Farnham, Paul G PhD*; Holtgrave, David R PhD†; Sansom, Stephanie L PhD*; Hall, H Irene PhD*
*Centers for Disease Control and Prevention, Atlanta, GA; †Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.
Revised April, 2010.
The findings and conclusions in this document are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
To the Editors:
The HIV epidemic has taken a terrible toll of human life lost and has imposed a substantial economic burden in the United States as measured by the lifetime medical costs of treating persons with HIV.1-4 These lifetime cost estimates are based on data on health care utilization by individuals in different stages of HIV disease, from sources including the AIDS Cost and Services Utilization Survey,1 the HIV Cost and Services Utilization Survey,5 and the HIV Research Network.6 The costs associated with health care utilization in each disease stage are summed across all disease stages from infection to death to calculate the costs expected to be incurred by a person infected with HIV. Disease progression models are used to predict the length of each disease stage and the efficacy of HIV treatment in slowing disease progression.
A rough measure of the effect of HIV prevention programs in the United States can be estimated by comparing the difference between the number of infections that have occurred with the number that might have occurred in the absence of these programs based on changes in HIV transmission rates.7-9 Plausible estimates of the number of infections averted can be derived and combined with estimates of life-time treatment costs to calculate the overall treatment cost savings from HIV infections averted. We present these calculations for the period 1991-2006.
Table 1 gives estimates of HIV incidence, prevalence, transmission rates, and infections averted from 1991-2006.7-11 I (x) is the number of new infections in year x, prevalence, P (x), is the number of persons living with HIV in year x, and the transmission rate for year x, T (x), is calculated as [I (x)/P (x)] × 100. We used updated incidence estimates from the Centers for Disease Control and Prevention calculated in multiyear time blocks that were chosen based on prior information about the likely shape of the incidence curve and the evaluation of models with different assumptions. Although this approach may have caused some discontinuities in the incidence data, the results are similar to other estimates.10 Prevalence was estimated as the number of existing cases of HIV plus the number of new cases [I (x)] minus the number of deaths among persons with HIV. Deaths for persons with AIDS were derived from Centers for Disease Control and Prevention's HIV surveillance database, whereas deaths for persons with HIV (not AIDS) were calculated by multiplying the number of persons living with HIV (not AIDS) in a given year by a general population mortality rate of 427.1 per 100,000 persons 45-54 years old.9
Infections averted by US prevention programs were calculated in the base case under the assumption that the US HIV annual transmission rate would not have dropped below 8.2, the 2007 global transmission rate across all nations,12-13 in the absence of these programs. This global transmission rate is based on Joint United Nations Program on HIV/AIDS/World Health Organization estimates of global HIV incidence and prevalence that draw on all pertinent available data and incorporate substantial revisions in estimation methodology resulting in more accurate and refined estimates.14,15 These estimates, which are derived from countries with both generalized and concentrated epidemics, apply to a period in which there has been substantial investment in HIV prevention and treatment by the Global Fund and PEPFAR. Thus, we are using a comparison transmission rate that is relatively conservative. The difference between actual and projected HIV incidence in the United States occurred from 1991 to 2006 and resulted in an estimate of approximately 362,000 infections averted (Table 1), which is within the range of previously published estimates.7
Life-time HIV treatment cost estimates developed at different points during the epidemic have incorporated varying assumptions about awareness of infection, life expectancy, and treatment regimens for persons with HIV. Guinan et al2 used the cost estimates by Hellinger1 that reflected the pre-antiretroviral therapy (ART) treatment regimens in the early 1990s. Hellinger1 calculated HIV treatment costs by disease stage with an assumed life expectancy of 12.4 years from the time of infection. Guinan et al2 adjusted these costs by assuming that patients with CD4 counts greater than 500 were unaware of their disease for 6 years, aware but asymptomatic for the next 3 years (CD4 count between 200 and 500), symptomatic for 1 more year before AIDS (CD4 count less than 200), and had AIDS for 2 years. Using a 5% discount rate to calculate the present value of these costs, their estimate of these treatment costs was $55,640 (1992 dollars).
Holtgrave and Pinkerton3 developed a range of treatment cost estimates that reflected initial treatment with ART regimens in the mid-1990s. Their intermediate cost scenario represented current treatment in 1996-1997 and was used as their base case. They defined disease stage by awareness of infection, CD4 count, and type of drug therapy and assumed a life expectancy of 16 years from the time of infection. Using a 3% discount rate, their present-value treatment cost estimate was $195,188 (1996 dollars).
Schackman et al4 used HIV Research Network utilization data, cost data from different sources, and the Cost-Effectiveness of Preventing AIDS Complications disease progression model to estimate HIV treatment costs in the era of established ART regimens. Using a 3% discount rate and a life expectancy of 32.1 years from the time of infection, they estimated the present value of HIV treatment costs to be $303,100 (2004 dollars).
Because these estimates were developed with similar methodologies, they were applied to the number of infections averted from 1991 to 2006. The Guinan et al2 costs were first re-estimated with a 3% discount rate for consistency with the other 2 estimates. All estimates were updated to 2009 dollars using the medical care component of the Consumer Price Index.16
The base case estimate of $129.9 billion for overall medical costs saved was derived by applying the Holtgrave and Pinkerton3 estimate of $321,276 (2009 dollars) to the years 1991-1995, whereas the Schackman et al4 estimate of $367,134 (2009 dollars) was applied to the years 1996-2006 (Table 1). Because Guinan et al2 assumed that HIV-infected persons would be unaware of their status for 6 years, we assumed individuals infected in 1991-1995 would begin treatment in the late 1990s with the ART regimens reflected in the Holtgrave and Pinkerton3 cost estimates. Given the Schackman et al4 assumption of 8.1 years from time of infection to diagnosis, persons infected from 1996 to 2006 would begin treatment with the regimens included in the Schackman et al4 costs. The overall estimate of $129.9 billion is conservative because some HIV patients in the earlier period would have adopted the more expensive treatment regimens as they were developed over the course of their lives. A lower-bound estimate for overall medical costs saved of approximately $49 billion was derived by applying the updated Guinan et al2 estimate of $134,928 (2009 dollars) to all 16 years of infections prevented, whereas an upper-bound estimate of approximately $133 billion was derived by applying the Schackman et al4 estimate of $367,134 (2009 dollars) to the data. This range of estimates shows the cost differences associated with treatments that evolved from single drug therapy in the early 1990s to the multidrug regimens used a decade later.
We also conducted sensitivity analyses in which we assumed that the annual transmission rate would not have dropped below the US 1990 rate of 11.7 or below the US 1997 rate of 7.5. In the former case, there would have been approximately 1.4 million infections and $516 billion (2009 dollars) of costs averted, whereas in the latter case, 196,000 infections and $70 billion (2009 dollars) in costs would have been averted. The impact of the changes in transmission rates over the course of the epidemic on the costs averted is substantial.
The $129.9 billion base case estimate reflects only medical costs averted, not net savings, because we did not subtract the cost of HIV prevention programs. We also did not include any costs of lost productivity in the estimates. One estimate of these costs is $353.2 billion (2009 dollars) based on the cost-per-case mortality-related productivity loss estimates of Hutchinson et al.17 The decline in transmission rates and the estimated infections averted in the analysis could have been influenced by clinical therapies that reduced the viral load of infected persons, although the rise in the HIV transmission rate in 1997 indicates one should be cautious about accepting this hypothesis at the national level. Our base case estimate of medical costs alone shows the substantial impact these cases of HIV would have had on the US economy had they not been averted.
Paul G. Farnham, PhD*
David R. Holtgrave, PhD†
Stephanie L. Sansom, PhD*
H. Irene Hall, PhD*
*Centers for Disease Control and Prevention, Atlanta, GA
†Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
1. Hellinger FJ. The lifetime cost of treating a person with HIV. JAMA
2. Guinan ME, Farnham PG, Holtgrave DR. Estimating the value of preventing a human immunodeficiency virus infection. Am J Prev Med
3. Holtgrave DR, Pinkerton SD. Updates of cost of illness and quality of life estimates for use in economic evaluations of HIV prevention programs. J Acquir Immune Defic Syndr
4. Schackman BR, Gebo KA, Walensky RP, et al. The lifetime cost of current human immunodeficiency virus care in the United States. Medical Care
5. Bozzette SA, Joyce G, McCaffrey DF, et al. Expenditures for the care of HIV-infected patients in the era of highly active antiretroviral therapy. N Engl J Med
6. HIV Research Network. Hospital and outpatient health services utilization among HIV-infected patients in care in 1999. J Acquire Immune Defic Syndr
7. Holtgrave DR. Estimating the effectiveness and efficiency of US HIV prevention efforts using scenario and cost-effectiveness analysis. AIDS
8. Holtgrave DR. Written testimony on HIV/AIDS incidence and prevention for the U.S. House of Representatives Committee on Oversight and Government Reform, September 16, 2008. Available at: http://oversight.house.gov/documents/20080916115223.pdf
. Accessed September 16, 2009.
9. Holtgrave DR, Hall HI, Rhodes PH, et al. Updated annual HIV transmission rates in the United States, 1977-2006. J Acquir Immune Defic Syndr
10. Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA
11. Campsmith ML, Rhodes PH, Hall HI, et al. Undiagnosed HIV prevalence among adults and adolescents in the United States at the end of 2006. J Acquir Immune Defic Syndr
12. Henry J. Kaiser Family Foundation. The Global HIV/AIDS Epidemic. HIV/AIDS Policy Fact Sheet
. July 2008. Available at: http://www.kff.org/hivaids
. Accessed October 22, 2009.
13. Holtgrave DR. Apparent declines in the global HIV transmission rate. Int J STD AIDS
14. Joint United Nations Programme on HIV/AIDS/World Health Organization. Q&A on HIV/AIDS estimates: understanding the latest estimates of the 2008 Report on the global AIDS epidemic. July 2008. Availabe at: http://www.unaids.org
. Accessed October 25, 2009.
15. Ghys PD, Walker N, McFarland W, et al. Improved data, methods and tools for the 2007 HIV and AIDS estimates and projections. Sex Transm Infect
. 2008;84(Suppl I):i1-i4.
16. U.S. Bureau of Labor Statistics. Consumer Price Index. CPI Detailed Report Tables. Available at: http://www.bls.gov/cpi/
. Accessed November 15, 2009.
17. Hutchinson AB, Farnham PG, Dean HD, et al. The economic burden of HIV in the United States in the era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr
© 2010 Lippincott Williams & Wilkins, Inc.