Abstract
Abstract 1164
The three new oral anticoagulants (NOAC) dabigatran 110mg bid and 150mg bid, investigated in the RE-LY trial, rivaroxaban 20mg od of the ROCKET trial, and apixaban 5mg bid of the ARISTOTLE trial showed equivalent or superior efficacy and safety compared to warfarin in these patients. We performed a cost-effectiveness analysis for these NOACs in Germany and compared the quality of life (QALY), incremental cost effectiveness ratios (ICER), and total costs across those countries form where these data are published.
The base case population was a hypothetical cohort of patients 65 years or older with AF who were at increased risk for stroke (CHADS2-score >1) and had no contraindications to anticoagulation. The time horizon was based on the life expectancy of the German population. The QALYs, health insurance costs, and ICER for the NOACs compared with warfarin were calculated for each study. The sensitivity analysis was performed for different base case prices. The Markov decision model was adopted using the TreeAge Pro 2011 program. The data of the outcomes of ischemic stroke and cerebral embolism, major and intracerebral haemorrhage, myocardial infarction, and mortality were taken from the 3 studies comparing the NOAC with INR-adjusted warfarin. Prices for clinical events and for outpatient care were taken from the institute for payment regulations in German hospitals (InEK).
The base-case analysis of a 65 years old person with a >2 CHADS2 score using the data from the RE-LY study resulted in 11.41 QALYs for warfarin, 11.53 QALYs for dabigatran 110mg bid, 11.66 QALYs for dabigatran 150 mg bid. ICERs per QALY were 49640€ for dabigatran 110 mg bid and 49590€ for dabigatran 150 mg bid versus warfarin. The same base-case analysis using the data from the ROCKET AF study resulted in 10.79 QALYs for warfarin versus 11.05 QALYs for rivaroxaban. ICERs per QALY were 48980€ for rivaroxaban versus warfarin. The base-case analysis using the data from the ARISTOTLE study resulted and 11.04 QALYs for warfarin versus 11.38 QALYs for apixaban. ICERs per QALY were 49720€ for apixaban versus warfarin. According the Markov Model the daily value based daily prices were 1.25€ for dabigatran 110 mg bid, 2.50€ for dabigatran 150 mg bid, 2.60€ for rivaroxaban, and 3.10€ for apixaban in Germany. The model was highly sensitive to the daily costs for the NOACs but relatively insensitive to other model inputs. Calculating the range of NOAC prices from 0.2 to 10 Euro versus the ICERs dabigatran 110 mg bid produced the highest increase of ICERs over this range of daily costs (Tukey-Kramer test, p<0.05) Comparing these data across countries using the published data shows that the willingness to pay per QALY as well as differences in treatment costs between substantially influences the daily prices of the NOACs.
The data demonstrate the necessity to analyse the cost-effectiveness separately for every study due to differences in the INR-adjusted warfarin treated control group. The better the outcome during treatment with warfarin the lower is the benefit of the NOAC. The tendency of the cost-effectiveness calculated by the Markov model is comparable across countries.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
