RATIONALE: Roflumilast is an oral, once-daily phosphodiesterase IV (PDE4) inhibitor under investigation for chronic obstructive pulmonary disease (COPD). This study investigated the cost effectiveness of roflumilast in patients with severe to very severe COPD from the perspective of the UK society and UK NHS. METHODS: The analysis was conducted alongside a 1-year, randomised, double-blind, placebo-controlled, multinational trial. The trial included 1514 COPD patients aged >or=40 years with a post-bronchodilator forced expiratory volume in 1 second (FEV1) % predicted <or=50% who were randomised to receive either roflumilast 500microg once daily (n = 761) or placebo (n = 753). Patients in both treatment groups were allowed to receive active treatment with a short-acting bronchodilator (salbutamol or anticholinergic) as needed. About 62% of patients in both groups were using an inhaled corticosteroid at trial entry. They were allowed to continue this on a stable dosage. Direct healthcare and productivity costs were calculated. Resource utilisation was recorded at every scheduled visit in health economics case report forms (HE-CRFs). Trial-wide resource use was combined with UK unit cost (2004 values). Roflumilast was assumed to cost euro1 per day. Incremental costs were related to the differences in the number of moderate to severe exacerbations and the net proportion of patients with an improvement of at least 4 units on the total score of the St George's Respiratory Questionnaire (SGRQ). An intention-to-treat analysis was conducted. Costs and health outcomes that were missing after withdrawal of patients from the trial were imputed using multiple imputation with the propensity score method. Various sensitivity analyses were conducted to test the robustness of the data. RESULTS: In the total group, annual COPD-related costs from a societal perspective were euro1637 in the roflumilast group and euro1401 in the placebo group. From an NHS perspective, this was euro1418 and euro1242, respectively. The rate of moderate to severe COPD exacerbations per patient was low, and no statistically significant difference existed between roflumilast (0.96) and placebo (1.06). The net proportion of patients with a relevant improvement on SGRQ total score was higher in the roflumilast group (0.19) than in the placebo group (0.14), but the difference was not statistically significant. From a societal perspective, COPD-related costs were euro2356 per exacerbation avoided and euro4712 per net additional patient with a relevant improvement on the SGRQ. The probability that roflumilast was cost effective exceeded 70% at a willingness to pay of euro5000 to avoid an exacerbation. In a subgroup of patients with very severe COPD (n = 223), the placebo group had a high exacerbation rate (1.7 per patient per year) whereas roflumilast recipients showed 35% fewer exacerbations (1.1 per patient per year). This resulted in roflumilast dominating placebo. In a subgroup of patients with high healthcare utilisation prior to the study (n = 549) roflumilast recipients showed 19% fewer exacerbations than those receiving placebo, which translated into an ICER of euro804 per exacerbation avoided. CONCLUSION:Roflumilast increased the overall treatment costs of COPD, although the increase was partly offset by reductions in other forms of healthcare use. Roflumilast has the potential to be cost saving in patients with very severe COPD, due to a statistically significant reduction of exacerbations.
RCT Entities:
RATIONALE: Roflumilast is an oral, once-daily phosphodiesterase IV (PDE4) inhibitor under investigation for chronic obstructive pulmonary disease (COPD). This study investigated the cost effectiveness of roflumilast in patients with severe to very severe COPD from the perspective of the UK society and UK NHS. METHODS: The analysis was conducted alongside a 1-year, randomised, double-blind, placebo-controlled, multinational trial. The trial included 1514 COPDpatients aged >or=40 years with a post-bronchodilator forced expiratory volume in 1 second (FEV1) % predicted <or=50% who were randomised to receive either roflumilast 500microg once daily (n = 761) or placebo (n = 753). Patients in both treatment groups were allowed to receive active treatment with a short-acting bronchodilator (salbutamol or anticholinergic) as needed. About 62% of patients in both groups were using an inhaled corticosteroid at trial entry. They were allowed to continue this on a stable dosage. Direct healthcare and productivity costs were calculated. Resource utilisation was recorded at every scheduled visit in health economics case report forms (HE-CRFs). Trial-wide resource use was combined with UK unit cost (2004 values). Roflumilast was assumed to cost euro1 per day. Incremental costs were related to the differences in the number of moderate to severe exacerbations and the net proportion of patients with an improvement of at least 4 units on the total score of the St George's Respiratory Questionnaire (SGRQ). An intention-to-treat analysis was conducted. Costs and health outcomes that were missing after withdrawal of patients from the trial were imputed using multiple imputation with the propensity score method. Various sensitivity analyses were conducted to test the robustness of the data. RESULTS: In the total group, annual COPD-related costs from a societal perspective were euro1637 in the roflumilast group and euro1401 in the placebo group. From an NHS perspective, this was euro1418 and euro1242, respectively. The rate of moderate to severe COPD exacerbations per patient was low, and no statistically significant difference existed between roflumilast (0.96) and placebo (1.06). The net proportion of patients with a relevant improvement on SGRQ total score was higher in the roflumilast group (0.19) than in the placebo group (0.14), but the difference was not statistically significant. From a societal perspective, COPD-related costs were euro2356 per exacerbation avoided and euro4712 per net additional patient with a relevant improvement on the SGRQ. The probability that roflumilast was cost effective exceeded 70% at a willingness to pay of euro5000 to avoid an exacerbation. In a subgroup of patients with very severe COPD (n = 223), the placebo group had a high exacerbation rate (1.7 per patient per year) whereas roflumilast recipients showed 35% fewer exacerbations (1.1 per patient per year). This resulted in roflumilast dominating placebo. In a subgroup of patients with high healthcare utilisation prior to the study (n = 549) roflumilast recipients showed 19% fewer exacerbations than those receiving placebo, which translated into an ICER of euro804 per exacerbation avoided. CONCLUSION: Roflumilast increased the overall treatment costs of COPD, although the increase was partly offset by reductions in other forms of healthcare use. Roflumilast has the potential to be cost saving in patients with very severe COPD, due to a statistically significant reduction of exacerbations.