Missing data is common in most clinical trials because some subjects withdraw from the trial before the crucial final measurements. Without those final measurements, our picture of the new treatment is imperfect, and it could actually be wrong. What if most of the withdrawals from the new treatment arm dropped out because they felt they were getting worse, or because of strong side effects?
The healthcare community have for some time been looking afresh at how to strengthen the credibility of results from clinical trials with missing data. In 2010, the FDA-sponsored National Research Council report Prevention and treatment of missing data in clinical trials gave a number of recommendations that can be summarised as follows:
1. Minimise the problem of missing data by preventing it as far as possible.
2. Choose assumptions about missing data that make sense to clinicians and patients.
3. The report argued that, for any clinical trial, there is no single “right answer” to missing data, because we simply don’t know what the missing data might have been – so the report recommended providing sensitivity analyses that would allow regulators, clinicians and patients to assess for themselves the robustness of a study’s results to missing data.
In this webinar:
- Statistician Michael O’Kelly will introduce the current missing data landscape.
- Jessica Cooper will show how missing data can be prevented, describing how she and the study teams used both common-sense and “outside the box” methods to minimise loss to follow-up in two recent megatrials.
- Returning to statistics, Ilya Lipkovich will discuss the framework for handling missing data known as multiple imputation. Ilya gives examples of how multiple imputation can provide reasonable approaches to analysing missing data.
- Michael O’Kelly finishes the webinar by describing a particular use of multiple imputation called “control-based” multiple imputation, and shows how historic data could motivate the use of “control-based” imputation for primary analyses and sensitivity analyses, using as an example a real study of major depressive disorder.
Speakers
Michael O’Kelly, Ph.D., Senior Director, Centre for Statistics in Drug Development (CSDD), Innovation, Quintiles
Michael O’Kelly, Ph.D., Senior Director, Centre for Statistics in Drug Development (CSDD), Innovation, Quintiles
Michael O’Kelly is Senior Director in the Centre for Statistics in Drug Development (CSDD), Innovation, Quintiles. He has over twenty years experience in clinical statistics: ten as a project statistician, followed by twelve in research and consulting. His received his degree from University College Dublin in the area of applied statistical modeling, and this is reflected in two of his current research areas – missing data in clinical trials, and model based drug development (MBDD).
In the area of missing data, with colleagues O’Kelly has developed new methods for missing data that are now widely used in clinical trials. He works with colleagues and customers on a wide variety of missing data issues in clinical trials. His book “Clinical trials with missing data: a guide for practitioners” was published in March 2014 by Wiley. O’Kelly is co-ordinator of the DIA Scientific Working Group on Missing Data.
O’Kelly was founder and chair of the Special Interest Group (SIG) for Modelling and Simulation of the UK-based professional statisticians’ group PSI. The SIG has linked to the European body EFSPI, and membership is now European-wide.
More generally, O’Kelly is a resource for Quintiles project statisticians, where a wide variety of queries often provide him and his CSDD colleagues with a stimulus to further research.
Message Presenter
Jessica Cooper, Clinical Project Manager, Quintiles
Jessica Cooper is a Senior Clinical Project Manager in the Cardiovascular Metabolic Therapeutic Delivery Unit. She has over 7 years of experience working in project management at Quintiles.
In regards to the prevention of missing data, Cooper spent the past two years working as the Global Retention Lead of a Phase 3 study that recruited over 8,000 subjects. The end result was record level retention metrics (less than .01% LTFU and .08% withdrawn consent).
Message Presenter
Ilya Lipkovich, Ph.D., Senior Director, Center for Statistics in Drug Development (CSDD), Innovation, Quintiles
Ilya Lipkovich, Ph.D., Senior Director, Center for Statistics in Drug Development (CSDD), Innovation, Quintiles
Ilya Lipkovich, Ph.D., is Senior Director, Center for Statistics in Drug Development (CSDD), Innovation, Quintiles. He has over 15 years of statistical consulting experience working in various areas including econometrics, manufacturing and quality control, and Pharmaceutical industry (last 12 years).
Dr. Lipkovich received his Ph.D. in Applied Statistics from Virginia Polytechnic Institute and State University in 2002. His research interests include Bayesian model averaging, clustering, predictive modeling and subgroup identification in clinical data using recursive partitioning, missing data and multiple imputation, and causal inference for observational data including marginal structural models and propensity-based estimation.
Dr. Lipkovich is a co-developer of novel subgroup identification methods (SIDES and SIDEScreen). He also chairs the QSPI Subgroup Analysis Working Group sponsored by the Society of Clinical Trials.
Message PresenterWho Should Attend?
EVPs, VPs, Directors, Heads, Managers, Principal Scientists of:
- Clinical R&D
- Clinical Operations (Clinical Project Leads, Clinical Program Managers, Therapeutic Heads, Clinical Data Managers, Pharmacovigilance Scientist)
- Regulatory Affairs
- Medical Affairs
- Outsourcing Management
Xtalks Partner
Quintiles
Quintiles is the only fully integrated biopharmaceutical services company offering clinical, commercial, and consulting solutions worldwide. The Quintiles network of more than 27,000 engaged professionals in 59 countries works with an unwavering commitment to patients, safety and ethics. Quintiles helps biopharmaceutical companies navigate risk and seize opportunities in an environment where change is constant. For more information, please visit www.quintiles.com.
Media Partner
Related Webinars
You Must Login To Register for this Free Webinar
Already have an account? LOGIN HERE. If you don’t have an account you need to create a free account.
Create Account