Comparing the cost-effectiveness of a wide range of COPD interventions using a stochastic population model for COPD

Objectives: To develop a stochastic population model of disease progression in COPD that includes the impact of COPD exacerbations on health-related quality of life, costs, disease progression and mortality and allows assessment of a wide range of different interventions. Methods: The model is a multistate Markov model with time varying transition rates depending on age, sex, smoking status, COPD disease severity, and/or exacerbation type. The model simulates COPD incidence, prevalence, (severe) exacerbations, disease progression (annual decline in FEV % predicted) and mortality. Main outcome variables are quality-adjusted life years (QALYs) and COPD-related healthcare costs. The exacerbation-related input parameters were based on quantitative meta-analysis. All important model parameters are entered into the model as probability distributions. To illustrate potential use of the model, 10year costs and effects were projected for three different COPD interventions, one pharmacological (ICS+LABA), one on smoking cessation and one on pulmonary rehabilitation. Results: Compared to minimal intervention, the 10-year cost/QALY ratio of 3year implementation was €8,300 for ICS+LABA, €10,800 for the smoking cessation therapy and €17,200 for pulmonary rehabilitation. If the maximum willingness to pay would be €20,000/QALY, the probability that the cost/QALY was lower than that varied from 58% for pulmonary rehabilitation to 100% for ICS+LABA. Conclusions: The COPD model provides policy makers with comparable information on long-term costs and effects of interventions ranging from primary prevention to care for very severe COPD and includes uncertainty around the outcomes