|
Very low-dose warfarin prophylaxis to prevent thromboembolism in women with metastatic breast cancer receiving chemotherapy: an economic evaluation |
Rajan R, Gafni A, Levine M, Hirsh J, Gent 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. Health technology Very low-dose warfarin prophylaxis.
Economic study type Cost-effectiveness analysis.
Study population Women with metastatic stage 4 breast cancer receiving first or second-line chemotherapy for 4 weeks or less. The population had a mean age of 57.1 years in the intervention group and 56.1 years in the placebo group.
Setting Tertiary care hospital in Hamilton, Canada.
Dates to which data relate The effectiveness and resource data from the trial were taken from 1989 to 1992.1992 prices were used.
Source of effectiveness data Link between effectiveness and cost data Costs were collected prospectively on the same patient sample as that used in the trial, apart from data concerning the outpatient management of patients which was modelled on a typical patient with a thromboembolic episode.
Study sample An a priori sample size calculation was performed. 152 patients were randomised to the very low-dose warfarin group (VLDW), and 159 patients to the placebo group. Patient were excluded if their Eastern Cooperative Oncology Group performance status was greater than or equal to 3, if they had an underlying bleeding disorder or active peptic ulcer disease, or overt brain metastases.
Study design A multi-centre, stratified, double-blind, randomised, controlled trial.11 centres were involved: 8 in Canada, 2 in the US and 1 in Italy. Patients were randomised to either VLDW or placebo. Stratification was by centre and the presence or absence of a central venous catheter. Loss to follow-up was 17%.
Analysis of effectiveness Analysis was based on intention to treat. Primary health outcomes were: thromboembolic events, minor and major bleeding complications and mean time at risk for thrombosis. At baseline the placebo and intervention groups were comparable in age and other characteristics.
Effectiveness results The mean time at risk from thrombosis was 199 days for warfarin-treated patients and 188 days for placebo recipients (p = 0.95). There were 7 thromboembolic events: 6 deep-vein thrombosis (DVT) and 1 pulmonary embolism (PE) in the placebo group, and 1 PE in the warfarin group, representing a relative risk reduction of 85% (p = 0.031). There were 8 (53%) bleeding events in warfarin-treated patients compared with 5 (3.1%) in placebo recipients (p = 0.4; 95% CI: -6.2%, 2.8%).87 (57%) warfarin treated patients and 99 (63%) placebo-group patients died (p = 0.55). There was no statistically significant difference in mean survival durations (442 days vs 456 days in the VLDW and placebo groups, respectively).
Clinical conclusions VLDW was a safe, effective method for prevention of thromboembolism in patients with metastatic breast cancer who are receiving chemotherapy.
Measure of benefits used in the economic analysis Thromboembolic events, bleeding episodes and out-patient anticoagulant dosage (follow-up care).
Direct costs The analysis was performed primarily from the viewpoint of the Ministry of Health. The patient viewpoint was included only so far as including the direct patient cost of the warfarin tablets. Health care costs common to both intervention and control groups were excluded. The cost of VLDW included: cost of medication, prescription fees, prothrombin times. The cost of complications and follow-up care were added (ie DVT, PE, and major and minor bleeding episodes). The unit prices of the resources consumed through outpatient and inpatient care were obtained using a simultaneously fully allocated costing model in place at McMaster University Medical Centre in Hamilton, Canada. Doses of VLDW and frequency of PTs were assumed to follow the pattern in the trial (i.e. estimation of quantities were based on actual data). As data regarding the outpatient management of patients who experienced a DVT or PEwas not collected during the trial, the oral anticoagulant management of a typical patient with a thromboembolic episode was modelled. Discounting was not performed, as all costs and benefits were expected to manifest within one year.
Currency Canadian dollars (Can$). Conversion in 1992 was: Can$ 1 = US$ 0.83.
Sensitivity analysis One-way threshold sensitivity analysis. Parameters were: cost of VLDW, the incidence and cost of managing venous thromboembolism (VTE) episodes and the frequency of bleeding episodes.
Estimated benefits used in the economic analysis 6 patients with DVT, and 1 patient with PE in the placebo group vs 1 patient with PE in the warfarin-treated group.2 patients had major bleeds and 3 had minor bleeds in the placebo group, while 1 patient had major bleed and 7 had minor bleeds in the warfarin-treated group.
Cost results The cost of providing VLDW consisted of prescription fees, cost of medication and prothrombin times and was $21,854 per 100 patients. The additional cost of complications, i.e. thromboembolic events and bleeding as well as follow-up care brought the total cost in the warfarin group to $33,170 per 100 patients. The cost of complication and follow-up care in the placebo group was $35,613 per 100 patients. The cost savings resulting from treating 100 patients with warfarin were therefore $2,443 due to the reduction in complications, particularly deep-vein thrombosis.
Synthesis of costs and benefits VLDW was the dominant strategy, i.e. was cheaper and produced better health outcomes. Results were robust to sensitivity analysis.
Authors' conclusions Very low-dose warfarin prophylaxis was cost-effective in the prevention of thromboembolism in women with metastatic breast cancer receiving chemotherapy: particularly in the Canadian healthcare setting.
CRD Commentary The trial methodology was rigorous, however, since this was a multi-centre study, it would have been useful to check whether or not costs differed significantly between sites (the authors stated the assumption that differences in unit prices between centres did not influence the pattern of resource use). The costing methodology was also rigorous. A sensitivity analysis was performed to allow for uncertainty in the measurement of certain costs and consequences. Generalisability of results to other settings (for example, the UK) is not clear, especially since costs and quantities were not reported separately.
Bibliographic details Rajan R, Gafni A, Levine M, Hirsh J, Gent M. Very low-dose warfarin prophylaxis to prevent thromboembolism in women with metastatic breast cancer receiving chemotherapy: an economic evaluation. Journal of Clinical Oncology 1995; 13(1): 42-46 Indexing Status Subject indexing assigned by NLM MeSH Breast Neoplasms /complications /drug therapy; Canada; Female; Health Care Costs; Humans; Thromboembolism /prevention & Warfarin /administration & control; dosage /economics AccessionNumber 21995000142 Date bibliographic record published 30/11/1997 Date abstract record published 30/11/1997 |
|
|
|