|
A cost-effectiveness model comparing rivaroxaban and dabigatran etexilate with enoxaparin sodium as thromboprophylaxis after total hip and total knee replacement in the Irish healthcare setting |
McCullagh L, Tilson L, Walsh C, Barry M |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. CRD summary This study examined the cost-effectiveness of rivaroxaban, dabigatran etexilate, and enoxaparin sodium, for the primary prevention of venous thromboembolism in patients undergoing elective total hip or knee replacement. Rivaroxaban was less costly and more effective than dabigatran and enoxaparin, following either hip or knee replacement, but the probabilistic analysis showed uncertainty in rivaroxaban being more cost-effective than dabigatran etexilate, when both were compared with enoxaparin. Methodologically, the study was well conducted and the authors’ conclusions appear to be robust. Type of economic evaluation Cost-effectiveness analysis, cost-utility analysis Study objective This economic evaluation examined the cost-effectiveness of rivaroxaban and dabigatran etexilate versus enoxaparin sodium, for the primary prevention of venous thromboembolism, in adult patients undergoing elective total hip or knee replacement. Interventions The three preventive treatments were rivaroxaban, dabigatran etexilate, and enoxaparin sodium. Dabigatran was given at a dosage of 220mg daily, enoxaparin at 40mg daily, and rivaroxaban at 10mg daily. Rivaroxaban treatment lasted 31 to 39 days for patients undergoing total hip replacement and 10 to 14 days for total knee replacement; enoxaparin treatment lasted for 10 to 14 days for both hip and knee replacement; dabigatran treatment lasted for 28 to 35 days for hip and 6 to 10 days for knee replacement. Methods Analytical approach:The analysis was based on a static decision tree, with a time horizon of 180 days after surgery. Separate models were produced for total hip and total knee replacement. The authors stated that the analysis was conducted from the perspective of the Irish health service payer.
Effectiveness data:The clinical data appear to have been identified in two ways. The key inputs, which were the treatment efficacy (defined by the rate of venous thromboembolisms) and the bleeding rate, were both from selected pivotal randomised controlled trials (RCTs). For the comparison between dabigatran etexilate and enoxaparin sodium, two phase III head-to-head RCTs were used and for the comparison between rivaroxaban and enoxaparin sodium, another two phase III RCTs were used. The remaining probabilities for the model were identified by a systematic search of the English-language literature. The key methods of this review, such as the databases searched and the keywords used, were reported.
Monetary benefit and utility valuations:The utility values were identified by a literature review and from additional references that were in the articles retrieved. Various published sources were used and their key information was given. For example, the utilities for venous thromboembolism were from a sample of 106 US healthy females, derived using the standard gamble instrument, while the values for other health states were from the Irish population, derived using the time trade-off instrument.
Measure of benefit:The benefit measures were life-years (LYs) and quality-adjusted life-years (QALYs).
Cost data:The economic analysis included the costs of the orthopaedic procedure, the prevention of venous thromboembolisms, the diagnosis and treatment of venous thromboembolisms, and prophylaxis-related major bleeding. The following items were considered: drugs (acquisition and administration), hospital stay, blood tests, Doppler ultrasound, chest X-ray, electrocardiogram (ECG), computed tomography pulmonary angiogram, warfarin therapy, and emergency department visits. A breakdown of cost items was given. Most of the costs were derived from the Irish Health Service Executive. The drug costs were from the Irish Monthly Index of Medical Specialties. The data on resource consumption were from published sources. Diagnosis-related group data were mainly used for hospitalisations. All costs were in Euros (EUR) and the price year was 2008.
Analysis of uncertainty:Both a deterministic and a probabilistic sensitivity analysis were carried out. The former was run for all probabilities in the model, using published confidence intervals or arbitrary ranges of values (±50%). The latter was a second-order Monte Carlo simulation that used a cost-effectiveness threshold of EUR 45,000 per QALY. Conventional probability distributions were assigned to the model inputs. Results In the hip model, the expected costs were EUR 13,342.13 with rivaroxaban, EUR 13,359.51 with enoxaparin, and EUR 13,452.10 with dabigatran. The LYs were 0.939 with rivaroxaban, 0.929 with enoxaparin, and 0.934 with dabigatran. The QALYs were 0.999 with rivaroxaban, 0.992 with enoxaparin, and 0.996 with dabigatran. Rivaroxaban dominated both dabigatran and enoxaparin, as it was less expensive and more effective.
The incremental cost per LY gained with dabigatran versus enoxaparin was EUR 23,934 and the incremental cost per QALY gained was EUR 17,835. These findings were sensitive to variations in some of the probability inputs, but rivaroxaban remained the preferred strategy. The probability of being cost-effective at a threshold of EUR 45,000 per QALY was 39% for rivaroxaban, 32% for dabigatran, and 29% for enoxaparin.
In the knee model, rivaroxaban dominated both enoxaparin and dabigatran, and dabigatran dominated enoxaparin. The results were robust to changes in the model parameters. The probability of being cost-effective at a threshold of EUR 45,000 per QALY was 46% for rivaroxaban, 30% for dabigatran, and 24% for enoxaparin. Authors' conclusions The authors concluded that rivaroxaban was less costly and more effective than either dabigatran etexilate or enoxaparin, after total hip or knee replacement. The probabilistic analysis showed uncertainty around rivaroxaban being more cost-effective than dabigatran etexilate, when both were compared with enoxaparin. CRD commentary Interventions:The authors provided a clear justification for their selection of the comparators. Rivaroxaban and dabigatran were two oral anticoagulants that were newly licensed in Ireland for the primary prevention of venous thromboembolism in adult patients. Enoxaparin was selected as the comparator to mimic clinical practice and because it was used in the available phase III clinical trials.
Effectiveness/benefits:The methods used to identify the primary sources of data were reported and appear to have been valid. The use of RCTs for the efficacy data is generally valid, given the strengths of their design. The other clinical inputs and utility estimates were found by a literature review and its key methods and conduct were reported. This should have identified all the relevant data. The authors acknowledged that the efficacy of dabigatran versus rivaroxaban was an indirect comparison given the lack of published head-to-head trials. This could have introduced some uncertainty into the analysis. Both benefit measures (LYs and QALYs) were appropriate and will permit cross-disease comparisons. The sources for the utility values were reported and they used valid instruments, but with different populations.
Costs:The cost items and data sources reflected the perspective of the study. Detailed unit costs and resource quantities were reported, enhancing the transparency of the economic analysis. Appropriate sources were used for the resource quantities and they reflected the Irish context. In general, the economic analysis was satisfactorily carried out and the assumptions made were explicitly reported.
Analysis and results:The analytic approach of an incremental analysis was appropriate and allowed the identification of the most cost-effective strategy. The study results were clearly reported and a clear description of the decision tree and the key transition patterns was given. An appropriate time horizon was considered. The issue of uncertainty was satisfactorily investigated and the methods of the sensitivity analyses were clearly described and appear to have been valid.
Concluding remarks:Methodologically, the study was well conducted and the authors’ conclusions appear to be robust. Bibliographic details McCullagh L, Tilson L, Walsh C, Barry M. A cost-effectiveness model comparing rivaroxaban and dabigatran etexilate with enoxaparin sodium as thromboprophylaxis after total hip and total knee replacement in the Irish healthcare setting. PharmacoEconomics 2009; 27(10): 829-846 Indexing Status Subject indexing assigned by NLM MeSH Anticoagulants /economics /therapeutic use; Arthroplasty, Replacement, Hip /economics /methods; Arthroplasty, Replacement, Knee /economics /methods; Benzimidazoles /economics /therapeutic use; Cost-Benefit Analysis; Dabigatran; Decision Trees; Enoxaparin /economics /therapeutic use; Fibrinolytic Agents /economics /therapeutic use; Humans; Ireland; Models, Economic; Monte Carlo Method; Morpholines /economics /therapeutic use; Pyridines /economics /therapeutic use; Quality-Adjusted Life Years; Rivaroxaban; Thiophenes /economics /therapeutic use; Venous Thromboembolism /economics /prevention & control AccessionNumber 22009103787 Date bibliographic record published 04/08/2010 Date abstract record published 13/10/2010 |
|
|
|