Definition of Terms

The goal of this analysis is to model two outcomes simultaneously based on the generation of discounted costs (2014 US dollars) and discounted quality adjusted life years (QALY) that accrue to our cohort over time. Each model input which governs a patient's natural history is a probability and beyond that there is a distribution of values (uncertainty) for most clinical variables—e.g. confidence intervals or ranges of values in the literature. For this reason, the model is called a microsimulation—which follows the stochastic movements of individual subjects through the chances of developing clinical outcomes—and a probabilistic analytic model which analyzes 10,000 random samples within each parameter’s distribution. Accordingly, this method repeats the analysis with 10,000 permutations of the variable inputs, thereby performing sensitivity analyses for each variable within its distribution. The end result is the probability of cost-effectiveness for a given strategy in the overall set of simulations. The relative cost-effectiveness of the strategies, in turn, are adjudicated with reference to a society’s willingness-to-pay (WTP) threshold. The WTP threshold is the amount of money per person that society is willing to pay to adopt a new clinical strategy for an additional QALY over the current acceptable strategy. Finally, we calculated a metric called the population ‘expected value of perfect information’ (EVPI). The population EVPI is a reflection of the benefit derived from further research and is therefore a measure of the uncertainty in this analysis.