Background: Considerable advances have been made in modeling Alzheimer's disease (AD), with a move towards individual-level rather than cohort models and simulations that consider multiple dimensions when evaluating disease severity. However, the possibility that disease-modifying agents (DMAs) may emerge requires an update of existing modeling frameworks.
Objectives: The aim of this study was to develop a simulation allowing for economic evaluation of DMAs in AD.
Methods: The model was developed based on a previously published, well-validated, discrete event simulation which measures disease severity on the basis of cognition, behaviour, and function, and captures the interrelated changes in these measures for individuals. The updated model adds one more domain, patient dependence, in addition to cognition, behaviour, and function to better characterize disease severity. Furthermore, the model was modified to have greater flexibility in assessing the impact of various important assumptions, such as the long-term effectiveness of DMAs and their impact on survival, on model outcomes. A validation analysis was performed to examine how well the model predicted change in disease severity among patients not receiving DMA treatment by comparing model results to those observed in two recent phase III clinical trials of bapineuzumab. In addition, various hypothetical scenarios were tested to demonstrate the improved features of the model.
Results: Validation results show that the model closely predicts the mean changes in disease severity over 18 months. Results from different hypothetical scenarios show that the model allows for credible assessment of those major uncertainties surrounding the long-term effectiveness of DMAs, including the potential impact of improved survival with DMA treatment. They also indicate that varying these assumptions could have a major impact on the value of DMAs.
Conclusions: The updated economic model has good predictive power, but validation against longer-term outcomes is still needed. Our analyses also demonstrate the importance of designing a model with sufficient flexibility such that the model allows for assessment of the impact of key sources of uncertainty on the value of DMAs.