Volume 24
Number 4
2003

[Table of Contents]

Which cancer clinical trials should be considered for economic evaluation? Selection criteria from the National Cancer Institute of Canada's Working Group on Economic Analysis

William K Evans, Douglas Coyle, Amiram Gafni, Hugh Walker and the National Cancer Institute of Canada Clinical Trials Group Working Group on Economic Analysis

Abstract

Rising health care costs, expensive new health care technologies and increasing patient expectations are placing huge pressures on the publicly funded health care system in Canada. As a result, policy makers need information on the cost and cost-effectiveness of new therapies in addition to their clinical benefits. In response to this need, the National Cancer Institute of Canada Clinical Trials Group (NCIC CTG) established a Working Group on Economic Analysis (WGEA) to provide advice on the economic evaluation of new cancer therapies. This article describes the WGEA's recommendations on which trials should be considered for concurrent analysis of economic, as well as related issues, such as the number of patients required for an economic analysis within a prospective clinical trial and the selection of participating centres. The recommendations in this document are meant to be pragmatic, as the WGEA recognizes that both the research funds and human resource capacity for this type of research in Canada are limited. These recommendations are currently guiding priority setting with regard to trials for economic evaluation in NCIC trials. Examples of how these recommendations have been applied to actual trials are presented.

Key Words: clinical trials; cost; economic evaluation

Introduction

Rising health care costs, expensive new health care technologies and rising patient expectations are all creating pressure on provincial governments, as the principal payer in the Canadian universal access health care system.¹ Cancer contributes significantly to this health care burden, and its impact can be expected to increase as the population ages and as new diagnostic and therapeutic approaches emerge. In 1998, the economic burden of cancer care in Canada was estimated to be $14.2 billion, direct costs accounting for $2.46 billion and indirect costs for $11.76 billion.²

Decision-makers within government and agencies managing health care resources increasingly need information on the cost as well as the benefits of new interventions. However, there have been relatively few economic analyses of medical interventions to assist decision-makers in allocating resources for cancer treatments or any other health care interventions.^3-5 In 1998, the National Cancer Institute of Canada Clinical Trials Group (NCIC CTG) established a Working Group on Economic Analysis (WGEA) in response to the need to provide economic data on new cancer therapies. This article describes the WGEA's recommendations on which trials should be considered for concurrent analysis of economic and related issues, such as the number of patients required for an economic analysis within a prospective clinical trial and the selection of participating centres. It does not attempt to describe how to conduct an economic evaluation. Readers are directed to resources such as the Guidelines for Economic Evaluation of Pharmaceuticals from the Canadian Coordinating Office for Health Technology Assessment.⁶

The role of economic evaluations

Economic analysis may assist in the choice of one therapeutic intervention over another or help to estimate the total impact of a new therapy on a health system. For these reasons, some government regulatory bodies require economic analyses as part of new drug submissions from the pharmaceutical industry;^7-9 but cost is just one of a number of factors to be considered in determining the value of a new diagnostic or treatment approach. The clinical benefits of the therapy are the most important consideration and include outcomes such as improved survival, delayed tumour progression, reduced toxicity and improved quality of life. As well, the decision-maker's personal values and specific notions of equity influence the decision-making process.¹⁰

Economic analyses are of greatest benefit when there is a comparison of both the incremental benefit and the incremental cost. There are four possible outcomes that can occur when benefits and costs are measured concurrently (Table 1).

1. Improved outcome and decreased cost. This type of strategy is referred to as a dominant strategy and, in principle, should always be adopted.

2. Improved outcome and increased cost. There is a clinical advantage but an incremental cost over the current standard treatment.

3. Poorer outcome and decreased cost. There is a decrease in the clinical benefit but savings to the health care system.

4. Poorer outcome and increased cost. With a worse clinical outcome and increased costs, such therapies should not be adopted.

TABLE 1
Potential outcomes of an economic analysis

In the context of a clinical trial, resource utilization data and measures of health state preference can be collected prospectively with the same rigour as the clinical data, enabling sophisticated analyses to be done that will stand up to scientific scrutiny.

The resource utilization data (cost) and clinical outcome data can be analyzed to provide an estimate of the cost-effectiveness or cost-utility of the therapeutic intervention. For example, in a cost-effectiveness analysis, the primary outcome measure is most commonly the cost of an additional life year gained.¹¹ As survival differences are often small in cancer trials, measures of disease and treatment-related morbidity are important in deciding about the value of a new therapy. In a cost-utility analysis, information is collected on the health state(s) experienced by the patients during treatment, using methods such the Standard Gamble or the Time Trade Off, and is incorporated into the analysis. This provides a measure of the quality of the life gained through the treatment intervention. Cost-utility is usually expressed as the incremental cost per quality-adjusted life year gained (QALY).¹²

There is no sharp definition of what constitutes a cost-effective treatment intervention. The figure of $50,000 (US) per QALY is commonly used to describe a level of expenditure that is believed to be acceptable to society. This is based on the level of cost-effectiveness of hemodialysis when the Congress of the United States voted on its coverage under Medicare. In Canada, Laupacis et al. have suggested that less than $20,000 per QALY should be considered cost-effective and between $20,000 and 40,000 per QALY should be considered moderately cost-effective.¹³ As the authors acknowledge, these boundaries are arbitrary but are felt to reflect the "gut feeling" about the cost-effectiveness of new technologies.¹⁴

Factors limiting the conduct of economic evaluations

Although the inclusion of an economic evaluation alongside a clinical trial adds value, it also adds to the burden and cost of conducting the trial.¹¹ This burden includes the cost of collecting additional data on the resources used to provide treatment, descriptions of the quality of life, and the health state preferences of patients. Data collection may require extraction of information from source documents, interviews of patients and the use of patient diaries. In the early stages of adding economic analyses to clinical trials, clinical research assistants need to be trained and an infrastructure for data collection and analysis developed. The limited availability of research funds to support health services research and the small number of health economists interested in cancer in Canada are important limiting factors to the conduct of economic analysis alongside clinical trials.

Finally to have economic information of value from a Canadian perspective, the clinical trial must have resource utilization data on a sufficient number of Canadian patients. This requirement may become an increasingly important barrier to economic evaluations as more trials are conducted as international cooperative group studies with only a small number of Canadian patients.

In determining whether an economic analysis should be performed alongside a clinical trial, the incremental burden of performing the economic analysis must be weighed against other alternative methods of addressing the economic question. If the burden of data collection on investigators and patients is too high, this could jeopardize the recruitment of patients and affect the completeness and quality of the clinical trial data collected. For these reasons, it is necessary to have a practical approach to determining which cancer trials should have economic evaluations together with clear criteria for selecting the most appropriate clinical trials for economic evaluation.

Selecting appropriate clinical trials for economic evaluation

Some clinical trial designs are not suitable to answer economic questions.¹¹ Trials must be at least partly pragmatic and relate to actual clinical practice if they are to have an economic analysis.¹⁵ The NCIC's WGEA recommends that at least one of the following criteria be met before an economic analysis is undertaken alongside an NCIC CTG clinical trial.

1.    The new intervention is anticipated to have only a modest therapeutic benefit in a potentially large population. 

An example of such a trial is the randomized trial of anastrozole versus tamoxifen in postmenopausal women with early breast cancer.¹⁶ After a median follow-up of 47 months, anastrozole provided approximately a 2% (p = 0.03) absolute risk reduction in disease-free survival. Given the large number of patients who are potentially eligible to receive this treatment, an economic evaluation would be informative to policy makers. A large incremental cost might not justify the modest benefits.

2.    The new therapy is potentially very costly. 

If the treatment intervention is known to be very expensive and is expected to be used frequently enough to produce a large aggregate cost, then an economic evaluation alongside a clinical trial may be helpful in determining whether the new treatment is sufficiently cost-effective to warrant adoption. An example would be the use of high dose interleukin-2 (IL-2) in patients with stage IV melanoma. The requirement for hospitalization to manage the substantial treatment-related toxic effects and the high cost of IL-2 are important cost drivers.¹⁷ In addition, the clinical benefit is small, yielding only a low rate of tumour regression. These factors are compelling reasons for undertaking an economic analysis in a trial of IL-2 in melanoma, in order to inform a policy decision about whether to fund the intervention.

It should be noted that an economic analysis is unlikely to be required to evaluate a high-cost but infrequently used therapy. Similarly, a high-cost but highly effective treatment (the treatment cures a high proportion of patients) is unlikely to require an economic analysis.

3.    There is a high degree of uncertainty about the economic impact of the treatment of interest. 

A new treatment may appear to produce health benefits but be associated with significant side effects or other impacts that make it uncertain whether the net economic impact is positive or negative. In this situation, the economic analysis should ideally take the form of a cost-utility study, because this is the best way to capture the impact of side effects on the economic profile of a new treatment. The evaluation of chemotherapy regimens in advanced non small-cell lung cancer is a good example. There are multiple regimens, which are comparable in terms of tumour response and overall survival but unique in their side effect profile. Patient utilities, (information on the health states experienced by the patients) captured during a comparative trial would enable the determination of the cost per QALY gained relative to the current standard.

4.    An economic evaluation associated with equivalence trials may yield information of importance in the determination of routine practice. 

In the case of an equivalence trial, the economic evaluation has the potential to yield important information when considered from different perspectives, including that of the patient, the provider, the government or society as a whole. Side effects, ease of administration and cost then become the major parameters that guide policy development.

A recent example is the use of zoledronic acid as an alternative to pamidronate for the prevention of skeletal related events (SREs) in advanced breast cancer and multiple myeloma.¹⁸ A clinical trial demonstrated that zoledronic acid is equivalent to pamidronate in the prevention of SREs, but zoledronic acid can be infused over 15 to 30 minutes as compared with two to four hours for pamidronate. However, zoledronic acid is approximately twice the cost of pamidronate. From the perspective of the government as payer, an economic evaluation would be of value to determine whether the increased cost of the zoledronic acid is offset by reduced treatment administration costs. An economic evaluation from the patient perspective may show reduced out-of-pocket expenses as a result of reduced parking and care provider costs. Full economic data and data collected prospectively on patient preferences would be of value to policy makers in this situation.

5.    Economic data will assist future economic evaluations of new therapies. 

For some studies, adding an economic analysis in the form of a cost-of-illness study or estimating the cost of side effects will provide resource utilization data and cost information that can be used in future studies, including modeling studies. Resource utilization data captured in the course of conducting a trial that failed to yield a significant therapeutic benefit may still be useful for future studies.

These five criteria are now used to assess new NCIC CTG trials for the appropriateness of an economic analysis. More than one criterion may apply to a particular trial.

When not to do economic analyses

Given the need to set priorities for the use of funds for economic analysis, it may not be worthwhile to do an economic analysis in a number of clinical circumstances, such as when an expensive therapy works very well in a small number of patients.¹⁹ The use of cisplatin for testicular cancer is a good example. Although cisplatin was very expensive when first introduced, it greatly increased the cure rate for patients with metastatic testicular cancer. Similarly, some therapies differ in cost only marginally and have similar clinical outcomes in common diseases. An economic evaluation may also be unsuitable if the sample size in the clinical trial is not large enough to capture sufficient resource and cost variables or if the length of clinical follow-up is inadequate for the economic evaluation. Furthermore, the primary clinical endpoints may not be suitable effectiveness measures for economic evaluation. For example, in a clinical trial of cancer therapy, local tumour control may be the primary outcome of interest. However, for an economic evaluation, length of survival would be a more appropriate outcome measure.

Selecting the sample size for economic analysis

The sample size for a clinical trial is normally determined by the number of patients required to answer the clinical question(s). However, once the need for an economic analysis has been established, it is important to determine the required sample size for the economic component, as that sample size will determine the ability to precisely measure the economic outcomes of interest.

It is impossible to make general statements as to whether the sample size for an economic evaluation should be less than, equal to or more than the sample size for the clinical question, as this will vary from study to study. However, the following considerations go into the design of the economic component of the clinical trial.

1. For cost minimization studies, sample size relates to the ability to precisely measure the cost difference between the two therapies of interest. A quantitatively important cost difference could be determined in advance, and the necessary sample size calculated according to standard methods. However, what constitutes a quantitatively important cost difference is unclear. Furthermore, the need to demonstrate statistical significance in economic studies has been questioned, especially as it is clinical efficacy that has to be proven.²⁰

2. In cost-effectiveness and cost-utility analyses, the outcome of interest is a ratio of two differences. There are no generally recognized methods for determining the statistical significance of such ratios and, again, the relevance of this practice has been questioned.^21,22 Estimating sample size calculations requires agreement on the definition of the maximum acceptable cost per outcome gained. Decision-makers have understandably shown reluctance to determine such limits.

Therefore, there are important methodological and practical difficulties in determining the sample size required when conducting an economic analysis alongside a clinical trial. To estimate sample size according to a standard frequentist approach requires estimates of the costs, benefits and the quality of life of the control therapy and the expected benefits of the new therapy. In reality, however, this information is not usually available at the time the trial is being designed and, if it were available, there might be an argument that a full economic analysis was not necessary and that a modeling study would be sufficient.²² Recently, Bayesian approaches have been suggested for determining sample size, although they also rely on the availability of the same information on cost, benefit and quality of life.²¹

From the research perspective, the ideal situation would be to have the optimal sample size to answer both the economic and clinical questions. However, there are two specific concerns that may require the economic analysis to be based on a smaller sample size.²²

First, there may be an additional burden on patients who participate in the economic analysis if they are required to keep diaries or complete measures of quality of life. Investigators may be hesitant to enroll some patients in studies requiring this increased level of participation.

Second, there may be an increased burden of data collection on the participating centres. To ensure that the clinical trial gets started smoothly, it may be useful to delay the implementation of data capture for the economic component of the trial until the procedures for recruitment of patients to the clinical trial are running smoothly. In a current trial of regional radiation therapy in early breast cancer conducted by the NCIC, a reduced sample size for the economic component was accepted, which allowed accrual to the economic component to be delayed until the trial was well under way.

Therefore, consideration of sample size in clinical trials with an economic analysis must balance pragmatic, ethical and scientific considerations. When the additional burden of participating in the economic study is low from the perspective of both the patient and the participating centre, the sample size should be equal to that required to answer the clinical question in the trial. When the burden of conducting the economic component of the study is of concern, sample size should relate to the degree of precision that can be obtained with a reduced sample size. Given these issues, sensitivity analyses should always be considered more important than statistical analysis.

Selecting centres for economic analysis

It is generally assumed that there is a large variation in the costs of care between treatment centres participating in multicentre economic evaluations. The costs of goods and services (unit costs) are likely to vary from one institution to another and from one geographical location to another because of different supply contracts, salaries and other factors.²³ Variations may also arise in the resources used in the treatment of patients at centres because of different clinical practices.²⁴ Multicentre clinical trials are typically undertaken in order to recruit sufficient patient numbers to answer a clinically important question, and data are pooled from across centres on the assumption that the clinical effects of the intervention are generalizable to all centres in the study. It is not clear, however, whether economic data collected from a number of diverse settings with different cost structures can be pooled in the same way. Therefore, it is necessary to consider how centres are selected for cost estimation.

A common approach is to pool the efficacy and resource utilization data from all patients from all of the centres but to choose one centre to obtain the unit cost estimates. The choice of the centre is typically made on the basis of convenience. The extent to which unit costs from the one centre are applicable to the unit costs of all participating centres (i.e., internal applicability) or to all centres where the intervention would typically take place (i.e., external applicability) is not clear and is usually ignored. For internal applicability, unit costs should ideally be obtained from all participating centres. However, the lack of standardized and credible cost information systems at most health care facilities means that this approach is not feasible. The following factors and options should be considered when selecting a sample of centres for unit cost estimation.

The first consideration is that the capture of resource utilization data has a cost associated with it. Therefore, the amount of funding available for the study will determine how many centres can be involved in the collection of the data. The second factor to be considered is the method of sampling centres for unit cost information. It could be a systematic approach (i.e., by geographical area, centre size) or a method involving random selection. This consideration may be influenced by the extent of subanalysis that is considered desirable. For example, if there is a need to describe the extent of geographical variation in costs, institutions from different geographical areas will need to be selected. A further consideration is the availability of good costing data and the ease of access to this cost information. The availability of an institutional costing framework is a powerful determinant of whether a centre is included in the sampling frame. If there is interest in reporting external applicability, a centre could be chosen that is not a participant in the study.

The choice of which centres are selected for unit cost estimation and how these centres are stratified for analysis in a multicentre trial can have a significant impact on the results of the cost analysis. There is currently not enough information to guide decisions on which sampling strategy is optimal, but the strategy is likely to vary depending on the goals(s) of the study.

Case studies

To illustrate the application of the guidelines presented in this paper, clinical trials that have been considered by the WGEA are presented. The first of these was a large multicentre study, which compared trastuzumab and placebo taken for one or two years by women with HER-2 positive primary breast cancers who had completed adjuvant chemotherapy. Given the high acquisition cost of trastuzumab, the large number of potentially eligible patients and the long duration of use of the drug in the event of a positive trial, the impact on the Canadian health care budget was anticipated to be large. Therefore, application of the current guidelines dictated that an economic analysis be conducted alongside this clinical trial. Resource utilization data would need to be collected from large representative cancer centres throughout Canada. Unit costs would need to be obtained from one centre and sensitivity analysis used to evaluate the impact of regional variations in unit costs.

A second trial considered by the WGEA was a study involving women who had undergone mastectomy for stage II breast cancer and were at risk of recurrence because of axillary lymph node involvement. The intervention in the experimental arm of the trial was radiotherapy, and the outcome of interest was its impact on overall survival. A review of the status of the trial revealed that patient accrual was slow, and this was compromising the ability of the trial to test the primary hypothesis in a timely manner. The nature of the intervention lent itself to computer modeling techniques for the economic evaluation. In this case, the WGEA recommended that an economic evaluation alongside the clinical trial not be undertaken.

The third clinical trial that the WGEA considered for an economic evaluation was a randomized placebo-controlled trial of adjuvant therapy with ZD-1839 (Iressa®) in patients with non small-cell lung cancer (NSCLC) who had undergone a complete surgical resection. Patients were to be randomly assigned to receive one year of adjuvant ZD-1839 or placebo. The main outcomes for this trial were disease recurrence rates and overall survival. There is evidence from two Phase II trials (IDEAL 1 and 2) that ZD-1839 (250 mg/day) can palliate patients with NSCLC refractory to chemotherapy.^25,26 Since the adjuvant use of this agent in lung cancer would affect a large population and be administered for a long duration, an economic analysis to determine the incremental cost per life year gained was felt to be necessary. This economic analysis would be highly valued by health policy makers as it would allow the cost of ZD-1839 per life year gained to be compared with other anticancer therapies currently being used in Canada.

In summary, three clinical trials considered by the NCIC's WGEA have been presented to illustrate how the proposed guidelines for the selection and design of economic evaluations in association with NCIC CTG clinical trials have been applied.

Conclusions

Economic analyses alongside NCIC CTG randomized controlled trials are of increasing importance in the face of a proliferation of new treatment approaches for cancer and concerns about the sustainability of the publicly funded health care system in Canada. Given the current availability of resources, we have suggested a practical approach to determining which trials are appropriate for economic analysis. This article also provides guidance on the issues of sample size and the selection of centres for participation in the capture of resource utilization and unit cost data. However, it does not address the practical issues that relate to the conduct of these analyses within a given trial.

Currently the NCIC CTG has established a process whereby the WGEA reviews trials brought forward from the Disease Site Groups through their liaison representatives to the WGEA. If a trial is felt to meet the criteria discussed here and the NCIC CTG approves of the scientific merit of the clinical trial, the WGEA identifies a health economist to work with the principal investigators. Together, they identify the economic endpoints of the study, independent of the clinical endpoints, and the data elements needed to undertake the economic analysis.

As concerns over budgetary restrictions in the Canadian health care system increase, the need to demonstrate the value for money of new and more costly technologies is paramount. Given this new reality, the NCIC CTG's approach to economic evaluations and clinical trials will provide Canadian data that will help to inform decisions on the efficient allocation of scarce health resources.

References

1. Lewis SJ, Donaldson C, Mitton C, Currie G. The future of health care in Canada. BMJ 2001;323:926-29.

2. Health Canada. Economic burden of illness in Canada, 1998. Ottawa: Health Canada, Cat No. H21-136/1998E.

3. Earle CC, Coyle D, Evans WK. Cost-effectiveness analysis in oncology. Ann Oncol 1998;9:475-82.

4. Goddard M, Drummond M. The economic evaluation of cancer treatments and programmes. Eur J Cancer 1991;27:1191-6.

5. Bennett CL, Armitage JL, Buchner D, Gulati S. Economic analysis in phase III clinical cancer trials. Cancer Invest 1994; 12:336-42.

6. Canadian Coordinating Office for Health Technology Assessment. Guidelines for Economic Evaluation of Pharmaceuticals: Canada. Ottawa: The Canadian Coordinating Office for Health Technology Assessment, 1997.

7. Commonwealth Department of Human Services and Health. Guidelines for the pharmaceutical industry on preparation of submissions to the Pharmaceutical Benefits Advisory Committee including major submissions involving economic analyses. Canberra: Australian Government Publishing Service, 1995.

8. National Institute for Clinical Excellence. Technical guidance for manufacturers and sponsors. URL: http://www.nice.org.uk/pdf/technicalguidance
   formanufacturersandsponsors.pdf. 

9. Detsky AS. Guidelines for economic analysis of pharmaceutical products: a draft for Ontario and Canada. Pharmacoeconomics 1993;3:354-61.

10. Bennett CL, Armitage JL, LeSage S, Gulati SC, Armitage JO, Gorin NC. Economic analyses of clinical trials in cancer: are they helpful to policy makers? Stem Cells 1994;12:424-9.

11. Drummond MF, O'Brien B, Stoddart G, Torrance G. Methods for the economic evaluation of health care programmes. Oxford: Oxford University Press, 1997.

12. Bennett CL, Westerman IL. Economic analysis during phase III clinical trials: who, what, when, where, and why? Oncology 1995;9:69-75.

13. Laupacis A, Feeny D, Detsky AS, Tugwell PX. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. Can Med Assoc J 1992;146:473-81.

14. Laupacis A, Feeny D, Detsky AS, Tugwell PX. Tentative guidelines for using clinical and economic evaluations revisited. Can Med Assoc J 1993;148:927-29.

15. Schwartz D, Lellouch J. Explanatory and pragmatic attitudes in therapeutic trials. J Chron Dis 1967;20:637-48.

16. ATC Trialists' Group. Anastrozole alone or in combination with tamoxifen alone for the adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomized trial. Lancet 2002;359:2131-9

17. Atkins MD, Lotze MT, Dutcher JP, et al. High dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol 1999;17: 2105-16.

18. Rosen LS, Gordon D, Antonio BS, et al. Zoledronic acid versus pamidronate in the treatment of skeletal metastases in patients with breast cancer or osteolytic lesions of multiple myeloma: a phase III, double-blind, comparative trial. Cancer J 2001; 7:377-87.

19. Claxton K, Posnett J. An economic approach to clinical trial design and research priority-setting. J Health Econ 1996; 5:513-24.

20. Claxton K. The irrelevance of inference: a decision-making approach to the stochastic evaluation of health care technologies. J Health Econ 1999;18:341-64.

21. Chaudhary MA, Stearns SC. Estimating confidence intervals for cost-effectiveness ratios: an example from a randomized trial. Stat Med 1996;15:1447-58.

22. Coyle D, Davies L, Drummond MF. Trials and tribulations. Emerging issues in designing economic evaluations alongside clinical trials. Int J Technol Assess Health Care 1998;14:135-44.

23. Goeree R, Gafni A, Hannah M, Myhr T, Blackhouse G. Hospital selection for unit cost estimates in multicentre trials: Does the choice of hospitals make a difference? Pharmacoeconomics 1999;15: 561-72.

24. Coyle D, Drummond MF. Analyzing differences in the costs of treatment across centers within economic evaluations. Int J Technol Assess Health Care 2001;17:155-63.

25. Fukuoka M, Yano S, Giaccone G, et al. Final results from a phase II trial of ZD1839 ('Iressa') for patients with advanced non-small-cell lung cancer (IDEAL 1). Proc Am Soc Clin Oncol 2002;21:Abst 1188.

26. Baselga J, Kris M, Yano S, et al. Phase II trials (IDEAL 1 and IDEAL 2) of ZD1839 in locally advanced or metastatic non-small cell lung cancer (NSCLC) patients. Ann Oncol 2002;13(S5):131 (abstract 481).

Author References

William K Evans, Cancer Care Ontario, Toronto, Ontario, Canada

Douglas Coyle, Clinical Epidemiology Unit, Ottawa Civic Hospital, Ottawa, Ontario, Canada

Amiram Gafni, Centre for Health Economics and Policy Analysis, McMaster University, Hamilton, Ontario, Canada

Hugh Walker, Radiation Oncology Research Unit, Kingston General Hospital, Kingston, Ontario, Canada

The National Cancer Institute of Canada, Clinical Trials Group Working Group on Economic Analysis (Douglas Coyle, Ottawa Civic Hospital, Eva Grunfeld, Ottawa Regional Cancer Centre, M Neil Reaume, Ottawa Regional Cancer Centre, Kathryn Roche, Toronto-Sunnybrook Regional Cancer Centre, Bev Koski, Queen's University, Carole Chambers, Tom Baker Cancer Centre, Heather-Jane Au, Cross Cancer Institute, Amin Haiderali, AstraZaneca Canada, William Evans, Cancer Care Ontario, Jeff Hoch, University of Western Ontario, Steve Morgan, University of British Columbia)

Correspondence: Dr. WK Evans, Chief Medical Officer, Cancer Care Ontario, 620 University Ave., Toronto, ON Canada M5G 2L7; Fax: (416) 217-1235; E-mail: bill.evans@cancercare.on.ca

[Previous] [Table of Contents] [Next]