The module aims to further students’ understanding of statistical methods used on records data and introduce them to a number of specific techniques for dealing with the challenges of interpreting and drawing inferences from electronic healthcare records.
Module code
CHME0015
UCL credits
15
Course Length
9 Weeks
Face to Face Dates
Jan: 07; 14; 21; 28. Feb: 04; 11; 25. Mar: 03; 10; 17.
Assessment Dates
11th May 2020
Module organisers
Dr Michail Katsoulis Please direct queries to courses-IHI@ucl.ac.uk
Content
The course will cover a range of advanced statistical techniques used in records research. In particular:
- More advanced regression techniques, including Cox modelling, logistic regression and Poisson regression.
- Methods for handling missing data with a focus on multiple imputation.
- Challenges in using observational data to assess causality approaches.
- Dealing with clustering in regression either through multilevel modelling or generalized estimating equation approaches.
- Overview of risk prediction modelling.
Teaching and learning methods
This 15 credit module lasts for 6 weeks and should represent roughly 150 hours of learning time. The module will use a mixture of lectures and computer-based practicals using Stata. There will be private reading and materials will be made available via Moodle, with some online activities.
Assessment
The final assessment will involve analysing an example data set and writing a report to answer the given research questions.