A Cost-Effectiveness And Budget Impact Analysis Of Fidaxomicin For Treating Clostridium Difficile Patients In Germany

Clostridium difficile infection (CDI) is a debilitating illness. In two phase III trials fidaxomicin produced comparable initial cure rates, significantly lower recurrence rates (p<0.005) and significantly higher sustained cure rates (p=0.001) versus vancomycin. This cost-effectiveness and budget impact model analysed the costs and outcomes of vancomycin and fidaxomicin for treatment of CDI in Germany. The model was a Markov cohort simulation, with 10-day cycle length. The analysis timescale was either 40 days (hospital perspective) or 1 year (payer perspective). Clinical inputs included : 30-day CDI-attributable mortality, probability of clinical cure and 30-day probability of recurrence after end of treatment. The first-line treatment is either Dificlir (€1300) or vancomycin (€61). Second line treatment is user defined, and in the base case is vancomycin. Third line treatment is a rescue treatment (€1500), assumed to have 100% cure rate. The model is populated with cost data for Germany. Drug costs are based on list prices, and the costs of hospitalisation are DRG tariff rates; Cost per day of CDI treated on a general ward (€348). A deterministic sensitivity analysis was carried out to test the robustness of the model outcomes. The cost-effectiveness of six CDI patient subgroups was also analysed based on the two fidaxomicin clinical trials’ results. The outcomes of the model for All Patient group: incremental cost per QALY gained €40,807, cost per recurrence avoided €2,068, and cost per bed-day saved €110. For the All Patient group fidaxomicin reduced the number of recurrences by 49%. Fewer recurrences led to a reduction in attributable deaths, a gain in life years; an improvement in quality of life; and a reduction in the number of bed-days. First-line fidaxomicin is likely to be a cost-effective treatment option, compared to vancomycin at a willingness to pay threshold of €50,000 per QALY gained.