Competing Events Modeling and Simulation: a Potential Approach to Benefit Risk Assessment

• Activity Number: 566
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Risk Analysis
• Author(s): Eric Frimpong*+ and Victor Crentsil
• Companies: FDA and FDA
• Keywords: risk-benefit ; competing events ; hazards ; mixture model ; survival analysis

Abstract

The approach to benefit-risk assessment (B-R) in regulatory decision is of great interest to reviewers, practitioners and patients. When the data is time-to-event, competing events may preclude realization of benefit or risk. In general, analytic approaches to handling such scenarios are lacking. In this simulation study, a mixture model approach, motivated by a competing risk framework, is developed to perform a drug B-R assessment. To assess the utility of our simulation approach, we compared our results to analysis of data obtained from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE). Comparisons of the simulation results to the CATIE data show that the model based approach can potentially enhance the drug regulatory review process by leveraging a novel analytic approach to increase yield when data is sparse, while introducing more objectivity into the decision process.