|
Cost-effectiveness of the bird's nest filter for preventing pulmonary embolism among patients with malignant brain tumors and deep venous thrombosis of the lower extremities |
Chau Q, Cantor S B, Caramel E, Hicks M, Kurtin D, Grover T, Elting L S |
|
|
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 The use of intravenous bird's nest filter (BNF) with anticoagulation (heparin) for the prevention of pulmonary embolism (PE) in patients with malignant brain tumour, who were at high risk of developing deep venous thrombosis (DVT) in the lower extremities.
Study population The study referred to patients with malignant brain tumour who were at risk of DVT in the lower extremities. Patients with upper extremity DVT were excluded.
Setting The setting was secondary care. The economic study was carried out in Texas, USA.
Dates to which data relate The resource use and effectiveness data were collected from January 1994 to September 1997. The price year was 1999.
Source of effectiveness data The effectiveness data were derived from a single study and one authors' assumption.
Link between effectiveness and cost data The costing was carried out retrospectively on the same patients who were included in the effectiveness study.
Study sample Power calculations, if performed, were not reported. A sample of 24 consecutive patients with DVT and brain tumours, who were hospitalised at the authors' institution, was identified between January 1994 and September 1997. The mean age was 50 years (range: 27 - 54). Fourteen patients had a BNF inserted and formed the BNF group. The remaining 10 patients received only anticoagulation and represented the no BNF group.
Study design This was a retrospective cohort study, based on a review of medical charts, which was carried out in a single hospital. A radiologist and a general internist retrospectively confirmed the diagnosis of PE from diagnostic data. The length of follow-up was not reported. The authors stated that only 3 patients were alive at the final follow-up examination, 21 patients having died from a brain tumour.
Analysis of effectiveness It appears that all the patients included in the initial study sample were considered in the effectiveness study. The health outcomes used in the analysis were:
the number of patients achieving adequate anticoagulation,
the average and total number of PE-free months,
the rate of PE per 100 person-months, and
survival.
The baseline comparability of the two groups was not discussed.
Effectiveness results Three of the 14 patients in the BNF group achieved adequate anticoagulation versus 5 of the 10 in the no BNF group, (p=0.15).
The average number of PE-free months was 11.6 in the BNF group and 9.4 in the no BNF group, (p=0.24).
The total number of PE-free months was 175.09 (rate of PE: 0.57 per 100 person-months) in the BNF group and 94.76 (rate of PE: 1.05 per 100 person months) in the no BNF group, (p=0.24).
The mean survival was 12.1 months in the BNF group and 9.5 months in the no BNF group, (p=0.13).
Clinical conclusions The rates derived from the single study were converted into monthly probability values and used as inputs in the decision model.
Modelling A decision model based on a Markov process was used to represent all outcomes that might have occurred from the two study strategies. The time horizon of the model was not reported. The length of the model cycles was one month. Patients developing PE in the no BNF group could die from PE, or survive and move to the BNF group. Those who did not develop PE, or die from other causes, remained in the model because they were still at risk of PE. A similar pattern was observed for patients in the BNF group.
Methods used to derive estimates of effectiveness The authors made an assumption concerning the utility value of the health state associated with PE. This value, which was estimated from the literature, was assumed to be comparable to that of other thromboembolic health states.
Estimates of effectiveness and key assumptions The utility value of the health state associated with PE was 0.7.
Measure of benefits used in the economic analysis The summary benefit measure used in the economic analysis was the quality-adjusted life-years (QALYs). The QALYs were obtained by combining survival and quality of life data estimated from the observational study and from the authors' assumption. No discounting was performed due to the short life expectancy of the patients considered in the study.
Direct costs It was unclear whether discounting would have been relevant, as the length of follow-up was not reported. However, the authors stated that an annual rate of 3% was applied. The unit costs and the quantities of resources used were not presented separately. The cost categories considered were BNF procedure, anticoagulation and the treatment of PE. The BNF procedural costs covered professional services, supplies, recovery and room, laboratory evaluations and medications. The costs of treating PE covered the hospital room, diagnostic procedures, medications, laboratory evaluations and professional services. The perspective of the provider was adopted. The unit costs were estimated from the hospital accounting system. The costs of the BNF procedure and PE treatment were reported. Resource consumption was determined from the sample of patients who were considered in the effectiveness study. The price year was 1999.
Statistical analysis of costs No statistical tests of the costs were carried out.
Indirect Costs The indirect costs were not included.
Sensitivity analysis One-way sensitivity analyses were conducted to address the issue of variability in the data. The factors varied were the rate of PE, the cost of PE and the 5-year mortality rate. The 5-year mortality rate was estimated from three sources. More specifically, a hypothetical cohort of lung cancer patients (mortality rate of 85%), a hypothetical cohort of breast cancer patients (mortality rate of 15%), and a sample of 68 breast cancer patients at the authors' institution (mortality rate of 57%).
Estimated benefits used in the economic analysis The estimated QALYs were 2.33 with no BNF and 2.34 with BNF. The difference in QALYs was 0.01.
Cost results The estimated discounted costs were $4,730 with no BNF and $7,502 with BNF. The difference in costs was $2,772.
Synthesis of costs and benefits The incremental cost per QALY gained with BNF over no BNF was $198,852. This value became $185,079 when using mortality rates from the hypothetical cohort of patients with lung cancer, and $21,149 when using mortality rates from the institutional cohort of patients with breast cancer. However, when mortality data from the hypothetical cohort of patients with breast cancer were used, BNF dominated no BNF. If the rate of PE exceeded 1.51 per 100 person-months, then the incremental cost per QALY was below the threshold of $50,000. Finally, the cost of treating PE would have to exceed $34,531 before the filter would be considered a cost-effectiveness strategy.
Authors' conclusions The bird's nest filter (BNF) was not cost-effective among patients with brain tumour, although it was effective in reducing the rate of PE. For patients with longer life expectancy, the BNF strategy was cost-effective relative to no BNF.
CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparator was clear. Anticoagulation alone was selected so as to reflect the standard treatment pattern at the authors' institution. You should decide whether it is a valid comparator in your own setting.
Validity of estimate of measure of effectiveness The measure of effectiveness was derived from a retrospective cohort study. However, the use of a prospective randomised trial would have been more appropriate for the study question. It was not stated whether the two groups were comparable at baseline. In addition, it was unclear whether the study sample was representative of the study population. Data were identified from a retrospective review of medical charts. The sample size was small and power calculations were not carried out. Due to the design of the study, confounding factors and assessment bias might have affected the conclusions. The authors noted that selection bias was likely to have occurred.
Validity of estimate of measure of benefit QALYs were used as the summary benefit measure. These appear to have been appropriate to reflect the impact of the disease and the treatment on both quantity and quality of life. Discounting was not applied but it was unclear whether it would have been relevant. The utility weights were derived from an authors' assumption, which was not tested in the sensitivity analysis.
Validity of estimate of costs The authors explicitly stated the perspective of the study. It appears that all the relevant categories of costs have been included. However, the unit costs were not presented separately from the quantities of resources used. The price year was reported. Statistical tests were not performed, but cost estimates were varied in the sensitivity analysis. Discounting was carried out but it was unclear whether it was relevant.
Other issues Some caution is required when extrapolating the study results to other populations of cancer patients, as the study referred to brain tumour patients who usually have a very short survival. The authors did not compare their results with those from other studies.
Implications of the study BNF could be cost-effective for cancer patients with an expected long survival.
Bibliographic details Chau Q, Cantor S B, Caramel E, Hicks M, Kurtin D, Grover T, Elting L S. Cost-effectiveness of the bird's nest filter for preventing pulmonary embolism among patients with malignant brain tumors and deep venous thrombosis of the lower extremities. Supportive Care in Cancer 2003; 11(12): 795-799 Indexing Status Subject indexing assigned by NLM MeSH Anticoagulants /economics /therapeutic use; Benchmarking; Brain Neoplasms /complications; Cost-Benefit Analysis; Humans; Lower Extremity /pathology; Markov Chains; Pulmonary Embolism /etiology /prevention & Quality-Adjusted Life Years; Treatment Outcome; Vena Cava Filters /economics; Venous Thrombosis /complications; control AccessionNumber 22004000000 Date bibliographic record published 31/08/2004 Date abstract record published 31/08/2004 |
|
|
|