Applied Statistics for Observational Studies & Clinical Trials module (GM52041)

Observational studies in public health involve observing and analysing health-related data. This happens without intervening or manipulating the subjects or their environment.

Clinical trials are research studies. They assess the safety and effectiveness of new medical interventions.

In this module you will learn about using statistics in both of these situations.

You’ll learn how to use advanced statistical methods with routine observational data linked to health records. You’ll create clinical prediction rules (CPR) using logistic regression.

You’ll also build your understanding of statistics in clinical trials. This will help you during the planning, analysis, and reporting stages of a trial.

What you will learn

In observational studies, you will learn about:

• multiple regression models; linear and logistic
• using and evaluating different research designs, using data from cases
• deriving drug adherence measures from prescribing databases
• developing and validating clinical prediction models

In clinical trials, you will learn about:

• the phases and elements of clinical trials
• how to design an appropriate clinical trial, assessing strengths and weaknesses
• randomisation methods, clustering, dealing with missing data
• sample size and power for particular designs
• applying statistical methods for trial data

By the end of this module, you will be able to:

• reflect on the principles, theories, concepts, and practices of advanced statistical methods
• develop statistical models, leading to clinical prediction rules, using routine data in SPSS
• adjust for biases and confounding in routine data
• assess novel designs, such as case-only studies, to reduce confounding
• analyse clinical trial data according to the design, and deal with missing data
• design a randomised trial and estimate sample size needed
• analyse trial outcomes appropriately and present results
• use appropriate software, including Excel, Word, PowerPoint, and SPSS

Assignments / assessment

• written assignment on observational studies (50%)
• written assignment on clinical trials (50%)

Teaching methods / timetable

• initial workshop introduction to use of SPSS for analysing health data
• weekly lectures with live demonstration of SPSS analyses
• tutorials