Rationale
The primary goal of the ROAR study is to develop personalised, accurate models to predict the risk of aneurysm rupture helping patients and doctors make better-informed treatment decisions. The current cohort includes detailed baseline data, allowing for validation of existing models like PHASES and investigation of vascular risk factors. However, by linking ROAR to GP records, we can track how key variables change over time, including new diagnoses, medications, blood pressure, and lifestyle factors.
This will allow a more detailed analysis of important risk factors such as blood pressure, ACE inhibitors, aspirin, anti-platelets, anticoagulants, cholesterol, statins, alcohol, and smoking. It will also enable a closer look at factors not included in the original ROAR dataset—such as obesity, biologics, the oral contraceptive pill (OCP), hormone replacement therapy (HRT), and menopause—which may be particularly important in understanding why aneurysms are more common in women. These enhancements will significantly improve the precision and relevance of our risk models.
Aims
1. Predict rupture risk – by assessing risk factors that change over time.
2. Reduce rupture risk – through modifiable risk factors and the influence of medications.
3. Understand female risk – by analysing hormonal factors including menopause and hormone replacement therapy.
Study design
The UK is the ideal setting for this project, with its comprehensive electronic GP record systems (with EMIS Web used by around 60% of practices, SystmOne by approximately 39%) enabling detailed linkage to primary care data and providing rich, population-wide insights. We are currently exploring several secure options for GP record linkage which provide access to anonymised health records from a large proportion of UK practices. Using this data, we will conduct time-varying survival analyses to model how evolving patient characteristics influence the risk of aneurysm rupture. This approach will enable us to build more personalised and accurate risk prediction tools for doctors and patients.
