Handling missing outcome data in randomised trials

The aims of this course are to:

• Review some popular ways to analyse trials with missing data, focussing on the assumptions underlying them
• Discuss what assumptions may be plausible and how one should decide this in particular trials
• Describe the theory for mixed models analysis of incomplete data, how they should be implemented, and what are the pitfalls
• Demonstrate how to analyse data using mixed models and multiple imputation in Stata, R and SAS
• Discuss the advantages and disadvantages of the two methods
• Discuss the different issues involved in handling missing baselines
• Demonstrate ways to perform sensitivity analyses to departures from assumptions

The course will provide practising statisticians with the necessary practical skills to handle missing data in their analyses, and in particular to move beyond the use of complete-case analysis and last observation carried forward analysis. We will focus on trials with quantitative outcomes, and also consider binary outcomes but not time-to-event outcomes.

• Statisticians who analyse clinical trials with missing outcome data.
• Participants are asked to bring their own laptops with their preferred statistical software (R, SAS or Stata).

• Our course on Practical use of multiple imputation to handle missing data in Stata 12.
• Our course on Missing data in mental health research: a practical approach using Stata .
• Other BSU short courses.

• Ian White, MRC Biostatistics Unit

• Simon Bond, Cambridge Clinical Trials Unit & MRC Biostatistics Unit
• Shaun Seaman, MRC Biostatistics Unit
• Ian White, MRC Biostatistics Unit

• James Wason, MRC Biostatistics Unit