Medicines regulators and guideline groups work hard to ensure that medicines we use are as safe and as effective as possible. The only way to get this ‘gold standard’ evidence is through randomised controlled trials (RCT). These are useful because they reduce bias and allow us to assess if a medicine is truly effective for a condition. However, there are conditions for which RCTs cannot be carried out, e.g. rare disease where there may not be enough participants, diseases of ageing where it may be too expensive to carry out an RCT over many years. There are also circumstances where RCTs are done but they do not provide enough information to help us understand the longer-term safety of a drug. This is because RCTs only take place over a few years, whereas the drug may be a lifelong treatment. In addition, RCTs do not always take account of more detailed information like genetic make-up of an individual. This might affect how an individual responds to a medicine (pharmacogenetics) or how groups that share genetic markers might respond to a treatment (stratified or personalise medicine).
In this project we will aim to develop and use statistical methods to address these questions. We will work with partners across the NHS, the UK medicines regulator the Medicines and Healthcare product Regulatory Agency (MHRA) to develop use cases. We will then use UK Biobank data to develop methods to answer these questions to help us understand how effectively medicines work and if they are safe, or if other medicines could also be developed and/ or used for a condition. These use cases will develop a framework of how different types of data can be combined using statistical methods to enhance how drugs work and how they can be made safer.