Modelling the 5-year cost effectiveness of tiotropium, salmeterol and ipratropium for the treatment of chronic obstructive pulmonary disease in Spain

Abstract

Our objective was to assess the 5-year cost effectiveness of bronchodilator therapy with tiotropium, salmeterol or ipratropium for chronic obstructive pulmonary disease (COPD) from the perspective of the Spanish National Health System (NHS). A probabilistic Markov model was designed wherein patients moved between moderate, severe or very severe COPD and had the risk of exacerbation and death. Probabilities were derived from clinical trials. Spanish healthcare utilisation, costs and utilities were estimated for each COPD and exacerbation state. Outcomes were exacerbations, exacerbation-free months, quality-adjusted life years (QALYs), and cost(-effectiveness). The mean (SE) 5-year number of exacerbations was 3.50 (0.14) for tiotropium, 4.16 (0.40) for salmeterol and 4.71 (0.54) for ipratropium. The mean (SE) number of QALYs was 3.15 (0.08), 3.02 (0.15) and 3.00 (0.20), respectively. Mean (SE) 5-year costs were 6,424 euro (305 euro) for tiotropium, 5,869 euro (505 euro) for salmeterol, and 5,181 euro (682 euro) for ipratropium (2005 values). Ipratropium and tiotropium formed the cost-effectiveness frontier, with tiotropium being preferred when willingness to pay (WTP) exceeded 639 euro per exacerbation-free month and 8,157 euro per QALY. In Spain, tiotropium demonstrated the highest expected net benefit for ratios of the willingness to pay per QALY, well within accepted limits.

Fig. 1

Graphical presentation of the Markov model. Backward transitions, i.e. from very severe to severe COPD and from severe to moderate COPD, were allowed during the first year, but not thereafter. The circles on the top left of each COPD severity state indicate that the patients can stay in the same state

Fig. 2

Cost-effectiveness acceptability curves and frontier of the costs per exacerbation-free month and the cost per QALY. Ceiling ratio: threshold value for the cost-effectiveness ratio in euros. The curves in grey represent the probability that a treatment is cost effective for a given value of the ceiling ratio. In the case of skewed distributions of the ratio, the treatment with the highest probability of being cost effective is not always the treatment with the highest expected net benefit. The acceptability frontier (in black) shows which treatment is associated with the highest expected net benefit for each value of the ceiling ratio