Scientific intelligence platform for AI-powered data management and workflow automation
In this free tutorial, we have explored some common challenges faced when designing clinical trials with survival endpoints, and the solutions for these problems.
Challenges in Survival Analysis
Non-Proportional Hazards and Solutions
Stratified effects in Survival Analysis
Survival analysis is an important and common statistical area of interest for oncology trials. Due to the complex nature of these trials, some unique challenges are faced.
Many of these issues are encountered when the proportional hazards assumption is not met and methods like the Log-Rank test and Cox Proportional Hazards model can not be directly applied.
Treatments that exhibit delayed effects or other non-proportional effect patterns can be analysed using weighted rank tests and concepts such as relative time. Interaction effects can also be identified and accounted for in stratified designs.
Clinical prediction models are another helpful tool to enable healthcare providers to make better risk evaluations for diagnostic or prognostic purposes. However, the quality of these models is often under scrutiny and it is important to ensure that these models meet certain metrics for quality and robustness.
In this free tutorial, we have examined the methods used to analyse survival trials that encounter non-proportional hazards, as well as assessing the challenges in constructing a prediction model with survival endpoints.
Who is this for?
This will be highly beneficial if you're a biostatistician, scientist, or clinical trial professional that is involved in sample size calculation and the optimization of clinical trials in:
Try for free and upgrade as your team grows