|Cost-utility analysis of melphalan plus prednisone with or without interferon-alpha2b in newly diagnosed multiple myeloma: results from a randomised controlled trial
|Nord E, Wisloff F, Hjorth M, Westin J
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.
Melphalan plus prednisone with interferon-Alpha2b (MP-IFN) in patients with newly-diagnosed multiple myeloma.
Patients with newly diagnosed, symptomatic multiple myeloma. Patients were included if they were within the age range 55-80 years (inclusive), were not being considered for intensive chemotherapy protocols, did not have any psychiatric disease, active malignant disease, severe heart disease, or other severe coincident illness and were not terminally ill.
Hospital. The economic study was carried out in Norway, Denmark and Sweden (Nordic Myeloma Study Group).
Dates to which data relate
The effectiveness and resource use data corresponded to patients enrolled from 1 June 1990 to 3 November 1992. The price year was1995.
Source of effectiveness data
Effectiveness data were derived from a single study.
Link between effectiveness and cost data
The costing was prospectively undertaken on the same patient sampleas that used in the effectiveness analysis.
Power calculations were used in determining the sample size, the authors estimating that 580 patients would be required for the trial, assuming a survival analysis with a statistical power of 80% and a significance level of 5%. Ultimately, 583 patients were enrolled in the study: 297 with a median age of 68 years being randomly allocated to the intervention group, and 286, with a median age of 67 years, to the comparator group.
The study was a randomised, multicentre, phase III, controlled trial, carried out in 107 hospitals. The duration of follow-up was from 24 to 53 months. No patients were lost to follow-up.
Analysis of effectiveness
The analysis was carried out under the intention to treat principle. The primary health outcomes were response duration, plateau-phase duration, and median survival time. The patients were seen at maximum 6-week intervals for clinical and laboratory evaluations. Results were further adjusted by difference in prognostic factors between groups in terms of age, World Health Organization performance status, and serum B2 microglobulin level.
The response duration was 15 months in the MP group versus 21 months in the MP-IFN group, (p=0.005). The plateau-phase duration was 14 and 19 months, respectively, (p=0.03). The median survival time was 29 and 32.5 months, respectively, (NS). The risk ratio of death in the MP group against that of the intervention was 1.02 (95% CI: 0.83- 1.23). The corresponding figure after adjusting for differences in prognostic factors was 1.12, (95% CI: 0.89 - 1.40). The corresponding unadjusted and adjusted differences in median survival time were 3 and 7 months.
The study data showed that administering MP-IFN to all patients with newly diagnosed, symptomatic multiple myeloma yields a gain in terms of quality-adjusted life years (QALYs). This is true even if the mean survival gain is assumed to be as low as 1 month.
Measure of benefits used in the economic analysis
Quality-adjusted life years (QALYs) gained were used as the measure of benefit in the economic analysis. This outcome was calculated from the net difference between quality of life gains from th eadditional 3-month mean survival assumed and the losses arising from the reduced quality of life (relative to the comparator) for the assumed period of 12 months and by using the observed differences at 6 months. A questionnaire was used to obtain the estimates of the measure of benefit. The core questionnaire (QLQ-C30), developed for the European Organization for Research and Treatment of Cancer(EORTC) Study on QoL, was mailed directly to the patients after 1,6, 12, 24, 36 and 48 months. In order to transform the data on health profiles to obtain single estimates of utility values as general indices of health states (within the 0 to 1 range), a two-step backtracing and mapping procedure was carried out to convert numerical scores from the QLQ C-30 into the descriptive systems of other instruments. These were the EuroQol (using time trade-off values), the Index of Health-Related Quality of Life (IHQL) (using the standard gamble method) and a 15-dimensional measure of health-related quality of life (15-D) (using rating-scale values).
The costs were discounted. The resource quantities were reported separately from the prices. The costs measured were operating costs and costs of complications. The boundary adopted was hospital. The estimation of resource use was based on actual data collected from 1 June 1990 to November 1994. The source of drug costs was the Felleskatalogen of Norway - market prices. 1995 prices were used. The costs associated with physician consultations were excluded because they were thought to be common (same number in both groups). Marginal costs were measured. Additional costs due to increased survival were not included in the cost analysis.
The costs were discounted. The resource quantities were reported separately from the prices. The costs measured were those associated with productivity losses. The boundary adopted was the patient. The estimation of resource use was based on actual data collected from 1 June 1990 to November 1994. Productivity compensation was estimated based on a study conducted in the Dutch economy in 1988. The average daily salary in Norway was used in the calculation of indirect cost. 1995 price data were used.
Norwegian kroner (NKr), converted to US dollars. The exchange rate used was $1 = NKr6.40 (1995 prices).
The effects explored were those arising from the variation of the following parameters: hospital days, days lost from production, hospital day cost, sick leave, additional survival time achieved with the intervention, value of the loss of quality of life (difference between the quality of life (QoL) with comparator and the reduced QoL with the intervention). A one-way simple sensitivity analysis was performed.
Estimated benefits used in the economic analysis
The number of QALYs gained with the intervention was 0.125 per patient.
The costs were discounted at a 5% rate. The total direct cost per patient in the intervention was NKr63,779 ($10,000). The total direct cost per patient in the control group was NKr138,799 ($21,700). Accounting for NKr12,600 ($2,000) of indirect cost per patient, the total incremental cost from the societal perspective amounted to NKr87,600 ($13,700).
Synthesis of costs and benefits
The costs and benefits were combined by using an incremental discounted cost per QALY gained. The discount rate for costs was 5%. The expected incremental discounted cost per QALY gained from using the intervention, relative to the comparator, was NKr700,000($110,000). By changing the difference in hospital days between the groups by one standard error (SE, 3.7 days), the cost per QALY changed from $101,000 to $119,000. Changing the difference in lost days from production between groups by one SE (7 days), resulted in a cost per QALY ranging from $105,000 to $115,000. The increase and decrease by NKr500 from a value of NKr2,000 per day in hospital day cost made the outcome measure range from $105,000 to $115,000. Increasing and decreasing the sick leave by NKr2,000 from a level of NKr500 per day resulted in the cost-utility ratio ranging from $104,000 to $116,000. When the assumptions of survival gain obtained with the intervention relative to the comparator were changed to 6 and 12 months instead of 3, the cost-utility ratio changed to $46,000 and $21,000, respectively. The change in the quality loss value to 0.01 and 0.10 from 0.05 resulted in cost-utility ratio values of $84,000 and $185,000, respectively.
The cost per QALY gained by adding interferon-Alpha2b was conservatively estimated at $110,000. Potentially better cost-effectiveness may be found in different treatment regimes or in certain patient subgroups.
CRD COMMENTARY - Selection of comparators
The comparator chosen was reported to be the "conventional treatment" in patients with multiple myeloma.
Validity of estimate of measure of benefit
The estimate of measure of benefit is likely to be internally valid.
Validity of estimate of costs
The resource quantities were reported separately from the prices and adequate details of methods of cost estimation were given.
No discussion was reported with respect to the potential generalisability of the results to other countries or regions of the world. Comparisons with other studies were reported in terms of the estimate of measure of benefit used in the economic analysis; these results were in accordance with the present study findings. The results were not presented selectively.
Source of funding
Supported by a grant from Schering-Plough International.
Nord E, Wisloff F, Hjorth M, Westin J. Cost-utility analysis of melphalan plus prednisone with or without interferon-alpha2b in newly diagnosed multiple myeloma: results from a randomised controlled trial. PharmacoEconomics 1997; 12(1): 89-103
Subject indexing assigned by NLM
Aged; Cost-Benefit Analysis /economics; Drug Combinations; Female; Humans; Interferon-alpha /economics /therapeutic use; Male; Melphalan /economics /therapeutic use; Middle Aged; Multiple Myeloma /drug therapy /economics; Prednisone /economics /therapeutic use; Quality of Life
Date bibliographic record published
Date abstract record published