmaceutical companies. For his work in best practice and in missing data, Michael received the RSS/PSI award for Excellence in Pharmaceutical Statistics in 2017; he is Senior Director with IQVIA’s Centre for Statistics in Drug Development.

Even a “treatment policy” estimand may have missing data: how can we take account of this?

Under the “treatment policy” estimand, outcomes contribute to the estimate of effect irrespective of treatment actually taken after randomisation, and all subjects are to be followed up for the full scheduled follow-up period. It is hoped that under this estimand missing data would be limited. However, in practice, subjects may avail of their right to discontinue completely from a trial. Even under a “treatment policy” estimand, this leaves the analyst with missing data. How is the analyst to take account of this residual missing data under the “treatment policy” estimand? The new ICH E9 R1 Addendum suggests “for subjects who discontinue treatment without further data being collected, a model may use data from other subjects who discontinued treatment but for whom data collection has continued”. However, numbers of appropriate subjects with data available may be small, and compromises or assumptions may be required to do the kind of modelling suggested by ICH E9 R1. This presentation evaluates some of the options that use statistical modelling to take account of this residual missing data.

Daniel Bratton cropped
Daniel Bratton


Dan has worked as a statistician within the Respiratory Clinical Statistics group at GSK since January 2016. During this time he has worked on RCTs investigating the effect of various treatments in severe asthma, COPD and nasal polyps. Dan started his career as a medical statistician at the Medical Research Council Clinical Trials Unit (MRC CTU) working on RCTs in respiratory diseases and dermatology. He then completed a PhD at the same unit through UCL investigating statistical issues in the design of multi-arm multi-stage clinical trials which are increasingly being used in practice to accelerate the evaluation of new therapies. Prior to starting at GSK, Dan worked as a statistician in the Department of Pulmonology at University Hospital Zurich for one year, focusing mainly on network meta-analyses of RCTs investigating treatments for sleep apnoea.

Treatment policy estimands for recurrent event data using data collected after cessation of randomised treatment.

Trials now increasingly collect data from subjects following premature discontinuation of study treatment, as this event is irrelevant for the purposes of a treatment policy estimand. However, despite efforts to keep subjects in a trial, some will still choose to withdraw. Publications for sensitivity analyses of recurrent event data have often focused on reference‐based imputation methods, more commonly applied to continuous outcomes, where imputation for the missing data for one treatment arm is based on the observed outcomes in another arm. The existence of data following premature discontinuation of treatment now raises the opportunity to impute missing data for subjects who withdraw from study using this observed ‘off-treatment’ data, potentially allowing more plausible assumptions for the missing post‐study‐withdrawal data than other reference‐based approaches. In this poster, we describe a recent imputation method (1) for recurrent event data in which the missing post‐study‐withdrawal event rate for a subject is assumed to reflect the rate observed from subjects during the off‐treatment period. The method is illustrated in a trial in chronic obstructive pulmonary disease (COPD; GSK funded NCT02105961) where the primary endpoint was the rate of exacerbations, analysed using a negative binomial model.

James cropped
James Roger
James Roger has retired after a career spanning university lecturing, international agricultural and medical research and an extensive period employed in the pharmaceutical industry. He is an honorary member of PSI and a honorary professor at LSHTM. Jump-to-Reference (J2R) versus Copy-Reference (CR) imputation. What's the difference and does it matter?

Multivariate Gaussian repeated measures is often used for quantitative data and can be fitted directly by maximum likelihood even when some observations are MCAR or MAR. When the missing data is monotone in nature, such as after trial withdrawal, alternatively a stepwise approach can be used, imputing values based on a series of univariate models. Typically each step involves regressing on the baseline covariates and the previous observations, both observed and imputed. This is the method used in the MONOTONE REG facility of proc MI in SAS. Although proc MI does not allow it, one could equally regress on the previous residuals rather than the previous observed values, as these approaches are mathematically equivalent. However this equivalence only holds when the same model is used at each step for baseline covariates, including treatment assignment. We explore the implications of changing the treatment allocation half way through on these two algorithmic approaches.
Mouna cropped
Mouna Akacha

Mouna Akacha is the Group Head of the Statistical Methodology group of Novartis Pharma AG, based in Basel, Switzerland. She and her team provide internal advice for clinical projects in all development phases and therapeutic areas. She is engaged in developing and implementing innovative statistical methods for clinical projects covering estimand discussions and approaches for missing data, longitudinal data, and recurrent event data. Before joining Novartis, Akacha studied mathematics at the University of Oldenburg in Germany. She holds a PhD in statistics from the University of Warwick in the United Kingdom. Discussion and Update on DIA Missing Data Working Group.

In this discussion, I will share some thoughts on the work which was presented throughout the day. In addition, I will give an overview of the available material and tools developed by the DIA Missing Data Working Group.

Upcoming Events

Latest Jobs