Methodological evaluation of the impact of survival costs in oncology modelling

Abstract:
Objectives: Economic evaluations typically include all costs relevant to a disease. This is particularly relevant to oncology modelling, as costs are assigned to each health state in the model, and, therefore, extending survival also increases costs. Because patients often incur higher healthcare costs in the post-progressed state of disease where costs of disease management are high, extending survival and increasing a patient’s time in the post-progressed stage can be particularly costly.  The objective of this research was to investigate the methodology used in oncology modelling, and to determine the effect that this has on predicted cost-effectiveness. 

Methods:  A simple three-state economic model was produced with with ten key parameters to calculate the ICERs associated with various combinations of inputs. Extensive scenario and multiway sensitivity analyses were carried out to document informative patterns and relationships between parameters that affected the results.  Specifically, the model tested the impact of: (i) the relative duration of progression-free survival and post-progression survival, (ii) the shape and scale of parametric coefficients for survival, (iii) the impact of treatment duration and (iv) the time-dependency of post-progression costs. 

Results: The paper presents the concept of a ‘natural ICER’, the value towards which the results tend as survival is indefinitely increased.  Results showed that the ‘natural ICER’ is independent of the model design and the choice of survival inputs, and is driven purely by the cost and utility of the post-progressed state.   In some cases with higher post-progression costs, the likelihood of a treatment being cost-effective decreased as the effectiveness of the treatment improved.  The results demonstrate circumstances in which no matter how effective a treatment is and how low the price is, it will not be cost-effective. 

Conclusions: The results demonstrate that when a treatment is not cost-effective, it is not always due to the pricing or effectiveness of the treatment. These results are due to the disease area (high post-progression background costs and low post-progression utility). For many oncology treatments whose primary aim is to extend survival, this impact can be prohibitive to an intervention’s probability of being cost-effective.