Taking account of future technology in cost effectiveness analysis
http://www.100md.com
《英国医生杂志》
1 Department of Population and International Health, Harvard School of Public Health, Harvard Center for Population and Development Studies, 9 Bow Street, Cambridge MA 02138, USA, 2 Department of Health Policy and Management, Harvard School of Public Health, Boston MA 02115, USA
Correspondence to: J A Salomon jsalomon@hsph.harvard.edu
Economic evaluations in health and medicine usually ignore the possibility of future advances in treatment. But when technological innovation is rapid such considerations can have major implications
Introduction
An estimated 2.7 million Americans and 6.7 million Europeans have chronic HCV infection and are at risk of cirrhosis, end stage liver disease, and liver cancer.7-9 Treatment decisions are complicated by variable progression rates and require difficult trade-offs between costs, side effects, and uncertain clinical benefits.10-12 Various treatments have emerged in recent years, and a series of decision analyses have examined their cost effectiveness (table 1). Analyses typically have found that each new treatment has an attractive cost effectiveness ratio compared with many common interventions. Most ratios, in fact, have fallen below $10 000 (￡5607, 8218)/QALY compared with the next most effective option. Given the rapid evolution of treatments for HCV infection, this example offers an illustration of how anticipated technological changes can be used to enrich conventional cost effectiveness analyses.
Table 1 Evolving treatment regimens for chronic hepatitis C virus and findings on cost effectiveness
Modelling the natural course of infection
We begin by travelling back in time to 1996, to revisit the decision problem confronting a patient with chronic HCV infection considering interferon monotherapy, the only approved treatment at that time. Adopting the conventional assumption that the "no treatment" strategy excludes the downstream potential for improved therapies, we calculated the incremental cost effectiveness ratio of monotherapy as $8700/QALY gained compared with no treatment (see bmj.com for details of the model). With the benefit of hindsight, however, we might ask whether no treatment was the only relevant comparator. In other words, should the alternatives to immediate treatment also have included deferred treatment?
A previous study considered one important element of this question, investigating watchful waiting with periodic liver biopsy versus immediate empirical treatment for chronic HCV infection.17 The authors found that immediate treatment was cost effective compared with biopsy management. We focus on an additional facet of this question—waiting and watching for technological innovation. For a patient in 1996, how would anticipation of imminent advances in treatment have changed the decision problem?
Perfect foresight scenario
The above analysis is based on an unrealistic assumption of perfect knowledge of future treatments. However, uncertainty about emerging treatments can be incorporated in a decision analysis framework in the same way that other uncertain outcomes, such as developing cirrhosis, are captured. Define p as the probability of a new treatment being available in three years (with the specifications of combination therapy). In the deferred treatment strategy, we assume that individuals receive combination therapy in year 3 with probability p, or will fall back on monotherapy if no new treatment emerges that year, with probability (1-p). In the immediate treatment strategy, individuals receive monotherapy now, but may (with probability p) have access to combination therapy in year 3, in the event of non-response or relapse. Allowing for this uncertainty, immediate treatment remains dominated by deferred treatment provided that p > 50%. The probability of a new treatment must be 30% or lower for the incremental cost effectiveness of immediate monotherapy to fall below $50 000/QALY (table 3).
Table 3 Incremental cost effectiveness of immediate therapy under different prospects for improved therapy
Summary points
Cost effectiveness analyses normally do not take account of possible future advances in treatment
Accounting for such possibilities can alter the conclusions of a cost effectiveness analysis greatly
Uncertainties about new treatment can be reflected in a decision analysis in a way that is comparable with the modelling of other uncertain events
We have examined different scenarios regarding the timing and likelihood of better treatment separately, but we may also combine these two dimensions. For example, imagine that the cumulative probability of an improved treatment rises linearly over time so that there is a 10% chance of the new treatment emerging in one year, 20% in two years, and so on, up to 50% in five years. In this scenario, the incremental costs of immediate treatment compared with deferred treatment would be $800, with incremental benefits of 0.007 QALYs (less than 3 days), implying an incremental cost effectiveness ratio of more than $115 000/QALY—that is, more than a 13-fold increase over the assumption of no potential improvements in treatment.
Conclusion
Weinstein MC, Stason WB. Foundations of cost effectiveness analysis for health and medical practices. N Engl J Med 1977;296: 716-21.
Sculpher M, Drummond M, O'Brien B. Effectiveness, efficiency, and NICE. BMJ 2001;322: 943-4.
Birch S, Gafni A. The NICE approach to technology assessment: an economics perspective. Health Care Manage Sci 2004;7: 35-41.
Hammer SM. Increasing choices for HIV therapy. N Engl J Med 2002;346: 2022-3.
Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost effectiveness in health and medicine. New York: Oxford University Press, 1996.
Evans C, Tavakoli M, Crawford B. Use of quality adjusted life years and life years gained as benchmarks in economic evaluations: a critical appraisal. Health Care Manage Sci 2004;7: 43-9.
Alter MJ, Kruszon-Moran D, Nainan OV, McQuillan GM, Gao F, Moyer LA, et al. The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. N Engl J Med 1999;341: 556-62.
Ray KW. Global epidemiology and burden of hepatitis C. Microbes Infect 2002;4: 1219-25.
Tong MJ, el Farra NS, Reikes AR, Co RL. Clinical outcomes after transfusion-associated hepatitis C. N Engl J Med 1995;332: 1463-6.
Seeff LB. Natural history of hepatitis C. Am J Med 1999;107: 5-10S.
Alter HJ, Seeff LB. Recovery, persistence, and sequelae in hepatitis C virus infection: a perspective on long-term outcome. Semin Liver Dis 2000;20: 17-35.
Salomon JA, Weinstein MC, Hammitt JK, Goldie SJ. Cost effectiveness of treatment for chronic hepatitis C infection in an evolving patient population. JAMA 2003;290: 228-37.
Bennett WG, Inoue Y, Beck JR, Wong JB, Pauker SG, Davis GL. Estimates of the cost effectiveness of a single course of interferon- alpha 2b in patients with histologically mild chronic hepatitis C. Ann Intern Med 1997;127: 855-65.
The French METAVIR Cooperative Study Group. Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. Hepatology 1994;20: 15-20.
Briggs A, Sculpher M. An introduction to Markov modelling for economic evaluation. Pharmacoeconomics 1998;13: 397-409.
Drummond M, McGuire A, eds. Economic evaluation in health care: merging theory with practice. New York: Oxford University Press, 2001.
Wong JB, Koff RS. Watchful waiting with periodic liver biopsy versus immediate empirical therapy for histologically mild chronic hepatitis C: a cost effectiveness analysis. Ann Intern Med 2000;133: 665-75.(Joshua A Salomon, assista)
Correspondence to: J A Salomon jsalomon@hsph.harvard.edu
Economic evaluations in health and medicine usually ignore the possibility of future advances in treatment. But when technological innovation is rapid such considerations can have major implications
Introduction
An estimated 2.7 million Americans and 6.7 million Europeans have chronic HCV infection and are at risk of cirrhosis, end stage liver disease, and liver cancer.7-9 Treatment decisions are complicated by variable progression rates and require difficult trade-offs between costs, side effects, and uncertain clinical benefits.10-12 Various treatments have emerged in recent years, and a series of decision analyses have examined their cost effectiveness (table 1). Analyses typically have found that each new treatment has an attractive cost effectiveness ratio compared with many common interventions. Most ratios, in fact, have fallen below $10 000 (￡5607, 8218)/QALY compared with the next most effective option. Given the rapid evolution of treatments for HCV infection, this example offers an illustration of how anticipated technological changes can be used to enrich conventional cost effectiveness analyses.
Table 1 Evolving treatment regimens for chronic hepatitis C virus and findings on cost effectiveness
Modelling the natural course of infection
We begin by travelling back in time to 1996, to revisit the decision problem confronting a patient with chronic HCV infection considering interferon monotherapy, the only approved treatment at that time. Adopting the conventional assumption that the "no treatment" strategy excludes the downstream potential for improved therapies, we calculated the incremental cost effectiveness ratio of monotherapy as $8700/QALY gained compared with no treatment (see bmj.com for details of the model). With the benefit of hindsight, however, we might ask whether no treatment was the only relevant comparator. In other words, should the alternatives to immediate treatment also have included deferred treatment?
A previous study considered one important element of this question, investigating watchful waiting with periodic liver biopsy versus immediate empirical treatment for chronic HCV infection.17 The authors found that immediate treatment was cost effective compared with biopsy management. We focus on an additional facet of this question—waiting and watching for technological innovation. For a patient in 1996, how would anticipation of imminent advances in treatment have changed the decision problem?
Perfect foresight scenario
The above analysis is based on an unrealistic assumption of perfect knowledge of future treatments. However, uncertainty about emerging treatments can be incorporated in a decision analysis framework in the same way that other uncertain outcomes, such as developing cirrhosis, are captured. Define p as the probability of a new treatment being available in three years (with the specifications of combination therapy). In the deferred treatment strategy, we assume that individuals receive combination therapy in year 3 with probability p, or will fall back on monotherapy if no new treatment emerges that year, with probability (1-p). In the immediate treatment strategy, individuals receive monotherapy now, but may (with probability p) have access to combination therapy in year 3, in the event of non-response or relapse. Allowing for this uncertainty, immediate treatment remains dominated by deferred treatment provided that p > 50%. The probability of a new treatment must be 30% or lower for the incremental cost effectiveness of immediate monotherapy to fall below $50 000/QALY (table 3).
Table 3 Incremental cost effectiveness of immediate therapy under different prospects for improved therapy
Summary points
Cost effectiveness analyses normally do not take account of possible future advances in treatment
Accounting for such possibilities can alter the conclusions of a cost effectiveness analysis greatly
Uncertainties about new treatment can be reflected in a decision analysis in a way that is comparable with the modelling of other uncertain events
We have examined different scenarios regarding the timing and likelihood of better treatment separately, but we may also combine these two dimensions. For example, imagine that the cumulative probability of an improved treatment rises linearly over time so that there is a 10% chance of the new treatment emerging in one year, 20% in two years, and so on, up to 50% in five years. In this scenario, the incremental costs of immediate treatment compared with deferred treatment would be $800, with incremental benefits of 0.007 QALYs (less than 3 days), implying an incremental cost effectiveness ratio of more than $115 000/QALY—that is, more than a 13-fold increase over the assumption of no potential improvements in treatment.
Conclusion
Weinstein MC, Stason WB. Foundations of cost effectiveness analysis for health and medical practices. N Engl J Med 1977;296: 716-21.
Sculpher M, Drummond M, O'Brien B. Effectiveness, efficiency, and NICE. BMJ 2001;322: 943-4.
Birch S, Gafni A. The NICE approach to technology assessment: an economics perspective. Health Care Manage Sci 2004;7: 35-41.
Hammer SM. Increasing choices for HIV therapy. N Engl J Med 2002;346: 2022-3.
Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost effectiveness in health and medicine. New York: Oxford University Press, 1996.
Evans C, Tavakoli M, Crawford B. Use of quality adjusted life years and life years gained as benchmarks in economic evaluations: a critical appraisal. Health Care Manage Sci 2004;7: 43-9.
Alter MJ, Kruszon-Moran D, Nainan OV, McQuillan GM, Gao F, Moyer LA, et al. The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. N Engl J Med 1999;341: 556-62.
Ray KW. Global epidemiology and burden of hepatitis C. Microbes Infect 2002;4: 1219-25.
Tong MJ, el Farra NS, Reikes AR, Co RL. Clinical outcomes after transfusion-associated hepatitis C. N Engl J Med 1995;332: 1463-6.
Seeff LB. Natural history of hepatitis C. Am J Med 1999;107: 5-10S.
Alter HJ, Seeff LB. Recovery, persistence, and sequelae in hepatitis C virus infection: a perspective on long-term outcome. Semin Liver Dis 2000;20: 17-35.
Salomon JA, Weinstein MC, Hammitt JK, Goldie SJ. Cost effectiveness of treatment for chronic hepatitis C infection in an evolving patient population. JAMA 2003;290: 228-37.
Bennett WG, Inoue Y, Beck JR, Wong JB, Pauker SG, Davis GL. Estimates of the cost effectiveness of a single course of interferon- alpha 2b in patients with histologically mild chronic hepatitis C. Ann Intern Med 1997;127: 855-65.
The French METAVIR Cooperative Study Group. Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. Hepatology 1994;20: 15-20.
Briggs A, Sculpher M. An introduction to Markov modelling for economic evaluation. Pharmacoeconomics 1998;13: 397-409.
Drummond M, McGuire A, eds. Economic evaluation in health care: merging theory with practice. New York: Oxford University Press, 2001.
Wong JB, Koff RS. Watchful waiting with periodic liver biopsy versus immediate empirical therapy for histologically mild chronic hepatitis C: a cost effectiveness analysis. Ann Intern Med 2000;133: 665-75.(Joshua A Salomon, assista)