Sample sizing for survival trials specifies a clinically significant proportional hazard, which if obtained, implies the adoption of the new therapy. Such adoption decisions are often evaluated using health economic decision models. We develop methods to formally integrate decision modelling with sample size for survival studies using the expected value of sample information approach.

Method,

The method requires characterization of prior uncertainty around the proportional hazard, together with a model of survival, quality of life benefits and economic costs. EVSI computation uses Monte Carlo sampling to produce new simulated data-sets with specified size and follow-up. Each simulated data-set is synthesised with existing prior information. Because Bayesian updating for Weibull parameters is not analytically tractable, we use a novel form of Laplace approximation to estimate the decision model outputs. A case study builds on a classic text book example. We examine 1st and 2nd order versions of the Laplace approximation formula, comparing EVSI estimates. Expected benefits from additional information are compared against a cost function for different proposed study designs.

Results,

The approach is more efficient than standard EVSI computation because it requires neither Markov Chain Monte Carlo to obtain posterior densities nor Monte Carlo sampling to quantify the effect of each simulated data-set on decisions.

Conclusion.

The results show how the formal integration of economic considerations is both feasible and potentially profound in the design of survival trials. This methodology provides a new and valuable alternative approach for design of survival studies.