|A pilot study to evaluate the feasibility of using willingness to pay as a measure of value in cancer supportive care: an assessment of amifostine cytoprotection
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.
The use of amifostine in reducing the risk of chemotherapy-induced toxicity in the treatment of advanced ovarian cancer.
Healthy Canadian taxpayers.
Hospital. This study was carried out at the Princess Margaret Hospital, Toronto, Canada.
Dates to which data relate
The effectiveness data were collected from 1988 to 1993. Resource consumption data secondary to neutropenic fever in patients who were not protected by amifostine were collected during the clinical study. Other costs were calculated after the clinical trial had finished. The price year was 1997.
Source of effectiveness data
Effectiveness data were derived from a single study.
Link between effectiveness and cost data
The costing was based on the same patient sample as that used in the effectiveness study and was undertaken prospectively.
The study sample comprised 242 patients who had a histological diagnosis of grades III/IV epithelial ovarian cancer and who had undergone surgical debulking and been left with residual disease. They were randomised in a 1:1 ratio to receive amifostine before chemotherapy or to a control group.
Randomised controlled trial.
Analysis of effectiveness
The analysis of the clinical study was based on intention to treat. The primary health outcomes used in the clinical study were the incidence (%) and absolute risk reduction (ARR)(%) in febrile neutropenia, renal toxicity and neurotoxicity, and the number of days in hospital.
An absolute risk reduction in febrile neutropenia of 11% (p=0.02) was observed (incidence rate: 21% without amifostine, 10% with amifostine). Renal toxicity was 36% without amifostine compared to 10% with amifostine (p=0.003). This implied an absolute risk reduction of 26%. Neurotoxicity fell from 42% without amifostine to 31% with amifostine (p=0.029, absolute risk reduction of 11%). The total number of days in hospital with febrile neutropenia was 226 for controls and 89 for the amifostine patients (p=0.02).
Pretreatment with amifostine reduces the cumulative hematologic, renal and neurologic toxicities associated with the chemotherapy regimen.
Measure of benefits used in the economic analysis
The measure of benefit used in the economic analysis was willingness to pay (WTP). The WTP scenarios were structured in the form of a hypothetical taxation question. Severity rating scales and the payment-card method were used to elicit WTP estimates. WTP estimates were derived for each of the 3 risk reductions. The greatest of the three WTP responses was used as the maximum amount that a Canadian tax payer would be willing to pay per year. This maximum WTP estimate was multiplied by the life expectancy and discounted at a rate of 3% to come up with an age-adjusted WTP estimate.50 healthy Canadian taxpayers were used to derive the WTP measures. A random multistage sampling technique was utilised to identify potential subjects. To be eligible for the survey, participants had to be 18 years of age or older, have permanent residence status in Ontario or Quebec, indirectly support the Canadian health care system through tax contributions and give informed consent to participate in the review. There was equal representation between Ontario and Quebec. The median age of the respondents was 37.5 (range: 18 - 60) years, with a balanced distribution for gender and marital status. Approximately two-thirds of the subjects did not have children and had attended post-secondary school education. Most (86%) of the sample were employed either full- or part-time, but only 20% had household incomes that exceeded Can$70,000. About one-third of participants reported no prior family history of cancer.
The analytic time period for the investigation was 6 months. Direct costs included the cost of six cycles of amifostine, expenditures for drug preparation, administration, supplies and patient monitoring during the amifostine infusion, hospital savings as a result of an ARR for febrile neutropenia, renal and neurological toxicity. The cost of antiemetics was not incorporated into the analysis. Costs of chemotherapy-induced renal and neurological damage were not included in the analysis, because hospital resource consumption data for these events were not collected in the clinical study. Costs were not discounted. Quantities and costs were not reported separately. The quantity/cost boundary adopted was that of the hospital.
The estimation of costs and quantities was based on actual data. The average costs of a febrile neutropenic event were determined on the basis of costs statistics from the Princess Margaret hospital. The acquisition cost of amifostine was obtained from Eli Lilly, Canada. Costs of daily hospitalisation for a teaching hospital were obtained from the Ontario Hospital Association. Costs for dose preparation, administration and patient monitoring were obtained from the Nursing and Pharmacy Departments, PMH. Laboratory test costs were obtained from the Department of Biochemistry, Hematology and Microbiology, PMH. The cost of physician fees for service was obtained from Schedule of Benefits: Physician Services Under the Health Insurance Act, Ontario Ministry of Health, October 1 version, 1992.1997 price data were used.
Statistical analysis of costs
Friedman's nonparametric ANOVA was used to test the significance of the difference between the three risk reductions. A multiple regression model was employed to measure the association between WTP and subject characteristics.
No indirect costs were included. This implies that indirect societal costs due to lost productivity were excluded from the analysis.
Canadian dollars (Can$), with Can$1=US$0.75 as at March, 1997.
A comprehensive sensitivity analysis was conducted to test the robustness of the results. A sensitivity analysis was performed on the method of combining WTP estimates, the cost of febrile neutropenia, the dosage of amifostine, the replacement of amifostine by G-CSF and the discount rate.
Estimated benefits used in the economic analysis
Respondents reported that they were willing to pay an average of Can$141 (95% CI: 92 - 189) for a drug that reduces the risk of febrile neutropenia from 21% to 10%. They were less inclined to pay for a risk reduction of 26% in renal toxicity (Can$71; 95% CI: 38 -104)and a risk reduction of 11% in neurotoxicity (Can$86; 95% CI: 46 - 127). These latter WTP values were significantly lower than what subjects were willing to pay to reduce the risk of fever and neutropenia (p<0.001). Age-adjusted WTP estimates showed that taxpayers were willing to pay an average of Can$3,476 (95% CI: 2,275 - 4,676) for reducing the risk of neutropenic fever, as against only Can$1,756 (95% CI: 941 - 2,571) and Can$2,134 (95% CI: 1,141 - 3,127) for renal and neurological toxicity, respectively. The results of the regression procedure determined that family income, marital status and province of residence were strongly associated with the maximum WTP.
When the cost of hospitalisation and supportive care were considered, the total cost per patient was Can$7,004 (+/- 1,877). This estimate and the ARR for neutropenic fever associated with adjuvant amifostine were multiplied to determine the expected hospital savings (Can$770). This figure was then subtracted from the total cost of amifostine administration (Can$4,596). Therefore, total hospital costs of amifostine cytoprotection were Can$3,826 per patient.
Synthesis of costs and benefits
The age-adjusted average value of respondents' maximum WTP and the overall cost of adjunct amifostine therapy were combined in a cost-benefit measure of an incremental societal cost of Can$350 per patient (95% CI: -850 to 1,551). The suggestion of cost neutrality was still supported when the method of combining the three WTP estimates was changed. The results were also relatively insensitive to variations in the cost of febrile neutropenia. Lowering the dosage of amifostine to 740 mg/m2, societal gain would be Can$597 (95% CI: -604 to 1,797). Replacing amifostine by 7 days of G-CSF would lead to an incremental societal cost of Can$1,984 per patient (95% CI: 784 - 3,185). When the discount rate was set at 5%, the incremental societal cost was Can$1,234 (95% CI: 338 - 2,129).
WTP as a measure of value in cancer supportive care is feasible and can easily be extended to other disease sites. Prophylactic amifostine in ovarian cancer patients receiving cyclophosphamide and cisplatin produces a situation of cost neutrality. This outcome was stable despite extremes in the cost of treating febrile neutropenia. However, the baseline results were sensitive to variations in respondents' maximum WTP, amifostine dosage and discount rate.
CRD COMMENTARY - Selection of comparators
The rationale for the choice of the comparator was clear.
Validity of estimate of measure of benefit
The authors must be applauded for applying a new technique to the field of oncology. However, from a general point of view, it is unclear how the WTP estimates for the three risk reductions should be combined. This is important since the baseline results were sensitive to variations in respondents' maximum WTP. Some more methodological work needs to be done before this technique can be safely applied to oncology.
Validity of estimate of costs
Some direct and indirect costs were excluded from the analysis due to lack of available data. In this study, this suggests that the net benefit of amifostine protection was underestimated.
Future studies should attempt to use larger samples which are more representative of the entire country to increase the generalisability of the results. The results of this study are unlikely to be generalisable to paclitaxel plus cisplatin which has now become the treatment standard.
Implications of the study
Future studies should apply WTP techniques to paclitaxel plus cisplatin which is now considered to be the treatment standard for patients with advanced ovarian cancer.
Source of funding
Funded by Eli Lilly Canada Inc.
Dranitsaris G. A pilot study to evaluate the feasibility of using willingness to pay as a measure of value in cancer supportive care: an assessment of amifostine cytoprotection. Supportive Care in Cancer 1997; 5(6): 489-499
Other publications of related interest
Kemp G, Rose P, Lurain J, Berman M, et al. Amifostine pretreatment for protection against cyclophosphamide-induced and cisplatin-induced toxicities: results of a randomised controlled trial in patients with advanced ovarian cancer. Journal of Clinical Oncology 1996;14:2101-2112.
Subject indexing assigned by NLM
Adolescent; Adult; Amifostine /economics; Antineoplastic Agents /adverse effects; Attitude to Health; Canada; Cost Savings; Cost-Benefit Analysis; Drug Costs; Feasibility Studies; Female; Hospital Costs; Humans; Income; Male; Middle Aged; Ovarian Neoplasms /drug therapy; Pilot Projects; Surveys and Questionnaires; Taxes /economics
Date bibliographic record published
Date abstract record published