Session (CTI)

ISSUE: Often when conducting meta-analyses (MA) of time-to-event (TTE) results, especially in Health Technology Assessment (HTA), a hazard ratio (HR) scale is used due to the ease with which a treatment effect can be incorporated into a cost-effectiveness decision model. However, issues arise when there is evidence of non-proportional hazards in some of the trials/studies included. A number of methods have been advocated, including; fractional polynomial models, flexible M-splines, parametric models, flexible parametric models, piecewise exponential models, Restricted Mean Survival Time (RMST) models, and more recently a time-varying HR approach. However, the use of some methods has been limited by either their complexity and/or the ease of incorporating the results into cost-effectiveness decision models. This issue panel introduces, debates and reviews both more flexible and simpler methods, and discusses the relative merits of the most commonly used and alternative approaches. OVERVIEW: Keith Abrams will introduce and provide an overview on the use of methods for dealing with non-proportional hazards in HTA (10 minutes – including audience participation via the ISPOR app), and the three panellists will then consider the different methods in HTA using illustrative examples (10 minutes each) – David Phillippo will discuss the use of spline-based approaches to dealing with non-proportional hazards and how these capture time-varying effects, Rhiannon Owen will discuss the use of simpler time-varying hazard ratio approaches and how these can be more useful in a HTA setting, and Shannon Cope will discuss the relative merits of all the methods available, especially regarding implications for integrating the results into modelling and Joint Clinical Assessment (JCA). Keith Abrams will then moderate a discussion involving the audience and panellists on the relative merits of using the different methods available in a HTA setting, before eliciting the audience’s final views via the ISPOR app.