|
Low-molecular-weight heparins compared with unfractionated heparin for treatment of acute deep venous thrombosis: a cost-effectiveness analysis |
Gould M K, Dembitzer A D, Sanders G D, Garber A 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 Low molecular weight heparin preparations (LMWH) and unfractionated heparin (UH) used in the treatment of established deep vein thrombosis.
Economic study type Cost-effectiveness analysis and cost-utility analysis.
Study population Hypothetical cohorts of 10,000 60 year old 75 kg men with acute proximal, lower extremity deep venous thrombosis.
Setting Hospital and community. The economic study was carried out in the USA.
Dates to which data relate Effectiveness and resource data were taken from literature published between 1991 and 1997. Additional resource data were obtained from 1995 and 1996 publications. 1997 prices were used.
Source of effectiveness data Effectiveness data were derived from a meta-analysis of previously published studies
Modelling A decision tree analysis model was used to combine information on clinical outcomes with cost estimations in order to determine costs per life year gained and costs per quality-adjusted life year gained. Early complications were assumed to occur during the first 6 months following treatment and late complications (post 6 months) were also taken into consideration. The duration of the model was the patient lifetime.
Outcomes assessed in the review The review assessed the probabilities of clinical complications such as major and minor bleeding, thrombocytopenia, recurrent deep vein thrombosis, pulmonary embolisms, mild and severe post-phlebitic syndrome and mortality rates.
Study designs and other criteria for inclusion in the review Randomised controlled trials comparing LMWH with UH in the treatment of deep vein thrombosis. Studies using patients with calf vein thrombosis were excluded from meta-analysis of estimates for recurrent deep venous thrombosis, pulmonary embolisms and mortality. Mild and severe post-phlebitic syndrome data were not taken from a meta-analysis but from a separate published long term observational study.
Sources searched to identify primary studies The Medline database was searched between January 1985 and September 1997. In addition references of papers identified through Medline were also checked, conference abstracts were examined and investigators in pharmaceutical companies were contacted.
Criteria used to ensure the validity of primary studies Study quality criteria developed by Schultz et al (1995) were adopted. These criteria assessed whether treatment allocation had been generated through a random numbers table or computer program, if the allocation sequence had been properly concealed, whether studies were double blinded and how many patients were lost to follow up.
Methods used to judge relevance and validity, and for extracting data Two independent reviewers were used to identify relevant data. Individual data from studies were extracted by the reviewers.
Number of primary studies included 11 randomised controlled trials were included although 5 studies were excluded from meta-analysis of estimates for recurrent deep venous thrombosis, pulmonary embolisms and mortality.
Methods of combining primary studies Meta-analysis using fixed effects and random effects models to estimate summary treatment effects.
Investigation of differences between primary studies The Mantel-Haenszel test was used to identify heterogeneity between studies. No statistically significant differences were identified.
Results of the review The probabilities of early complications for the UH and LMWH patients respectively were estimated to be:
minor bleeding 3.8% (UH) and 3.9% (MWH);
major bleeding 1.90% and 1.24%;
thrombocytopenia 1.80% and 1.19%;
recurrent deep venous thrombosis 4.4% and 3.04%;
pulmonary embolism 2.0% and 1.91%;
death 6.70% and 5.10%.
Late complication probabilities extracted from a single paper (using the same rates for both LMWH and UH groups respectively) were estimated to be:
mild post phlebitic syndrome (after 1 year) 14.70%;
mild post phlebitic syndrome (after 5 years) 18.70%;
severe post phlebitic syndrome (after 1 year) 2.60%
and severe post phlebitic syndrome (after 5 years) 9.30%.
Measure of benefits used in the economic analysis The benefit measures were life years gained and quality-adjusted life years (QALYs) gained. A decision tree analysis model was used to synthesise life years/QALYs gained using effectiveness data from published sources. QALY weights were taken from two previously published studies. The first, a 1993 study, which used age and sex weighted time trade-offs for different health states for a community based population sample, was used to determine utility for a health state with no complications following deep venous thrombosis. The second, 1997, study obtained utilities from healthy individuals for post-phlebitic states using a computerised interview and standard gamble technique. Utility lost due to early complications was calculated as days lost from quality-adjusted life expectancy due to hospitalisation using published US data from 1997 on length of stay.
Direct costs The costs of initial inpatient treatment, including 6 day hospitalisation (excluding pharmacy and supplies) were estimated using 1995 US Health Care Financing Administration Medicare DRG reimbursement rates for deep venous thrombosis. Physicians' service costs were extracted from 1996 Medicare reimbursement rates quoted in 'Physicians Current Procedural Terminology'. Average wholesale prices for LMWH and UH were taken from the 1997 Drug Red Book. Supply and ancillary resources were estimated using a proprietary cost accounting system - Transition I, Transition Systems Inc, Boston, Massachusetts. Outpatient costs including physician visits, telephone calls to physicians and formal homecare were also estimated, based on estimates reported in a 1996 US study and also using Medicare reimbursement rates for 1995 and 1996 and median nurse earnings for 1994. Informal home care provided by family members was assumed to be four hours per day at the 1997 seasonally adjusted earnings rate for production workers. The costs of treating early complications were based on average Medicare DRG reimbursement rates for 1995 and 1996. Lifetime costs related to late complications were taken from a 1995 Swedish observational study. Future health care costs for patients were calculated over a patient's lifetime using a 1997 Bureau of Labour published estimate of US annual age related health care expenditures (less expenses for late complications). Costs and benefits were discounted at a rate of 3% per annum and 1997 prices were used. Where necessary costs were converted to 1997 US dollars using the gross domestic product deflator. A societal perspective for costs was used in the analysis.
Sensitivity analysis One-way and multivariate sensitivity analyses were conducted varying a wide range of cost and probability parameters between 95% confidence intervals, where reported, or alternatively by 25% from the base case scenario.
Estimated benefits used in the economic analysis Expected life years in the LMWH cohort were 0.022 greater than in the UH cohort (9.429 years compared with 9.406 years) and similarly QALYs in the two groups were 8.00 and 7.98 respectively, (0.020 difference).
Cost results Average lifetime total costs per patient, assuming all treatment was provided on an inpatient basis in the LMWH and UH groups, were $26,516 and $26,361 respectively. Incremental costs for patients in the LMWH group were $155. In the scenario where 30% of patients were assumed to be treated on an outpatient basis and a further 25% discharged after 3 days, rather than 6, the average total costs per patient were $25,599 and $26,361 respectively, with costs in the LMWH being $762 less than for those treated in the UH group. These costs include those for treating adverse events and complications as well as additional lifetime healthcare costs.
Synthesis of costs and benefits In the first scenario where all treatment is on an inpatient basis, the incremental cost per life year gained using LMWH rather than UH was estimated to be $6,910 and the incremental cost per quality-adjusted life years gained was $7,820. In the second scenario with some treatment on an outpatient basis LMWH treatment was dominant compared with the UH option, having both lower costs and improved clinical outcomes. Costs and benefits were discounted at a rate of 3% per annum. The conclusions of the model were robust in sensitivity analysis with the exception of uncertainty on the effectiveness of LMWH in preventing late complications. In the base case scenario, if the probability of late complications decreased by 25% in patients receiving UH, the incremental cost per QALY gained increased to approximately $75,000 and a decrease in 25% in the probability of late term complications associated with LMWH patients would lead to the intervention being dominant over UH treatment.
Authors' conclusions LMWH is highly cost effective in the treatment of acute deep venous thrombosis in comparison to UH treatment in the inpatient setting and can lead to cost savings compared with UH treatment if some patients are treated on an outpatient basis.
CRD COMMENTARY - Selection of comparators A justification was provided for the comparator used, as UH is a well accepted treatment for acute deep venous thrombosis.
Validity of estimate of measure of benefit The estimate of benefits were based on a well conducted systematic literature review and meta-analysis. The authors noted, however, that their assumption that clinical benefits were independent of treatment setting was a limitation. Insufficient information is available to confirm this hypothesis, and inpatient treatment may be more effective than that in an outpatient setting. It is also noted that it may not be the case that all five interventions are equally safe and efficacious and any differences in effectiveness and safety between the five LMWHs used in the meta-analysis have yet to be evaluated.
Validity of estimate of costs Sufficient details were provided of sources and estimates of costs used in the analysis, including costs to family caregivers. In any future analysis the costs to patients themselves from a loss of earnings and loss of their leisure time might also be included.
Other issues The results of the analysis may not be generalisable outside the United States.
Implications of the study Further well conducted economic and clinical evaluations comparing the different LMWHs for different population groups are required.
Source of funding Supported by grant HS00028-11 from the Agency for Health Care Policy and Research, Rockville, Maryland.
Bibliographic details Gould M K, Dembitzer A D, Sanders G D, Garber A M. Low-molecular-weight heparins compared with unfractionated heparin for treatment of acute deep venous thrombosis: a cost-effectiveness analysis. Annals of Internal Medicine 1999; 130(10): 789-799 Other publications of related interest Comment in: Annals of Internal Medicine 1999;130(10):857-8.
Gould M K, Dembitzer A D, Doyle R L, Hastie T J, Garber A M. Low molecular weight heparins compared with unfractioned heparin for acute deep venous thrombosis. A meta-analysis of randomised controlled trials. Annals of Internal Medicine 1999;130(10):800-809.
Indexing Status Subject indexing assigned by NLM MeSH Anticoagulants /economics /therapeutic use; Cost-Benefit Analysis; Decision Support Techniques; Heparin /economics /therapeutic use; Heparin, Low-Molecular-Weight /economics /therapeutic use; Hospital Costs; Humans; Life Expectancy; Male; Middle Aged; Multivariate Analysis; Pulmonary Embolism /etiology; Quality-Adjusted Life Years; Randomized Controlled Trials as Topic; Recurrence; Sensitivity and Specificity; Venous Thrombosis /complications /drug therapy /economics AccessionNumber 21999008145 Date bibliographic record published 29/02/2000 Date abstract record published 29/02/2000 |
|
|
|