Abeloff's Clinical Oncology, 4th Edition

Part I – Science of Clinical Oncology

Section D – Preventing and Treating Cancer

Chapter 24 – Economic Analysis of Cancer Treatment

Charles L. Bennett,Karen A. Fitzner

SUMMARY OF KEY POINTS

  

   

Cancer care accounts for the largest number of dollars spent on any medical condition in the United States.

  

   

Cancer accounts for 10% of all Medicare expenditures.

  

   

Fewer than half of the states have passed legislation mandating coverage of clinical trials.

  

   

Cost-benefit, cost-effectiveness, cost-utility, and cost-identification analyses are the main methods for evaluating the economics of clinical interventions.

  

   

Empirical studies have found that results of cost-effectiveness studies are associated with pharmaceutical versus not-for-profit funding, although quality is not.

  

   

Few economic studies have been incorporated into clinical trials sponsored by the National Cancer Institute (NCI).

  

   

To date, only seven NCI-sponsored clinical trials have included economic assessments.

CANCER CARE IS EXPENSIVE

The costs of treating patients with cancer contribute substantially to the United States’ $1.6 trillion annual health care spending. In fact, the economic burden of cancer is the largest imposed by any medical illness. The high expense associated with cancer is related to a number of factors, including an increase in the prevalence of cancer as people live longer, the high rate of comorbid medical illness in cancer patients, and high costs associated with diagnosis, treatment, and end-of-life and palliative care as new, more expensive treatments become available. The direct costs of cancer care (spent for prevention, screening, diagnosis, treatment, and palliation) accounted for $61 billion in 2002. [1] [2] Pharmaceutical costs, which have been growing rapidly, account for a large amount of cancer-related expenditure.[3]

Cost is a major determinant of the type and intensity of cancer care, particularly related to reimbursement of high-tech and high-cost procedures and pharmaceutical products for cancer patients.[4] The physician has a large role in determining the medical costs incurred by individual patients. Medical costs are categorized as (1) direct medical costs; (2) direct nonmedical costs, that is, amounts spent for caregivers and travel; (3) indirect costs, that is, the economic value of lost productivity due to illness, disability, and death (mortality); and (4) intangible costs, that is, costs associated with pain, suffering, and grief. Although several studies have evaluated the direct costs of various types of cancers, few include estimates of direct nonmedical, indirect, and intangible costs. Pilot studies indicate that these costs may be as high as 75% of the total cost of cancer care. [5] [6]

In 2006, the most costly cancers for men were prostate cancer followed by lung and colorectal cancer; for women, the most costly cancers were breast cancer followed by colorectal and lung cancer.[7] In addition to considering total costs, it is important to consider the benefit (or value) provided by the intervention in exchange for the amount expended. The number of articles addressing cost-effectiveness of cancer treatments has increased dramatically in the past few years, a large number of these studies addressing supportive care agents. [8] [9] [10] Oncologists, insurers, and policymakers increasingly are addressing questions of costs, cost-effectiveness, and quality of life. Many changes have been made in this area since the late 1990s, particularly in relation to the costs of clinical studies. Previously, the national mandate to evaluate the effectiveness of health care, including cancer care, was designed to incorporate cost-effectiveness assessments. The U.S. Healthcare Research and Quality Act of 1999 gave a mandate to the Agency for Healthcare Research and Quality to focus on evidence-based technology assessments, which could then be incorporated by relevant people in the medical community into guidelines produced by physicians or medical societies. Unfortunately, cost considerations are not addressed explicitly in the current mandate. A few of the assessments will, however, provide information that may be applicable to cost-effectiveness analyses that can be conducted by others.

Clinical Trials and Their Reimbursement

The costs of cancer clinical trials have become an important issue related to cancer treatment over the past decade. Findings from pilot studies of about 1300 patients in phase II/III clinical trials estimate that costs ranged from 10% lower to 23% higher for clinical trials in comparison to control groups that received standard medical care. [8] [9] [10] [11] [12] [13] [14] On the basis of these findings, nearly half of U.S. states, Medicare, and several private insurers now cover the costs of patient care in “qualifying” clinical trials. Because of economic and other considerations, federal policies were not supportive of clinical trial reimbursement during the 1990s; Medicare excluded coverage of routine costs of care associated with clinical trial participation on the basis that such treatment was experimental or investigational.[10] However, a 1993 review found that Medicare was being billed millions of dollars for patients who received care in clinical trials.[12] The U.S. General Accounting Office estimated that Medicare had paid 50% to 90% of routine patient care costs in clinical trials because fewer than 4% of claims for clinical trial costs incurred by Medicare beneficiaries were denied and oncologists often submitted bills for components of complex treatments without specifying the procedure itself.[14]

In the late 1990s, there were several unsuccessful proposals, such as the Medicare Cancer Clinical Trial Coverage Act and the Medicare Cancer Clinical Trial Demonstration Act, that would have allocated considerable funds to cover cancer clinical trials sponsored by the National Institutes of Health, the Department of Defense, and the Department of Veterans Affairs. Following years of lobbying by individuals, patient groups, health care workers, and organizations who were concerned about reimbursement denials of clinical trial costs and the low rates of accrual to clinical trials, a 2000 executive order authorized Medicare to cover the costs of any cancer clinical trial that satisfied qualifying criteria. Medicare has paid routine care costs (e.g., office visits and lab tests) for patients enrolled in federally funded clinical trials since 2000. The Centers for Medicare and Medicaid Services revised its National Coverage Determination for off-label use in cancer clinical trials in early 2005. Medicare now pays for both routine and nonroutine costs associated with the patient's care as well as the off-label use of some anticancer drugs. [15] [16] [17]

At the state level, the question put before some state legislatures has been whether insurance barriers to clinical research are best removed through voluntary action of health insurers or formal legislation. As of December 2006, 22 states had passed laws mandating coverage of patient care costs associated with treatment provided in specified categories of cancer clinical trials.[18] In the mid-1990s, Rhode Island became the first state to legislate insurance coverage for phase II clinical trials. Since then, Georgia mandated insurance for selected pediatric cancer trials in 1998, and Maryland and Virginia mandated insurance for cancer trials conducted in in-state academic institutions in 1999. Maine requires managed care organizations and private insurers to cover trials approved and funded by the NIH, a NIH-sponsored cooperative group or center, or the U.S. Department of Health and Human Services. Louisiana requires coverage of Phase II, III, and IV clinical trials that are approved by entities such as the Food and Drug Administration (FDA), the Department of Defense, the Veterans Administration, and the Coalition of National Cancer Cooperative Groups. The New Jersey Association of Health Plans agreement is unique in that all private insurers in a single state have voluntarily agreed to provide cancer clinical trial coverage. These coverage mandates are important because without third-party coverage, most patients are unable to bear the cost of participating in a clinical trial.

Today, public payers and several large private health insurers reimburse for medical care costs associated with clinical trials. Concern over variable scientific quality has led many state legislatures to limit reimbursement to trials funded by federal agencies. Moreover, many states have mandated clinical trial coverage by private and public payers for cancer trials but not for other medical conditions. This is in part due to the national infrastructure surrounding cancer trials, which is the most established and comprehensive of any disease clinical trials.

Economic Assessments in Cancer Care

Economic issues in cancer care are paramount in planning for optimal use of scarce resources and in responding to the increased economic pressures faced by the health care system. Physicians and health policymakers need to make judgments on the effectiveness and cost-effectiveness of cancer care based on explicit assessments of the costs and benefits of alternative management strategies. In other words, payers want assurance that expensive treatments are cost-effective. To understand the financial impact of alternative cancer management strategies more clearly, it is essential to understand the basic terminology of economic studies in health care: costs, benefits, cost-effectiveness, cost utility, and cost minimization. It also is important to explain the methods that policymakers and oncology researchers use when they evaluate the costs of cancer care.

When Are Economic Assessments Likely to Be Helpful (or Unhelpful)?

Assessments of the costs of cancer care can be considered when significant resources are being used, when resource considerations have a direct impact on patient care, and when resource allocation decisions are likely to be made.[19] Economic analyses also help to identify the most efficient option when two or more clinically efficacious interventions are available. Examples of situations in which a large amount of resources are used include use of supportive care agents such as antiemetics, erythropoietin, granulocyte colony-stimulating factor (GCSF), and genetic predictors. Resource considerations played a prominent role in the decision to support high-dose chemotherapy with stem cell transplantation for breast cancer before reports of unfavorable results were received from several randomized clinical trials. Economic analyses are unlikely to be helpful in cases in which a treatment works well but only a small number of individuals are affected, thereby limiting the overall cost impact. For example, advanced testicular cancer routinely is treated with chemotherapy. The cure rate is high, but the number of cases is fewer than 10,000 per year. Similarly, data have been collected on the economic implications of high-dose chemotherapy with stem cell transplantation for breast cancer, but they will not be useful to policymakers because this procedure has not been found to be clinically effective.

Types of Economic Analyses

Economic evaluations provide information on the value of an intervention or therapy in relation to its costs when compared to a competing alternative when, ideally, the alternative is the current standard of care. Economic evaluations of cancer interventions can range from decision analytic models to retrospective database comparisons to analyses conducted alongside clinical trials. However, all of the economic analyses of clinical interventions can be grouped into four categories based on the measure of effectiveness used in the analysis: cost-benefit (costs and benefits are measured in the same terms, usually monetary), cost effectiveness (costs measured in monetary terms, effectiveness in clinical terms), cost utility (costs measured in monetary terms, utility measured in terms of utility), and cost identification (costs measured in monetary terms, no effectiveness measurement).[20]

Cost-benefit analysis compares the incremental cost of a medical intervention with its incremental benefit, with both terms measured in monetary units. Therefore, interventions with positive net benefits, in which the value of the incremental benefits is greater than the incremental costs, are cost-beneficial compared to the alternative. In theory, cost-benefit analysis allows for the comparison of health interventions with other programs or interventions from non–health care sectors that may be competing for the same dollars. That is, with a cost-benefit analysis, a government could compare whether to spend monies on a new after-school program for children or use the same monies to fund a breast cancer screening program. However, difficulties in valuing all the relevant factors (e.g., years of life lost and quality of life) in monetary terms limit the use of cost-benefit analysis.

More commonly, economic evaluations in cancer care involve cost-effectiveness analysis. Cost-effectiveness analysis provides information about the value of an intervention or therapy in relation to its costs compared to a competing alternative when effectiveness is measured in clinical terms. The analysis compares two or more interventions and provides information about the differences in costs and effects between comparators. The results are summarized into a ratio that provides the results in terms of the costs per unit of effect. This ratio is referred to as the incremental cost-effectiveness ratio, because it is assessing incremental differences between alternative treatments. Because it is a cost-effectiveness analysis, the denominator of the ratio is valued in natural units, such as years of life, and is calculated by finding the difference in the effectiveness measure between the alternatives. For example, many cost-effectiveness analyses in cancer treatments report the incremental cost per year of life saved.

Cost-utility analysis is a subset of cost-effectiveness analysis in which the measure of effectiveness is a utility or value. Utilities provide a measure of overall quality of life and are applicable across different types of cancer. The measure is intended to incorporate both positive and negative aspects of treatment. The utility is combined with information on survival to estimate quality-adjusted life years, which are used as the measure of effectiveness in cost-utility analyses. Thus, cost-utility analyses provide an estimate of the cost per quality-adjusted life year gained. In principle, this method can be used to compare the value of screening programs to new chemotherapeutic agents to gauge the relative value of an alternative.

The final type of clinical economic analysis used in cancer studies is cost identification. This technique reports the total types and amounts of resources used in providing medical care, without formal assessments of the clinical benefits of the treatment. This technique is an integral part of the other economic evaluations in that it provides the cost estimate used in the numerator. However, with this method, the benefits are not compared formally among alternatives.

It is important to emphasize that in all types of economic analyses, the way monetary units are assigned to treatments can make important differences. Specifically, costs differ markedly from charges.[21]Costs represent the true opportunity cost of a resource, whereas charges represent the amount that is billed for that resource. There might be little connection between costs and charges, charges typically being much greater than the actual costs. Therefore, it is important to be aware of whether costs or charges are being used to derive the economic estimates.

CANCER COSTS: ESTIMATES FROM MEDICARE POPULATIONS

Total medical care expenditures for oncology account for 10% of all Medicare expenditures.[22] Medicare Part A covers inpatient cancer care. Cancer-screening services are covered for cervical, breast, colorectal, and prostate cancer, and many types of chemotherapy and related treatments are covered under Part B. As of 2006, Medicare Part D provides a drug benefit that covers cancer drugs but requires significant financial contribution from patients who have prescription expenses more than $2,250.[23] In the future, prescription drug data will also be captured via Medicare part D.

Recent studies have incorporated economic analyses for various cancers experienced by the Medicare population. [1] [2] [24] Costs have been evaluated for the initial phase, the primary course of therapy, and any adjuvant therapy, continuing care, including surveillance activities for detecting recurrences and new cancers, and the terminal care phase. Data for these analyses derive from the linked Surveillance Epidemiology and End Results (SEER) and Medicare database that includes detailed financial and clinical data elements for cancer patients who received care in 11 geographic regions of the country (seehttp://seer.cancer.gov). The database provides detailed information for inpatient services (Part A) and payments for outpatient services (Part B). The Medicare-SEER cost files were generated by reviewing monthly cost files for each cancer patient identified in the SEER database. Prices are adjusted using the Medicare per capita index and Medicare-based price indices that account for differences in health care purchasing power over time and location. These data are used to provide estimates of national expenditures according to type of cancer and gender.

Quality Assessment of Economic Analyses

Controversy exists over the quality of economic analyses of medical treatments and has led to the development of grading systems for economic evaluations. [25] [26] The quality of the data and the models that are used in these analyses are the major determinants of the results. Additional items that are considered in the quality of economic evaluations include the objective and perspective of the analysis, data sources, type of comparison, handling of uncertainty, the time horizon of the analysis, and the discount rate. Other important considerations are the outcome measure that are used in the analysis, measurement of costs, measurement of effectiveness, and the overall transparency of the analysis (i.e., were the assumptions explicitly stated?).

Study sponsorship is gaining attention as concern has been raised over the potential for conflict of interest to bias the research, affect the design, and skew the interpretation of economic analyses of medical therapies. [27] [28] [29] The newness of pharmacoeconomics research and its potential effects on pharmaceutical company revenue make it particularly vulnerable to financial conflicts of interest. Researchers have examined the effects of conflict of interest on pharmacoeconomic research in breakthrough areas in oncology, such as hematopoietic colony-stimulating factors, antiemetics, and taxanes oncology. Published articles were classified according to qualitative conclusion, quantitative result, timing of study initiation, and funding source.

The first study found correlations between funding source and qualitative cost assessment, timing of study initiation, and discrepancies between qualitative conclusions and quantitative results. Favorable conclusions were reached by 81% of the pharmaceutical company-sponsored studies and 48% of the nonprofit-sponsored studies (P < 0.009). All of the studies that reached unfavorable conclusions had been sponsored by nonprofit organizations. Although nearly one fourth of the studies gave qualitative conclusions that overstated their quantitative results, this was not significantly greater for pharmaceutical company-sponsored studies than for nonprofit-sponsored studies.[27] More than 80% of economic studies funded by all sources were conducted after favorable clinical trial results were known. The findings of the study demonstrated a strong association between pharmaceutical company sponsorship and favorable economic assessments. Although there is no evidence of bias in individual articles, the results raised concerns about potential bias in pharmacoeconomic studies. Unfavorable cost profiles probably are underreported, and qualitative overstatements about the cost-effectiveness of new agents are not uncommon.

The second study addressed variations in study quality when the 44 pharmaceutical and not-for-profit-funded cost-effectiveness studies of the six breakthrough drugs in oncology were compared.[28]Investigators rated specific aspects of study reporting based on criteria from the U.S. Public Health Service Panel on Cost-effectiveness in Health and Medicine. Dissemination strategies were evaluated by using impact factor scores from the Science Citation Index. Operational aspects of pharmaceutical manufacturer–sponsored study reporting were better overall than were those associated with nonprofit-sponsored studies with respect to the following criteria: The results were more likely to be reported based on data obtained from randomized clinical trials or detailed cost models (90% versus 70%); to include descriptions of the source of cost differences (90% versus 79%); to state whether the study was carried out from a societal, governmental, or insurer perspective (70% versus 42%); and to indicate clearly the time period over which costs were evaluated (65% versus 50%). Nonprofit-sponsored studies were more likely than pharmaceutical-sponsored studies to report the generalizability of the findings to other clinical settings (58% versus 35%), to include statements on the statistical significance of the findings (38% versus 20%), and to clearly outline the cost per unit and data sources for the cost analyses (67% versus 45%). Most studies were published in low-impact-factor, peer-reviewed journals, and journal impact factor scores were similar between pharmaceutical- and nonprofit-sponsored studies. Overall, the study found differences in study reporting but not in types of journals where studies were published. These results, particularly with respect to differences in generalizability, may account in part for the finding that pharmaceutical manufacturer-sponsored studies were less likely to report unfavorable conclusions.

Strategies for Conducting Economic Analyses in Oncology

Because economic assessments are potentially useful as secondary endpoints in clinical trials of new cancer therapies or technologies,[19] cost analyses have been proposed alongside pivotal trials or new technologies. These are appropriate when investigational therapies are resource-intensive or when new technologies or treatments are likely to be used by large numbers of cancer patients. In most cases, these analyses are associated with phase III clinical trials, although inclusion in phase II trials can facilitate the collection of pilot data that can assist in study design for phase III trials. Phase IV studies also are possible sources of data for clinical and economic analyses. Estimates of the potential difference in costs between treatment arms can be helpful to coverage determinations.

The perspective of the economic analysis varies in many of the reported cost-effectiveness analyses. Most commonly, the perspective is that of the third-party payer; however, the societal perspective should also be considered.[21] Many clinical trials report economic data that are based on a “modified” societal perspective; that is, direct medical costs are quantified as societal costs, while direct nonmedical and productivity costs are not evaluated. Data collection can be prospective or retrospective. Prospective studies allow for timely assessment of clinical and economic outcomes at the end of the study. Detailed information on direct medical, direct nonmedical, productivity, and intangible costs can be obtained if careful planning is done before the study begins. Retrospective economic assessments are far less costly to conduct, but they include information on only a limited perspective (generally the third-party payer perspective).

The time interval from randomization to follow-up can influence the findings of the economic analysis, if follow-up is short. Ideally, the time horizon should be the same for the clinical and the economic analysis and should allow for realistic outcome measurement. Some economic analyses consider intermediate outcomes (e.g., response rate); other cost analyses are based on a review of final outcomes (e.g., survival in months or years). In either case, the outcomes should be identified prospectively. Despite increasing discussion about methodologies and practical approaches to including economic analyses in randomized clinical trials, remarkably few of these assessments have been reported in the literature. The National Cancer Institute (NCI)–sponsored Cooperative Trials Groups reported only one prospective and six retrospective economic analyses. [30] [31] [32] [33] [34] [35]

One can argue that cost data should be considered as important as clinical data and should be subjected to quality control assessments. Most of the literature related to cost analyses in cancer has been based on retrospective assessments of clinical trials, and these retrospective reviews generally have been funded by the pharmaceutical industry. Combining economic studies and clinical trials requires targeted funding, staff, and cooperation between clinical and economic researchers and analysts. Few independent nonprofit groups have the funds to supported economic analyses. Hence, concerns exist about researcher independence. One privately funded economic assessment of autologous stem cell transplant studies for breast cancer, for example, did not reveal the economic findings until after the clinical trial results were reported as negative. The savvy reader of economic analyses needs to think about the sponsor of a study and that group's motivation for undertaking a study and reporting its findings.

Economic Analyses Conducted by Cancer Clinical Trials Groups

During the 1990s, the NCI supported efforts to integrate economic analyses into cancer clinical trials. [19] [20] In 1994, the NCI sponsored a conference with representatives of cancer centers and cooperative groups that addressed the importance, appropriateness, and complexity of these evaluations. The next year, the American Society of Clinical Oncology (ASCO) established a Health Outcomes Working Group, which was charged with developing specific guidelines for implementing economic evaluations in cancer clinical trials.[20] In 1996, the NCI and ASCO convened a second meeting, attended by experts from the NCI-sponsored cooperative groups, NCI staff, and experts in the field of health economics, to consider the practical implementation of economic evaluation in cancer clinical trials. In 1998, a workbook that served as a developing guide designed to be used as practical reference for subsequent economic analyses of cancer clinical trials was published.[17] The first articles describing economic analyses alongside clinical trials conducted by the NCI-sponsored cooperative groups were published in 1997.

Table 24-1 provides information on seven NCI-sponsored studies that have been published to date. The following summary considers some of these real-life examples of the methods and operational considerations researchers face when conducting cancer clinical trials. The first cost-effectiveness study of an NCI-sponsored cooperative group trial was reported in 1997 by investigators who were affiliated with the Children's Cancer Study Group.[32] The clinical trial evaluated the clinical and cost-effectiveness of GCSF as an adjunctive therapy for children with acute lymphoblastic leukemia. The study randomized 164 children and found a reduction in the duration of neutropenia (5.3 versus 12.7 days) and duration of hospitalization (6 versus 10 days). Cost-minimization analyses indicated that GCSF did not add additional costs. Clinical and economic data were obtained directly from the clinical trial participants.


Table 24-1   -- Completed Economic Analyses Conducted by National Cancer Institute-Sponsored Clinical Trials Groups

Study

Treatment Arms

Effectiveness Time Frame

Economic Time Frame

Method

Outcome

Radiation Therapy Oncology Group 90-03 and 91-04[37]

Brain metastases; head and neck cancer

Not measured

Months

Cost estimation

90-03 costs estimated well; 91-04 costs not estimated very well

Eastern Cooperative Oncology Group 1490[35]

GM-CSF vs placebo for older

Weeks

Weeks

Cost minimization

GM-CSF was cost saving, and adults with AML decreased infections

Southwest Oncology Group 9031[33]

GCSF vs placebo for adults with AML

Weeks

Weeks

Cost minimization

GCSF did not add additional costs and decreased hospital stay

Southwest Oncology Group 9509[30]

Vinorelbine + cisplatin vs paclitaxel + carboplatin for non–small cell lung cancer

Months

24 months

Cost minimization

Cisplatin + vinorelbine is less costly

Children's Cancer Study Group[32]

GCSF vs placebo for children with leukemia

Weeks

Weeks

Cost minimization

GCSF did not add additional costs and was associated with shortened duration of neutropenia

Pediatrics Oncology Group[36]

GCSF vs control for children with leukemia

Weeks

Weeks

Cost minimization

GCSF did not add additional costs and was associated with shortened duration of neutropenia

Children's Cancer Study Group 1881, 1882, 1891, 1901, 1922, 1941[31]

Acute leukemia—various treatments

Years

Months

Cost effectiveness

Delayed intensification, augmented therapy, and dexamethasone therapy cost-effective vs treatment of first relapse

AML, acute myelogenous leukemia; GCSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte macrophage colony-stimulating factor.

 

 

 

The second economic analysis of an NCI-sponsored cooperative group trial was reported in 1999 by investigators who were affiliated with the Eastern Cooperative Oncology Group.[35] The randomized clinical trial evaluated the clinical effectiveness of the hematopoietic cytokine, granulocyte macrophage colony-stimulating factor (GM-CSF), as an adjunctive therapy for people 55 years of age or older who were receiving induction chemotherapy for acute myeloid leukemia. Patients receiving GM-CSF experienced a 72% reduction in severe infections, four fewer days with an absolute neutrophil count lower than 500 cells/mL, but no significant difference in the duration of hospitalization compared to the group taking the placebo. Decision analytic modeling was used to analyze the costs of GM-CSF use during induction therapy. Clinical probabilities of acquiring an infection were obtained from the clinical trial data; hospital costs per day for infected and uninfected patients were obtained from billing data from seven sites participating in the clinical trial. The significant improvements in rates of severe infections were associated with an estimated $2310 in cost savings per patient. The reduction in costs was particularly evident among individuals who received two cycles of induction chemotherapy. The study was conducted over a six-month period, with funding for the study provided by the pharmaceutical manufacturer.

A retrospective economic analysis of a randomized clinical trial conducted by the former Pediatrics Oncology Group[36] compared the costs of inpatient supportive care for pediatric patients (age 1 to 22 years) with T-cell leukemia and advanced lymphoblastic lymphoma. Patients received either GCSF (n = 45) or no GCSF (n = 43) following induction and two cycles of maintenance therapy. During maintenance therapy, the patients receiving GCSF had significantly fewer days to an absolute neutrophil count above 500 cells/mL and a trend toward fewer days of hospitalization. The study found that the total median costs of supportive care were similar for all patients in the study. Data on resource utilization were tabulated from case report forms, and costs were derived from national data on hospitalization costs, average wholesale prices of pharmaceuticals, and patient billing information from a single institution. The four-month study was conducted with support from the Cooperative Clinical Trial Group and the pharmaceutical supplier of the study drug.

In 2001, the Radiation Therapy Oncology Group (RTOG)[37] initial pilot study addressed four aims: (1) measurement of radiation therapy treatment costs for patients treated in different arms of two randomized controlled clinical trials, (2) comparison of measured costs to those predicted by an economic model, (3) examination of the distribution of costs among patients treated on the same arm, and (4) assessment of the feasibility of retrospective data collection effort. The RTOG selected two phase III clinical trials to evaluate in this pilot effort. The first study, RTOG 91-04, compared standard treatment to a total dose of 30 Gy with a second arm of accelerated hyperfractionation to a total dose of 54.4 Gy for cancer patients with brain metastases. The second study, RTOG 90-03, was a phase III study of patients with squamous cell carcinomas of the head and neck who received either standard fractionation to a dose of 70 Gy, hyperfractionation to a dose of 81.6 Gy, accelerated fractionation with a split to a total dose of 67.2 Gy, or accelerated fractionation with a concomitant boost to a total dose of 72 Gy. Expected quantities of procedure codes and relative value units associated with Medicare billing efforts were modeled. The median and mean relative value units were within the range predicted by an economic model for all arms of the head and neck cancer study but were above the predicted range for the brain cancer study. Some of the researchers encountered considerable difficulties in collecting the retrospective economic data, suggesting that prospective data collection might be the better strategy for economic analysis of RTOG studies. Clinical trials with complex treatment protocols, such as the head and neck cancer study, appeared particularly difficult to include in retrospective economic analytic efforts.

Nonfinancial metrics can be reported. One study, for example, reported on the feasibility of using duration of hospitalization as a surrogate for cost and event-free survival as a measure of effectiveness to estimate cost-effectiveness ratios of various treatment regimens evaluated in clinical trials of children with acute lymphoblastic leukemia.[26] Metrics included marginal cost-effectiveness estimates of the number of days per patient for delayed intensification and for augmented therapy and relapse-adjusted “savings” associated with augmented therapy (16 days) and with dexamethasone-based therapy (82 days). The study was supported by grants to the office of the cooperative group's chairman and indicated that retrospective economic analyses were feasible, provided that the economic analyses focused on duration of hospitalization.

Researchers affiliated with the Southwest Oncology Group conducted the first prospective economic analysis of a randomized clinical trial conducted by an NCI-sponsored cooperative clinical trial group.[30] The clinical trial included patients who were randomized to receive cisplatin plus vinorelbine versus carboplatin plus paclitaxel. The findings from this study indicated that prospective economic analyses could be conducted alongside randomized clinical trials, although these efforts did require external funding and committed resources from the statistical operations center of the cooperative clinical trial group.

This review of the design, outcome metrics used, and sponsorship of these studies is instructive in identifying potential opportunities and obstacles for future research. Most studies were based on retrospective economic analyses. Funding sources were diverse, most of the analytic efforts being supported by pharmaceutical suppliers.

CONCLUSIONS

U.S. health care expenditure on cancer care is greater than expenditure for any other illness. It is important to consider both the clinical benefit and the economic value provided by cancer treatment interventions, particularly during their development. Economic tools (cost-benefit, cost-effectiveness, cost-utility, and cost-identification analyses) can provide such information. Despite increases in coverage for patients in “qualifying” clinical trials by Medicare, several private insurers, and many U.S. states, few robust economic analyses have been conducted in trials sponsored by the NCI. Clinical trials that incorporate cost analyses can contribute to an understanding of the value of an intervention and can do so from a variety of perspectives.

Costs of cancer care can be evaluated in a variety of clinical settings. Economic considerations have important practical implications for oncologists and should be addressed for most new cancer treatments and procedures. Continuing economic pressures facing the health care system in the United States serve as an important reminder of the importance of unbiased cost analyses, particularly for cancer treatments. Despite nearly universal support for economic information, the paucity of cost-effectiveness analyses, especially studies that are not funded by the pharmaceutical industry, is particularly worrisome. Policymakers and physicians strive to make informed decisions about rational allocations of cancer resources. Economic data, if properly obtained, can assist in these decisions.

REFERENCES

  1. Brown ML, Lipscomb J, Snyder C: The burden of illness of cancer: economic cost and quality of life.  Annu Rev Pub Health2001; 22:91-113.
  2. Brown ML, Riley GF, Schusser N, Etzoni R: Estimating health care costs related to cancer treatment from SEER-Medicare data.  Medical Care2002; 40S:IV-104-IV-117.
  3. American Cancer Society : Cost of cancer on the rise. ACS News Center.  Available from: http://www.cancer.org/docroot/NWS/content?NWS_1_1x_Cost_of_Cancer_On_the_rise
  4. Schulman KA, Glick HA, Yabroff R, Eisenberg JM: Introduction to clinical economics: assessment of cancer therapies.  J Natl Cancer Inst Monogr1995; 19:1-9.
  5. Calhoun EA, Chang C-H, Welshman E, et al: Evaluating the total costs of chemotherapy-induced neutropenia: results from a pilot study with ovarian cancer patients.  Oncologist2001; 6:441-445.
  6. Calhoun EA, Bennett CL: Evaluating the total costs of cancer: The Northwestern University Costs of Cancer Program.  Oncology2003; 17:109-114.
  7. Estimated New Cancer Cases and Deaths 2006. Table I1. SEER Cancer Review. Available from: seer.cancer.gov/csr/1975_2003/results_single/sect_01_table.01.pdf
  8. Wagner JL, Alberts SR, Sloan JA, et al: Incremental costs of enrolling patients in clinical trials: a population based study.  J Natl Cancer Inst1999; 91:847-853.Erratum in: J Natl Cancer Inst 2000;92:164–165.
  9. Bennett CL, Stinson TJ, Vogel V, et al: Evaluating the financial impact of clinical trials in oncology: results from a pilot study from the Association of American Cancer Institutes/Northwestern University Clinical Trials Costs and Charges Project.  J Clin Oncol2000; 18:2805-2810.
  10. Goldman DP, Berry SH, McCabe MH, et al: Incremental costs in National Cancer Institute sponsored clinical trials.  JAMA2003; 289:2970-2977.
  11. Quirk J, Schrag D, Radzyner M, et al: Clinical trial costs are similar to and may be less than standard care and inpatient charges at an academic medical center are similar to major, minor, and non-teaching hospitals.  Proc Am Soc Clin Oncol2000; 19:433a.
  12. Bennett CL, Adams JR, Knox KS, et al: Clinical trials: are they a good buy?.  J Clin Oncol2001; 19:4330-4339.
  13. In: Aaron HJ, Gelband H, ed. Institute of Medicine Institute Report: Extending Medicare Reimbursement in Clinical Trials,  Washington, DC: National Academy Press; 2000.
  14. U.S. General Accounting Office: NIH Clinical Trials : Various Factors Affect Patient Participation. Publication No. GAO/HEHS-99-1821,  Washington, DC, U.S. General Accounting Office, 1999.
  15. National Cancer Institute, U.S. National Institutes of Health: Clinical trials covered under the Medicare anticancer drugs national coverage decision. Available from:http://www.cancer.gov/clinicaltrials/developments/NCD179N
  16. Medicare coverage: Clinical trials. Provider Bulletin.  Available from: http://www.cms.hhs.gov/ClinicalTrialPolicies/Downloads/providerbulletin.pdf
  17. Stuart B, Briesacher BA, Shea DG, et al: Riding the rollercoaster: the ups and downs in out-of-pocket spending under the standard Medicare drug benefit.  Health Aff (Millwood)2005; 24:1022-1031.
  18. National Cancer Institute : States that require health plans to cover patient care costs in clinical trials.  Available from: http:/www.cancer.gov/clinicaltrials/developments/laws-about-clinical-trial-costs
  19. Brown M, McCabe M, Schulman KA: Integrating economic analysis into cancer clinical trials: The National Cancer Institute–American Society of Clinical Oncology Economics Workbook.  J Natl Cancer Inst Monogr1998; 24:1-84.
  20. Schulman KA, Glick HA, Yabroff R, Eisenberg JM: Introduction to clinical economics: assessment of cancer therapies.  J Natl Cancer Inst Monogr1995; 19:1-9.
  21. Finkler SA: The distinction between cost and charges.  Ann Intern Med1982; 96:102-109.
  22. Reeder CE, Gordon D: Managing oncology costs.  Am J Manag Care2006; 12(suppl):S3-S16.quiz S17–S9
  23. American Cancer Society : Making a Part D Plan Decision.  Available from: http://www.cancer.org/docroot/MIT/content?MIT_3_1_Making_a_Plan_Decision.asp?sit
  24. Warren JL, Brown ML, Fay MP, et al: Costs of treatment for elderly women with early stage breast cancer in fee-for-service settings.  J Clin Oncol2001; 20:307-316.
  25. In: Gold MR, Siegal JE, Russel LB, Weinstein MC, ed. Cost-Effectiveness in Health and Medicine,  New York: Oxford University Press; 1996.
  26. Chiou CF, Hay JW, Wallace JF, et al: Development and validation of a grading system for the quality of cost-effectiveness studies.  Med Care2003; 41:32-44.
  27. Friedberg M, Saffran B, Stinson TJ, et al: Evaluation of conflict of interest in economic analyses of new drugs used in oncology.  JAMA1999; 282:1453-1457.
  28. Knox KS, Adams JR, Djulbegovic B, et al: Quality and dissemination of industry sponsored economic analyses of six novel drugs used in oncology.  Ann Oncol2000; 11:1591-1595.
  29. Azimi NA, Welch G: The effectiveness of cost-effectiveness analysis in containing costs.  J Gen Intern Med1998; 12:664-669.
  30. Ramsey SD, Moinpour CM, Lovato LC, et al: Economic analysis of vinorelbine plus cisplatin versus paclitaxel plus carboplatin for advanced non-small cell lung cancer.  J Natl Cancer Inst2002; 94:291-297.
  31. Gaynon PS, Bostrom BC, Hutchinson RJ, et al: Duration of hospitalization as a measure of cost on Children's Cancer Group acute lymphoblastic leukemia studies.  J Clin Oncol2001; 19:1916-1925.
  32. Pui CH, Boyett JM, Hughes WT, et al: Human granulocyte colony stimulating factor after induction chemotherapy in children with acute lymphoblastic leukemia.  N Engl J Med1997; 336:1781-1787.
  33. Bennett CL, Hynes D, Godwin J, et al: Economic analysis of granulocyte colony stimulating as adjunct therapy for older patients with acute myelogenous leukemia (AML): estimates from a Southwest Oncology Group clinical trial.  Cancer Investigation2001; 9:603-610.
  34. Bennett CL, Golub R, Waters TM, et al: Economic analyses of phase III cooperative cancer group clinical trials: are they feasible?.  Cancer Investigation1997; 15:227-236.
  35. Bennett CL, Stinson TJ, Tallman MS, et al: Economic analysis of a randomized placebo-controlled phase III study of granulocyte macrophage colony stimulating factor in adult patients (>55 to 70 years of age) with acute myelogenous leukemia: Eastern Cooperative Oncology Group (E1490).  Ann Oncol1999; 10:177-182.
  36. Bennett CL, Stinson TJ, Lane D, et al: A cost-analysis of filgrastim for the prevention of neutropenia in pediatric T-cell leukemia and advanced lymphoma: a case for prospective economic analysis in cooperative group trials.  Med Pediatr Oncol2000; 34:92-96.
  37. Owen JB, Grigsby PW, Caldwell TM, et al: Can costs be measured and predicted by modeling within a cooperative clinical trials group: economic methodologic pilot studies of the Radiation Therapy Oncology Group studies 90-03 and 91-04.  Int J Radiat Oncol Biol Phys2001; 49:633-639.