Analytical approach:
The cost-effectiveness analysis used a state-transition model, which was developed to simulate the transition between different forced expiratory volume (FEV) states. Data to inform the model came from published literature. The analysis adopted an NHS perspective over a lifetime horizon. A short term (24 week within-trial timeframe) analysis was also undertaken, which required no extrapolation of data.
Effectiveness data:
The main treatment effectiveness data were the probability of moving between forced expiratory volume states and the probability of lung flare-ups (exacerbations). Additional medical history data included the probability of transplantation and mean survival. The treatment effectiveness data came from one clinical trial. The probability of a lung transplant was from the CF Registry and the US Cystic Fibrosis Foundation. The mean survival data came from a multi-centre cohort study; several survival curves were fitted and the best fit selected (Weibull). Several assumptions on the independence of transitions between states and stability of measurements were necessary for modelling. These were fully reported.
Monetary benefit and utility valuations:
The utility values for the different forced expiratory volume strata, exacerbations, and minor and major exacerbations were obtained from the literature.
Measure of benefit:
The measure of benefit was quality-adjusted life-years (QALYs) gained. Total QALYs were calculated as the total time spent in each health state weighted by the respective utility for that health state, less any QALY losses resulting from exacerbations. The measure of benefit was discounted at a rate of 3.5%.
Cost data:
Costs included in the model were those of drug acquisition, nebulisers and management of exacerbations. The prices of the drugs were obtained from the manufacturer. The cost of hospitalisation treatment for came from the 2010-2011 NHS Reference Costs; asthma complications were used as a proxy. The cost of nebulisers was based on expert opinion. Costs were reported in UK £. The price year was 2011. Costs were discounted at a rate of 3.5%.
Analysis of uncertainty:
A probabilistic sensitivity analysis was conducted. Uncertainty in every parameter was accounted for simultaneously and characterised in the cost-effectiveness results. Additional one-way sensitivity analyses were conducted changing parameter estimates because of uncertainty in their validity or structural impact.