PSI Training Course: Mixed Models and Repeated Measures
Dates & Times:
Monday 3rd February 2025: 08:30-13:00 BST
Tuesday 4th February 2025: 08:30 - 13:00 BST
Wednesday 5th February 2025: 08:30 - 13:00 BST
Thursday 6th February 2025: 08:30 - 13:00 BST
Location:
Online via Zoom
Speakers: James Carpenter (LSHTM), Linda Sharples (LSHTM)
Who is this event intended for? This course is suitable for statisticians working on clinical trials, who already have a good understanding of linear and generalised linear models and want to further their knowledge of repeated measures and other clustered data.
What is the benefit of attending? By the end of the course attendees will know how to analyse repeated measures of patients through time or other clustered data in randomised clinical trials and associated observational studies. Attendees will develop their knowledge on conditional models for continuous hierarchical and longitudinal data, GEE and discrete models. Practical exercises will allow hands-on experience when working with this type of data and presenters will be available during the course to answer any questions.
Cost
Early Bird PSI Members = £320+VAT
PSI Members = £360+VAT
Early Bird Non-Members = £430+VAT
Non-Members = £470+VAT
*Please note: Non-Member rates include PSI membership until 31 Dec. 2025.
Registration
To book your place, please click here.
Overview
Repeated measures of patients through time and other clustered data are common in randomised clinical trials and associated observational studies. Measurements taken from the same patient (or from the same cluster) are likely to be correlated, so that the assumption that all responses will be identically distributed and independent from each other will not hold. Ignoring within-cluster correlation will result in bias in the estimate of the treatment effect standard error and therefore, incorrect confidence intervals and hypothesis tests. In some situations it can also result in bias in the treatment estimate itself. Using a range of worked examples, this course will explain how to analyse repeated measures and other clustered data, with an emphasis on estimating treatment effects using the appropriate covariance structure between measurements.
This course is presented through lectures and practical sessions using SAS code. It is suitable for statisticians working on clinical trials, who already have a good understanding of linear and generalised linear models.
Topics covered include:
• Conditional models for continuous hierarchical data
• Conditional models for continuous longitudinal data
• Marginal models (GEE) for continuous longitudinal data
• Discrete data
Speaker details
Speaker
|
Biography
|
James Carpenter
|
James Carpenter is Professor of Medical Statistics at the London School of Hygiene & Tropical Medicine, with a 50% secondment to the MRC Clinical Trials Unit at UCL, where he is an MRC Investigator in Trials Methodology.
He has long-standing interests in missing data methodology (particularly multiple imputation) and longitudinal modelling; last year he co-authored the second edition of Multiple imputation and its Application (Wiley). More recently, he became the senior statistician in the ACORD collaboration, focussed on developing, running and reporting multi-arm multi-stage trials in neurodegenerative diseases.
|
Gemma Hodgson
|
After finishing a PhD in Bayesian Methods for Hierarchical Data and a short post-doctoral lectureship in Medical Statistics, Linda joined the MRC Biostatistics Unit in Cambridge. At Cambridge she led a programme of research in the evaluation of new health technologies using both experimental and observational designs. The work was applied to interventions to treat heart and lung diseases, particularly transplantation and other surgery. Methodological contributions have focussed on longitudinal data and the synthesis of trial and observational data in decision models. She was appointed Professor of Statistics in the Clinical Trials Research Unit at Leeds University in 2013, but could not resist relocating to the LSHTM when the opportunity arose in 2017. Her current interests focus on developing clinical prediction models for early diagnosis of cancers with non-specific symptoms and use of electronic health records for evaluation of service provision for (among others) prostate cancer.
|