top of page

​Improving the Analysis of Adverse Events

AE_Project1.png

Refining time-to-event models to detect non-constant hazards to signal adverse drug reactions (ADRs)

​

Team: Odile Sauzet, Victoria Cornelius

​

Pharmacovigilance is the process of monitoring the emergence of harm from a medicine once it has been licensed and is in use. The last decade has seen increased interest for the use of electronic health records (EHRs) in pharmacovigilance. The causal mechanism of an ADR will often result in the occurrence being time dependant. In this project we look at identifying signals for ADRs based on detecting a variation in hazard of an event using time-to-event models. Based on previous work we extend this study to examine generalised Weibull model-based approaches on simulated data and applied to a case study.

​

​

​

​

AE_Project2.png

Better Methods for the analysis of adverse events – The why and how

​

Team: Rachel Phillips, Victoria Cornelius

​

In this project we are writing a series of article with collaborators which include in the first batch: 

  • The “Why” of safety analyses and adverse events collection in clinical trials

  • A guide for optimal analysis when dichotomising continuous adverse event outcomes

  • The important use of regression analysis for adverse event outcomes

​

​

​

​

​

DoRAH image.PNG

Development of Resources to promote best practice for the Analysis of Harm outcomes in RCTs (DoRAH)

​

Team: Rachel Phillips, Victoria Cornelius, Siobhan Creanor, Catherine Hewitt and Carrol Gamble

​

The aim of this project is to promote the use of accepted good practice principled analysis approaches when analysing harm outcomes in RCTs via development of resources to guide and aid implementation.  

​

Specifically, we aim to:

  • Develop a menu of analysis templates for harm outcomes/adverse events (AEs) in statistical analysis plans, modelling good practice approaches with consideration to intervention types (e.g., drug, surgical, complex interventions) and trial designs (e.g., parallel arm, multi-arm).

  • Develop case studies with accompanying statistical code.

  • Develop a standard operating procedure (SOP) available to all CTUs to support adoption of good statistical analysis practices for harm outcomes.

​

Project steering group members: Professor Richard Hooper (PCTU, QMUL), Kaspar Rufibach (Roche Basel), Dianna Papaioannou (Sheffield CTRU) and Professor Rupert Pearse (QMUL & Barts Health NHS).

​

bottom of page