Development of an Economic Model to Assess the Cost Effectiveness of Asthma Management Strategies

Background and objective: Asthma is a chronic-episodic disease characterised by acute, symptomatic episodes of varying severity. We developed a Markov model that can be used to estimate the cost effectiveness of alternative asthma treatments. Because of the costs they incur, asthma exacerbations (`attacks') requiring intervention by a healthcare professional were a central consideration in the development of the model. Methods: Treatment success was assessed as asthma control, a composite measure based on goals defined in world-wide asthma management guidelines and in terms of quality-adjusted life-years (QALYs). The data from which the transition probabilities were derived came from patients with asthma who received either salmeterol/fluticasone propionate combination (SFC) 50/100mug or fluticasone propionate (FP) 100mug, administered twice daily via an inhaler, in a 12-week, randomised, double-blind, clinical trial. Costs were estimated from resource profiles defined for each of the model states. A key aspect of the model was the use of probabilistic sensitivity analysis techniques to examine the uncertainty in the cost-effectiveness results. Distributions were fitted to transition probabilities and to cost input parameters and values were sampled at random from these distributions using a second order Monte Carlo simulation technique. This produced a distribution for incremental cost effectiveness that was employed to construct 95% uncertainty intervals and to construct cost effectiveness acceptability curves. Results: In this analysis, the model was run over a 12-week period using transition probabilities derived from the trial data. The results showed that treatment with SFC resulted in a higher proportion of successfully controlled weeks per patient than treatment with FP (66 vs 47%), and higher mean weekly direct asthma management costs (Lstg 15.77 vs Lstg 11.83; 2000 values). The average incremental cost per successfully controlled week with SFC was Lstg 20.83. Probabilistic sensitivity analysis showed that the 95% uncertainty intervals for the incremental cost-effectiveness ratio was -Lstg 64.94 to Lstg 112.66. In approximately 25% of cases, SFC was dominant (more effective and less costly), but in the remaining cases, it was both more effective and more costly. It was shown that if decision makers are willing to pay approximately Lstg 45 for an additional successfully controlled week, SFC will be the more cost-effective strategy in this patient population for 80% of the time. Conclusions: This is one of the first decision-analytic models of asthma to incorporate probabilistic sensitivity analysis techniques to explore uncertainty. The model's flexible yet standardised framework permits the cost effectiveness of alternative asthma management strategies in different healthcare settings to be established.