Cost-Effectiveness of Screening for HIV
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Sanders et al.1 and Paltiel et al.,2 in their reports on the cost-effectiveness of screening for HIV infection (Feb. 10 issue), have added another level of evidence in support of broader HIV testing. It appears that the benefits of routine counseling and testing will be enhanced if screening somehow leads to a decreased incidence of new infections. However, the frequently used procedure of one-time counseling regarding HIV3 and the awareness of being HIV-infected4 do not necessarily reduce risky behavior. Furthermore, although the recent report of 90 percent acceptance of HIV testing in urgent care settings is encouraging,5 the reproducibility of the results needs to be ascertained in continuity (long-term care) clinics, where patients are not acutely ill and are probably less likely to accede to HIV testing. Given the benefits of broader HIV screening, these and other outstanding problems call for studies that will assist in the process of designing feasible testing procedures that can be used in different clinical and nonclinical settings to increase the early diagnosis of HIV infection and to help prevent new cases. Perhaps the programs should be called "HIV diagnosis and prevention" instead of "HIV counseling and testing."
Babafemi O. Taiwo, M.D.
Northwestern University Medical School
Chicago, IL 60611
b-taiwo@northwestern.edu
References
Sanders GD, Bayoumi AM, Sundaram V, et al. Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy. N Engl J Med 2005;352:570-585.
Paltiel AD, Weinstein MC, Kimmel AD, et al. Expanded screening for HIV in the United States -- an analysis of cost-effectiveness. N Engl J Med 2005;352:586-595.
Clark LR, Brasseux C, Richmond D, Getson P, D'Angelo LJ. Effect of HIV counseling and testing on sexually transmitted diseases and condom use in an urban adolescent population. Arch Pediatr Adolesc Med 1998;152:269-273.
Simoni JM, Pantalone DW. Secrets and safety in the age of AIDS: does HIV disclosure lead to safer sex? Top HIV Med 2004;12:109-118.
Walensky RP, Losina E, Malatesta L, et al. Effective HIV case identification through routine HIV screening at urgent care centers in Massachusetts. Am J Public Health 2005;95:71-73.
To the Editor: Routine HIV testing is an important population health strategy that is cost-effective and should be strongly promoted, as Bozzette1 points out in his editorial on the studies by Sanders et al. and Paltiel et al. Such broad-based approaches, however, require critical examination with respect to potential unintended consequences.2 For example, routine HIV testing may ultimately give rise to wider social acceptance of persons living with HIV or AIDS.1 Without appropriate safeguards or support systems, however, the stigma associated with HIV and AIDS may suppress the use of health care services1 and thus exacerbate health disparities in high-risk, marginalized populations.2,3 Such unintended consequences would limit the benefits of routine HIV testing.
Recommendations regarding routine HIV testing must be placed within the current context of recent policy shifts away from the primary prevention of HIV infection and in the context of waning federal support for social programs. The effects of these factors must be considered because they limit access to care and do little to reduce the incidence of infection among those at greatest risk. We encourage the adoption of comprehensive HIV–AIDS initiatives that include routine screening yet do not compromise nonclinical primary-prevention efforts and that are also concerned with reducing stigmatization.
Angela D. Thrasher, M.P.H.
Chandra L. Ford, M.P.H.
University of North Carolina School of Public Health
Chapel Hill, NC 27599-7440
adthrash@email.unc.edu
Kathryn A. Nearing, M.A.
University of Colorado Health Sciences Center
Denver, CO 80262
References
Bozzette SA. Routine screening for HIV infection -- timely and cost-effective. N Engl J Med 2005;352:620-621.
Mechanic D. Disadvantage, inequality, and social policy: major initiatives intended to improve population health may also increase health disparities. Health Aff (Millwood) 2002;21:48-59.
Aggleton P, Parker R. A conceptual framework and basis for action: HIV/AIDS stigma and discrimination. Geneva: Joint United Nations Programme on HIV/AIDS, 2002.
To the Editor: Although the methods used in the deterministic cost-effectiveness analysis by Sanders et al. do not appear contentious, at least two aspects of the incremental cost-effectiveness ratio (ICER) can threaten the validity of the results.
First, the end result of a fraction (ICER = change in cost ÷ change in effectiveness) becomes unstable in the presence of small denominators (range, 0.002 to 0.015 year). Consequently, a minimal error in estimating life expectancy or quality-adjusted life-years would have a substantial effect on the ICER and could even change the conclusions of the study. An analysis based on net health benefit is an alternative method of assessing cost-effectiveness that has several advantages over use of the ICER.1 Second, measurement of uncertainty in cost-effectiveness analysis permits statistical inference and helps policymakers decide whether further research is still needed.2,3
I agree that one-time screening for HIV is the optimal alternative, but only if the parameter estimates obtained by the investigators are extremely close to the true unknown values. Measurement of uncertainty permits the value of information to be estimated and should be routinely performed in cost-effectiveness analyses.
Eduardo da Silveira, M.D.
McGill University
Montreal, QC H2X 3R4, Canada
References
O'Hagan A, Stevens JW. Bayesian methods for design and analysis of cost-effectiveness trials in the evaluation of health care technologies. Stat Methods Med Res 2002;11:469-490.
Claxton K, Neumann PJ, Araki S, Weinstein MC. Bayesian value-of-information analysis: an application to a policy model of Alzheimer's disease. Int J Technol Assess Health Care 2001;17:38-55.
Fenwick EK, Claxton K, Sculpher M. Representing uncertainty: the role of cost-effectiveness acceptability curves. Health Econ 2001;10:779-787.
Dr. Sanders and colleagues reply: Dr. Taiwo accurately points out the importance of HIV counseling in reducing an HIV-infected patient's risk-related behavior. Although such programs improve the cost-effectiveness of screening by decreasing future HIV transmission, our analysis emphasizes that HIV screening can be economically favorable even when the costs and benefits associated with transmission are ignored. Ms. Thrasher and colleagues raise important points about stigma, discrimination, and subsequent health-seeking behavior. These complex issues were not included in our economic analysis, but we agree that they are important for policymaking. In addition, Thrasher et al. are correct in noting that we did not examine the question of whether HIV screening is cost-effective as compared with increased use of primary-prevention methods, since that was not the purpose of our analysis.
We thank Dr. da Silveira for his thoughtful comments about the methods and limitations of cost-effectiveness analysis. We agree that the cost-effectiveness ratio can be sensitive to small changes in the denominator when the incremental benefit is small. In our analysis, although the differences in quality-adjusted life-years were small for the entire screened population, the increase in quality-adjusted life-years for an individual HIV-infected patient was quite large (approximately 1.5 years) and robust. In addition, we believe that the suggested net-health-benefit method is an alternative method of expressing the results of a cost-effectiveness analysis, not an alternative method of assessing the cost-effectiveness of an intervention.
Although the net health benefit has many appealing features, it has two main limitations. First, reporting a net health benefit requires the analyst to choose a societal willingness-to-pay value for a quality-adjusted life-year — a necessity that remains controversial. Second, interpreting a net-health-benefit value is not intuitive for many readers. We also agree that estimation of the value of information can be quite useful, particularly with respect to screening.1 However, calculation of the value of information can be challenging, particularly when models are complex, and it can also be inaccurate when many variables in the model are collinear, as in our case. These methodologic issues need to be addressed for the value of information to be used routinely. Meanwhile, deterministic models such as ours remain valuable sources of evidence for policymakers and provide estimates of the incremental cost-effectiveness of interventions that can be used in guiding policy decisions.2
Gillian D. Sanders, Ph.D.
Duke University
Durham, NC 27705
gillian.sanders@duke.edu
Ahmed M. Bayoumi, M.D.
St. Michael's Hospital
Toronto, ON M5B 1W8, Canada
Douglas K. Owens, M.D.
Veterans Affairs Palo Alto Healthcare System
Palo Alto, CA 94304
References
Owens DK, Nease RF Jr. A normative analytic framework for development of practice guidelines for specific clinical populations. Med Decis Making 1997;17:409-426.
Claxton K. The irrelevance of inference: a decision-making approach to the stochastic evaluation of health care technologies. J Health Econ 1999;18:341-364.
Dr. Paltiel and colleagues reply: Dr. Taiwo calls for proof that expanded HIV-testing procedures can feasibly be implemented in a variety of real-world settings. Our research group and others have already mounted successful demonstration projects in both inpatient and outpatient settings showing that expanded HIV counseling and testing programs are feasible, affordable, and able to identify substantial numbers of new cases of infection.1,2,3,4,5 These studies set the stage for broader implementation of HIV testing and have already demonstrated that acceptance of the test increases with time.
Dr. Taiwo also notes that the attractiveness of HIV testing improves if it decreases the incidence of new infections. This is certainly true, but it is not a necessary condition. Our analysis establishes that routine HIV testing is cost-effective according to U.S. standards, even when viewed strictly from the perspective of the individual infected person. Demonstrating that expanded HIV testing reduces rates of HIV transmission would strengthen an already strong case.
Ms. Thrasher and colleagues express the concern that stigmas and cutbacks in funding for prevention and treatment programs may exacerbate health disparities in populations where routine HIV-testing services are expanded. Although we share this concern, there is little evidence that we are aware of either to support or to refute it. Routine HIV testing may very well serve to destigmatize both HIV testing and HIV disease. Whether or not the concern expressed by Thrasher et al. is borne out in practice, it is our view that the appropriate response to waning federal support for HIV prevention and treatment is not to persist in the pursuit of an outdated approach to HIV testing — an approach that ignores the availability of highly accurate, inexpensive screening tests and affordable, life-sustaining therapies. Rather, the appropriate response is to continue to press on all fronts for a coordinated, evidence-based, national policy on HIV and AIDS that links funding for expanded testing with funding for patient care that conforms to national guidelines and that embraces interventions — both preventive and therapeutic — that deliver good value.
A. David Paltiel, Ph.D.
Yale University
New Haven, CT 06520
david.paltiel@yale.edu
Rochelle P. Walensky, M.D., M.P.H.
Kenneth A. Freedberg, M.D.
Massachusetts General Hospital
Boston, MA 02114
References
Walensky RP, Losina E, Steger-Craven KA, Freedberg KA. Identifying undiagnosed human immunodeficiency virus: the yield of routine, voluntary inpatient testing. Arch Intern Med 2002;162:887-892.
Pincus JM, Crosby SS, Losina E, King ER, LaBelle C, Freedberg KA. Acute human immunodeficiency virus infection in patients presenting to an urban urgent care center. Clin Infect Dis 2003;37:1699-1704.
Kelen GD, Shahan JB, Quinn TC. Emergency department-based HIV screening and counseling: experience with rapid and standard serologic testing. Ann Emerg Med 1999;33:147-155.
Routinely recommended HIV testing at an urban urgent-care clinic -- Atlanta, Georgia, 2000. MMWR Morb Mortal Wkly Rep 2001;50:538-541.
Walensky RP, Losina E, Malatesta L, et al. Effective HIV case identification through routine HIV screening at urgent care centers in Massachusetts. Am J Public Health 2005;95:71-73.
Babafemi O. Taiwo, M.D.
Northwestern University Medical School
Chicago, IL 60611
b-taiwo@northwestern.edu
References
Sanders GD, Bayoumi AM, Sundaram V, et al. Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy. N Engl J Med 2005;352:570-585.
Paltiel AD, Weinstein MC, Kimmel AD, et al. Expanded screening for HIV in the United States -- an analysis of cost-effectiveness. N Engl J Med 2005;352:586-595.
Clark LR, Brasseux C, Richmond D, Getson P, D'Angelo LJ. Effect of HIV counseling and testing on sexually transmitted diseases and condom use in an urban adolescent population. Arch Pediatr Adolesc Med 1998;152:269-273.
Simoni JM, Pantalone DW. Secrets and safety in the age of AIDS: does HIV disclosure lead to safer sex? Top HIV Med 2004;12:109-118.
Walensky RP, Losina E, Malatesta L, et al. Effective HIV case identification through routine HIV screening at urgent care centers in Massachusetts. Am J Public Health 2005;95:71-73.
To the Editor: Routine HIV testing is an important population health strategy that is cost-effective and should be strongly promoted, as Bozzette1 points out in his editorial on the studies by Sanders et al. and Paltiel et al. Such broad-based approaches, however, require critical examination with respect to potential unintended consequences.2 For example, routine HIV testing may ultimately give rise to wider social acceptance of persons living with HIV or AIDS.1 Without appropriate safeguards or support systems, however, the stigma associated with HIV and AIDS may suppress the use of health care services1 and thus exacerbate health disparities in high-risk, marginalized populations.2,3 Such unintended consequences would limit the benefits of routine HIV testing.
Recommendations regarding routine HIV testing must be placed within the current context of recent policy shifts away from the primary prevention of HIV infection and in the context of waning federal support for social programs. The effects of these factors must be considered because they limit access to care and do little to reduce the incidence of infection among those at greatest risk. We encourage the adoption of comprehensive HIV–AIDS initiatives that include routine screening yet do not compromise nonclinical primary-prevention efforts and that are also concerned with reducing stigmatization.
Angela D. Thrasher, M.P.H.
Chandra L. Ford, M.P.H.
University of North Carolina School of Public Health
Chapel Hill, NC 27599-7440
adthrash@email.unc.edu
Kathryn A. Nearing, M.A.
University of Colorado Health Sciences Center
Denver, CO 80262
References
Bozzette SA. Routine screening for HIV infection -- timely and cost-effective. N Engl J Med 2005;352:620-621.
Mechanic D. Disadvantage, inequality, and social policy: major initiatives intended to improve population health may also increase health disparities. Health Aff (Millwood) 2002;21:48-59.
Aggleton P, Parker R. A conceptual framework and basis for action: HIV/AIDS stigma and discrimination. Geneva: Joint United Nations Programme on HIV/AIDS, 2002.
To the Editor: Although the methods used in the deterministic cost-effectiveness analysis by Sanders et al. do not appear contentious, at least two aspects of the incremental cost-effectiveness ratio (ICER) can threaten the validity of the results.
First, the end result of a fraction (ICER = change in cost ÷ change in effectiveness) becomes unstable in the presence of small denominators (range, 0.002 to 0.015 year). Consequently, a minimal error in estimating life expectancy or quality-adjusted life-years would have a substantial effect on the ICER and could even change the conclusions of the study. An analysis based on net health benefit is an alternative method of assessing cost-effectiveness that has several advantages over use of the ICER.1 Second, measurement of uncertainty in cost-effectiveness analysis permits statistical inference and helps policymakers decide whether further research is still needed.2,3
I agree that one-time screening for HIV is the optimal alternative, but only if the parameter estimates obtained by the investigators are extremely close to the true unknown values. Measurement of uncertainty permits the value of information to be estimated and should be routinely performed in cost-effectiveness analyses.
Eduardo da Silveira, M.D.
McGill University
Montreal, QC H2X 3R4, Canada
References
O'Hagan A, Stevens JW. Bayesian methods for design and analysis of cost-effectiveness trials in the evaluation of health care technologies. Stat Methods Med Res 2002;11:469-490.
Claxton K, Neumann PJ, Araki S, Weinstein MC. Bayesian value-of-information analysis: an application to a policy model of Alzheimer's disease. Int J Technol Assess Health Care 2001;17:38-55.
Fenwick EK, Claxton K, Sculpher M. Representing uncertainty: the role of cost-effectiveness acceptability curves. Health Econ 2001;10:779-787.
Dr. Sanders and colleagues reply: Dr. Taiwo accurately points out the importance of HIV counseling in reducing an HIV-infected patient's risk-related behavior. Although such programs improve the cost-effectiveness of screening by decreasing future HIV transmission, our analysis emphasizes that HIV screening can be economically favorable even when the costs and benefits associated with transmission are ignored. Ms. Thrasher and colleagues raise important points about stigma, discrimination, and subsequent health-seeking behavior. These complex issues were not included in our economic analysis, but we agree that they are important for policymaking. In addition, Thrasher et al. are correct in noting that we did not examine the question of whether HIV screening is cost-effective as compared with increased use of primary-prevention methods, since that was not the purpose of our analysis.
We thank Dr. da Silveira for his thoughtful comments about the methods and limitations of cost-effectiveness analysis. We agree that the cost-effectiveness ratio can be sensitive to small changes in the denominator when the incremental benefit is small. In our analysis, although the differences in quality-adjusted life-years were small for the entire screened population, the increase in quality-adjusted life-years for an individual HIV-infected patient was quite large (approximately 1.5 years) and robust. In addition, we believe that the suggested net-health-benefit method is an alternative method of expressing the results of a cost-effectiveness analysis, not an alternative method of assessing the cost-effectiveness of an intervention.
Although the net health benefit has many appealing features, it has two main limitations. First, reporting a net health benefit requires the analyst to choose a societal willingness-to-pay value for a quality-adjusted life-year — a necessity that remains controversial. Second, interpreting a net-health-benefit value is not intuitive for many readers. We also agree that estimation of the value of information can be quite useful, particularly with respect to screening.1 However, calculation of the value of information can be challenging, particularly when models are complex, and it can also be inaccurate when many variables in the model are collinear, as in our case. These methodologic issues need to be addressed for the value of information to be used routinely. Meanwhile, deterministic models such as ours remain valuable sources of evidence for policymakers and provide estimates of the incremental cost-effectiveness of interventions that can be used in guiding policy decisions.2
Gillian D. Sanders, Ph.D.
Duke University
Durham, NC 27705
gillian.sanders@duke.edu
Ahmed M. Bayoumi, M.D.
St. Michael's Hospital
Toronto, ON M5B 1W8, Canada
Douglas K. Owens, M.D.
Veterans Affairs Palo Alto Healthcare System
Palo Alto, CA 94304
References
Owens DK, Nease RF Jr. A normative analytic framework for development of practice guidelines for specific clinical populations. Med Decis Making 1997;17:409-426.
Claxton K. The irrelevance of inference: a decision-making approach to the stochastic evaluation of health care technologies. J Health Econ 1999;18:341-364.
Dr. Paltiel and colleagues reply: Dr. Taiwo calls for proof that expanded HIV-testing procedures can feasibly be implemented in a variety of real-world settings. Our research group and others have already mounted successful demonstration projects in both inpatient and outpatient settings showing that expanded HIV counseling and testing programs are feasible, affordable, and able to identify substantial numbers of new cases of infection.1,2,3,4,5 These studies set the stage for broader implementation of HIV testing and have already demonstrated that acceptance of the test increases with time.
Dr. Taiwo also notes that the attractiveness of HIV testing improves if it decreases the incidence of new infections. This is certainly true, but it is not a necessary condition. Our analysis establishes that routine HIV testing is cost-effective according to U.S. standards, even when viewed strictly from the perspective of the individual infected person. Demonstrating that expanded HIV testing reduces rates of HIV transmission would strengthen an already strong case.
Ms. Thrasher and colleagues express the concern that stigmas and cutbacks in funding for prevention and treatment programs may exacerbate health disparities in populations where routine HIV-testing services are expanded. Although we share this concern, there is little evidence that we are aware of either to support or to refute it. Routine HIV testing may very well serve to destigmatize both HIV testing and HIV disease. Whether or not the concern expressed by Thrasher et al. is borne out in practice, it is our view that the appropriate response to waning federal support for HIV prevention and treatment is not to persist in the pursuit of an outdated approach to HIV testing — an approach that ignores the availability of highly accurate, inexpensive screening tests and affordable, life-sustaining therapies. Rather, the appropriate response is to continue to press on all fronts for a coordinated, evidence-based, national policy on HIV and AIDS that links funding for expanded testing with funding for patient care that conforms to national guidelines and that embraces interventions — both preventive and therapeutic — that deliver good value.
A. David Paltiel, Ph.D.
Yale University
New Haven, CT 06520
david.paltiel@yale.edu
Rochelle P. Walensky, M.D., M.P.H.
Kenneth A. Freedberg, M.D.
Massachusetts General Hospital
Boston, MA 02114
References
Walensky RP, Losina E, Steger-Craven KA, Freedberg KA. Identifying undiagnosed human immunodeficiency virus: the yield of routine, voluntary inpatient testing. Arch Intern Med 2002;162:887-892.
Pincus JM, Crosby SS, Losina E, King ER, LaBelle C, Freedberg KA. Acute human immunodeficiency virus infection in patients presenting to an urban urgent care center. Clin Infect Dis 2003;37:1699-1704.
Kelen GD, Shahan JB, Quinn TC. Emergency department-based HIV screening and counseling: experience with rapid and standard serologic testing. Ann Emerg Med 1999;33:147-155.
Routinely recommended HIV testing at an urban urgent-care clinic -- Atlanta, Georgia, 2000. MMWR Morb Mortal Wkly Rep 2001;50:538-541.
Walensky RP, Losina E, Malatesta L, et al. Effective HIV case identification through routine HIV screening at urgent care centers in Massachusetts. Am J Public Health 2005;95:71-73.